From hughsient at gmail.com Tue Dec 1 14:58:41 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 1 Dec 2009 14:58:41 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file Message-ID: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Attached is the new widget I've just added to gnome-color-manager git master. It allows us to show the user a preview of the gamut range of the profile in a graphical way, much like OSX does. You should be able to see the lightly shaded section. At the moment I've hardcoded the CIE REC 709 x-y values, but of course it's easy to change when put into GcmPrefs. So, the point of this email: I wanted to get the CIE x and y coordinates (or XYZ of course) of the three primary illuminants and the x and y coordinates of the white point from the device ICC profile. Is there an easy way to extract these end-point values from the ICC file? I guess mediaWhitePointTag is the latter half of the problem, but I'm not sure how to deal with the LUT values. Ideas welcome. Thanks. Richard. -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot.png Type: image/png Size: 49528 bytes Desc: not available URL: From lars.tore at mulebakken.net Tue Dec 1 16:43:31 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Tue, 1 Dec 2009 17:43:31 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912010843re5f1d81mef40fd5d57e50310@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: > Attached is the new widget I've just added to gnome-color-manager git > master. It allows us to show the user a preview of the gamut range of > the profile in a graphical way, much like OSX does. > > You should be able to see the lightly shaded section. At the moment > I've hardcoded the CIE REC 709 x-y values, but of course it's easy to > change when put into GcmPrefs. > > So, the point of this email: I wanted to get the CIE x and y > coordinates (or XYZ of course) of the three primary illuminants and > the x and y coordinates of the white point from the device ICC > profile. > > Is there an easy way to extract these end-point values from the ICC > file? I guess mediaWhitePointTag is the latter half of the problem, > but I'm not sure how to deal with the LUT values. Ideas welcome. > Thanks. > > Richard. > What about iccdump -v3 -t wtpt -t bkpt -t rXYZ -t gXYZ -t bXYZ myprofile.icc Lars Tore Gustavsen From lars.tore at mulebakken.net Tue Dec 1 18:12:38 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Tue, 1 Dec 2009 19:12:38 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912011012q42a44f4cvd385a8b97ca5f5eb@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: I'm not sure how to deal with the LUT values. Ideas welcome. > Thanks. > > Richard. > > _ I think I missed the last part. Can something like this be used? Just thinking loud here: echo 1 0 0|icclu -ff lutprofile.icc echo 0 1 0|icclu -ff lutprofile.icc echo 0 0 1|icclu -ff lutprofile.icc Lars Tore Gustavsen From graeme2 at argyllcms.com Wed Dec 2 00:19:40 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Wed, 02 Dec 2009 11:19:40 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <4B15B29C.2090105@argyllcms.com> Richard Hughes wrote: > Attached is the new widget I've just added to gnome-color-manager git > master. It allows us to show the user a preview of the gamut range of > the profile in a graphical way, much like OSX does. Professor Hunt would be jumping up and down in the first row at this point saying "Why are you using an XYZ chromaticity diagram ?" :-) :-) [Those who have been to the Color Imaging Conference will understand this.] Actually if you're wanting to plot Yxy values then it's appropriate, but I'll note that it's also very misleading in indicating gamut, since an xy diagram is highly visually non-uniform. A u' v' plot is a better choice to illustrate gamut, while still being linearly additive. > So, the point of this email: I wanted to get the CIE x and y > coordinates (or XYZ of course) of the three primary illuminants and > the x and y coordinates of the white point from the device ICC > profile. > > Is there an easy way to extract these end-point values from the ICC > file? I guess mediaWhitePointTag is the latter half of the problem, > but I'm not sure how to deal with the LUT values. Ideas welcome. The white point is in the white point tag (but see my final note). For a matrix profile the D50 adapted primaries are in the red/green/blue ColorantTags. For a cLut based profile, you would have to (as Lars Tore Gustavsen suggested) run the primary values through the profile (or do the equivalent of looking up what the cLUT values are at the primaries). Note a complication - you probably want the absolute colorant values, not the D50 adapted white ones. You need to undo the chromatic adaptation applied to the colorant tags, and/or do the lookup in absolute colorimetric intent mode. Note also that while absolute colorimetric interpretation of display profiles is perfectly consistent using Argyll/icclib based tools (and is consistent with the original sRGB and current AdobeRGB profiles), it probably will not be so for all other libaries and tools and profiles. This is due to a disagreement about the interpretation and intent of the ICC specification in this regard. Some display profiles have a white point tag of D50 even though this is not the white point of the display, and encode the absolute<->relative transformation in the ChromaticAdapation tag, meaning that the absolute white point and colorant values are not obtained using a normal absolute colorimetric lookup on these profiles. This latter interpretation is the one that has been adopted in the changes made for ICC V4 profiles, even though it is out of sync with much original ICC V2 practice. Graeme Gill. From hughsient at gmail.com Wed Dec 2 08:37:43 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 08:37:43 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B15B29C.2090105@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> Message-ID: <15e53e180912020037o6168de9dm4a20f429e7d665c@mail.gmail.com> 2009/12/2 Graeme Gill : > Actually if you're wanting to plot Yxy values then it's appropriate, > but I'll note that it's also very misleading in indicating gamut, > since an xy diagram is highly visually non-uniform. A u' v' plot > is a better choice to illustrate gamut, while still being linearly > additive. Sure, converting the code to plot u' v' should be pretty easy. The graph is really just for a passing glance, not for any serious comparison. > This is due to a disagreement about the interpretation and intent > of the ICC specification in this regard. Some display profiles > have a white point tag of D50 even though this is not the white > point of the display, and encode the absolute<->relative > transformation in the ChromaticAdapation tag, meaning that > the absolute white point and colorant values are not obtained > using a normal absolute colorimetric lookup on these profiles. This is exactly the information I needed, thanks very much for such a comprehensive answer. Richard. From hughsient at gmail.com Wed Dec 2 15:12:39 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 15:12:39 +0000 Subject: Uncaught bounce notification In-Reply-To: References: Message-ID: <15e53e180912020712l4506d8e1ja991ca132c63ae85@mail.gmail.com> 2009/12/2 : > i'v followed g-c-m using Pascal de Bruijn ppa (thanks to him), and for > these tests I'v built myself the last git (for CIE widget). For some reason your mail got bounced, apologies. I've forwarded it to the list. > So : > 1/ The window for listing screens are too big looking at the rest > (left/right panel) That should be fixed in git master now. > 2/ It's ok that this tool must be kept as simple as possible, but I > would love to have the possibility to set white K? That's on the assumption you're not using a hardware calibration device then? > 3/ if you play with icc profils, size of the right panel change and the > square of CIE widget may be crushed > 4/ I would love to see more (bigger/more details) the CIE widget when > clicking on it Sure, I'm not completely happy with the location or size of this widget. It seems to me I'm trying to cram the "I want to compare different profiles" and "assign a profile with a device" functionality into the single tab. maybe we just need to break out another tab called "Profiles" and list all the profiles along with some more information about them. In this way we can get rid of some of the clutter in the first page. Comments welcome. Richard. From ubuntu at kagou.fr Wed Dec 2 16:20:03 2009 From: ubuntu at kagou.fr (Patrice Vetsel) Date: Wed, 02 Dec 2009 17:20:03 +0100 Subject: Some tests Message-ID: <4B1693B3.8000809@kagou.fr> Hi, i'v followed g-c-m using Pascal de Bruijn ppa (thanks to him), and for these tests I'v built myself the last git (for CIE widget). So : 1/ The window for listing screens are too big looking at the rest (left/right panel) 2/ It's ok that this tool must be kept as simple as possible, but I would love to have the possibility to set white K? 3/ if you play with icc profils, size of the right panel change and the square of CIE widget may be crushed 4/ I would love to see more (bigger/more details) the CIE widget when clicking on it Thanks for this tool Regards Patrice Vetsel Aka Kagou From hughsient at gmail.com Wed Dec 2 16:47:21 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 16:47:21 +0000 Subject: Some tests In-Reply-To: <4B1693B3.8000809@kagou.fr> References: <4B1693B3.8000809@kagou.fr> Message-ID: <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> 2009/12/2 Patrice Vetsel : > 4/ I would love to see more (bigger/more details) the CIE widget when > clicking on it What do you all think of the attached mockups? Richard. -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-1.png Type: image/png Size: 42154 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-2.png Type: image/png Size: 86945 bytes Desc: not available URL: From pmjdebruijn at pcode.nl Wed Dec 2 17:09:30 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:09:30 +0100 Subject: Some tests In-Reply-To: <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> Message-ID: <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> On Wed, Dec 2, 2009 at 5:47 PM, Richard Hughes wrote: > 2009/12/2 Patrice Vetsel : >> 4/ I would love to see more (bigger/more details) the CIE widget when >> clicking on it > > What do you all think of the attached mockups? @Richard: Looks pretty good... The CIE widget is very nice. Though the profile whitepoint isn't being displayed yet, I'm assuming this is still on the TODO list. I also agree it would be nice (but not critical) if double clicking on the CIE widget would create a dialog with just the profile name and the CIE widget, for detailed viewing. And dare I ask, possibly with a right-click "Save (CIE diagram) as PNG/JPEG/PDF" option? Also, I notice the CIE widget has two "squares" of "space" to the left of the CIE diagram. Does this have a purpose? Otherwise couldn't this be decreased to a single white square. I also think either centering the CIE widget, or placing the buttons to the right would look a bit "better" esthetically... Or maybe move the button onder the list view? Giving the CIE diagram a bit more room to "grow"? @Patrice: I rebuild my PPA between 1 to 4 times per two days, depending on how interesting Richard's changes are... All in all it's pretty hard to keep up, and he's doing all the hard work :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 2 17:15:31 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:15:31 +0100 Subject: Idea: GTK/Cario CIE widget as a seperate library Message-ID: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> Hi, The following is just a thought, and is by no means very important... The GTK/Cairo CIE widget written for GCM is beginning to kick ass... However I guess it's GCM specific. Maybe for the long term it's a thought to separate it into a small library, so other apps can use it as well? For example whenever GIMP/UFraw open a dialog to import a new profile it would be very cool if the GNOME File Chooser would "preview" the profile by showing a CIE diagram. Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 2 17:35:04 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 17:35:04 +0000 Subject: Idea: GTK/Cario CIE widget as a seperate library In-Reply-To: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> References: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> Message-ID: <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > The GTK/Cairo CIE widget written for GCM is beginning to kick ass... > However I guess it's GCM specific. Not really, it's a standard GObject so it could be reused. > Maybe for the long term it's a thought to separate it into a small > library, so other apps can use it as well? Sure, there is loads of other common code that might be useful, but I don't think the project is at any level of internal ABI stability. Maybe we can revisit this idea in a few months time? Richard. From pmjdebruijn at pcode.nl Wed Dec 2 17:36:50 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:36:50 +0100 Subject: Profile description for laptop displays In-Reply-To: <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> References: <4c2c37f90911290453x143c6820nb27f3a6b6188e3c7@mail.gmail.com> <15e53e180911291342o2b8755aavad083da3dcd849d2@mail.gmail.com> <4c2c37f90911291527v71d2cf9cy6bb41add8c7b3683@mail.gmail.com> <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> Message-ID: <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> On Mon, Nov 30, 2009 at 11:10 AM, Richard Hughes wrote: > 2009/11/29 Pascal de Bruijn : >> I noticed you included the product_version as well, at least on my >> laptop it's actually the BIOS revision, I'm not sure what this is on >> other laptops. It's probably not that relevant... > > Hmm, it seems the manufacturers were in a rush when they designed your > laptop! The bios version, logically, belongs in bios_version :-) > > Could you attach the output of "cat /sys/class/dmi/id/*" and we'll try > and work around bios issues like this. Thanks. It's attached. >> I noticed you're currently filling the ICC Make with the username, it >> seems rather redundant with the copyright. Besides the point that the >> field wasn't ment to be used like that. I'd stick the EDID info there >> (even with laptops). For the Model field as well, EDID is most likely >> best and most accurate here. > > I deliberately changed it, as the vendor is supposed to be the person > or company that designed or created the profile. In this sense, it > seemed wrong to say that the profile was created "by Lenovo" when it > was created by me, for my Lenovo display. The ICCv2 specification disagrees: 6.4.15 deviceMfgDescTag Tag Type: textDescriptionType Tag Signature: ?dmnd? (646D6E64h) Structure containing invariant and localizable versions of the device manufacturer for display. The content of this structure is described in 6.5.17. 6.4.16 deviceModelDescTag Tag Type: textDescriptionType Tag Signature: ?dmdd? (646D6464h) Structure containing invariant and localizable versions of the device model for display. The content of this structure is described in 6.5.17. >> For example: >> >> Make: SEC (don't lookup in pnp.ids for storage in the ICC profile, >> only lookup in the gcm-prefs GUI). >> Model: 154AT07-H01 >> >> Storing the EDID info as verbatim as possible in the ICC make/models >> fields, will make it easier to auto match displays and profiles in the >> future. > > Hmm, I think the fields have to be displayable. We need to look at > adding private fields if we want to add any machine parsable data into > the ICC file. > >> Storing the EDID verbatim without looking up in pnp.ids prevents false >> data from conflicting lookups (like my Samsung/Seiko issue) from >> ending up in the profile itself. > > I think the lookup is valuable as it works most of the time. Your > panel should have encoded SAM into the PNP field, not SEC, unless of > course the panel really is a Seiko panel that's been rebadged as a > Samsung. For what it's worth, my nice expensive LG panel is actually a > generic Goldstar chipset with LG additions and plastic wrapping. > Hence, it shows up as "Goldstar" as the vendor. :-( I'm not saying my laptop's EDID info is not semi-broken... But translating SEC into Seiko for embedding in the profile, is moving from semi-wrong to completely-wrong... And I'm afraid my laptop isn't the only one having this or similar issues. I do wonder how Samsung's own brand sold displays identify themselves. I wish we had some more EDID samples to compare... Maybe a call for sample EDID data on Planet GNOME would get us some more samples? Regards, Pascal de Bruijn -------------- next part -------------- /sys/class/dmi/id/bios_date 03/12/2009 /sys/class/dmi/id/bios_vendor Hewlett-Packard /sys/class/dmi/id/bios_version 68PZU Ver. F.0C /sys/class/dmi/id/board_asset_tag /sys/class/dmi/id/board_name 30E8 /sys/class/dmi/id/board_serial /sys/class/dmi/id/board_vendor Hewlett-Packard /sys/class/dmi/id/board_version KBC Version 94.1C /sys/class/dmi/id/chassis_asset_tag CNU9100JZH /sys/class/dmi/id/chassis_serial CNU9100JZH /sys/class/dmi/id/chassis_type 10 /sys/class/dmi/id/chassis_vendor Hewlett-Packard /sys/class/dmi/id/chassis_version /sys/class/dmi/id/modalias dmi:bvnHewlett-Packard:bvr68PZUVer.F.0C:bd03/12/2009:svnHewlett-Packard:pnHPCompaq6730s:pvrF.0C:rvnHewlett-Packard:rn30E8:rvrKBCVersion94.1C:cvnHewlett-Packard:ct10:cvr: /sys/class/dmi/id/power cat: /sys/class/dmi/id/power: Is a directory /sys/class/dmi/id/product_name HP Compaq 6730s /sys/class/dmi/id/product_serial CNU9100JZH /sys/class/dmi/id/product_uuid 4377676E-D605-DE11-B150-80487F19A84A /sys/class/dmi/id/product_version F.0C /sys/class/dmi/id/subsystem cat: /sys/class/dmi/id/subsystem: Is a directory /sys/class/dmi/id/sys_vendor Hewlett-Packard /sys/class/dmi/id/uevent MODALIAS=dmi:bvnHewlett-Packard:bvr68PZUVer.F.0C:bd03/12/2009:svnHewlett-Packard:pnHPCompaq6730s:pvrF.0C:rvnHewlett-Packard:rn30E8:rvrKBCVersion94.1C:cvnHewlett-Packard:ct10:cvr: From pmjdebruijn at pcode.nl Wed Dec 2 17:46:29 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:46:29 +0100 Subject: Idea: GTK/Cario CIE widget as a seperate library In-Reply-To: <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> References: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> Message-ID: <4c2c37f90912020946r1a55d1afi1f7752bae2d80f58@mail.gmail.com> On Wed, Dec 2, 2009 at 6:35 PM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> The GTK/Cairo CIE widget written for GCM is beginning to kick ass... >> However I guess it's GCM specific. > > Not really, it's a standard GObject so it could be reused. Nice. >> Maybe for the long term it's a thought to separate it into a small >> library, so other apps can use it as well? > > Sure, there is loads of other common code that might be useful, but I > don't think the project is at any level of internal ABI stability. > Maybe we can revisit this idea in a few months time? Of course, like I said, it's just a thought. Rushing it won't make it better :) Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 2 19:47:00 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 19:47:00 +0000 Subject: Some tests In-Reply-To: <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> Message-ID: <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > The CIE widget is very nice. Though the profile whitepoint isn't being > displayed yet, I'm assuming this is still on the TODO list. I've just added this to git master. > I also agree it would be nice (but not critical) if double clicking on > the CIE widget would create a dialog with just the profile name and > the CIE widget, for detailed viewing. And dare I ask, possibly with a > right-click "Save (CIE diagram) as PNG/JPEG/PDF" option? Sure, we could do that. I'm tempted to leave that feature until the CIE widget displays more useful data. > Also, I notice the CIE widget has two "squares" of "space" to the left > of the CIE diagram. Does this have a purpose? Otherwise couldn't this > be decreased to a single white square. Sure, we can scale this later. > @Patrice: I rebuild my PPA between 1 to 4 times per two days, > depending on how interesting Richard's changes are... All in all it's > pretty hard to keep up, and he's doing all the hard work :) I'll take that as a compliment. After Monday development will slow quite a bit, as I've got a ton of other stuff to do. Richard. From lars.tore at mulebakken.net Wed Dec 2 20:29:04 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Wed, 2 Dec 2009 21:29:04 +0100 Subject: Some tests In-Reply-To: <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> Message-ID: <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> On Wed, Dec 2, 2009 at 8:47 PM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> The CIE widget is very nice. Though the profile whitepoint isn't being >> displayed yet, I'm assuming this is still on the TODO list. > > I've just added this to git master. > It works for matrix profiles. Lut profiles are already displayed black so it is not possibly to see the tiny cross. See attachment. Lars Tore Gustavsen -------------- next part -------------- A non-text attachment was scrubbed... Name: whitepoint-lut.png Type: image/png Size: 40793 bytes Desc: not available URL: From hughsient at gmail.com Wed Dec 2 20:57:23 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 20:57:23 +0000 Subject: Some tests In-Reply-To: <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> Message-ID: <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> 2009/12/2 Lars Tore Gustavsen : > It works for matrix profiles. Lut profiles are already displayed black > so it is not possibly to see the tiny cross. Sure, we need to add support for LUT profiles as described by Graeme. If you're feeling handy I would love a patch, although I can give this a go next week if nobody starts it before me. Richard. From pmjdebruijn at pcode.nl Wed Dec 2 22:06:57 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 23:06:57 +0100 Subject: Some tests In-Reply-To: <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> Message-ID: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> On Wed, Dec 2, 2009 at 9:57 PM, Richard Hughes wrote: > 2009/12/2 Lars Tore Gustavsen : >> It works for matrix profiles. Lut profiles are already displayed black >> so it is not possibly to see the tiny cross. > > Sure, we need to add support for LUT profiles as described by Graeme. > If you're feeling handy I would love a patch, although I can give this > a go next week if nobody starts it before me. LUT profiles have a normal white point defined, I think Lars ment, it would be nice to invert the color of the whitepoint cross, so it's visible on the black CIE diagram. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 2 22:32:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 23:32:01 +0100 Subject: GCM git fails to build Message-ID: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> Hi, I just tried to build GCM from git on Ubuntu Karmic, and it failed: cc1: warnings being treated as errors gcm-profile.c: In function ?gcm_profile_parse?: gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? from incompatible pointer type /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? but argument is of type ?guint *? make[4]: *** [libgcmshared_a-gcm-profile.o] Error 1 make[4]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager/src' make[3]: *** [all] Error 2 make[3]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager/src' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager' make[1]: *** [all] Error 2 make[1]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager' make: *** [debian/stamp-makefile-build] Error 2 dpkg-buildpackage: error: debian/rules build gave error exit status 2 Regards, Pascal de Bruijn From lars.tore at mulebakken.net Thu Dec 3 06:19:58 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 07:19:58 +0100 Subject: Some tests In-Reply-To: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> Message-ID: <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> On Wed, Dec 2, 2009 at 11:06 PM, Pascal de Bruijn wrote: > LUT profiles have a normal white point defined, I think Lars ment, it > would be nice to invert the color of the whitepoint cross, so it's > visible on the black CIE diagram. > Yep, I meant so. Anyway I like the new tabbed interface. Since we now have the profile size displayed in kb, I think it would be of great benefit to also show the profiles filename. Regards Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 08:03:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 08:03:40 +0000 Subject: GCM git fails to build In-Reply-To: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> Message-ID: <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > cc1: warnings being treated as errors > gcm-profile.c: In function ?gcm_profile_parse?: > gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? > from incompatible pointer type > /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? > but argument is of type ?guint *? Should be fixed in git, thanks. Richard. From hughsient at gmail.com Thu Dec 3 08:04:37 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 08:04:37 +0000 Subject: Some tests In-Reply-To: <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> Message-ID: <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> 2009/12/3 Lars Tore Gustavsen : > Anyway I like the new tabbed interface. Since we now have the profile > size displayed in kb, I think it would be of great benefit to also > show the profiles filename. Filename or filename with path? The latter is quite long. Richard. From lars.tore at mulebakken.net Thu Dec 3 09:12:46 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 10:12:46 +0100 Subject: Some tests In-Reply-To: <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> Message-ID: <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> On Thu, Dec 3, 2009 at 9:04 AM, Richard Hughes wrote: > 2009/12/3 Lars Tore Gustavsen : >> Anyway I like the new tabbed interface. Since we now have the profile >> size displayed in kb, I think it would be of great benefit to also >> show the profiles filename. > > Filename or filename with path? The latter is quite long. > > Richard. > Filename would be nice. I also thinks that full path will be to long for the GUI. Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 10:07:01 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 10:07:01 +0000 Subject: Some tests In-Reply-To: <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> Message-ID: <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> 2009/12/3 Lars Tore Gustavsen : > Filename would be nice. I also thinks that full path will be to long > for the GUI. commit 813c39747c4a0c01a5ee97842e32ab38ea5d71fc Author: Richard Hughes Date: Thu Dec 3 10:06:08 2009 +0000 Add the filename basename to the profile view Richard. From lars.tore at mulebakken.net Thu Dec 3 10:15:16 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 11:15:16 +0100 Subject: Some tests In-Reply-To: <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> Message-ID: <58497f010912030215n7a5e9ed6i8234bce4b3d54a6b@mail.gmail.com> On Thu, Dec 3, 2009 at 11:07 AM, Richard Hughes wrote: > 2009/12/3 Lars Tore Gustavsen : >> Filename would be nice. I also thinks that full path will be to long >> for the GUI. > > commit 813c39747c4a0c01a5ee97842e32ab38ea5d71fc > Author: Richard Hughes > Date: ? Thu Dec 3 10:06:08 2009 +0000 > > ? ?Add the filename basename to the profile view > > Richard. > Excellent, and it works. Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 11:42:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 11:42:49 +0000 Subject: Some tests In-Reply-To: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> Message-ID: <15e53e180912030342q2bde3d13ncb616f24c7a3617f@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > LUT profiles have a normal white point defined, I think Lars ment, it > would be nice to invert the color of the whitepoint cross, so it's > visible on the black CIE diagram. commit 6c77c29086ab851bec34c5667d0963bb32ec9217 Author: Richard Hughes Date: Thu Dec 3 11:41:21 2009 +0000 Show the white point cross in white if the CIE chart is all black Richard. From pmjdebruijn at pcode.nl Thu Dec 3 18:31:29 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 19:31:29 +0100 Subject: GCM git fails to build In-Reply-To: <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> Message-ID: <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> On Thu, Dec 3, 2009 at 9:03 AM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> cc1: warnings being treated as errors >> gcm-profile.c: In function ?gcm_profile_parse?: >> gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? >> from incompatible pointer type >> /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? >> but argument is of type ?guint *? > > Should be fixed in git, thanks. Indeed, it's fixed. Some minor trivia: it seems the above only broke on amd64, not on i386... Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 3 18:34:48 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 19:34:48 +0100 Subject: Camera input profiling make/model handling Message-ID: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> Hi, I haven't had time yet to properly check this, but what is used by GCM for make/model info when profiling cameras/scanners? At least for camera's it seems wise to use Make/Model from EXIF. This way, software like UFRaw and DarkTable (and others) can use the EXIF make/model to match which profiles should be valid for a certain RAW file... Luckily EXIF is much better human readable (and generally less ambiguous) than EDID... Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 3 19:02:31 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:02:31 +0000 Subject: GCM git fails to build In-Reply-To: <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> Message-ID: <15e53e180912031102w2c2e16dcmadc15da0983fb24c@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > Some minor trivia: it seems the above only broke on amd64, not on i386... Right, gsize is a different memory size to guint on 64bit architectures. Richard. From hughsient at gmail.com Thu Dec 3 19:09:22 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:09:22 +0000 Subject: Camera input profiling make/model handling In-Reply-To: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> References: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> Message-ID: <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > I haven't had time yet to properly check this, but what is used by GCM > for make/model info when profiling cameras/scanners? Information from sysfs. > At least for camera's it seems wise to use Make/Model from EXIF. This > way, software like UFRaw and DarkTable (and others) can use the EXIF > make/model to match which profiles should be valid for a certain RAW > file... GCM doesn't treat the device like a storage device, it treats it as a USB endpoint device, which means we don't look at the files contained within. > Luckily EXIF is much better human readable (and generally less > ambiguous) than EDID... I think this is the wrong layer. What happens if the EXIF data for one file on the "device" has different colorspaces or even two different device models? The device can carry all number of different EXIF files, but the device can only have one profile. If you inserted a pendrive of photos, would you expect to assign a per-pendrive ICC profile? I think it's sane for the end application to open the file with EXIF metadata and then decided what to do. If it wants to just use the embedded profile, that's fine, or it can ask gcm for the default device profile from the volume (which it can get from udev). Richard. From info at justinsseasonaltreasures.com Thu Dec 3 18:20:37 2009 From: info at justinsseasonaltreasures.com (Unknown) Date: Thu, 03 Dec 2009 10:20:37 -0800 Subject: Support for B9100 Message-ID: <1259864437.2325.1.camel@localhost> Will there be support for color management of the B9100? Thanks, Jazbo From hughsient at gmail.com Thu Dec 3 19:30:15 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:30:15 +0000 Subject: Support for B9100 In-Reply-To: <1259864437.2325.1.camel@localhost> References: <1259864437.2325.1.camel@localhost> Message-ID: <15e53e180912031130k38517da1j65cc586154c34625@mail.gmail.com> 2009/12/3 Unknown : > Will there be support for color management of the B9100? I assume you mean the HP printer. At the moment ArgyllCMS has some support for generating print profiles, but CUPS does not have a way to set the ICC profile for the device. This needs much more work (for the session CUPS process to talk to GCM and get the current profile for the device) before it will be a case of plug and play. Apple did a lot of the CUPS work, but we need to flesh that out for Linux. For using hplip it's a bit different, and we probably need to support that too. Either way, more help is really welcome. If you just want to assign a profile to a device, and then select that in an application, that should work now. Richard. From pmjdebruijn at pcode.nl Thu Dec 3 20:54:23 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 21:54:23 +0100 Subject: Camera input profiling make/model handling In-Reply-To: <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> References: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> Message-ID: <4c2c37f90912031254i1dc63397g70e35c11a902f5d2@mail.gmail.com> On Thu, Dec 3, 2009 at 8:09 PM, Richard Hughes wrote: > 2009/12/3 Pascal de Bruijn : >> I haven't had time yet to properly check this, but what is used by GCM >> for make/model info when profiling cameras/scanners? > > Information from sysfs. Maybe I'll test this over the weekend... >> At least for camera's it seems wise to use Make/Model from EXIF. This >> way, software like UFRaw and DarkTable (and others) can use the EXIF >> make/model to match which profiles should be valid for a certain RAW >> file... > > GCM doesn't treat the device like a storage device, it treats it as a > USB endpoint device, which means we don't look at the files contained > within. > >> Luckily EXIF is much better human readable (and generally less >> ambiguous) than EDID... > > I think this is the wrong layer. What happens if the EXIF data for one > file on the "device" has different colorspaces or even two different > device models? The device can carry all number of different EXIF > files, but the device can only have one profile. If you inserted a > pendrive of photos, would you expect to assign a per-pendrive ICC > profile? Huh? A "Device" does not really have different color spaces... a camera's sensor (RAW file) has a native color space characterized by the profiling of the IT8 target (or another target, which is not really relevant, it's a means to an end). Any camera that produces sRGB or AdobeRGB, is likely to be not fit to be really profiled. Since a camera most likely does not statically convert sensor data to sRGB/AdobeRGB, it's likely a dynamic (image dependant) process. Anyway I filed a feature request with UFRaw, for "special" profiling mode, to convert a camera RAW file an image with linear gamma in the native camera color space: http://sourceforge.net/tracker/?func=detail&aid=2841222&group_id=127649&atid=709089 If one would have a pen drive of photos, the pen drive would be completely and utterly irrelevant from a profiling/profile assignment perspective. Each and every file on the drive could require a different profile (at least for proper results). And normally EXIF could be reasonably reliably used to "match" that. > I think it's sane for the end application to open the file with EXIF > metadata and then decided what to do. If it wants to just use the > embedded profile, that's fine, or it can ask gcm for the default > device profile from the volume (which it can get from udev). Again, with images, only the image color space is truely relevant (embedded or not). However, RAW files do not have assigned (embedded) profiles. UFRaw/DCRaw "solves" this by ripping generic profiles (Color Matrix) from Adobe DNG SDK, and applying them on a by EXIF make/model basis. Please do note, the above applies to camera's. I'm not really sure about scanners. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 3 22:23:48 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 23:23:48 +0100 Subject: ICC operating system identifiers Message-ID: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Hi, Currently there is no registered operating system identifier for Linux in general. Some examples: Microsoft 'msft' Apple 'appl' Sun Microsystems 'sunw' In the past I tried to mail the Linux Foundation about this, since they seem to be the best (least unfit) umbrella organisation to register such an identifier with the ICC. For example: Linux Foundation 'lnxf' However, it seems the mail was ignored. Richard, could you possibly attempt this, since you can mail from a gnome.org e-mail adres, the request might be more seriously considered. Or do you have other thoughts on how to orchestrate this? Regards, Pascal de Bruijn From anders at brander.dk Thu Dec 3 22:32:11 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 03 Dec 2009 23:32:11 +0100 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <1259879531.25715.4.camel@video64> Hi, On Thu, 2009-12-03 at 23:23 +0100, Pascal de Bruijn wrote: > Currently there is no registered operating system identifier for Linux > in general. Is this relevant in this day and age? Is there any CMM/CMS that does anything useful with this information? Regards, Anders Brander From pmjdebruijn at pcode.nl Thu Dec 3 22:46:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 23:46:01 +0100 Subject: ICC operating system identifiers In-Reply-To: <1259879531.25715.4.camel@video64> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> <1259879531.25715.4.camel@video64> Message-ID: <4c2c37f90912031446p58a57498i6cf869fb254962a1@mail.gmail.com> On Thu, Dec 3, 2009 at 11:32 PM, Anders Brander wrote: > Hi, > > On Thu, 2009-12-03 at 23:23 +0100, Pascal de Bruijn wrote: >> Currently there is no registered operating system identifier for Linux >> in general. > > Is this relevant in this day and age? Is there any CMM/CMS that does > anything useful with this information? I think this remains relevant until the ICC officially deprecates the fields... Regards, Pascal de Bruijn From graeme2 at argyllcms.com Fri Dec 4 00:03:34 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 04 Dec 2009 11:03:34 +1100 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <4B1851D6.8080801@argyllcms.com> Pascal de Bruijn wrote: > For example: > Linux Foundation 'lnxf' Hi, note that Argyll uses '*nix' as an unofficial platform signature for Linux and other Unix like platforms. See also this thread on the OpenICC list: Graeme Gill. From hughsient at gmail.com Fri Dec 4 09:22:59 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 09:22:59 +0000 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > Richard, could you possibly attempt this, since you can mail from a > gnome.org e-mail adres, the request might be more seriously > considered. I would suspect my @redhat email account might have more gravitas, although I'm not sure what it would achieve. Do programs actually do anything differently depending on the OSI information, or even display it at any point? If you think it may be useful, then I guess "*nix" makes most sense as Graeme has been using this for ages. If anything, Graeme's email would convey more weight than mine ever would. Richard. From graeme2 at argyllcms.com Fri Dec 4 11:20:47 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 04 Dec 2009 22:20:47 +1100 Subject: ICC operating system identifiers In-Reply-To: <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> Message-ID: <4B18F08F.3080306@argyllcms.com> Richard Hughes wrote: > If you think it may be useful, then I guess "*nix" makes most sense as > Graeme has been using this for ages. If anything, Graeme's email would > convey more weight than mine ever would. The issues are the following: 1) The platform signature is not of great importance, and it's difficult to know what the original intention of it was. ICC profiles are typically not platform specific. It's lack of importance means that it might be difficult to excite anybodies interest in this issue. The only reason I added '*nix' to my code was an interest in not being misleading. My other alternatives were to use one of the other two Unix based platform signatures (SGI or Sun), or to set it to Unknown (0). 2) Most of the other ICC signatures are in registries, so it's not that bigger deal to ask someone to add a signature to a registry - a single person will have the authority to just do it (hence 'argl' added as a CMM signature). The platform signatures only appear in the ICC spec. though, and they also seem to correlate with the founding members (Although I notice that Taligent has been dropped between V2 an V4). So to add a new signature means someone within the ICC has to propose a spec. change and it then has to be approved by the members at some point ! 3) I'm not sure how much consensus there is about a signature that would cover Linux. My thought process was that the existing signatures were both too vague and too specific. They are really operating system vendors, not specific operating system platforms, and so are fairly loose (there is a noticeable difference between Win3.1 and WinNT, between OS9 and OS X). They are also too specific, in that Sun and SGI were both Unix based systems, although SGI has switch to Windows, so what does it mean ? If the signatures were just a registry, there wouldn't be any problem being very specific (Linux, FreeBSD, OpenBSD, RedHatLinux, etc. etc.) So I went for something that would cover cover Linux, BSD etc. and would be enough to distinguish it from Apple and Microsoft. As I suggested on the OpenICC list, if there is sufficient consensus then I'm happy to mail a couple of people and request that a new platform signature be put forward as a modification to the ICC spec., but on the other hand I won't be surprised if action on it is a little slow :-) cheers, Graeme Gill. From lars.tore at mulebakken.net Fri Dec 4 11:53:06 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 4 Dec 2009 12:53:06 +0100 Subject: More verbose output in calibration? Message-ID: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> May I suggest that when I calibrate my monitor the details view are more verbose than now (like dispcal -v). I guess a beginner will think this is not working at all, when it measures all the black patches and almost noting change on the display. I think more experienced users also like to see the progress in the calibration and the deltaE report at the end. In a longer time frame some progress bar (1) or progress window (2) would be very nice eye candy in the calibration process. 1 http://developer.gnome.org/projects/gup/hig/2.0/controls-progress-bars.html 2 http://developer.gnome.org/projects/gup/hig/2.0/windows-progress.html Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 4 14:51:47 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 15:51:47 +0100 Subject: Bug: Profiles tab is empty as long as no profiles have been assigned. Message-ID: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> Hi, I think I found a small bug. When starting gcm-prefs after a clean install, it does list all the available profiles in the profiles dropdown box on the Devices tab. But the list on the Profiles tab is completely empty. When I restart gcm-prefs the list is still empty. However, when I assign a profile (for example blueish) to the screen, it seems to change nothing at first glance. However after restarting gcm-prefs again, suddenly the dialog is bigger, and the list on the Profiles tabs has been filled. I haven't been able to test this, but when a new profile is created, does this get properly added the the list on the Profiles tab as well? Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Fri Dec 4 14:58:25 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 15:58:25 +0100 Subject: gcm-prefs on netbook Message-ID: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> Hi, I managed to test-drive gcm-prefs on a friends netbook (which runs Ubuntu Karmic Netbook Edition). And when the list on the Profiles tab is filled, the dialog quickly grows to big. I've attached a screenshot. I personally don't particularly like netbooks, but they are popular, and thus I think it might be nice for gcm-prefs to work well on netbooks too. I'm quite aware netbooks aren't particularly well suited for "graphics-work", but that's not really the point here. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: gcm-prefs-netbook.png Type: image/png Size: 83037 bytes Desc: not available URL: From hughsient at gmail.com Fri Dec 4 15:00:38 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:00:38 +0000 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> Message-ID: <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > I personally don't particularly like netbooks, but they are popular, > and thus I think it might be nice for gcm-prefs to work well on > netbooks too. What about if the window size < 1024x768 then make the cie widget smaller? Richard. From hughsient at gmail.com Fri Dec 4 15:18:18 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:18:18 +0000 Subject: Bug: Profiles tab is empty as long as no profiles have been assigned. In-Reply-To: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> References: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> Message-ID: <15e53e180912040718m678446b7s3dd3797494ac05e9@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > I think I found a small bug. When starting gcm-prefs after a clean > install, it does list all the available profiles in the profiles > dropdown box on the Devices tab. But the list on the Profiles tab is > completely empty. Ahh, thanks. commit 902fbf2f4271d6be79985d40338fcf4b35bbc2a1 Author: Richard Hughes Date: Fri Dec 4 15:16:37 2009 +0000 Don't skip phase2 startup if we can't load the config file :100644 100644 2726f42... 792b729... M src/gcm-prefs.c Richard. From hughsient at gmail.com Fri Dec 4 15:18:39 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:18:39 +0000 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> Message-ID: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> 2009/12/4 Richard Hughes : > What about if the window size < 1024x768 then make the cie widget smaller? commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 Author: Richard Hughes Date: Fri Dec 4 15:11:19 2009 +0000 Reduce the size of the ICC widget on small displays :100644 100644 4a1551d... 2726f42... M src/gcm-prefs.c Richard. From pmjdebruijn at pcode.nl Fri Dec 4 15:49:27 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 16:49:27 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <4c2c37f90912040749m6c8cdea1vb5c4aa14b43555@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays > > :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c Excellent. The CIE widget isn't the most import thing ever, however nice it may be. The user chose to have a small (low resolution) screen, so having the buttons visible is much more important. Regards, Pascal de Bruijn From lars.tore at mulebakken.net Fri Dec 4 17:16:10 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 4 Dec 2009 18:16:10 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays > > :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c > > Richard. On my 1280x1024 monitor the CIE diagram is quite small now. I can live with it, but I'm not sure if 1280x1024 was meant to be a small display? Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 4 17:18:10 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 18:18:10 +0100 Subject: gcm-prefs on netbook In-Reply-To: <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> Message-ID: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> On Fri, Dec 4, 2009 at 6:16 PM, Lars Tore Gustavsen wrote: > On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: >> 2009/12/4 Richard Hughes : >>> What about if the window size < 1024x768 then make the cie widget smaller? >> >> commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 >> Author: Richard Hughes >> Date: ? Fri Dec 4 15:11:19 2009 +0000 >> >> ? ?Reduce the size of the ICC widget on small displays >> >> :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c >> >> Richard. > > On my 1280x1024 monitor the CIE diagram is quite small now. I can live > with it, but I'm not sure if 1280x1024 was meant to be a small > display? That wasn't intentional: + if (gdk_screen_get_width (screen) < 10240 || That's a typo :) Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Fri Dec 4 17:24:17 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 4 Dec 2009 20:24:17 +0300 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> Message-ID: <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> On Fri, Dec 4, 2009 at 8:18 PM, Pascal de Bruijn wrote: >> On my 1280x1024 monitor the CIE diagram is quite small now. I can live >> with it, but I'm not sure if 1280x1024 was meant to be a small >> display? > > That wasn't intentional: So rectangular shape is? :) http://img37.imagefra.me/img/img37/1/12/4/prokoudine/f_1m_444c758.png Alexandre From hughsient at gmail.com Fri Dec 4 17:44:41 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 17:44:41 +0000 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> Message-ID: <15e53e180912040944k21f00207qa486932397da75eb@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > + if (gdk_screen_get_width (screen) < 10240 || > > That's a typo :) Fixed, thanks. Richard. From hughsient at gmail.com Fri Dec 4 17:49:13 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 17:49:13 +0000 Subject: gcm-prefs on netbook In-Reply-To: <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> Message-ID: <15e53e180912040949p144a83b8gce479add4b1a829@mail.gmail.com> 2009/12/4 Alexandre Prokoudine : > So rectangular shape is? :) Wrong! commit e818364b44830c6178a2fcbd81a255f47ad01a6d Author: Richard Hughes Date: Fri Dec 4 17:47:31 2009 +0000 Ensure the CIE widget is square Richard. From pmjdebruijn at pcode.nl Sat Dec 5 12:29:07 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 13:29:07 +0100 Subject: Devices list sorting Message-ID: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> Hi, Since I'm _guessing_ GCM will most commonly be used to calibrate displays, it might be a thought to sort on device "class" first, and then alphabetically? Possibly displaying displays at the top of the list? I'm guessing this order makes most sense (to me): - Displays - Printers - Scanners - Camera's - Video Camera's Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Sat Dec 5 12:30:56 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 13:30:56 +0100 Subject: Feature: Nautilus Thumbnailer for profiles Message-ID: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> Hi, Another idea for the long term: Wouldn't it severely kick ass, if Nautilus were to thumbnail ICC profiles using the CIE widget? Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Sat Dec 5 13:11:25 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sat, 5 Dec 2009 16:11:25 +0300 Subject: Feature: Nautilus Thumbnailer for profiles In-Reply-To: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> References: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> Message-ID: <733f2c730912050511s6634231qf84b50fb000b148b@mail.gmail.com> On 12/5/09, Pascal de Bruijn wrote: > Another idea for the long term: > > Wouldn't it severely kick ass, if Nautilus were to thumbnail ICC > profiles using the CIE widget? +2^32 :) Alexandre From hughsient at gmail.com Sat Dec 5 15:52:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 5 Dec 2009 15:52:24 +0000 Subject: Devices list sorting In-Reply-To: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> Message-ID: <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> 2009/12/5 Pascal de Bruijn : > Since I'm _guessing_ GCM will most commonly be used to calibrate > displays, it might be a thought to sort on device "class" first, and > then alphabetically? The profile title or filename as the secondary sort? > Possibly displaying displays at the top of the list? I'm guessing this > order makes most sense (to me): > > - Displays > - Printers > - Scanners > - Camera's > - Video Camera's I'm not sure it matters, but sure, it would be better if they were sorted. Richard. From pmjdebruijn at pcode.nl Sat Dec 5 15:54:18 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 16:54:18 +0100 Subject: Devices list sorting In-Reply-To: <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> Message-ID: <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> On Sat, Dec 5, 2009 at 4:52 PM, Richard Hughes wrote: > 2009/12/5 Pascal de Bruijn : >> Since I'm _guessing_ GCM will most commonly be used to calibrate >> displays, it might be a thought to sort on device "class" first, and >> then alphabetically? > > The profile title or filename as the secondary sort? I'd use the profile description as secondary sort... >> Possibly displaying displays at the top of the list? I'm guessing this >> order makes most sense (to me): >> >> - Displays >> - Printers >> - Scanners >> - Camera's >> - Video Camera's > > I'm not sure it matters, but sure, it would be better if they were sorted. It's a convenience thing... For example, since video4linux support has been commited, on my laptop I get my Chicony webcam as the first entry, and below that my display... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Sun Dec 6 09:31:43 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:31:43 +0300 Subject: gcm-inspect segfaults Message-ID: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> (gdb) run -d Starting program: /usr/bin/gcm-inspect -d [Thread debugging using libthread_db enabled] Output profile 'VGA1': not set Output profile 'HDMI1': not set Output profile 'DP1': not set Output profile 'HDMI2': not set Output profile 'DP2': not set Program received signal SIGSEGV, Segmentation fault. 0x00007ffff3ffba52 in g_atomic_int_exchange_and_add () from /lib/libglib-2.0.so.0 (gdb) bt #0 0x00007ffff3ffba52 in g_atomic_int_exchange_and_add () from /lib/libglib-2.0.so.0 #1 0x00007ffff401162e in g_hash_table_unref () from /lib/libglib-2.0.so.0 #2 0x0000000000404ee7 in gcm_inspect_get_properties (argc=1, argv=0x7fffffffe3d8) at gcm-inspect.c:371 #3 main (argc=1, argv=0x7fffffffe3d8) at gcm-inspect.c:438 Should I be concerned about that sort of thing? :) Alexandre From alexandre.prokoudine at gmail.com Sun Dec 6 09:45:02 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:45:02 +0300 Subject: Yelp isn't happy Message-ID: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> Hi, When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI ?ghelp:gnome-color-manager?preferences? is incorrect". Alexandre From alexandre.prokoudine at gmail.com Sun Dec 6 09:51:56 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:51:56 +0300 Subject: yet another PO comment bug Message-ID: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> Hi, A comment for "Setting up device" says "TRANSLATORS: title, a profile is a ICC file" Alexandre From hughsient at gmail.com Sun Dec 6 10:09:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 10:09:49 +0000 Subject: gcm-inspect segfaults In-Reply-To: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> References: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> Message-ID: <15e53e180912060209i151c479gd985e4d60e4a829e@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > #1 ?0x00007ffff401162e in g_hash_table_unref () from /lib/libglib-2.0.so.0 Fixed in git, thanks. Richard. From hughsient at gmail.com Sun Dec 6 10:16:29 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 10:16:29 +0000 Subject: yet another PO comment bug In-Reply-To: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> References: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> Message-ID: <15e53e180912060216y4b0c3daeh2c850b6b15b8ce35@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > A comment for "Setting up device" says "TRANSLATORS: title, a profile > is a ICC file" Fixed, thanks. Richard. From pmjdebruijn at pcode.nl Sun Dec 6 11:39:55 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 6 Dec 2009 12:39:55 +0100 Subject: CIE widget drawing odd graphs Message-ID: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> Hi, I might have uncovered some minor issues with the CIE widget drawing code. I've attached an archive containing several non-standard profiles, which are drawn quite oddly. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: weirdciewidget.zip Type: application/zip Size: 31374 bytes Desc: not available URL: From pmjdebruijn at pcode.nl Sun Dec 6 12:33:52 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 6 Dec 2009 13:33:52 +0100 Subject: Icon Sizes Message-ID: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> Hi, I managed to test GCM before on a netbook with Ubuntu Karmic Netbook edition installed. I noticed somehow the GCM icon was very small in comparison to all the other application icons... While I'm not entirely sure about this, I think the Ubuntu Netbook edition uses the 64x64 icons (and does not fall back to scalable to save CPU cycles on low end CPUs), since GCM does not supply a 64x64 icon, it falls back to a smaller one. Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? Regards, Pascal de Bruijn From lars.tore at mulebakken.net Sun Dec 6 16:04:46 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Sun, 6 Dec 2009 17:04:46 +0100 Subject: Bluish.icc In-Reply-To: <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> References: <58497f010911110351o3d7df74dt8ac35ef7c8cb5e4@mail.gmail.com> <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> Message-ID: <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> On Sun, Dec 6, 2009 at 3:32 PM, Pascal de Bruijn wrote: > On Wed, Nov 11, 2009 at 12:51 PM, Lars Tore Gustavsen > wrote: >> The icc profile include with gnome-color-manager is very nice to test >> if everything works. However it advertise itself as "sRGB >> IEC61966-2-1" >> >> I have edited the attached profile and it now reads "bluish test" in gcm-prefs. > > Would it be possible for you to rename it to "Blueish Test" instead of > "blueish test"? > > Not extremely important, but it does look a bit better :) > > Regards, > Pascal de Bruijn > Here we go: It's done with Andrew Shepherd's ICC Profile Toolkit http://www.tlbtlb.com/links/ Regards Lars Tore Gustavsen -------------- next part -------------- A non-text attachment was scrubbed... Name: bluish.icc Type: application/x-icc Size: 3966 bytes Desc: not available URL: From hughsient at gmail.com Sun Dec 6 19:37:09 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 19:37:09 +0000 Subject: Bluish.icc In-Reply-To: <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> References: <58497f010911110351o3d7df74dt8ac35ef7c8cb5e4@mail.gmail.com> <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> Message-ID: <15e53e180912061137h57d3f969m2a57af3568e547e7@mail.gmail.com> 2009/12/6 Lars Tore Gustavsen : > Here we go: Applied, thanks. Richard From hughsient at gmail.com Sun Dec 6 19:46:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 19:46:40 +0000 Subject: Icon Sizes In-Reply-To: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> References: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> Message-ID: <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> 2009/12/6 Pascal de Bruijn : > Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? commit 29b66375ed40e10bfffba601eeca9f25bd8a46b7 Author: Richard Hughes Date: Sun Dec 6 19:45:42 2009 +0000 Add a 64x64 pixel application icon Richard. From hughsient at gmail.com Sun Dec 6 20:06:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 20:06:24 +0000 Subject: Devices list sorting In-Reply-To: <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> Message-ID: <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> 2009/12/5 Pascal de Bruijn : > For example, since video4linux support has been commited, on my laptop > I get my Chicony webcam as the first entry, and below that my > display... commit dfe5fc80cef25cac2deddb5687719ab298e786e0 Author: Richard Hughes Date: Sun Dec 6 20:05:18 2009 +0000 Make the list orders predictable by setting a sort string Richard. From hughsient at gmail.com Sun Dec 6 20:19:37 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 20:19:37 +0000 Subject: CIE widget drawing odd graphs In-Reply-To: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> Message-ID: <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> 2009/12/6 Pascal de Bruijn : > I might have uncovered some minor issues with the CIE widget drawing code. They look pretty major to me! > I've attached an archive containing several non-standard profiles, > which are drawn quite oddly. After the release I would like to look at the widget drawing code and the profile extraction code (there are still quite a lot of FIXME's and unimplemented stuff) and add a boat load more self tests and internal validations. I fear I might not have time in the next couple of weeks (lots of other work to do, which I've been putting off for ages) so if anyone is familiar with XYZ co-ordinates and ICC profiles I would appreciate any code review and possible patches. Thanks, Richard. From pmjdebruijn at pcode.nl Sun Dec 6 23:35:43 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 00:35:43 +0100 Subject: Devices list sorting In-Reply-To: <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> Message-ID: <4c2c37f90912061535n22c07324s60f6a5edbc7e5147@mail.gmail.com> On Sun, Dec 6, 2009 at 9:06 PM, Richard Hughes wrote: > 2009/12/5 Pascal de Bruijn : >> For example, since video4linux support has been commited, on my laptop >> I get my Chicony webcam as the first entry, and below that my >> display... > > commit dfe5fc80cef25cac2deddb5687719ab298e786e0 > Author: Richard Hughes > Date: ? Sun Dec 6 20:05:18 2009 +0000 > > ? ?Make the list orders predictable by setting a sort string Seems to work excellently here... Thanks, Pascal de Bruijn From hughsient at gmail.com Mon Dec 7 12:03:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 7 Dec 2009 12:03:28 +0000 Subject: gnome-color-manager 2.29.1 Message-ID: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> gnome-color-manager is a session program that makes it easy to manage, install and generate color profiles in the GNOME desktop. Version 2.29.1 ~~~~~~~~~~~~~~ Released: 2009-12-07 * Translations - Add Russian translation (Alexandre Prokoudine) - Add British English translation (Bruce Cowan) - Add Indonesian translation (Andika Triwidada) - Add French translation (Claude Paroz) - Add Swedish translation (Daniel Nylander) - Add Brazilian Portuguese translation (Flamarion Jorge) - Add Lithuanian translation (Gintautas Miliauskas) - Add German translation (Hendrik Brandt) - Add Danish translation (Joe Hansen) - Add Spanish translation (Jorge Gonz?lez) - Add Norwegian bokm?l translation (Kjartan Maraas) - Add Thai translation (Phondanai Khanti) - Add Polish translation (Piotr Dr?g) - Add Estonian translation (Priit Laes, Mattias P?ldaru) - Add Tamil translation (vasudeven) - Add Russian user guide translation (Alexandre Prokoudine) - Add Indonesian user guide translation (Andika Triwidada) * New Features: - Add gcm-import, a helper to allow double clicking on ICC profiles to import them (Richard Hughes) - Add gcm-prefs, a utility to assign profiles to devices, examine profiles, and set session-wide defaults (Richard Hughes) - Add gcm-session, a dbus-activated session daemon for applications to get the profiles for a device, or device class and to get session-wide defaults. It exits when no longer used to save resources. (Richard Hughes) - Add gcm-self-test, a self test framework that tests GCM functionality (Richard Hughes) - Add gcm-apply, a simple utility to just set (or reset) display profiles (Richard Hughes) - Add gcm-inspect, a debugging utility to inspect the profiles set in the session (Richard Hughes) - Add gcm-dump-edid, a utility to dump the EDID to disk for debugging (Richard Hughes) - Add gcm-dump-profile, a utility to dump the ICC profile to the screen (Richard Hughes) - Add some simple man pages and help document (Richard Hughes) - Add ArgyllCMS support to generate device profiles (Richard Hughes) - Add color calibration hardware auto-detection (Richard Hughes) - Add code to set the _ICC_PROFILE atom per-output and also per-screen (Richard Hughes) - Add some pre-calibration steps for external displays (Pascal de Bruijn) - Add hardware support for gphoto supported cameras (Richard Hughes) - Add hardware support for SANE suppoerted scanners (Richard Hughes) - Add hardware support for video4linux supported video devices (Richard Hughes) - Add hardware support for XRandR supported displays (Richard Hughes) - Add hardware support for hplip supported printers (Richard Hughes) - Add CIE widget to display visual data about different profiles (Richard Hughes) - Use the system DMI data to better itentify internal LCD panels (Richard Hughes) - Parse the EDID to get a better device description for displays (Richard Hughes) - Make the list orders predictable by setting a sort string (Richard Hughes) * Bugfix: - Use XDG directory to store data (Baptiste Mille-Mathias) - Remove markup from GTKBuilder translatable strings (Claude Paroz) - Update bluish.icc title (Lars Tore Gustavsen, Pascal de Bruijn) - Enable adding xrandr devices with no EDID (Martin Szulecki) - Avoid reporting a (false) failure on first import (Stephane Delcroix) - Fix the message-received cb signature (Stephane Delcroix) - Fix up numerous small bugs prior to first release (Richard Hughes) - Look for the debian-named argyllcms binaries first (Richard Hughes) - Set the brightness to 100% on internal LCD panels before we generate a output profile (Richard Hughes) Richard. From pmjdebruijn at pcode.nl Mon Dec 7 12:37:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 13:37:01 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> Message-ID: <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> On Mon, Dec 7, 2009 at 1:03 PM, Richard Hughes wrote: > gnome-color-manager is a session program that makes it easy to manage, install > and generate color profiles in the GNOME desktop. > > Version 2.29.1 > ~~~~~~~~~~~~~~ > Released: 2009-12-07 Excellent :) Packages are available from my PPA as usual. However, since I'll be tracking git again in my main repo, I've decided to start a gcm-release PPA: https://launchpad.net/~pmjdebruijn/+archive/gcm-release I've just submitted the builds, so they'll be ready in an hour or six... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Mon Dec 7 15:08:18 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 7 Dec 2009 18:08:18 +0300 Subject: metadata Message-ID: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Hi, It looks like there are problems writing unicode text into icc metadata or reading from it. http://img39.imagefra.me/img/img39/1/12/7/prokoudine/f_0m_948cb9f.png Exhibit 1 is description of the profile that is obviously RU translation of "Custom" Exhibit 2 is copyright field that seems to be suffering from more or less same thing. Any ideas? Alexandre From pmjdebruijn at pcode.nl Mon Dec 7 15:14:42 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 16:14:42 +0100 Subject: metadata In-Reply-To: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Message-ID: <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> On Mon, Dec 7, 2009 at 4:08 PM, Alexandre Prokoudine wrote: > Hi, > > It looks like there are problems writing unicode text into icc > metadata or reading from it. > > http://img39.imagefra.me/img/img39/1/12/7/prokoudine/f_0m_948cb9f.png > > Exhibit 1 is description of the profile that is obviously RU > translation of "Custom" > > Exhibit 2 is copyright field that seems to be suffering from more or > less same thing. This actually begs a bigger question: Should we even try to internationalize metadata contents? I'd prefer not... It would make exchanging ICC profiles internationally harder... Say, one day, we make a big repository of ICC profiles which are ready to go for people who don't have a colorimeter. I'd be pretty crappy for someone in Germany to see Russian in the metadata? I'd prefer to stick to English... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Mon Dec 7 15:47:16 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 7 Dec 2009 18:47:16 +0300 Subject: metadata In-Reply-To: <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> Message-ID: <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> On Mon, Dec 7, 2009 at 6:14 PM, Pascal de Bruijn wrote: > This actually begs a bigger question: Should we even try to > internationalize metadata contents? > > I'd prefer not... It would make exchanging ICC profiles > internationally harder... Quite so :) > Say, one day, we make a big repository of ICC profiles which are ready > to go for people who don't have a colorimeter. Are we talking about profiles for DSLRs? :) Alexandre From hughsient at gmail.com Mon Dec 7 17:29:00 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 7 Dec 2009 17:29:00 +0000 Subject: gnome-color-manager 2.29.1 In-Reply-To: <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> Message-ID: <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> 2009/12/7 Pascal de Bruijn : > I've just submitted the builds, so they'll be ready in an hour or six... I've just started an official build for Fedora 12: http://koji.fedoraproject.org/koji/taskinfo?taskID=1860405 Richard. From pmjdebruijn at pcode.nl Mon Dec 7 19:21:36 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 20:21:36 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> Message-ID: <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> On Mon, Dec 7, 2009 at 6:29 PM, Richard Hughes wrote: > 2009/12/7 Pascal de Bruijn : >> I've just submitted the builds, so they'll be ready in an hour or six... > > I've just started an official build for Fedora 12: > http://koji.fedoraproject.org/koji/taskinfo?taskID=1860405 Richard, my mail to the mailing list got rejected: "Message rejected by filter rule match" Regards, Pascal de Bruijn From hughsient at gmail.com Tue Dec 8 10:15:58 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 10:15:58 +0000 Subject: gnome-color-manager 2.29.1 In-Reply-To: <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> Message-ID: <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> 2009/12/7 Pascal de Bruijn : > Richard, my mail to the mailing list got rejected: I think gnome-announce-list at gnome.org bounced the mail (correctly) -- dropping the cc should make things work again. Richard. From hughsient at gmail.com Tue Dec 8 11:02:50 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 11:02:50 +0000 Subject: metadata In-Reply-To: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Message-ID: <15e53e180912080302x2f413450v3b99d884d709ed57@mail.gmail.com> 2009/12/7 Alexandre Prokoudine : > Exhibit 1 is description of the profile that is obviously RU > translation of "Custom" Right, I made the Custom localized without thinking that it would involve non-ascii chars. I've fixed that in cfebe91aef33254709bc22ca547563044f1fbdd8. > Exhibit 2 is copyright field that seems to be suffering from more or > less same thing. > > Any ideas? Well, for textType, the profile spec mandates 7 bit ASCII, so we're wrong to translate Custom. For the unicodeTextType we should probably decode UTF-16 properly, although I've not seen a profile that uses this type in the wild yet. Note: if you've got a profile that breaks GCM in one way or another (and it's freely redistributable, which excludes the ICC and Adobe profiles..) then we should probably check it into GCM and wire it into the self test code. If nothing else you can use gcm-dump-profile -v thefilename.icc and send me the output. Richard. From hughsient at gmail.com Tue Dec 8 11:05:57 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 11:05:57 +0000 Subject: metadata In-Reply-To: <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> Message-ID: <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> 2009/12/7 Alexandre Prokoudine : >> Say, one day, we make a big repository of ICC profiles which are ready >> to go for people who don't have a colorimeter. > > Are we talking about profiles for DSLRs? :) Just in case, I've got a repo waiting: http://github.com/hughsie/gnome-color-profiles I'm going to be checking in the Adobe and ICC profiles first, but alas, they can't be installed by default in distros like Fedora as they are non free [1]. Richard. [1] https://www.redhat.com/archives/fedora-legal-list/2009-December/msg00009.html and https://www.redhat.com/archives/fedora-legal-list/2009-December/msg00010.html From pmjdebruijn at pcode.nl Tue Dec 8 11:16:35 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 12:16:35 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> Message-ID: <4c2c37f90912080316w42894a61m7ad18ee019eba3dd@mail.gmail.com> On Tue, Dec 8, 2009 at 11:15 AM, Richard Hughes wrote: > 2009/12/7 Pascal de Bruijn : >> Richard, my mail to the mailing list got rejected: > > I think ?gnome-announce-list at gnome.org bounced the mail (correctly) -- > dropping the cc should make things work again. Oops, silly me... Yeah, the gnome-color-manager list accepted my message just fine... Regards, Pascal de Bruijn From graeme2 at argyllcms.com Tue Dec 8 12:38:17 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Tue, 08 Dec 2009 23:38:17 +1100 Subject: metadata In-Reply-To: <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> Message-ID: <4B1E48B9.1080408@argyllcms.com> Richard Hughes wrote: > I'm going to be checking in the Adobe and ICC profiles first, but > alas, they can't be installed by default in distros like Fedora as > they are non free [1]. Note that Argyll V1.1.0 includes a public domain implementation of an sRGB profiles as well as an AdobeRGB1998 equivalent. Graeme Gill. From hughsient at gmail.com Tue Dec 8 12:51:45 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 12:51:45 +0000 Subject: metadata In-Reply-To: <4B1E48B9.1080408@argyllcms.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> <4B1E48B9.1080408@argyllcms.com> Message-ID: <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> 2009/12/8 Graeme Gill : > Richard Hughes wrote: > >> I'm going to be checking in the Adobe and ICC profiles first, but >> alas, they can't be installed by default in distros like Fedora as >> they are non free [1]. > > Note that Argyll V1.1.0 includes a public domain implementation > of an sRGB profiles as well as an AdobeRGB1998 equivalent. That's really good to know, and I might pinch those. What license are those available as? Really public domain? (if so, that's fantastic). Can they be installed alongside the real adobe profiles? Richard. From graeme2 at argyllcms.com Tue Dec 8 13:11:07 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Wed, 09 Dec 2009 00:11:07 +1100 Subject: metadata In-Reply-To: <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> <4B1E48B9.1080408@argyllcms.com> <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> Message-ID: <4B1E506B.4080709@argyllcms.com> Richard Hughes wrote: > 2009/12/8 Graeme Gill : >> Note that Argyll V1.1.0 includes a public domain implementation >> of an sRGB profiles as well as an AdobeRGB1998 equivalent. > > That's really good to know, and I might pinch those. What license are > those available as? Really public domain? (if so, that's fantastic). Yes. There didn't seem any sensible reason to retain copyright on them. > Can they be installed alongside the real adobe profiles? I'm not sure what you mean by that. Since "Adobe" is a registered company name, I've given the Adobe compatible profile a different name ("ClayRGB1998.icm", the best I could come up with at the time). The sRGB profile is called "sRGB.icm", just like the original one from HP/Microsoft that is no longer available, and to which it is functionally identical. Graeme Gill. From pmjdebruijn at pcode.nl Tue Dec 8 17:56:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 18:56:33 +0100 Subject: Icon Sizes In-Reply-To: <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> References: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> Message-ID: <4c2c37f90912080956y6dcf9264padf8217b2c3a5ab4@mail.gmail.com> On Sun, Dec 6, 2009 at 8:46 PM, Richard Hughes wrote: > 2009/12/6 Pascal de Bruijn : >> Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? > > commit 29b66375ed40e10bfffba601eeca9f25bd8a46b7 > Author: Richard Hughes > Date: ? Sun Dec 6 19:45:42 2009 +0000 > > ? ?Add a 64x64 pixel application icon Today, I had a chance to test this on my colleagues HP netbook again... And this seems to fix the "small icon" issue on Ubuntu Karmic Netbook Edition. Thanks, Pascal de Bruijn From pmjdebruijn at pcode.nl Tue Dec 8 17:58:02 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 18:58:02 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <4c2c37f90912080958s1dac9704qdc99372d9c1a3401@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays I had a chance to test this on a colleagues HP netbook today, and it seems to work fine. The CIE widget looks a bit silly at 50x50, but again, this is a netbook's owner choice... Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 9 12:47:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 12:47:28 +0000 Subject: Yelp isn't happy In-Reply-To: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> References: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> Message-ID: <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI > ?ghelp:gnome-color-manager?preferences? is incorrect". commit ef84a7c5dca50e40bfaf38b6c884c8c86983f334 Author: Richard Hughes Date: Wed Dec 9 12:46:33 2009 +0000 Fix the help file installation so that yelp recognises our help file Richard. From alexandre.prokoudine at gmail.com Wed Dec 9 13:18:19 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Wed, 9 Dec 2009 16:18:19 +0300 Subject: Yelp isn't happy In-Reply-To: <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> References: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> Message-ID: <733f2c730912090518w1ed31992g2921881109c65f9b@mail.gmail.com> On Wed, Dec 9, 2009 at 3:47 PM, Richard Hughes wrote: > 2009/12/6 Alexandre Prokoudine : >> When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI >> ?ghelp:gnome-color-manager?preferences? is incorrect". > > commit ef84a7c5dca50e40bfaf38b6c884c8c86983f334 > Author: Richard Hughes > Date: ? Wed Dec 9 12:46:33 2009 +0000 > > ? ?Fix the help file installation so that yelp recognises our help file Many thanks :) Alexandre From hughsient at gmail.com Wed Dec 9 13:18:35 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 13:18:35 +0000 Subject: Profile description for laptop displays In-Reply-To: <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> References: <4c2c37f90911290453x143c6820nb27f3a6b6188e3c7@mail.gmail.com> <15e53e180911291342o2b8755aavad083da3dcd849d2@mail.gmail.com> <4c2c37f90911291527v71d2cf9cy6bb41add8c7b3683@mail.gmail.com> <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> Message-ID: <15e53e180912090518h5b6e9deas414d6c5e1f1ec594@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > I wish we had some more EDID samples to compare... Maybe a call for > sample EDID data on Planet GNOME would get us some more samples? Could you send me your EDID please. Thanks. Richard. From hughsient at gmail.com Wed Dec 9 13:19:56 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 13:19:56 +0000 Subject: More verbose output in calibration? In-Reply-To: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> References: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> Message-ID: <15e53e180912090519v6481e569q867923aa08984dd3@mail.gmail.com> 2009/12/4 Lars Tore Gustavsen : > May I suggest that when I calibrate my monitor the details view are > more verbose than now (like dispcal -v). I guess a beginner will think > this is not working at all, when it measures all the black patches and > almost noting change on the display. ?I think more experienced users > also like to see the progress in the calibration ?and the deltaE > report at the end. ?In a longer time frame some progress bar (1) or > progress window (2) would be very nice eye candy in the calibration > process. You can already get the deltaE values in the dropdown widget. I think it would be very hard to do a progress bar from the data Argyll gives us. Richard. From hughsient at gmail.com Wed Dec 9 17:04:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 17:04:24 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B15B29C.2090105@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> Message-ID: <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> 2009/12/2 Graeme Gill : > For a cLut based profile, you would have to (as Lars Tore Gustavsen > suggested) run the primary values through the profile (or > do the equivalent of looking up what the cLUT values are at > the primaries). I'm sort-of stuck here. I'm confused about which cLUT I should be using (TRC curves?), as different profiles are doing very different things. Could you please elaborate about how to run the primaries through the table? Thanks. Richard. From hughsient at gmail.com Wed Dec 9 19:51:53 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 19:51:53 +0000 Subject: shared-color-profiles 0.1.0 release Message-ID: <15e53e180912091151y1f2303acw48cedde3eb0231f@mail.gmail.com> Version 0.1.0 ~~~~~~~~~~~~~ Released: 2009-12-09 * New Features: - First release! - Add the free Argyll profiles that are in the public domain (Richard Hughes) * Bugfix: - Don't install README and LICENSE in /usr/share/color (Richard Hughes) - Require color-filesystem on Fedora (Richard Hughes) The next release of gnome-color-manager will depend on this package. See http://blogs.gnome.org/hughsie/2009/12/08/shared-color-profiles/ for more details. Richard. From alexandre.prokoudine at gmail.com Wed Dec 9 23:02:22 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 02:02:22 +0300 Subject: sadly, segfault Message-ID: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> Hi, On commit f04af5c3072983e9b5113628f7782dc346c75b18: http://pastebin.com/m18c9d8c9 Alexandre From graeme2 at argyllcms.com Wed Dec 9 23:04:54 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Thu, 10 Dec 2009 10:04:54 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> Message-ID: <4B202D16.1000901@argyllcms.com> Richard Hughes wrote: > I'm sort-of stuck here. I'm confused about which cLUT I should be > using (TRC curves?), as different profiles are doing very different > things. Could you please elaborate about how to run the primaries > through the table? Thanks. Well I can only talk specifically about my set of tools: Using icclib command line tools it would be something like this: icclu -ff -ia -px sRGB.icm 1 0 0 1.000000 0.000000 0.000000 [RGB] -> MatrixFwd -> 0.412385 0.212596 0.019306 [XYZ] 0 1 0 0.000000 1.000000 0.000000 [RGB] -> MatrixFwd -> 0.357573 0.715199 0.119206 [XYZ] 0 0 1 0.000000 0.000000 1.000000 [RGB] -> MatrixFwd -> 0.180500 0.072205 0.950546 [XYZ] Progromatically, one would do the equivalent, e.g. using icclib open the profile, create a lookup object to do forward absolute colorimetric to XYZ space, and then feed colorant only combinations through it. icclu.c will illustrate the details. I'm sure something similar is possible using Marti's lcms. I wouldn't recommend trying to re-create the wheel and do it from ICC tags as there are a number of details to get right. Graeme Gill. From hughsient at gmail.com Thu Dec 10 08:07:04 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 08:07:04 +0000 Subject: sadly, segfault In-Reply-To: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> References: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> Message-ID: <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> 2009/12/9 Alexandre Prokoudine : > On commit f04af5c3072983e9b5113628f7782dc346c75b18: > http://pastebin.com/m18c9d8c9 Thanks, commit 64037aacc1ba6fa7bfa53deabc00dca57cdd658d Author: Richard Hughes Date: Thu Dec 10 08:05:25 2009 +0000 trivial: fix up an error in the last commit, so that if filename is not set we skip the load There are some more changes happening today, so please yell if anything breaks. Thanks. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 14:54:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 17:54:09 +0300 Subject: sadly, segfault In-Reply-To: <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> References: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> Message-ID: <733f2c730912100654g6eafc1e9v5c51b2783e119437@mail.gmail.com> On 12/10/09, Richard Hughes wrote: > There are some more changes happening today, so please yell if > anything breaks. Thanks. Well, apart from the profiles I sent you before whose CIE diagram doesn't render nicely, so far so good :) Alexandre From hughsient at gmail.com Thu Dec 10 15:26:35 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 15:26:35 +0000 Subject: LCMS dependency Message-ID: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> How would everyone feel if I added lcms as a hard dependency to gnome-color-manager? Now how would you all feel if I added lcms2 as a hard dep? Comments please. I think using lcms allows us to nuke a lot of the custom (fragile) profile code, but only lcms2 has the functionality I need to decode the vcgt tags. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 15:30:18 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 18:30:18 +0300 Subject: LCMS dependency In-Reply-To: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> Message-ID: <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> On 12/10/09, Richard Hughes wrote: > How would everyone feel if I added lcms as a hard dependency to > gnome-color-manager? Fine with me. Most graphics apps I use already have it as requirement. > Now how would you all feel if I added lcms2 as a hard dep? Marti does amazing job with lcms2, and since now it has a separate header for v2, this really might work :) Alexandre From pmjdebruijn at pcode.nl Thu Dec 10 15:47:06 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 16:47:06 +0100 Subject: LCMS dependency In-Reply-To: <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> Message-ID: <4c2c37f90912100747i1a0a507au47db84306b64dd42@mail.gmail.com> On Thu, Dec 10, 2009 at 4:30 PM, Alexandre Prokoudine wrote: > On 12/10/09, Richard Hughes wrote: >> How would everyone feel if I added lcms as a hard dependency to >> gnome-color-manager? > > Fine with me. Most graphics apps I use already have it as requirement. Indeed... >> Now how would you all feel if I added lcms2 as a hard dep? > > Marti does amazing job with lcms2, and since now it has a separate > header for v2, this really might work :) Adding lcms2 as a hard dependancy seems less then desirable (at this moment), since lcms2 hasn't been released yet. And it requires additional packaging... Adding it as an optional dependancy seems ok though. Regards, Pascal de Bruijn From graeme2 at argyllcms.com Thu Dec 10 16:05:34 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 11 Dec 2009 03:05:34 +1100 Subject: LCMS dependency In-Reply-To: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> Message-ID: <4B211C4E.9060502@argyllcms.com> Richard Hughes wrote: > I think using lcms allows us to nuke a lot of the custom (fragile) > profile code, but only lcms2 has the functionality I need to decode > the vcgt tags. Well of course icclib has support for accessing the vcgt tag. Graeme Gill. From hughsient at gmail.com Thu Dec 10 16:18:36 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 16:18:36 +0000 Subject: LCMS dependency In-Reply-To: <4B211C4E.9060502@argyllcms.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> Message-ID: <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> 2009/12/10 Graeme Gill : > Well of course icclib has support for accessing the vcgt tag. Sure, that's also a possibility. Richard. From pmjdebruijn at pcode.nl Thu Dec 10 17:00:03 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:00:03 +0100 Subject: LCMS dependency In-Reply-To: <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> Message-ID: <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> On Thu, Dec 10, 2009 at 5:18 PM, Richard Hughes wrote: > 2009/12/10 Graeme Gill : >> Well of course icclib has support for accessing the vcgt tag. > > Sure, that's also a possibility. Well, lcms is used by most OSS graphical apps for color management, so lcms is a natural choice when we don't want to introduce new dependancies into most systems in general. But as long as lcms2 hasn't been released and packaged into most distro's, it makes testing gcm a bit harder, which means feedback on git versions might degrade. Another slightly nasty option would be, to use lcms2, but just (for now) include the code into the gcm repo until it has been released, and just statically compile it into the gcm binaries. And yes, I'll wash my mouth with some soap :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 10 17:11:56 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:11:56 +0100 Subject: The new TRC graphs Message-ID: <4c2c37f90912100911o6f9c4b23y4648e95b2390d82a@mail.gmail.com> Hi, Just thought I should mention this... The new TRC graphs genuinely kick-ass! Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 10 17:14:31 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:14:31 +0000 Subject: LCMS dependency In-Reply-To: <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> Message-ID: <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > Well, lcms is used by most OSS graphical apps for color management, so > lcms is a natural choice when we don't want to introduce new > dependancies into most systems in general. Right. > But as long as lcms2 hasn't been released and packaged into most > distro's, it makes testing gcm a bit harder, which means feedback on > git versions might degrade. Agreed. > Another slightly nasty option would be, to use lcms2, but just (for > now) include the code into the gcm repo until it has been released, > and just statically compile it into the gcm binaries. Ick, ick. > And yes, I'll wash my mouth with some soap :) Okay, I think after what everyone has said I'll add an lcms dep, but not a lcms2 dep. We can convert when lcms2 has had it's first public release. Richard. From pmjdebruijn at pcode.nl Thu Dec 10 17:15:23 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:15:23 +0100 Subject: Bug: All profiles are 256 bytes Message-ID: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> Hi, I just built git again, and I noticed that GCM (on the profiles tab), tells me all my ICC profiles are 256 bytes... There is definitely something fishy about this :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 10 17:19:10 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:19:10 +0100 Subject: CIE widget drawing odd graphs In-Reply-To: <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> Message-ID: <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> On Sun, Dec 6, 2009 at 9:19 PM, Richard Hughes wrote: > 2009/12/6 Pascal de Bruijn : >> I might have uncovered some minor issues with the CIE widget drawing code. > > They look pretty major to me! Right :) >> I've attached an archive containing several non-standard profiles, >> which are drawn quite oddly. > > After the release I would like to look at the widget drawing code and > the profile extraction code (there are still quite a lot of FIXME's > and unimplemented stuff) and add a boat load more self tests and > internal validations. I fear I might not have time in the next couple > of weeks (lots of other work to do, which I've been putting off for > ages) so if anyone is familiar with XYZ co-ordinates and ICC profiles > I would appreciate any code review and possible patches. I just built git again, and I noticed XYZ drawing has improved... But I didn't see any actual improvements to the CIE code in git... I assume the problem wasn't in the CIE XYZ widget, but in the data that was fed to it? Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 10 17:42:38 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:42:38 +0000 Subject: Bug: All profiles are 256 bytes In-Reply-To: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> References: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> Message-ID: <15e53e180912100942t39465d40rf8875000038e9378@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > I just built git again, and I noticed that GCM (on the profiles tab), > tells me all my ICC profiles are 256 bytes... There is definitely > something fishy about this :) commit 99e9f26fe78202515187dc14b93917664ad01dc0 Author: Richard Hughes Date: Thu Dec 10 17:41:49 2009 +0000 trivial: show the file size in the prefs dialog, not the clut size Richard. From hughsient at gmail.com Thu Dec 10 17:43:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:43:40 +0000 Subject: CIE widget drawing odd graphs In-Reply-To: <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> Message-ID: <15e53e180912100943o413617a4j3d75cc58a096540b@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > I just built git again, and I noticed XYZ drawing has improved... But > I didn't see any actual improvements to the CIE code in git... The profile code had a subtle signing bug that wasn't tested for. The change from gchar to guint8 fixed this bug (and possibly a few more). Richard. From web.kiddo at free.fr Thu Dec 10 20:56:13 2009 From: web.kiddo at free.fr (Jeff) Date: Thu, 10 Dec 2009 15:56:13 -0500 Subject: Incorrect parsing of .icm files? Message-ID: <1260478573.20024.33.camel@kusanagi> Hello, I wanted to give gnome-color-manager a try just out of curiosity. I own a Dell Inspiron 910 (a.k.a. Mini 9) netbook, which (to my untrained eye) seems to have good colors out of the box. I downloaded the Windows drivers for this netbook's monitor here: http://support.us.dell.com/support/downloads/download.aspx?releaseid=R229102&fileid=329762 It seems like the monitor is called ST2010, there's an ST2010B and ST2010F version of the driver... both contain the exact same ".icm" file (verified with md5sum), which is attached to this message. Now, here comes the actual question: loading this .icm file in GNOME Color Manager and using it creates obviously incorrect colors (as if someone punched contrast and saturation way up), compared to the default colors (no color profile loaded; colors appear neutral). With the ICC profile loaded, for example, gradients in the Clearlooks GTK theme lose definition: the slight gray gradients on the combobox widgets disappears completely into white. I'm just wondering if - this is a bug in GNOME Color Manager - this is a buggy .icm file - it's a PEBKAC. I wish I could take a screenshot, but screenshots don't work in this case (and photo cameras have trouble photographing LCD screens)... -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ST2010.icm Type: image/x-tga Size: 552 bytes Desc: URL: From hughsient at gmail.com Thu Dec 10 21:24:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 21:24:28 +0000 Subject: Incorrect parsing of .icm files? In-Reply-To: <1260478573.20024.33.camel@kusanagi> References: <1260478573.20024.33.camel@kusanagi> Message-ID: <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> 2009/12/10 Jeff : > Now, here comes the actual question: loading this .icm file in GNOME Color > Manager and using it creates obviously incorrect colors (as if someone > punched contrast and saturation way up), compared to the default colors (no > color profile loaded; colors appear neutral). With the ICC profile loaded, > for example, gradients in the Clearlooks GTK theme lose definition: the > slight gray gradients on the combobox widgets disappears completely into > white. > > I'm just wondering if > - this is a bug in GNOME Color Manager > - this is a buggy .icm file Both, I think. If you try gnome-power-manager in git master, it won't "do" anything, as the profile does not have a vcgt or mlut tag, and even using the linear tables doesn't do anything. Even trying using dispwin doesn't appear to do anything. If you use the 2.29.1 release there's a bug where we're using the xTRC values as a gamma table (eek). Richard. From hughsient at gmail.com Thu Dec 10 21:28:59 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 21:28:59 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B202D16.1000901@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> Message-ID: <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> 2009/12/9 Graeme Gill : > Progromatically, one would do the equivalent, e.g. using icclib > open the profile, create a lookup object to do forward absolute colorimetric > to XYZ space, and then feed colorant only combinations through it. > icclu.c will illustrate the details. I'm sure something similar is > possible using Marti's lcms. I've just added code to git master to do this, and it indeed works to get the gamut range for a few scanner and print (RGB) profiles I've got here. It doesn't seem to work for CYMK profiles, although I'm sure that's something crazy I'm doing (or not doing..). Anyway, I'll give it another go to get CYMK profiles working correctly. Richard. From graeme2 at argyllcms.com Thu Dec 10 21:51:35 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 11 Dec 2009 08:51:35 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> Message-ID: <4B216D67.2080307@argyllcms.com> Richard Hughes wrote: > I've just added code to git master to do this, and it indeed works to > get the gamut range for a few scanner and print (RGB) profiles I've > got here. It doesn't seem to work for CYMK profiles, although I'm sure > that's something crazy I'm doing (or not doing..). Subtractive media profiles usually have less easily defined gamut extents than additive (RGB) profiles. Unlike additive, you can't expect it to be a nice polygon in XYZ/xy/u'v' space. Something roughly accurate for a chromaticity diagram would involve sampling along the color girdle (ie. going through combinations between CMY: 1 0 0, 1 1 0, 0 1 0, 0 1 1, 0 0 1, 1 0 1, e.g. 1 0 0, 1 .1 0, 1 .2 0 .. 1 .9 0, 1 1 0, .9 1 0, .8 1 0, ... etc.) Graeme Gill. From hughsient at gmail.com Thu Dec 10 22:08:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 22:08:49 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B216D67.2080307@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> <4B216D67.2080307@argyllcms.com> Message-ID: <15e53e180912101408w4b7dca45o4dac5869574a39bd@mail.gmail.com> 2009/12/10 Graeme Gill : > 1 1 0, 0 1 0, 0 1 1, 0 0 1, 1 0 1, > e.g. 1 0 0, 1 .1 0, 1 .2 0 .. 1 .9 0, 1 1 0, .9 1 0, .8 1 0, ... etc.) Excellent, this makes sense. I'll give it a go tomorrow. Thanks again. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 22:35:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 01:35:09 +0300 Subject: Incorrect parsing of .icm files? In-Reply-To: <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> References: <1260478573.20024.33.camel@kusanagi> <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> Message-ID: <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> On 12/11/09, Richard Hughes wrote: > Both, I think. If you try gnome-power-manager in git master, it won't > "do" anything, as the profile does not have a vcgt or mlut tag, and > even using the linear tables doesn't do anything. Even trying using > dispwin doesn't appear to do anything. If you use the 2.29.1 release > there's a bug where we're using the xTRC values as a gamma table > (eek). BTW, what's new regarding restoration of applied profile after dehibernation and leaving a screensaver? Alexandre From hughsient at gmail.com Fri Dec 11 08:55:43 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 08:55:43 +0000 Subject: Incorrect parsing of .icm files? In-Reply-To: <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> References: <1260478573.20024.33.camel@kusanagi> <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> Message-ID: <15e53e180912110055s2a1bc938ob58b1ef4508ef83a@mail.gmail.com> 2009/12/10 Alexandre Prokoudine : > BTW, what's new regarding restoration of applied profile after > dehibernation and leaving a screensaver? If you're using a KMS driver (the future) then the "thaw" should re-apply the per-output gamma tables. With the screensaver I want to work on a patch that either: * saves the gamma table before the table is destroyed * calls gcm-apply after the screensaver comes back from black I don't use a screensaver myself, but if nobody does the gnome-screensaver work I guess I'll have to add that to my list. Richard. From knizek.confy at volny.cz Fri Dec 11 09:46:18 2009 From: knizek.confy at volny.cz (Milan =?UTF-8?Q?Kn=C3=AD=C5=BEek?=) Date: Fri, 11 Dec 2009 10:46:18 +0100 Subject: shared-color-profiles 0.1.0 release In-Reply-To: <15e53e180912101003s2a013080qf839379565a40964@mail.gmail.com> References: <15e53e180912091151y1f2303acw48cedde3eb0231f@mail.gmail.com> <1260467822.3147.14.camel@athlon> <15e53e180912101003s2a013080qf839379565a40964@mail.gmail.com> Message-ID: <1260524778.2203.3.camel@athlon> (I am forwarding your response to the emailing list, since my original intention was not to email just you but the list and I did not notice the "reply" address is just yours.) Possibly someone knows about other pre-packaged profiles? regards, Milan Richard Hughes p??e v ?t 10. 12. 2009 v 18:03 +0000: > 2009/12/10 Milan Kn??ek : > > Will the project make "icc-profiles" and "oyranos-icc" obsolete or > > rather complementary? > > I hope we can standardize on one -- I must admit I had not heard of > icc-profiles until you pointed me at them. > > In the future, we'll be distributing shared-color-targets as well (8.7 > targets) to make all this stuff just work out of the box. > > Richard. > Milan Knizek knizek (dot) confy (at) volny (dot) cz http://www.milan-knizek.net - About linux and photography (Czech language only) From lars.tore at mulebakken.net Fri Dec 11 15:11:33 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 11 Dec 2009 16:11:33 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912110711n7d6a6e0ub9a3bf5b9f817b2f@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: > the x and y coordinates of the white point from the device ICC > profile. > The white point cross is nice, but I wonder if it would be nice to add a more human readable text string with the whitepoint somewhere: The wtpt program included in liblcms-utils outputs the correlated temperature or the locus temperature, so I guess it's easy to this with lcms. Here are some sample output: profile 1: White point near 4937K profile 2: WhitePoint : D65 (daylight) And I think this is more readable to most people than xy values. Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 11 15:51:36 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 11 Dec 2009 16:51:36 +0100 Subject: LCMS dependency In-Reply-To: <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> Message-ID: <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> On Thu, Dec 10, 2009 at 6:14 PM, Richard Hughes wrote: > 2009/12/10 Pascal de Bruijn : >> Well, lcms is used by most OSS graphical apps for color management, so >> lcms is a natural choice when we don't want to introduce new >> dependancies into most systems in general. > > Right. I see a lot of commits regarding lcms in git... Has all the ICC parsing code already been removed from gcm? I'm just asking this out of curiosity... Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Fri Dec 11 17:47:18 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 11 Dec 2009 18:47:18 +0100 Subject: Regression: the TRC graphs disappear on working space/input profiles Message-ID: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> Hi, With today's git checkout the TRC graphs disappear with working space profiles and input profiles. Is this intentional? Or just transitory during the lcms migration? Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Fri Dec 11 17:47:30 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 20:47:30 +0300 Subject: apply/session Message-ID: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> Hi, So, to get things right, what exactly is the difference between gcm-session and gcm-apply? Alexandre From hughsient at gmail.com Fri Dec 11 17:50:06 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:50:06 +0000 Subject: LCMS dependency In-Reply-To: <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> Message-ID: <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> 2009/12/11 Pascal de Bruijn : > I see a lot of commits regarding lcms in git... > Has all the ICC parsing code already been removed from gcm? About half. To remove the other half we need lcms2, which may be some months away. Richard. From hughsient at gmail.com Fri Dec 11 17:51:47 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:51:47 +0000 Subject: apply/session In-Reply-To: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> Message-ID: <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> 2009/12/11 Alexandre Prokoudine : > So, to get things right, what exactly is the difference between > gcm-session and gcm-apply? gcm-session is the dbus-activated session service, that provides an interface applications like gimp can use. gcm-apply is just a quick hack that sets (or resets) all the CLUT values to what they should be, in case something destroyed them (e.g. hibernate, screensaver dim, etc). Richard From hughsient at gmail.com Fri Dec 11 17:52:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:52:49 +0000 Subject: Regression: the TRC graphs disappear on working space/input profiles In-Reply-To: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> References: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> Message-ID: <15e53e180912110952s3f10d3dmc6305f350a17548@mail.gmail.com> 2009/12/11 Pascal de Bruijn : > With today's git checkout the TRC graphs disappear with working space > profiles and input profiles. > Is this intentional? Or just transitory during the lcms migration? Intentional, but they'll be coming back hopefully on Monday when I merge proper TRC support. Richard From alexandre.prokoudine at gmail.com Fri Dec 11 18:02:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 21:02:09 +0300 Subject: apply/session In-Reply-To: <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> Message-ID: <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> On Fri, Dec 11, 2009 at 8:51 PM, Richard Hughes wrote: > 2009/12/11 Alexandre Prokoudine : >> So, to get things right, what exactly is the difference between >> gcm-session and gcm-apply? > > gcm-session is the dbus-activated session service, that provides an > interface applications like gimp can use. And it's not currently used by anything, if I get it right? > gcm-apply is just a quick > hack that sets (or resets) all the CLUT values to what they should be, > in case something destroyed them (e.g. hibernate, screensaver dim, > etc). So vcgt part and setting the atom is done entirely by gcm-prefs? Alexandre From alexandre.prokoudine at gmail.com Fri Dec 11 18:06:41 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 21:06:41 +0300 Subject: LCMS dependency In-Reply-To: <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> Message-ID: <733f2c730912111006s640088c6g6b11a1266c54e4c8@mail.gmail.com> On Fri, Dec 11, 2009 at 8:50 PM, Richard Hughes wrote: >> Has all the ICC parsing code already been removed from gcm? > > About half. To remove the other half we need lcms2, which may be some > months away. Is this why TRC curve has (temp.) gone? :) Alexandre From claude at 2xlibre.net Fri Dec 11 22:36:40 2009 From: claude at 2xlibre.net (Claude Paroz) Date: Fri, 11 Dec 2009 23:36:40 +0100 Subject: Why not using strftime? Message-ID: <1260571000.4145.121.camel@shuttle> Hi, As I didn't find any bug tracker now, I'm reporting this here. I wonder why you're using a complicated machinery with gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of simply using strftime call in gcm-profile.c Something like (not tested code): ret = cmsTakeCreationDateTime (&created, priv->lcms_profile); if (ret) - priv->datetime = gcm_utils_format_date_time (1900 +created.tm_year, created.tm_mon+1, created + /* Translators: this is the profile creation date strftime format */ + strftime (priv->datetime, sizeof(priv->datetime), _("%d %B %Y, %H:%M:%S"), created); Months names should not be translated again, this is a waste of time for translators. Claude P.S. Please cc me on answer From hughsient at gmail.com Sat Dec 12 09:01:19 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 12 Dec 2009 09:01:19 +0000 Subject: apply/session In-Reply-To: <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> Message-ID: <15e53e180912120101n74bc1e92i31b4359ccaab961e@mail.gmail.com> 2009/12/11 Alexandre Prokoudine : >> gcm-session is the dbus-activated session service, that provides an >> interface applications like gimp can use. > > And it's not currently used ?by anything, if I get it right? Correct. It's the egg part of the chicken-and-egg question. >> gcm-apply is just a quick >> hack that sets (or resets) all the CLUT values to what they should be, >> in case something destroyed them (e.g. hibernate, screensaver dim, >> etc). > > So vcgt part and setting the atom is done entirely by gcm-prefs? No, when gcm "set the CLUT" it actually: * Applies the CLUT to each adaptor * Sets the _ICC_PROFILE atom on each output * Sets the _ICC_PROFILE atom on the screen for the primary output To apply the clut we extract the VCGT and then push it into the video card. Settign the CLUT is done by gcm-apply and gcm-prefs. You could argue the latter is redundant, unless you're changing the profile, and I would agree. Richard. From hughsient at gmail.com Sat Dec 12 09:21:57 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 12 Dec 2009 09:21:57 +0000 Subject: Why not using strftime? In-Reply-To: <1260571000.4145.121.camel@shuttle> References: <1260571000.4145.121.camel@shuttle> Message-ID: <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> 2009/12/11 Claude Paroz : > As I didn't find any bug tracker now, I'm reporting this here. I wonder > why you're using a complicated machinery with > gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of > simply using strftime call in gcm-profile.c Originally because we got some complicated date structure to decode, now we're using lcms we just get a nice sane tm struct. I've fixed things up in the following commit: commit 6607487433a1bf074638d6833bd7c493ef1d75ff Author: Richard Hughes Date: Sat Dec 12 09:19:57 2009 +0000 Use strftime rather than our own hand-rolled function :100644 100644 a6abf96... 0abfb35... M src/gcm-profile.c :100644 100644 203466c... 76e5b24... M src/gcm-utils.c :100644 100644 bc3ad9d... a570d6b... M src/gcm-utils.h Thanks Richard. From claude at 2xlibre.net Sat Dec 12 09:48:28 2009 From: claude at 2xlibre.net (Claude Paroz) Date: Sat, 12 Dec 2009 10:48:28 +0100 Subject: Why not using strftime? In-Reply-To: <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> References: <1260571000.4145.121.camel@shuttle> <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> Message-ID: <1260611308.3833.38.camel@shuttle> Le samedi 12 d?cembre 2009 ? 09:21 +0000, Richard Hughes a ?crit : > 2009/12/11 Claude Paroz : > > As I didn't find any bug tracker now, I'm reporting this here. I wonder > > why you're using a complicated machinery with > > gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of > > simply using strftime call in gcm-profile.c > > Originally because we got some complicated date structure to decode, > now we're using lcms we just get a nice sane tm struct. I've fixed > things up in the following commit: > > commit 6607487433a1bf074638d6833bd7c493ef1d75ff > Author: Richard Hughes > Date: Sat Dec 12 09:19:57 2009 +0000 > > Use strftime rather than our own hand-rolled function > > :100644 100644 a6abf96... 0abfb35... M src/gcm-profile.c > :100644 100644 203466c... 76e5b24... M src/gcm-utils.c > :100644 100644 bc3ad9d... a570d6b... M src/gcm-utils.h > > Thanks Great, thanks ! Claude From pmjdebruijn at pcode.nl Mon Dec 14 14:44:19 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 14 Dec 2009 15:44:19 +0100 Subject: Patch: Debian binary detection fix In-Reply-To: <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> Message-ID: <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> On Thu, Nov 26, 2009 at 10:07 PM, Richard Hughes wrote: > 2009/11/26 Pascal de Bruijn : >> Therefore the detect should _first_ detect if the Debian renamed >> binaries exist, and if not, fallback to the original binary names. > > Agreed. I've attached the patch I've committed. Thanks. Maybe it's a thought to (last) search /usr/local/bin as well for the standard binary names? Regards, Pascal de Bruijn From hughsient at gmail.com Mon Dec 14 15:13:51 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 14 Dec 2009 15:13:51 +0000 Subject: Patch: Debian binary detection fix In-Reply-To: <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> Message-ID: <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> 2009/12/14 Pascal de Bruijn : > Maybe it's a thought to (last) search /usr/local/bin as well for the > standard binary names? Sure. I'll do a deal: You code it, and I'll apply it. If you need instructions, I can explain how. Richard. From hughsient at gmail.com Wed Dec 16 11:57:19 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 16 Dec 2009 11:57:19 +0000 Subject: PackageKit integration Message-ID: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> gnome-color-manager in git master now integrates with PackageKit in the following ways: * Offers to install argyllcms if the user tries to calibrate without argyllcms installed * Offers to install shared-color-targets if they are not installed, and the user tries to calibrate a non-display device This means gnome-color-manager can be shipped on live-cd's without bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) shared-color-targets, just in case the user wants to calibrate a device. If people want to turn off the PackageKit bits, then I can add a GConf key or #ifdef as required. Richard. From pmjdebruijn at pcode.nl Wed Dec 16 12:23:52 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 16 Dec 2009 13:23:52 +0100 Subject: PackageKit integration In-Reply-To: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> Message-ID: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> On Wed, Dec 16, 2009 at 12:57 PM, Richard Hughes wrote: > gnome-color-manager in git master now integrates with PackageKit in > the following ways: > > * Offers to install argyllcms if the user tries to calibrate without > argyllcms installed > * Offers to install shared-color-targets if they are not installed, > and the user tries to calibrate a non-display device > > This means gnome-color-manager can be shipped on live-cd's without > bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) > shared-color-targets, just in case the user wants to calibrate a > device. > > If people want to turn off the PackageKit bits, then I can add a GConf > key or #ifdef as required. Since not all distro's don't use PackageKit that would be very welcome. A ./configure option would probably be best? Since this is a distro specific setting, it needs to be set in the packaging process. And GConf is very user specific, so it would seem odd to use GConf for this particular purpose. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 16 12:32:06 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 16 Dec 2009 13:32:06 +0100 Subject: PackageKit integration In-Reply-To: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> Message-ID: <4c2c37f90912160432r5cc68f8as29bc4d7419eeecd8@mail.gmail.com> On Wed, Dec 16, 2009 at 1:23 PM, Pascal de Bruijn wrote: > On Wed, Dec 16, 2009 at 12:57 PM, Richard Hughes wrote: >> gnome-color-manager in git master now integrates with PackageKit in >> the following ways: >> >> * Offers to install argyllcms if the user tries to calibrate without >> argyllcms installed >> * Offers to install shared-color-targets if they are not installed, >> and the user tries to calibrate a non-display device >> >> This means gnome-color-manager can be shipped on live-cd's without >> bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) >> shared-color-targets, just in case the user wants to calibrate a >> device. >> >> If people want to turn off the PackageKit bits, then I can add a GConf >> key or #ifdef as required. > > Since not all distro's don't use PackageKit that would be very welcome. > > A ./configure option would probably be best? Since this is a distro > specific setting, it needs to be set in the packaging process. > > And GConf is very user specific, so it would seem odd to use GConf for > this particular purpose. Hmm doesn't totem do similar things with codecs and the gstreamer packages? GCM should probably follow a similar approach... Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 16 12:42:07 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 16 Dec 2009 12:42:07 +0000 Subject: PackageKit integration In-Reply-To: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> Message-ID: <15e53e180912160442v2f95d4rb4b670389fb556d9@mail.gmail.com> 2009/12/16 Pascal de Bruijn : > Since not all distro's don't use PackageKit that would be very welcome. commit 494ebc7e84d21d2626e06aa7e009cc90625e0fd8 Author: Richard Hughes Date: Wed Dec 16 12:41:02 2009 +0000 Add a configure option --enable-packagekit (default yes) to choose if auto-install functionality should be included Richard. From joedalton2 at yahoo.dk Wed Dec 16 22:16:18 2009 From: joedalton2 at yahoo.dk (Joe Dalton) Date: Wed, 16 Dec 2009 14:16:18 -0800 (PST) Subject: bugreport gnome-color-manager translation Message-ID: <977326.61727.qm@web24408.mail.ird.yahoo.com> "Report-Msgid-Bugs-To: http://bugzilla.gnome.org/enter_bug.cgi?product=gnome-" "color-manager\n" but https://bugzilla.gnome.org/enter_bug.cgi?product=gnome-color-manager gives this message Sorry, either the product gnome-color-manager does not exist or you aren't authorized to enter a bug into it. Please choose a valid product. What does intent mean here. Is the sentence broken in some way/missing something? #: ../data/gnome-color-manager.schemas.in.h:7 msgid "The output intent to use, e.g. 'perceptual'." Hi, a google did not help me, and i have only found one danish translation, which maybe is correct mayby not. Please give some more info. #. TRANSLATORS: rendering intent: you probably want to google this #: ../src/gcm-prefs.c:2029 msgid "Perceptual" bye Joe Danish translation ___________________________________________________________ Skal du k?be ny bil? Sammenlign priser p? brugte biler med Kelkoo og find et godt tilbud! - Se mere her http://dk.yahoo.com/r/pat/mmb From hughsient at gmail.com Thu Dec 17 09:31:22 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 17 Dec 2009 09:31:22 +0000 Subject: bugreport gnome-color-manager translation In-Reply-To: <977326.61727.qm@web24408.mail.ird.yahoo.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> Message-ID: <15e53e180912170131p5c35354dv6e3df58be1ee6782@mail.gmail.com> 2009/12/16 Joe Dalton : > Sorry, either the product gnome-color-manager ?does not exist or you aren't authorized to enter a bug into it. Please choose a valid product. I've opened a bug to have the bugzilla product created: https://bugzilla.gnome.org/show_bug.cgi?id=604811 > What does intent mean here. Is the sentence broken in some way/missing something? > > #: ../data/gnome-color-manager.schemas.in.h:7 > msgid "The output intent to use, e.g. 'perceptual'." The output intent is the way that one range of colors is mapped to another range, for instance, do we clip the extremes of color, or do we try to scale it so they are all represented. Intent in English means "what you would do if possible" although in this technical speak it means more of "what method do you want me to use when transforming the color" > Hi, a google did not help me, and i have only found one danish translation, which maybe is correct mayby not. Please give some more info. > #. TRANSLATORS: rendering intent: you probably want to google this > #: ../src/gcm-prefs.c:2029 > msgid "Perceptual" This is how I understand perceptual rendering intent: '''This rendering intent maps color "smoothly", preserving relationships between similar colors. This prevents "gamut clipping" with its potential loss of detail and "tonal banding" problems. Gamut clipping occurs when colors that are different in the input image appear the same when printed. Perceptual rendering intent makes small compromises throughout the entire color space in order to preserve color relationships. It sacrifices some precision of in-gamut colors in order to ensure pleasing results.''' Does that make things clearer? Feel free to change the translator comments to something more sensible -- you know what you are doing. :-) Richard. From alexandre.prokoudine at gmail.com Thu Dec 17 12:29:54 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 17 Dec 2009 15:29:54 +0300 Subject: bugreport gnome-color-manager translation In-Reply-To: <977326.61727.qm@web24408.mail.ird.yahoo.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> Message-ID: <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> On 12/17/09, Joe Dalton wrote: > What does intent mean here. Is the sentence broken in some way/missing > something? > > #: ../data/gnome-color-manager.schemas.in.h:7 > msgid "The output intent to use, e.g. 'perceptual'." > > > Hi, a google did not help me, and i have only found one danish translation, > which maybe is correct mayby not. I'm pretty sure that both GIMP, Inkscape and Scribus have Danish translations. Check them re. "Color management" tab in Preferences of all these applications. I also highly recommend "Real world color management" book. It was translated into many languages, maybe Danish was among them. You will understand a lot more about color management after reading it, trust me on this :) Alexandre From anders at brander.dk Thu Dec 17 12:35:12 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 17 Dec 2009 13:35:12 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> Message-ID: <1261053312.7873.1.camel@smutogtab> Hi, On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: > I also highly recommend "Real world color management" book. It was > translated into many languages, maybe Danish was among them. You will > understand a lot more about color management after reading it, trust > me on this :) I second that, go read "Real world color management" now, if you haven't already. It doesn't have a danish translation by the way. Regards, Anders Brander From pmjdebruijn at pcode.nl Thu Dec 17 12:38:42 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 17 Dec 2009 13:38:42 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <1261053312.7873.1.camel@smutogtab> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> Message-ID: <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> On Thu, Dec 17, 2009 at 1:35 PM, Anders Brander wrote: > Hi, > > On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: >> I also highly recommend "Real world color management" book. It was >> translated into many languages, maybe Danish was among them. You will >> understand a lot more about color management after reading it, trust >> me on this :) > > I second that, go read "Real world color management" now, if you haven't > already. It doesn't have a danish translation by the way. I third that, the first 75% is written completely generic and os/application independent. It's an excellent book... Generally for most photographical work you'll want Perceptual... Relative Colorimetric is usually most useful when you have very high end equipment, thus limiting clipping... Regards, Pascal de Bruijn From lars.tore at mulebakken.net Thu Dec 17 15:11:12 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 17 Dec 2009 16:11:12 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> Message-ID: <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> On Thu, Dec 17, 2009 at 1:38 PM, Pascal de Bruijn wrote: > On Thu, Dec 17, 2009 at 1:35 PM, Anders Brander wrote: >> Hi, >> >> On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: >>> I also highly recommend "Real world color management" book. It was >>> translated into many languages, maybe Danish was among them. You will >>> understand a lot more about color management after reading it, trust >>> me on this :) >> >> I second that, go read "Real world color management" now, if you haven't >> already. It doesn't have a danish translation by the way. > > I third that, the first 75% is written completely generic and > os/application independent. It's an excellent book... > > Generally for most photographical work you'll want Perceptual... > > Relative Colorimetric is usually most useful when you have very high > end equipment, thus limiting clipping... > > Regards, > Pascal de Bruijn > _______________________________________________ > gnome-color-manager-list mailing list > gnome-color-manager-list at gnome.org > http://mail.gnome.org/mailman/listinfo/gnome-color-manager-list > I think the book Understanding color managment by Abhay Sharma is a better book, and I have both :-) About the translation, in Norwegian (it might be close to danish) the correct word is "Perseptuell". But I think we don't have to translate all that specific terms. Since most of the written literature are in english, I guess most people who use color managment have no problems with the english terms on the rendering intets. Actually I think at least in my language that translated word will create a lot of confusion since the words are very seldom used, and everyone use the English terms. I guess we can blame photoshop for this. Lars Tore Gustavsen From anders at brander.dk Thu Dec 17 15:20:42 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 17 Dec 2009 16:20:42 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> Message-ID: <1261063242.15170.1.camel@smutogtab> Hi, On Thu, 2009-12-17 at 16:11 +0100, Lars Tore Gustavsen wrote: > About the translation, in Norwegian (it might be close to danish) the > correct word is "Perseptuell". I just looked it up in the danish Rawstudio translation, it's translated as "Opfattelse" - but I don't think that's correct (and I'm danish). Regards, Anders Brander From hughsient at gmail.com Sat Dec 19 17:32:08 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 19 Dec 2009 17:32:08 +0000 Subject: Bugzilla account created Message-ID: <15e53e180912190932x47e70876xcb147ce2718e3bff@mail.gmail.com> I've arranged for a bugzilla component to be created. If you want to file a bug, use https://bugzilla.gnome.org/enter_bug.cgi?product=gnome-color-manager but be sure to check for duplicates first, using https://bugzilla.gnome.org/buglist.cgi?product=gnome-color-manager At the moment, there are no bugs, which means GCM must be feature complete and perfect in every way. I can't imagine this situation will last long :-) Richard. From robert.ancell at gmail.com Sun Dec 20 22:34:24 2009 From: robert.ancell at gmail.com (Robert Ancell) Date: Mon, 21 Dec 2009 09:34:24 +1100 Subject: Color management and Simple Scan Message-ID: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Hi all, I've been developing a basic scanning application called Simple Scan ( http://launchpad.net/simple-scan) and one request is to add color management to it: https://bugs.edge.launchpad.net/simple-scan/+bug/498029 I don't know much about color management so my question is: - What do I need to do to make Simple Scan work with GNOME Color Manager? - Is there a GPL project I can lift some code from? Thanks, --Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughsient at gmail.com Mon Dec 21 08:28:27 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 21 Dec 2009 08:28:27 +0000 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Message-ID: <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> 2009/12/20 Robert Ancell : > I don't know much about color management so my question is: > - What do I need to do to make Simple Scan work with GNOME Color Manager? > - Is there a GPL project I can lift some code from? Well, first you need to ask gnome-color-manager for a list of suitable profiles using the DBus interface: http://git.gnome.org/browse/gnome-color-manager/plain/src/org.gnome.ColorManager.xml If you know the device you're using (you probably should) then you can call GetProfilesForDevice() with the sysfs path, and a list of profiles will be returned. If more than one profile is returned you probably need to show it in a combo-box or something, but it would be quite unusual for more than one profile to be returned. If you get no results, then it's probably because the user has never assigned a profile for that device (i.e. there is no a-priori mapping). In this case you can either default to not doing color conversion yourself, or instead call the GetProfilesForType() dbus method, using the type=scanner. This will return a list of profiles that are suitable for use with scanners. Now, when you have a filename of the profile you want to use it's an easy (!) case of acquiring the image like usual, and then you can either: * Just save the image with an embedded color profile (see bug https://bugzilla.gnome.org/show_bug.cgi?id=604610 -- I'm working as fast as I can on the infrastructure bits) * Convert the image from device-mapping to something like sRGB using lcms (there are examples in the lcms tarball) I think it's probably best you do the first one, unless you know you're going to be using the images in a non-color-aware application. Anyway, I hope that helps, Richard. From alexandre.prokoudine at gmail.com Mon Dec 21 16:01:01 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 21 Dec 2009 19:01:01 +0300 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Message-ID: <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> On Mon, Dec 21, 2009 at 1:34 AM, Robert Ancell wrote: > - Is there a GPL project I can lift some code from? XSane sounds like an obvious source :) Alexandre From hughsient at gmail.com Mon Dec 21 16:08:05 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 21 Dec 2009 16:08:05 +0000 Subject: Color management and Simple Scan In-Reply-To: <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> Message-ID: <15e53e180912210808y5fe054c7p1f2d41f4fdb97ee1@mail.gmail.com> 2009/12/21 Alexandre Prokoudine : > XSane sounds like an obvious source :) Well, depends on how you like your sanity. Richard. From robert.ancell at gmail.com Mon Dec 21 22:57:54 2009 From: robert.ancell at gmail.com (Robert Ancell) Date: Tue, 22 Dec 2009 09:57:54 +1100 Subject: Color management and Simple Scan In-Reply-To: <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> Message-ID: <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> Thanks very much Richard! Very interested in the GDK patch - I need to make a patch there too to embed resolution information: https://bugs.edge.launchpad.net/simple-scan/+bug/484616 I think I need to do the conversion regardless so the preview is displayed correctly. Thanks for the lcms link. --Robert 2009/12/21 Richard Hughes > 2009/12/20 Robert Ancell : > > I don't know much about color management so my question is: > > - What do I need to do to make Simple Scan work with GNOME Color Manager? > > - Is there a GPL project I can lift some code from? > > Well, first you need to ask gnome-color-manager for a list of suitable > profiles using the DBus interface: > > http://git.gnome.org/browse/gnome-color-manager/plain/src/org.gnome.ColorManager.xml > > If you know the device you're using (you probably should) then you can > call GetProfilesForDevice() with the sysfs path, and a list of > profiles will be returned. If more than one profile is returned you > probably need to show it in a combo-box or something, but it would be > quite unusual for more than one profile to be returned. If you get no > results, then it's probably because the user has never assigned a > profile for that device (i.e. there is no a-priori mapping). In this > case you can either default to not doing color conversion yourself, or > instead call the GetProfilesForType() dbus method, using the > type=scanner. This will return a list of profiles that are suitable > for use with scanners. > > Now, when you have a filename of the profile you want to use it's an > easy (!) case of acquiring the image like usual, and then you can > either: > > * Just save the image with an embedded color profile (see bug > https://bugzilla.gnome.org/show_bug.cgi?id=604610 -- I'm working as > fast as I can on the infrastructure bits) > * Convert the image from device-mapping to something like sRGB using > lcms (there are examples in the lcms tarball) > > I think it's probably best you do the first one, unless you know > you're going to be using the images in a non-color-aware application. > > Anyway, I hope that helps, > > Richard. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughsient at gmail.com Tue Dec 22 09:13:09 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 22 Dec 2009 09:13:09 +0000 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> Message-ID: <15e53e180912220113n329e8f81j982c21203734945f@mail.gmail.com> 2009/12/21 Robert Ancell : > Very interested in the GDK patch - I need to make a patch there too to embed > resolution information: > https://bugs.edge.launchpad.net/simple-scan/+bug/484616 Sure, looks like you just need to add a new pixbuf-option. It's not that hard, as long as you're comfortable just implementing a spec in otherwise fairly hairy code. :-) If you need a hand, give me a shout on or off list. Richard. From bcowan at fastmail.co.uk Thu Dec 24 20:06:29 2009 From: bcowan at fastmail.co.uk (Bruce Cowan) Date: Thu, 24 Dec 2009 20:06:29 +0000 Subject: String fixes Message-ID: <1261685189.7547.4.camel@localhost.localdomain> I've found a few string errors. Here's a patch to fix them. I thought I better post it here rather than just commit straight to the repository, because there is a potential point of disagreement. I've changed the strformat format to the en_US one, which could be a bit contentious. -- Bruce Cowan -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Minor-string-fixes.patch Type: text/x-patch Size: 2223 bytes Desc: not available URL: From hughsient at gmail.com Fri Dec 25 11:26:32 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 25 Dec 2009 11:26:32 +0000 Subject: Happy Christmas Message-ID: <15e53e180912250326n7d35ca86tc59fb282fd940a1c@mail.gmail.com> Hey guys. I just wanted to send to you all a Happy Christmas, and my thanks for the hard work you've put in this year. It's appreciated. Richard. From hughsient at gmail.com Fri Dec 25 11:28:10 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 25 Dec 2009 11:28:10 +0000 Subject: String fixes In-Reply-To: <1261685189.7547.4.camel@localhost.localdomain> References: <1261685189.7547.4.camel@localhost.localdomain> Message-ID: <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> 2009/12/24 Bruce Cowan : > I've found a few string errors. Here's a patch to fix them. I thought I > better post it here rather than just commit straight to the repository, > because there is a potential point of disagreement. The first two are obvious, so please commit those. > I've changed the strformat format to the en_US one, which could be a bit > contentious. I agree it's probably the correct thing to do, but it does however want me to rip my eyes out every time I see that format. :-) So, please commit the entire patch. Thanks. Richard. From pmjdebruijn at pcode.nl Fri Dec 25 12:28:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 25 Dec 2009 13:28:33 +0100 Subject: String fixes In-Reply-To: <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> References: <1261685189.7547.4.camel@localhost.localdomain> <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> Message-ID: <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> On Fri, Dec 25, 2009 at 12:28 PM, Richard Hughes wrote: > 2009/12/24 Bruce Cowan : >> I've found a few string errors. Here's a patch to fix them. I thought I >> better post it here rather than just commit straight to the repository, >> because there is a potential point of disagreement. > > The first two are obvious, so please commit those. > >> I've changed the strformat format to the en_US one, which could be a bit >> contentious. > > I agree it's probably the correct thing to do, but it does however > want me to rip my eyes out every time I see that format. :-) /me Shudders... We're introducing AM/PM time... oh well seems correct for the en_US locale... However, if one we're to switch to the en_UK locale, would AM/PM time be gone again? Regards, Pascal de Bruijn From bcowan at fastmail.co.uk Fri Dec 25 14:37:16 2009 From: bcowan at fastmail.co.uk (Bruce Cowan) Date: Fri, 25 Dec 2009 14:37:16 +0000 Subject: String fixes In-Reply-To: <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> References: <1261685189.7547.4.camel@localhost.localdomain> <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> Message-ID: <1261751836.1631.0.camel@localhost.localdomain> On Fri, 2009-12-25 at 13:28 +0100, Pascal de Bruijn wrote: > On Fri, Dec 25, 2009 at 12:28 PM, Richard Hughes wrote: > > 2009/12/24 Bruce Cowan : > >> I've found a few string errors. Here's a patch to fix them. I thought I > >> better post it here rather than just commit straight to the repository, > >> because there is a potential point of disagreement. > > > > The first two are obvious, so please commit those. > > > >> I've changed the strformat format to the en_US one, which could be a bit > >> contentious. > > > > I agree it's probably the correct thing to do, but it does however > > want me to rip my eyes out every time I see that format. :-) > > /me Shudders... > > We're introducing AM/PM time... oh well seems correct for the en_US locale... > > However, if one we're to switch to the en_UK locale, would AM/PM time > be gone again? Yup, I'll update the en_GB translation later as well, meaning we'll see 24 hour time still. -- Bruce Cowan From joedalton2 at yahoo.dk Fri Dec 25 17:59:33 2009 From: joedalton2 at yahoo.dk (Joe Dalton) Date: Fri, 25 Dec 2009 17:59:33 +0000 (GMT) Subject: bugreport gnome-color-manager translation In-Reply-To: <15e53e180912170131p5c35354dv6e3df58be1ee6782@mail.gmail.com> Message-ID: <381861.61449.qm@web24403.mail.ird.yahoo.com> thanx --- Den tors 17/12/09 skrev Richard Hughes : > Fra: Richard Hughes > Emne: Re: bugreport gnome-color-manager translation > Til: "Joe Dalton" > Cc: gnome-color-manager-list at gnome.org > Dato: torsdag 17. december 2009 10.31 > 2009/12/16 Joe Dalton : > > Sorry, either the product gnome-color-manager ?does > not exist or you aren't authorized to enter a bug into it. > Please choose a valid product. > > I've opened a bug to have the bugzilla product created: > https://bugzilla.gnome.org/show_bug.cgi?id=604811 > > > What does intent mean here. Is the sentence broken in > some way/missing something? > > > > #: ../data/gnome-color-manager.schemas.in.h:7 > > msgid "The output intent to use, e.g. 'perceptual'." > > The output intent is the way that one range of colors is > mapped to > another range, for instance, do we clip the extremes of > color, or do > we try to scale it so they are all represented. Intent in > English > means "what you would do if possible" although in this > technical speak > it means more of "what method do you want me to use when > transforming > the color" > > > Hi, a google did not help me, and i have only found > one danish translation, which maybe is correct mayby not. > Please give some more info. > > #. TRANSLATORS: rendering intent: you probably want to > google this > > #: ../src/gcm-prefs.c:2029 > > msgid "Perceptual" > > This is how I understand perceptual rendering intent: > > '''This rendering intent maps color "smoothly", preserving > relationships between similar colors. This prevents "gamut > clipping" > with its potential loss of detail and "tonal banding" > problems. Gamut > clipping occurs when colors that are different in the input > image > appear the same when printed. Perceptual rendering intent > makes small > compromises throughout the entire color space in order to > preserve > color relationships. It sacrifices some precision of > in-gamut colors > in order to ensure pleasing results.''' > > Does that make things clearer? Feel free to change the > translator > comments to something more sensible -- you know what you > are doing. > :-) > > Richard. > Find din nye laptop p? kelkoo.dk. Se de gode tilbud her - http://dk.yahoo.com/r/pat/mm From pmjdebruijn at pcode.nl Sun Dec 27 18:33:43 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 27 Dec 2009 19:33:43 +0100 Subject: Patch: Debian binary detection fix In-Reply-To: <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> Message-ID: <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> On Mon, Dec 14, 2009 at 4:13 PM, Richard Hughes wrote: > 2009/12/14 Pascal de Bruijn : >> Maybe it's a thought to (last) search /usr/local/bin as well for the >> standard binary names? > > Sure. I'll do a deal: You code it, and I'll apply it. If you need > instructions, I can explain how. I've submitted it has an enhancement on GNOME Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=605552 Regards, Pascal de Bruijn From hughsient at gmail.com Mon Dec 28 10:26:15 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 28 Dec 2009 10:26:15 +0000 Subject: Patch: Debian binary detection fix In-Reply-To: <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> Message-ID: <15e53e180912280226r2b1d3265x103a12fd65174192@mail.gmail.com> 2009/12/27 Pascal de Bruijn : > I've submitted it has an enhancement on GNOME Bugzilla: Committed, thanks. Richard. From pmjdebruijn at pcode.nl Mon Dec 28 18:20:54 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 28 Dec 2009 19:20:54 +0100 Subject: Documentation Images: Display OSD Message-ID: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Hi, I took a few shots of my display's OSD, mainly for my own purposes, but maybe they'll come in handy for GNOME Color Manager's documentation. These shots aren't perfect (at all), but it's a start. I can put these under any license you like... I'm fine with GPL and CC-BY-AC. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: lowres-osd-jpegs.zip Type: application/zip Size: 213749 bytes Desc: not available URL: From niklas.park at gmail.com Tue Dec 29 10:44:50 2009 From: niklas.park at gmail.com (Niklas Park) Date: Tue, 29 Dec 2009 11:44:50 +0100 Subject: Documentation Images: Display OSD In-Reply-To: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> References: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Message-ID: A little warning seems to be in place here. I can't see how you could put these screenshots under CC, FDL or any other license, since the copyright clearly belongs to the manufacturer of the device, or the person or entity the manufacturer licenses the OSD from. I can't really see that a screengrab could constitute original art under any jurisdiction. The best one could hope for is that using these images constitutes fair use, of course with correct attribution to the copyright holder, but IANAL. Kindly /Niklas Park 2009/12/28 Pascal de Bruijn > Hi, > > I took a few shots of my display's OSD, mainly for my own purposes, > but maybe they'll come in handy for GNOME Color Manager's > documentation. > > These shots aren't perfect (at all), but it's a start. > > I can put these under any license you like... I'm fine with GPL and > CC-BY-AC. > > Regards, > Pascal de Bruijn > > _______________________________________________ > gnome-color-manager-list mailing list > gnome-color-manager-list at gnome.org > http://mail.gnome.org/mailman/listinfo/gnome-color-manager-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pmjdebruijn at pcode.nl Tue Dec 29 14:41:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 29 Dec 2009 15:41:33 +0100 Subject: Documentation Images: Display OSD In-Reply-To: References: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Message-ID: <4c2c37f90912290641w8d6dff1y675a3a07844da84d@mail.gmail.com> On Tue, Dec 29, 2009 at 11:44 AM, Niklas Park wrote: > A little warning seems to be in place here. > I can't see how you could put these screenshots under CC, FDL or any other > license, since the copyright clearly belongs to the manufacturer of the > device, or the person or entity the manufacturer licenses the OSD from. I > can't really see that a screengrab could constitute original art under any > jurisdiction. Well, you've got a fair point there... In this case that would be HP... > The best one could hope for is that using these images constitutes fair use, > of course with correct attribution to the copyright holder, but IANAL. Since it's not a 1:1 (pixel-perfect) copy, and doesn't contain any brand logo's, it probably does... Regards, Pascal de Bruijn From hughsient at gmail.com Tue Dec 1 14:58:41 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 1 Dec 2009 14:58:41 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file Message-ID: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Attached is the new widget I've just added to gnome-color-manager git master. It allows us to show the user a preview of the gamut range of the profile in a graphical way, much like OSX does. You should be able to see the lightly shaded section. At the moment I've hardcoded the CIE REC 709 x-y values, but of course it's easy to change when put into GcmPrefs. So, the point of this email: I wanted to get the CIE x and y coordinates (or XYZ of course) of the three primary illuminants and the x and y coordinates of the white point from the device ICC profile. Is there an easy way to extract these end-point values from the ICC file? I guess mediaWhitePointTag is the latter half of the problem, but I'm not sure how to deal with the LUT values. Ideas welcome. Thanks. Richard. -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot.png Type: image/png Size: 49528 bytes Desc: not available URL: From lars.tore at mulebakken.net Tue Dec 1 16:43:31 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Tue, 1 Dec 2009 17:43:31 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912010843re5f1d81mef40fd5d57e50310@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: > Attached is the new widget I've just added to gnome-color-manager git > master. It allows us to show the user a preview of the gamut range of > the profile in a graphical way, much like OSX does. > > You should be able to see the lightly shaded section. At the moment > I've hardcoded the CIE REC 709 x-y values, but of course it's easy to > change when put into GcmPrefs. > > So, the point of this email: I wanted to get the CIE x and y > coordinates (or XYZ of course) of the three primary illuminants and > the x and y coordinates of the white point from the device ICC > profile. > > Is there an easy way to extract these end-point values from the ICC > file? I guess mediaWhitePointTag is the latter half of the problem, > but I'm not sure how to deal with the LUT values. Ideas welcome. > Thanks. > > Richard. > What about iccdump -v3 -t wtpt -t bkpt -t rXYZ -t gXYZ -t bXYZ myprofile.icc Lars Tore Gustavsen From lars.tore at mulebakken.net Tue Dec 1 18:12:38 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Tue, 1 Dec 2009 19:12:38 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912011012q42a44f4cvd385a8b97ca5f5eb@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: I'm not sure how to deal with the LUT values. Ideas welcome. > Thanks. > > Richard. > > _ I think I missed the last part. Can something like this be used? Just thinking loud here: echo 1 0 0|icclu -ff lutprofile.icc echo 0 1 0|icclu -ff lutprofile.icc echo 0 0 1|icclu -ff lutprofile.icc Lars Tore Gustavsen From graeme2 at argyllcms.com Wed Dec 2 00:19:40 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Wed, 02 Dec 2009 11:19:40 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <4B15B29C.2090105@argyllcms.com> Richard Hughes wrote: > Attached is the new widget I've just added to gnome-color-manager git > master. It allows us to show the user a preview of the gamut range of > the profile in a graphical way, much like OSX does. Professor Hunt would be jumping up and down in the first row at this point saying "Why are you using an XYZ chromaticity diagram ?" :-) :-) [Those who have been to the Color Imaging Conference will understand this.] Actually if you're wanting to plot Yxy values then it's appropriate, but I'll note that it's also very misleading in indicating gamut, since an xy diagram is highly visually non-uniform. A u' v' plot is a better choice to illustrate gamut, while still being linearly additive. > So, the point of this email: I wanted to get the CIE x and y > coordinates (or XYZ of course) of the three primary illuminants and > the x and y coordinates of the white point from the device ICC > profile. > > Is there an easy way to extract these end-point values from the ICC > file? I guess mediaWhitePointTag is the latter half of the problem, > but I'm not sure how to deal with the LUT values. Ideas welcome. The white point is in the white point tag (but see my final note). For a matrix profile the D50 adapted primaries are in the red/green/blue ColorantTags. For a cLut based profile, you would have to (as Lars Tore Gustavsen suggested) run the primary values through the profile (or do the equivalent of looking up what the cLUT values are at the primaries). Note a complication - you probably want the absolute colorant values, not the D50 adapted white ones. You need to undo the chromatic adaptation applied to the colorant tags, and/or do the lookup in absolute colorimetric intent mode. Note also that while absolute colorimetric interpretation of display profiles is perfectly consistent using Argyll/icclib based tools (and is consistent with the original sRGB and current AdobeRGB profiles), it probably will not be so for all other libaries and tools and profiles. This is due to a disagreement about the interpretation and intent of the ICC specification in this regard. Some display profiles have a white point tag of D50 even though this is not the white point of the display, and encode the absolute<->relative transformation in the ChromaticAdapation tag, meaning that the absolute white point and colorant values are not obtained using a normal absolute colorimetric lookup on these profiles. This latter interpretation is the one that has been adopted in the changes made for ICC V4 profiles, even though it is out of sync with much original ICC V2 practice. Graeme Gill. From hughsient at gmail.com Wed Dec 2 08:37:43 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 08:37:43 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B15B29C.2090105@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> Message-ID: <15e53e180912020037o6168de9dm4a20f429e7d665c@mail.gmail.com> 2009/12/2 Graeme Gill : > Actually if you're wanting to plot Yxy values then it's appropriate, > but I'll note that it's also very misleading in indicating gamut, > since an xy diagram is highly visually non-uniform. A u' v' plot > is a better choice to illustrate gamut, while still being linearly > additive. Sure, converting the code to plot u' v' should be pretty easy. The graph is really just for a passing glance, not for any serious comparison. > This is due to a disagreement about the interpretation and intent > of the ICC specification in this regard. Some display profiles > have a white point tag of D50 even though this is not the white > point of the display, and encode the absolute<->relative > transformation in the ChromaticAdapation tag, meaning that > the absolute white point and colorant values are not obtained > using a normal absolute colorimetric lookup on these profiles. This is exactly the information I needed, thanks very much for such a comprehensive answer. Richard. From hughsient at gmail.com Wed Dec 2 15:12:39 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 15:12:39 +0000 Subject: Uncaught bounce notification In-Reply-To: References: Message-ID: <15e53e180912020712l4506d8e1ja991ca132c63ae85@mail.gmail.com> 2009/12/2 : > i'v followed g-c-m using Pascal de Bruijn ppa (thanks to him), and for > these tests I'v built myself the last git (for CIE widget). For some reason your mail got bounced, apologies. I've forwarded it to the list. > So : > 1/ The window for listing screens are too big looking at the rest > (left/right panel) That should be fixed in git master now. > 2/ It's ok that this tool must be kept as simple as possible, but I > would love to have the possibility to set white K? That's on the assumption you're not using a hardware calibration device then? > 3/ if you play with icc profils, size of the right panel change and the > square of CIE widget may be crushed > 4/ I would love to see more (bigger/more details) the CIE widget when > clicking on it Sure, I'm not completely happy with the location or size of this widget. It seems to me I'm trying to cram the "I want to compare different profiles" and "assign a profile with a device" functionality into the single tab. maybe we just need to break out another tab called "Profiles" and list all the profiles along with some more information about them. In this way we can get rid of some of the clutter in the first page. Comments welcome. Richard. From ubuntu at kagou.fr Wed Dec 2 16:20:03 2009 From: ubuntu at kagou.fr (Patrice Vetsel) Date: Wed, 02 Dec 2009 17:20:03 +0100 Subject: Some tests Message-ID: <4B1693B3.8000809@kagou.fr> Hi, i'v followed g-c-m using Pascal de Bruijn ppa (thanks to him), and for these tests I'v built myself the last git (for CIE widget). So : 1/ The window for listing screens are too big looking at the rest (left/right panel) 2/ It's ok that this tool must be kept as simple as possible, but I would love to have the possibility to set white K? 3/ if you play with icc profils, size of the right panel change and the square of CIE widget may be crushed 4/ I would love to see more (bigger/more details) the CIE widget when clicking on it Thanks for this tool Regards Patrice Vetsel Aka Kagou From hughsient at gmail.com Wed Dec 2 16:47:21 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 16:47:21 +0000 Subject: Some tests In-Reply-To: <4B1693B3.8000809@kagou.fr> References: <4B1693B3.8000809@kagou.fr> Message-ID: <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> 2009/12/2 Patrice Vetsel : > 4/ I would love to see more (bigger/more details) the CIE widget when > clicking on it What do you all think of the attached mockups? Richard. -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-1.png Type: image/png Size: 42154 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-2.png Type: image/png Size: 86945 bytes Desc: not available URL: From pmjdebruijn at pcode.nl Wed Dec 2 17:09:30 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:09:30 +0100 Subject: Some tests In-Reply-To: <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> Message-ID: <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> On Wed, Dec 2, 2009 at 5:47 PM, Richard Hughes wrote: > 2009/12/2 Patrice Vetsel : >> 4/ I would love to see more (bigger/more details) the CIE widget when >> clicking on it > > What do you all think of the attached mockups? @Richard: Looks pretty good... The CIE widget is very nice. Though the profile whitepoint isn't being displayed yet, I'm assuming this is still on the TODO list. I also agree it would be nice (but not critical) if double clicking on the CIE widget would create a dialog with just the profile name and the CIE widget, for detailed viewing. And dare I ask, possibly with a right-click "Save (CIE diagram) as PNG/JPEG/PDF" option? Also, I notice the CIE widget has two "squares" of "space" to the left of the CIE diagram. Does this have a purpose? Otherwise couldn't this be decreased to a single white square. I also think either centering the CIE widget, or placing the buttons to the right would look a bit "better" esthetically... Or maybe move the button onder the list view? Giving the CIE diagram a bit more room to "grow"? @Patrice: I rebuild my PPA between 1 to 4 times per two days, depending on how interesting Richard's changes are... All in all it's pretty hard to keep up, and he's doing all the hard work :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 2 17:15:31 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:15:31 +0100 Subject: Idea: GTK/Cario CIE widget as a seperate library Message-ID: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> Hi, The following is just a thought, and is by no means very important... The GTK/Cairo CIE widget written for GCM is beginning to kick ass... However I guess it's GCM specific. Maybe for the long term it's a thought to separate it into a small library, so other apps can use it as well? For example whenever GIMP/UFraw open a dialog to import a new profile it would be very cool if the GNOME File Chooser would "preview" the profile by showing a CIE diagram. Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 2 17:35:04 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 17:35:04 +0000 Subject: Idea: GTK/Cario CIE widget as a seperate library In-Reply-To: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> References: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> Message-ID: <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > The GTK/Cairo CIE widget written for GCM is beginning to kick ass... > However I guess it's GCM specific. Not really, it's a standard GObject so it could be reused. > Maybe for the long term it's a thought to separate it into a small > library, so other apps can use it as well? Sure, there is loads of other common code that might be useful, but I don't think the project is at any level of internal ABI stability. Maybe we can revisit this idea in a few months time? Richard. From pmjdebruijn at pcode.nl Wed Dec 2 17:36:50 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:36:50 +0100 Subject: Profile description for laptop displays In-Reply-To: <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> References: <4c2c37f90911290453x143c6820nb27f3a6b6188e3c7@mail.gmail.com> <15e53e180911291342o2b8755aavad083da3dcd849d2@mail.gmail.com> <4c2c37f90911291527v71d2cf9cy6bb41add8c7b3683@mail.gmail.com> <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> Message-ID: <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> On Mon, Nov 30, 2009 at 11:10 AM, Richard Hughes wrote: > 2009/11/29 Pascal de Bruijn : >> I noticed you included the product_version as well, at least on my >> laptop it's actually the BIOS revision, I'm not sure what this is on >> other laptops. It's probably not that relevant... > > Hmm, it seems the manufacturers were in a rush when they designed your > laptop! The bios version, logically, belongs in bios_version :-) > > Could you attach the output of "cat /sys/class/dmi/id/*" and we'll try > and work around bios issues like this. Thanks. It's attached. >> I noticed you're currently filling the ICC Make with the username, it >> seems rather redundant with the copyright. Besides the point that the >> field wasn't ment to be used like that. I'd stick the EDID info there >> (even with laptops). For the Model field as well, EDID is most likely >> best and most accurate here. > > I deliberately changed it, as the vendor is supposed to be the person > or company that designed or created the profile. In this sense, it > seemed wrong to say that the profile was created "by Lenovo" when it > was created by me, for my Lenovo display. The ICCv2 specification disagrees: 6.4.15 deviceMfgDescTag Tag Type: textDescriptionType Tag Signature: ?dmnd? (646D6E64h) Structure containing invariant and localizable versions of the device manufacturer for display. The content of this structure is described in 6.5.17. 6.4.16 deviceModelDescTag Tag Type: textDescriptionType Tag Signature: ?dmdd? (646D6464h) Structure containing invariant and localizable versions of the device model for display. The content of this structure is described in 6.5.17. >> For example: >> >> Make: SEC (don't lookup in pnp.ids for storage in the ICC profile, >> only lookup in the gcm-prefs GUI). >> Model: 154AT07-H01 >> >> Storing the EDID info as verbatim as possible in the ICC make/models >> fields, will make it easier to auto match displays and profiles in the >> future. > > Hmm, I think the fields have to be displayable. We need to look at > adding private fields if we want to add any machine parsable data into > the ICC file. > >> Storing the EDID verbatim without looking up in pnp.ids prevents false >> data from conflicting lookups (like my Samsung/Seiko issue) from >> ending up in the profile itself. > > I think the lookup is valuable as it works most of the time. Your > panel should have encoded SAM into the PNP field, not SEC, unless of > course the panel really is a Seiko panel that's been rebadged as a > Samsung. For what it's worth, my nice expensive LG panel is actually a > generic Goldstar chipset with LG additions and plastic wrapping. > Hence, it shows up as "Goldstar" as the vendor. :-( I'm not saying my laptop's EDID info is not semi-broken... But translating SEC into Seiko for embedding in the profile, is moving from semi-wrong to completely-wrong... And I'm afraid my laptop isn't the only one having this or similar issues. I do wonder how Samsung's own brand sold displays identify themselves. I wish we had some more EDID samples to compare... Maybe a call for sample EDID data on Planet GNOME would get us some more samples? Regards, Pascal de Bruijn -------------- next part -------------- /sys/class/dmi/id/bios_date 03/12/2009 /sys/class/dmi/id/bios_vendor Hewlett-Packard /sys/class/dmi/id/bios_version 68PZU Ver. F.0C /sys/class/dmi/id/board_asset_tag /sys/class/dmi/id/board_name 30E8 /sys/class/dmi/id/board_serial /sys/class/dmi/id/board_vendor Hewlett-Packard /sys/class/dmi/id/board_version KBC Version 94.1C /sys/class/dmi/id/chassis_asset_tag CNU9100JZH /sys/class/dmi/id/chassis_serial CNU9100JZH /sys/class/dmi/id/chassis_type 10 /sys/class/dmi/id/chassis_vendor Hewlett-Packard /sys/class/dmi/id/chassis_version /sys/class/dmi/id/modalias dmi:bvnHewlett-Packard:bvr68PZUVer.F.0C:bd03/12/2009:svnHewlett-Packard:pnHPCompaq6730s:pvrF.0C:rvnHewlett-Packard:rn30E8:rvrKBCVersion94.1C:cvnHewlett-Packard:ct10:cvr: /sys/class/dmi/id/power cat: /sys/class/dmi/id/power: Is a directory /sys/class/dmi/id/product_name HP Compaq 6730s /sys/class/dmi/id/product_serial CNU9100JZH /sys/class/dmi/id/product_uuid 4377676E-D605-DE11-B150-80487F19A84A /sys/class/dmi/id/product_version F.0C /sys/class/dmi/id/subsystem cat: /sys/class/dmi/id/subsystem: Is a directory /sys/class/dmi/id/sys_vendor Hewlett-Packard /sys/class/dmi/id/uevent MODALIAS=dmi:bvnHewlett-Packard:bvr68PZUVer.F.0C:bd03/12/2009:svnHewlett-Packard:pnHPCompaq6730s:pvrF.0C:rvnHewlett-Packard:rn30E8:rvrKBCVersion94.1C:cvnHewlett-Packard:ct10:cvr: From pmjdebruijn at pcode.nl Wed Dec 2 17:46:29 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:46:29 +0100 Subject: Idea: GTK/Cario CIE widget as a seperate library In-Reply-To: <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> References: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> Message-ID: <4c2c37f90912020946r1a55d1afi1f7752bae2d80f58@mail.gmail.com> On Wed, Dec 2, 2009 at 6:35 PM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> The GTK/Cairo CIE widget written for GCM is beginning to kick ass... >> However I guess it's GCM specific. > > Not really, it's a standard GObject so it could be reused. Nice. >> Maybe for the long term it's a thought to separate it into a small >> library, so other apps can use it as well? > > Sure, there is loads of other common code that might be useful, but I > don't think the project is at any level of internal ABI stability. > Maybe we can revisit this idea in a few months time? Of course, like I said, it's just a thought. Rushing it won't make it better :) Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 2 19:47:00 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 19:47:00 +0000 Subject: Some tests In-Reply-To: <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> Message-ID: <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > The CIE widget is very nice. Though the profile whitepoint isn't being > displayed yet, I'm assuming this is still on the TODO list. I've just added this to git master. > I also agree it would be nice (but not critical) if double clicking on > the CIE widget would create a dialog with just the profile name and > the CIE widget, for detailed viewing. And dare I ask, possibly with a > right-click "Save (CIE diagram) as PNG/JPEG/PDF" option? Sure, we could do that. I'm tempted to leave that feature until the CIE widget displays more useful data. > Also, I notice the CIE widget has two "squares" of "space" to the left > of the CIE diagram. Does this have a purpose? Otherwise couldn't this > be decreased to a single white square. Sure, we can scale this later. > @Patrice: I rebuild my PPA between 1 to 4 times per two days, > depending on how interesting Richard's changes are... All in all it's > pretty hard to keep up, and he's doing all the hard work :) I'll take that as a compliment. After Monday development will slow quite a bit, as I've got a ton of other stuff to do. Richard. From lars.tore at mulebakken.net Wed Dec 2 20:29:04 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Wed, 2 Dec 2009 21:29:04 +0100 Subject: Some tests In-Reply-To: <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> Message-ID: <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> On Wed, Dec 2, 2009 at 8:47 PM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> The CIE widget is very nice. Though the profile whitepoint isn't being >> displayed yet, I'm assuming this is still on the TODO list. > > I've just added this to git master. > It works for matrix profiles. Lut profiles are already displayed black so it is not possibly to see the tiny cross. See attachment. Lars Tore Gustavsen -------------- next part -------------- A non-text attachment was scrubbed... Name: whitepoint-lut.png Type: image/png Size: 40793 bytes Desc: not available URL: From hughsient at gmail.com Wed Dec 2 20:57:23 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 20:57:23 +0000 Subject: Some tests In-Reply-To: <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> Message-ID: <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> 2009/12/2 Lars Tore Gustavsen : > It works for matrix profiles. Lut profiles are already displayed black > so it is not possibly to see the tiny cross. Sure, we need to add support for LUT profiles as described by Graeme. If you're feeling handy I would love a patch, although I can give this a go next week if nobody starts it before me. Richard. From pmjdebruijn at pcode.nl Wed Dec 2 22:06:57 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 23:06:57 +0100 Subject: Some tests In-Reply-To: <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> Message-ID: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> On Wed, Dec 2, 2009 at 9:57 PM, Richard Hughes wrote: > 2009/12/2 Lars Tore Gustavsen : >> It works for matrix profiles. Lut profiles are already displayed black >> so it is not possibly to see the tiny cross. > > Sure, we need to add support for LUT profiles as described by Graeme. > If you're feeling handy I would love a patch, although I can give this > a go next week if nobody starts it before me. LUT profiles have a normal white point defined, I think Lars ment, it would be nice to invert the color of the whitepoint cross, so it's visible on the black CIE diagram. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 2 22:32:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 23:32:01 +0100 Subject: GCM git fails to build Message-ID: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> Hi, I just tried to build GCM from git on Ubuntu Karmic, and it failed: cc1: warnings being treated as errors gcm-profile.c: In function ?gcm_profile_parse?: gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? from incompatible pointer type /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? but argument is of type ?guint *? make[4]: *** [libgcmshared_a-gcm-profile.o] Error 1 make[4]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager/src' make[3]: *** [all] Error 2 make[3]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager/src' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager' make[1]: *** [all] Error 2 make[1]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager' make: *** [debian/stamp-makefile-build] Error 2 dpkg-buildpackage: error: debian/rules build gave error exit status 2 Regards, Pascal de Bruijn From lars.tore at mulebakken.net Thu Dec 3 06:19:58 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 07:19:58 +0100 Subject: Some tests In-Reply-To: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> Message-ID: <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> On Wed, Dec 2, 2009 at 11:06 PM, Pascal de Bruijn wrote: > LUT profiles have a normal white point defined, I think Lars ment, it > would be nice to invert the color of the whitepoint cross, so it's > visible on the black CIE diagram. > Yep, I meant so. Anyway I like the new tabbed interface. Since we now have the profile size displayed in kb, I think it would be of great benefit to also show the profiles filename. Regards Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 08:03:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 08:03:40 +0000 Subject: GCM git fails to build In-Reply-To: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> Message-ID: <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > cc1: warnings being treated as errors > gcm-profile.c: In function ?gcm_profile_parse?: > gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? > from incompatible pointer type > /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? > but argument is of type ?guint *? Should be fixed in git, thanks. Richard. From hughsient at gmail.com Thu Dec 3 08:04:37 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 08:04:37 +0000 Subject: Some tests In-Reply-To: <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> Message-ID: <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> 2009/12/3 Lars Tore Gustavsen : > Anyway I like the new tabbed interface. Since we now have the profile > size displayed in kb, I think it would be of great benefit to also > show the profiles filename. Filename or filename with path? The latter is quite long. Richard. From lars.tore at mulebakken.net Thu Dec 3 09:12:46 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 10:12:46 +0100 Subject: Some tests In-Reply-To: <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> Message-ID: <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> On Thu, Dec 3, 2009 at 9:04 AM, Richard Hughes wrote: > 2009/12/3 Lars Tore Gustavsen : >> Anyway I like the new tabbed interface. Since we now have the profile >> size displayed in kb, I think it would be of great benefit to also >> show the profiles filename. > > Filename or filename with path? The latter is quite long. > > Richard. > Filename would be nice. I also thinks that full path will be to long for the GUI. Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 10:07:01 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 10:07:01 +0000 Subject: Some tests In-Reply-To: <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> Message-ID: <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> 2009/12/3 Lars Tore Gustavsen : > Filename would be nice. I also thinks that full path will be to long > for the GUI. commit 813c39747c4a0c01a5ee97842e32ab38ea5d71fc Author: Richard Hughes Date: Thu Dec 3 10:06:08 2009 +0000 Add the filename basename to the profile view Richard. From lars.tore at mulebakken.net Thu Dec 3 10:15:16 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 11:15:16 +0100 Subject: Some tests In-Reply-To: <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> Message-ID: <58497f010912030215n7a5e9ed6i8234bce4b3d54a6b@mail.gmail.com> On Thu, Dec 3, 2009 at 11:07 AM, Richard Hughes wrote: > 2009/12/3 Lars Tore Gustavsen : >> Filename would be nice. I also thinks that full path will be to long >> for the GUI. > > commit 813c39747c4a0c01a5ee97842e32ab38ea5d71fc > Author: Richard Hughes > Date: ? Thu Dec 3 10:06:08 2009 +0000 > > ? ?Add the filename basename to the profile view > > Richard. > Excellent, and it works. Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 11:42:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 11:42:49 +0000 Subject: Some tests In-Reply-To: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> Message-ID: <15e53e180912030342q2bde3d13ncb616f24c7a3617f@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > LUT profiles have a normal white point defined, I think Lars ment, it > would be nice to invert the color of the whitepoint cross, so it's > visible on the black CIE diagram. commit 6c77c29086ab851bec34c5667d0963bb32ec9217 Author: Richard Hughes Date: Thu Dec 3 11:41:21 2009 +0000 Show the white point cross in white if the CIE chart is all black Richard. From pmjdebruijn at pcode.nl Thu Dec 3 18:31:29 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 19:31:29 +0100 Subject: GCM git fails to build In-Reply-To: <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> Message-ID: <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> On Thu, Dec 3, 2009 at 9:03 AM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> cc1: warnings being treated as errors >> gcm-profile.c: In function ?gcm_profile_parse?: >> gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? >> from incompatible pointer type >> /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? >> but argument is of type ?guint *? > > Should be fixed in git, thanks. Indeed, it's fixed. Some minor trivia: it seems the above only broke on amd64, not on i386... Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 3 18:34:48 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 19:34:48 +0100 Subject: Camera input profiling make/model handling Message-ID: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> Hi, I haven't had time yet to properly check this, but what is used by GCM for make/model info when profiling cameras/scanners? At least for camera's it seems wise to use Make/Model from EXIF. This way, software like UFRaw and DarkTable (and others) can use the EXIF make/model to match which profiles should be valid for a certain RAW file... Luckily EXIF is much better human readable (and generally less ambiguous) than EDID... Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 3 19:02:31 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:02:31 +0000 Subject: GCM git fails to build In-Reply-To: <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> Message-ID: <15e53e180912031102w2c2e16dcmadc15da0983fb24c@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > Some minor trivia: it seems the above only broke on amd64, not on i386... Right, gsize is a different memory size to guint on 64bit architectures. Richard. From hughsient at gmail.com Thu Dec 3 19:09:22 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:09:22 +0000 Subject: Camera input profiling make/model handling In-Reply-To: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> References: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> Message-ID: <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > I haven't had time yet to properly check this, but what is used by GCM > for make/model info when profiling cameras/scanners? Information from sysfs. > At least for camera's it seems wise to use Make/Model from EXIF. This > way, software like UFRaw and DarkTable (and others) can use the EXIF > make/model to match which profiles should be valid for a certain RAW > file... GCM doesn't treat the device like a storage device, it treats it as a USB endpoint device, which means we don't look at the files contained within. > Luckily EXIF is much better human readable (and generally less > ambiguous) than EDID... I think this is the wrong layer. What happens if the EXIF data for one file on the "device" has different colorspaces or even two different device models? The device can carry all number of different EXIF files, but the device can only have one profile. If you inserted a pendrive of photos, would you expect to assign a per-pendrive ICC profile? I think it's sane for the end application to open the file with EXIF metadata and then decided what to do. If it wants to just use the embedded profile, that's fine, or it can ask gcm for the default device profile from the volume (which it can get from udev). Richard. From info at justinsseasonaltreasures.com Thu Dec 3 18:20:37 2009 From: info at justinsseasonaltreasures.com (Unknown) Date: Thu, 03 Dec 2009 10:20:37 -0800 Subject: Support for B9100 Message-ID: <1259864437.2325.1.camel@localhost> Will there be support for color management of the B9100? Thanks, Jazbo From hughsient at gmail.com Thu Dec 3 19:30:15 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:30:15 +0000 Subject: Support for B9100 In-Reply-To: <1259864437.2325.1.camel@localhost> References: <1259864437.2325.1.camel@localhost> Message-ID: <15e53e180912031130k38517da1j65cc586154c34625@mail.gmail.com> 2009/12/3 Unknown : > Will there be support for color management of the B9100? I assume you mean the HP printer. At the moment ArgyllCMS has some support for generating print profiles, but CUPS does not have a way to set the ICC profile for the device. This needs much more work (for the session CUPS process to talk to GCM and get the current profile for the device) before it will be a case of plug and play. Apple did a lot of the CUPS work, but we need to flesh that out for Linux. For using hplip it's a bit different, and we probably need to support that too. Either way, more help is really welcome. If you just want to assign a profile to a device, and then select that in an application, that should work now. Richard. From pmjdebruijn at pcode.nl Thu Dec 3 20:54:23 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 21:54:23 +0100 Subject: Camera input profiling make/model handling In-Reply-To: <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> References: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> Message-ID: <4c2c37f90912031254i1dc63397g70e35c11a902f5d2@mail.gmail.com> On Thu, Dec 3, 2009 at 8:09 PM, Richard Hughes wrote: > 2009/12/3 Pascal de Bruijn : >> I haven't had time yet to properly check this, but what is used by GCM >> for make/model info when profiling cameras/scanners? > > Information from sysfs. Maybe I'll test this over the weekend... >> At least for camera's it seems wise to use Make/Model from EXIF. This >> way, software like UFRaw and DarkTable (and others) can use the EXIF >> make/model to match which profiles should be valid for a certain RAW >> file... > > GCM doesn't treat the device like a storage device, it treats it as a > USB endpoint device, which means we don't look at the files contained > within. > >> Luckily EXIF is much better human readable (and generally less >> ambiguous) than EDID... > > I think this is the wrong layer. What happens if the EXIF data for one > file on the "device" has different colorspaces or even two different > device models? The device can carry all number of different EXIF > files, but the device can only have one profile. If you inserted a > pendrive of photos, would you expect to assign a per-pendrive ICC > profile? Huh? A "Device" does not really have different color spaces... a camera's sensor (RAW file) has a native color space characterized by the profiling of the IT8 target (or another target, which is not really relevant, it's a means to an end). Any camera that produces sRGB or AdobeRGB, is likely to be not fit to be really profiled. Since a camera most likely does not statically convert sensor data to sRGB/AdobeRGB, it's likely a dynamic (image dependant) process. Anyway I filed a feature request with UFRaw, for "special" profiling mode, to convert a camera RAW file an image with linear gamma in the native camera color space: http://sourceforge.net/tracker/?func=detail&aid=2841222&group_id=127649&atid=709089 If one would have a pen drive of photos, the pen drive would be completely and utterly irrelevant from a profiling/profile assignment perspective. Each and every file on the drive could require a different profile (at least for proper results). And normally EXIF could be reasonably reliably used to "match" that. > I think it's sane for the end application to open the file with EXIF > metadata and then decided what to do. If it wants to just use the > embedded profile, that's fine, or it can ask gcm for the default > device profile from the volume (which it can get from udev). Again, with images, only the image color space is truely relevant (embedded or not). However, RAW files do not have assigned (embedded) profiles. UFRaw/DCRaw "solves" this by ripping generic profiles (Color Matrix) from Adobe DNG SDK, and applying them on a by EXIF make/model basis. Please do note, the above applies to camera's. I'm not really sure about scanners. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 3 22:23:48 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 23:23:48 +0100 Subject: ICC operating system identifiers Message-ID: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Hi, Currently there is no registered operating system identifier for Linux in general. Some examples: Microsoft 'msft' Apple 'appl' Sun Microsystems 'sunw' In the past I tried to mail the Linux Foundation about this, since they seem to be the best (least unfit) umbrella organisation to register such an identifier with the ICC. For example: Linux Foundation 'lnxf' However, it seems the mail was ignored. Richard, could you possibly attempt this, since you can mail from a gnome.org e-mail adres, the request might be more seriously considered. Or do you have other thoughts on how to orchestrate this? Regards, Pascal de Bruijn From anders at brander.dk Thu Dec 3 22:32:11 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 03 Dec 2009 23:32:11 +0100 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <1259879531.25715.4.camel@video64> Hi, On Thu, 2009-12-03 at 23:23 +0100, Pascal de Bruijn wrote: > Currently there is no registered operating system identifier for Linux > in general. Is this relevant in this day and age? Is there any CMM/CMS that does anything useful with this information? Regards, Anders Brander From pmjdebruijn at pcode.nl Thu Dec 3 22:46:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 23:46:01 +0100 Subject: ICC operating system identifiers In-Reply-To: <1259879531.25715.4.camel@video64> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> <1259879531.25715.4.camel@video64> Message-ID: <4c2c37f90912031446p58a57498i6cf869fb254962a1@mail.gmail.com> On Thu, Dec 3, 2009 at 11:32 PM, Anders Brander wrote: > Hi, > > On Thu, 2009-12-03 at 23:23 +0100, Pascal de Bruijn wrote: >> Currently there is no registered operating system identifier for Linux >> in general. > > Is this relevant in this day and age? Is there any CMM/CMS that does > anything useful with this information? I think this remains relevant until the ICC officially deprecates the fields... Regards, Pascal de Bruijn From graeme2 at argyllcms.com Fri Dec 4 00:03:34 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 04 Dec 2009 11:03:34 +1100 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <4B1851D6.8080801@argyllcms.com> Pascal de Bruijn wrote: > For example: > Linux Foundation 'lnxf' Hi, note that Argyll uses '*nix' as an unofficial platform signature for Linux and other Unix like platforms. See also this thread on the OpenICC list: Graeme Gill. From hughsient at gmail.com Fri Dec 4 09:22:59 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 09:22:59 +0000 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > Richard, could you possibly attempt this, since you can mail from a > gnome.org e-mail adres, the request might be more seriously > considered. I would suspect my @redhat email account might have more gravitas, although I'm not sure what it would achieve. Do programs actually do anything differently depending on the OSI information, or even display it at any point? If you think it may be useful, then I guess "*nix" makes most sense as Graeme has been using this for ages. If anything, Graeme's email would convey more weight than mine ever would. Richard. From graeme2 at argyllcms.com Fri Dec 4 11:20:47 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 04 Dec 2009 22:20:47 +1100 Subject: ICC operating system identifiers In-Reply-To: <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> Message-ID: <4B18F08F.3080306@argyllcms.com> Richard Hughes wrote: > If you think it may be useful, then I guess "*nix" makes most sense as > Graeme has been using this for ages. If anything, Graeme's email would > convey more weight than mine ever would. The issues are the following: 1) The platform signature is not of great importance, and it's difficult to know what the original intention of it was. ICC profiles are typically not platform specific. It's lack of importance means that it might be difficult to excite anybodies interest in this issue. The only reason I added '*nix' to my code was an interest in not being misleading. My other alternatives were to use one of the other two Unix based platform signatures (SGI or Sun), or to set it to Unknown (0). 2) Most of the other ICC signatures are in registries, so it's not that bigger deal to ask someone to add a signature to a registry - a single person will have the authority to just do it (hence 'argl' added as a CMM signature). The platform signatures only appear in the ICC spec. though, and they also seem to correlate with the founding members (Although I notice that Taligent has been dropped between V2 an V4). So to add a new signature means someone within the ICC has to propose a spec. change and it then has to be approved by the members at some point ! 3) I'm not sure how much consensus there is about a signature that would cover Linux. My thought process was that the existing signatures were both too vague and too specific. They are really operating system vendors, not specific operating system platforms, and so are fairly loose (there is a noticeable difference between Win3.1 and WinNT, between OS9 and OS X). They are also too specific, in that Sun and SGI were both Unix based systems, although SGI has switch to Windows, so what does it mean ? If the signatures were just a registry, there wouldn't be any problem being very specific (Linux, FreeBSD, OpenBSD, RedHatLinux, etc. etc.) So I went for something that would cover cover Linux, BSD etc. and would be enough to distinguish it from Apple and Microsoft. As I suggested on the OpenICC list, if there is sufficient consensus then I'm happy to mail a couple of people and request that a new platform signature be put forward as a modification to the ICC spec., but on the other hand I won't be surprised if action on it is a little slow :-) cheers, Graeme Gill. From lars.tore at mulebakken.net Fri Dec 4 11:53:06 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 4 Dec 2009 12:53:06 +0100 Subject: More verbose output in calibration? Message-ID: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> May I suggest that when I calibrate my monitor the details view are more verbose than now (like dispcal -v). I guess a beginner will think this is not working at all, when it measures all the black patches and almost noting change on the display. I think more experienced users also like to see the progress in the calibration and the deltaE report at the end. In a longer time frame some progress bar (1) or progress window (2) would be very nice eye candy in the calibration process. 1 http://developer.gnome.org/projects/gup/hig/2.0/controls-progress-bars.html 2 http://developer.gnome.org/projects/gup/hig/2.0/windows-progress.html Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 4 14:51:47 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 15:51:47 +0100 Subject: Bug: Profiles tab is empty as long as no profiles have been assigned. Message-ID: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> Hi, I think I found a small bug. When starting gcm-prefs after a clean install, it does list all the available profiles in the profiles dropdown box on the Devices tab. But the list on the Profiles tab is completely empty. When I restart gcm-prefs the list is still empty. However, when I assign a profile (for example blueish) to the screen, it seems to change nothing at first glance. However after restarting gcm-prefs again, suddenly the dialog is bigger, and the list on the Profiles tabs has been filled. I haven't been able to test this, but when a new profile is created, does this get properly added the the list on the Profiles tab as well? Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Fri Dec 4 14:58:25 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 15:58:25 +0100 Subject: gcm-prefs on netbook Message-ID: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> Hi, I managed to test-drive gcm-prefs on a friends netbook (which runs Ubuntu Karmic Netbook Edition). And when the list on the Profiles tab is filled, the dialog quickly grows to big. I've attached a screenshot. I personally don't particularly like netbooks, but they are popular, and thus I think it might be nice for gcm-prefs to work well on netbooks too. I'm quite aware netbooks aren't particularly well suited for "graphics-work", but that's not really the point here. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: gcm-prefs-netbook.png Type: image/png Size: 83037 bytes Desc: not available URL: From hughsient at gmail.com Fri Dec 4 15:00:38 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:00:38 +0000 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> Message-ID: <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > I personally don't particularly like netbooks, but they are popular, > and thus I think it might be nice for gcm-prefs to work well on > netbooks too. What about if the window size < 1024x768 then make the cie widget smaller? Richard. From hughsient at gmail.com Fri Dec 4 15:18:18 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:18:18 +0000 Subject: Bug: Profiles tab is empty as long as no profiles have been assigned. In-Reply-To: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> References: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> Message-ID: <15e53e180912040718m678446b7s3dd3797494ac05e9@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > I think I found a small bug. When starting gcm-prefs after a clean > install, it does list all the available profiles in the profiles > dropdown box on the Devices tab. But the list on the Profiles tab is > completely empty. Ahh, thanks. commit 902fbf2f4271d6be79985d40338fcf4b35bbc2a1 Author: Richard Hughes Date: Fri Dec 4 15:16:37 2009 +0000 Don't skip phase2 startup if we can't load the config file :100644 100644 2726f42... 792b729... M src/gcm-prefs.c Richard. From hughsient at gmail.com Fri Dec 4 15:18:39 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:18:39 +0000 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> Message-ID: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> 2009/12/4 Richard Hughes : > What about if the window size < 1024x768 then make the cie widget smaller? commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 Author: Richard Hughes Date: Fri Dec 4 15:11:19 2009 +0000 Reduce the size of the ICC widget on small displays :100644 100644 4a1551d... 2726f42... M src/gcm-prefs.c Richard. From pmjdebruijn at pcode.nl Fri Dec 4 15:49:27 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 16:49:27 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <4c2c37f90912040749m6c8cdea1vb5c4aa14b43555@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays > > :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c Excellent. The CIE widget isn't the most import thing ever, however nice it may be. The user chose to have a small (low resolution) screen, so having the buttons visible is much more important. Regards, Pascal de Bruijn From lars.tore at mulebakken.net Fri Dec 4 17:16:10 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 4 Dec 2009 18:16:10 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays > > :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c > > Richard. On my 1280x1024 monitor the CIE diagram is quite small now. I can live with it, but I'm not sure if 1280x1024 was meant to be a small display? Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 4 17:18:10 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 18:18:10 +0100 Subject: gcm-prefs on netbook In-Reply-To: <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> Message-ID: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> On Fri, Dec 4, 2009 at 6:16 PM, Lars Tore Gustavsen wrote: > On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: >> 2009/12/4 Richard Hughes : >>> What about if the window size < 1024x768 then make the cie widget smaller? >> >> commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 >> Author: Richard Hughes >> Date: ? Fri Dec 4 15:11:19 2009 +0000 >> >> ? ?Reduce the size of the ICC widget on small displays >> >> :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c >> >> Richard. > > On my 1280x1024 monitor the CIE diagram is quite small now. I can live > with it, but I'm not sure if 1280x1024 was meant to be a small > display? That wasn't intentional: + if (gdk_screen_get_width (screen) < 10240 || That's a typo :) Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Fri Dec 4 17:24:17 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 4 Dec 2009 20:24:17 +0300 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> Message-ID: <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> On Fri, Dec 4, 2009 at 8:18 PM, Pascal de Bruijn wrote: >> On my 1280x1024 monitor the CIE diagram is quite small now. I can live >> with it, but I'm not sure if 1280x1024 was meant to be a small >> display? > > That wasn't intentional: So rectangular shape is? :) http://img37.imagefra.me/img/img37/1/12/4/prokoudine/f_1m_444c758.png Alexandre From hughsient at gmail.com Fri Dec 4 17:44:41 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 17:44:41 +0000 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> Message-ID: <15e53e180912040944k21f00207qa486932397da75eb@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > + if (gdk_screen_get_width (screen) < 10240 || > > That's a typo :) Fixed, thanks. Richard. From hughsient at gmail.com Fri Dec 4 17:49:13 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 17:49:13 +0000 Subject: gcm-prefs on netbook In-Reply-To: <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> Message-ID: <15e53e180912040949p144a83b8gce479add4b1a829@mail.gmail.com> 2009/12/4 Alexandre Prokoudine : > So rectangular shape is? :) Wrong! commit e818364b44830c6178a2fcbd81a255f47ad01a6d Author: Richard Hughes Date: Fri Dec 4 17:47:31 2009 +0000 Ensure the CIE widget is square Richard. From pmjdebruijn at pcode.nl Sat Dec 5 12:29:07 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 13:29:07 +0100 Subject: Devices list sorting Message-ID: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> Hi, Since I'm _guessing_ GCM will most commonly be used to calibrate displays, it might be a thought to sort on device "class" first, and then alphabetically? Possibly displaying displays at the top of the list? I'm guessing this order makes most sense (to me): - Displays - Printers - Scanners - Camera's - Video Camera's Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Sat Dec 5 12:30:56 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 13:30:56 +0100 Subject: Feature: Nautilus Thumbnailer for profiles Message-ID: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> Hi, Another idea for the long term: Wouldn't it severely kick ass, if Nautilus were to thumbnail ICC profiles using the CIE widget? Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Sat Dec 5 13:11:25 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sat, 5 Dec 2009 16:11:25 +0300 Subject: Feature: Nautilus Thumbnailer for profiles In-Reply-To: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> References: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> Message-ID: <733f2c730912050511s6634231qf84b50fb000b148b@mail.gmail.com> On 12/5/09, Pascal de Bruijn wrote: > Another idea for the long term: > > Wouldn't it severely kick ass, if Nautilus were to thumbnail ICC > profiles using the CIE widget? +2^32 :) Alexandre From hughsient at gmail.com Sat Dec 5 15:52:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 5 Dec 2009 15:52:24 +0000 Subject: Devices list sorting In-Reply-To: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> Message-ID: <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> 2009/12/5 Pascal de Bruijn : > Since I'm _guessing_ GCM will most commonly be used to calibrate > displays, it might be a thought to sort on device "class" first, and > then alphabetically? The profile title or filename as the secondary sort? > Possibly displaying displays at the top of the list? I'm guessing this > order makes most sense (to me): > > - Displays > - Printers > - Scanners > - Camera's > - Video Camera's I'm not sure it matters, but sure, it would be better if they were sorted. Richard. From pmjdebruijn at pcode.nl Sat Dec 5 15:54:18 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 16:54:18 +0100 Subject: Devices list sorting In-Reply-To: <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> Message-ID: <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> On Sat, Dec 5, 2009 at 4:52 PM, Richard Hughes wrote: > 2009/12/5 Pascal de Bruijn : >> Since I'm _guessing_ GCM will most commonly be used to calibrate >> displays, it might be a thought to sort on device "class" first, and >> then alphabetically? > > The profile title or filename as the secondary sort? I'd use the profile description as secondary sort... >> Possibly displaying displays at the top of the list? I'm guessing this >> order makes most sense (to me): >> >> - Displays >> - Printers >> - Scanners >> - Camera's >> - Video Camera's > > I'm not sure it matters, but sure, it would be better if they were sorted. It's a convenience thing... For example, since video4linux support has been commited, on my laptop I get my Chicony webcam as the first entry, and below that my display... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Sun Dec 6 09:31:43 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:31:43 +0300 Subject: gcm-inspect segfaults Message-ID: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> (gdb) run -d Starting program: /usr/bin/gcm-inspect -d [Thread debugging using libthread_db enabled] Output profile 'VGA1': not set Output profile 'HDMI1': not set Output profile 'DP1': not set Output profile 'HDMI2': not set Output profile 'DP2': not set Program received signal SIGSEGV, Segmentation fault. 0x00007ffff3ffba52 in g_atomic_int_exchange_and_add () from /lib/libglib-2.0.so.0 (gdb) bt #0 0x00007ffff3ffba52 in g_atomic_int_exchange_and_add () from /lib/libglib-2.0.so.0 #1 0x00007ffff401162e in g_hash_table_unref () from /lib/libglib-2.0.so.0 #2 0x0000000000404ee7 in gcm_inspect_get_properties (argc=1, argv=0x7fffffffe3d8) at gcm-inspect.c:371 #3 main (argc=1, argv=0x7fffffffe3d8) at gcm-inspect.c:438 Should I be concerned about that sort of thing? :) Alexandre From alexandre.prokoudine at gmail.com Sun Dec 6 09:45:02 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:45:02 +0300 Subject: Yelp isn't happy Message-ID: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> Hi, When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI ?ghelp:gnome-color-manager?preferences? is incorrect". Alexandre From alexandre.prokoudine at gmail.com Sun Dec 6 09:51:56 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:51:56 +0300 Subject: yet another PO comment bug Message-ID: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> Hi, A comment for "Setting up device" says "TRANSLATORS: title, a profile is a ICC file" Alexandre From hughsient at gmail.com Sun Dec 6 10:09:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 10:09:49 +0000 Subject: gcm-inspect segfaults In-Reply-To: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> References: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> Message-ID: <15e53e180912060209i151c479gd985e4d60e4a829e@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > #1 ?0x00007ffff401162e in g_hash_table_unref () from /lib/libglib-2.0.so.0 Fixed in git, thanks. Richard. From hughsient at gmail.com Sun Dec 6 10:16:29 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 10:16:29 +0000 Subject: yet another PO comment bug In-Reply-To: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> References: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> Message-ID: <15e53e180912060216y4b0c3daeh2c850b6b15b8ce35@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > A comment for "Setting up device" says "TRANSLATORS: title, a profile > is a ICC file" Fixed, thanks. Richard. From pmjdebruijn at pcode.nl Sun Dec 6 11:39:55 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 6 Dec 2009 12:39:55 +0100 Subject: CIE widget drawing odd graphs Message-ID: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> Hi, I might have uncovered some minor issues with the CIE widget drawing code. I've attached an archive containing several non-standard profiles, which are drawn quite oddly. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: weirdciewidget.zip Type: application/zip Size: 31374 bytes Desc: not available URL: From pmjdebruijn at pcode.nl Sun Dec 6 12:33:52 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 6 Dec 2009 13:33:52 +0100 Subject: Icon Sizes Message-ID: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> Hi, I managed to test GCM before on a netbook with Ubuntu Karmic Netbook edition installed. I noticed somehow the GCM icon was very small in comparison to all the other application icons... While I'm not entirely sure about this, I think the Ubuntu Netbook edition uses the 64x64 icons (and does not fall back to scalable to save CPU cycles on low end CPUs), since GCM does not supply a 64x64 icon, it falls back to a smaller one. Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? Regards, Pascal de Bruijn From lars.tore at mulebakken.net Sun Dec 6 16:04:46 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Sun, 6 Dec 2009 17:04:46 +0100 Subject: Bluish.icc In-Reply-To: <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> References: <58497f010911110351o3d7df74dt8ac35ef7c8cb5e4@mail.gmail.com> <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> Message-ID: <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> On Sun, Dec 6, 2009 at 3:32 PM, Pascal de Bruijn wrote: > On Wed, Nov 11, 2009 at 12:51 PM, Lars Tore Gustavsen > wrote: >> The icc profile include with gnome-color-manager is very nice to test >> if everything works. However it advertise itself as "sRGB >> IEC61966-2-1" >> >> I have edited the attached profile and it now reads "bluish test" in gcm-prefs. > > Would it be possible for you to rename it to "Blueish Test" instead of > "blueish test"? > > Not extremely important, but it does look a bit better :) > > Regards, > Pascal de Bruijn > Here we go: It's done with Andrew Shepherd's ICC Profile Toolkit http://www.tlbtlb.com/links/ Regards Lars Tore Gustavsen -------------- next part -------------- A non-text attachment was scrubbed... Name: bluish.icc Type: application/x-icc Size: 3966 bytes Desc: not available URL: From hughsient at gmail.com Sun Dec 6 19:37:09 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 19:37:09 +0000 Subject: Bluish.icc In-Reply-To: <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> References: <58497f010911110351o3d7df74dt8ac35ef7c8cb5e4@mail.gmail.com> <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> Message-ID: <15e53e180912061137h57d3f969m2a57af3568e547e7@mail.gmail.com> 2009/12/6 Lars Tore Gustavsen : > Here we go: Applied, thanks. Richard From hughsient at gmail.com Sun Dec 6 19:46:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 19:46:40 +0000 Subject: Icon Sizes In-Reply-To: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> References: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> Message-ID: <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> 2009/12/6 Pascal de Bruijn : > Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? commit 29b66375ed40e10bfffba601eeca9f25bd8a46b7 Author: Richard Hughes Date: Sun Dec 6 19:45:42 2009 +0000 Add a 64x64 pixel application icon Richard. From hughsient at gmail.com Sun Dec 6 20:06:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 20:06:24 +0000 Subject: Devices list sorting In-Reply-To: <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> Message-ID: <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> 2009/12/5 Pascal de Bruijn : > For example, since video4linux support has been commited, on my laptop > I get my Chicony webcam as the first entry, and below that my > display... commit dfe5fc80cef25cac2deddb5687719ab298e786e0 Author: Richard Hughes Date: Sun Dec 6 20:05:18 2009 +0000 Make the list orders predictable by setting a sort string Richard. From hughsient at gmail.com Sun Dec 6 20:19:37 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 20:19:37 +0000 Subject: CIE widget drawing odd graphs In-Reply-To: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> Message-ID: <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> 2009/12/6 Pascal de Bruijn : > I might have uncovered some minor issues with the CIE widget drawing code. They look pretty major to me! > I've attached an archive containing several non-standard profiles, > which are drawn quite oddly. After the release I would like to look at the widget drawing code and the profile extraction code (there are still quite a lot of FIXME's and unimplemented stuff) and add a boat load more self tests and internal validations. I fear I might not have time in the next couple of weeks (lots of other work to do, which I've been putting off for ages) so if anyone is familiar with XYZ co-ordinates and ICC profiles I would appreciate any code review and possible patches. Thanks, Richard. From pmjdebruijn at pcode.nl Sun Dec 6 23:35:43 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 00:35:43 +0100 Subject: Devices list sorting In-Reply-To: <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> Message-ID: <4c2c37f90912061535n22c07324s60f6a5edbc7e5147@mail.gmail.com> On Sun, Dec 6, 2009 at 9:06 PM, Richard Hughes wrote: > 2009/12/5 Pascal de Bruijn : >> For example, since video4linux support has been commited, on my laptop >> I get my Chicony webcam as the first entry, and below that my >> display... > > commit dfe5fc80cef25cac2deddb5687719ab298e786e0 > Author: Richard Hughes > Date: ? Sun Dec 6 20:05:18 2009 +0000 > > ? ?Make the list orders predictable by setting a sort string Seems to work excellently here... Thanks, Pascal de Bruijn From hughsient at gmail.com Mon Dec 7 12:03:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 7 Dec 2009 12:03:28 +0000 Subject: gnome-color-manager 2.29.1 Message-ID: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> gnome-color-manager is a session program that makes it easy to manage, install and generate color profiles in the GNOME desktop. Version 2.29.1 ~~~~~~~~~~~~~~ Released: 2009-12-07 * Translations - Add Russian translation (Alexandre Prokoudine) - Add British English translation (Bruce Cowan) - Add Indonesian translation (Andika Triwidada) - Add French translation (Claude Paroz) - Add Swedish translation (Daniel Nylander) - Add Brazilian Portuguese translation (Flamarion Jorge) - Add Lithuanian translation (Gintautas Miliauskas) - Add German translation (Hendrik Brandt) - Add Danish translation (Joe Hansen) - Add Spanish translation (Jorge Gonz?lez) - Add Norwegian bokm?l translation (Kjartan Maraas) - Add Thai translation (Phondanai Khanti) - Add Polish translation (Piotr Dr?g) - Add Estonian translation (Priit Laes, Mattias P?ldaru) - Add Tamil translation (vasudeven) - Add Russian user guide translation (Alexandre Prokoudine) - Add Indonesian user guide translation (Andika Triwidada) * New Features: - Add gcm-import, a helper to allow double clicking on ICC profiles to import them (Richard Hughes) - Add gcm-prefs, a utility to assign profiles to devices, examine profiles, and set session-wide defaults (Richard Hughes) - Add gcm-session, a dbus-activated session daemon for applications to get the profiles for a device, or device class and to get session-wide defaults. It exits when no longer used to save resources. (Richard Hughes) - Add gcm-self-test, a self test framework that tests GCM functionality (Richard Hughes) - Add gcm-apply, a simple utility to just set (or reset) display profiles (Richard Hughes) - Add gcm-inspect, a debugging utility to inspect the profiles set in the session (Richard Hughes) - Add gcm-dump-edid, a utility to dump the EDID to disk for debugging (Richard Hughes) - Add gcm-dump-profile, a utility to dump the ICC profile to the screen (Richard Hughes) - Add some simple man pages and help document (Richard Hughes) - Add ArgyllCMS support to generate device profiles (Richard Hughes) - Add color calibration hardware auto-detection (Richard Hughes) - Add code to set the _ICC_PROFILE atom per-output and also per-screen (Richard Hughes) - Add some pre-calibration steps for external displays (Pascal de Bruijn) - Add hardware support for gphoto supported cameras (Richard Hughes) - Add hardware support for SANE suppoerted scanners (Richard Hughes) - Add hardware support for video4linux supported video devices (Richard Hughes) - Add hardware support for XRandR supported displays (Richard Hughes) - Add hardware support for hplip supported printers (Richard Hughes) - Add CIE widget to display visual data about different profiles (Richard Hughes) - Use the system DMI data to better itentify internal LCD panels (Richard Hughes) - Parse the EDID to get a better device description for displays (Richard Hughes) - Make the list orders predictable by setting a sort string (Richard Hughes) * Bugfix: - Use XDG directory to store data (Baptiste Mille-Mathias) - Remove markup from GTKBuilder translatable strings (Claude Paroz) - Update bluish.icc title (Lars Tore Gustavsen, Pascal de Bruijn) - Enable adding xrandr devices with no EDID (Martin Szulecki) - Avoid reporting a (false) failure on first import (Stephane Delcroix) - Fix the message-received cb signature (Stephane Delcroix) - Fix up numerous small bugs prior to first release (Richard Hughes) - Look for the debian-named argyllcms binaries first (Richard Hughes) - Set the brightness to 100% on internal LCD panels before we generate a output profile (Richard Hughes) Richard. From pmjdebruijn at pcode.nl Mon Dec 7 12:37:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 13:37:01 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> Message-ID: <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> On Mon, Dec 7, 2009 at 1:03 PM, Richard Hughes wrote: > gnome-color-manager is a session program that makes it easy to manage, install > and generate color profiles in the GNOME desktop. > > Version 2.29.1 > ~~~~~~~~~~~~~~ > Released: 2009-12-07 Excellent :) Packages are available from my PPA as usual. However, since I'll be tracking git again in my main repo, I've decided to start a gcm-release PPA: https://launchpad.net/~pmjdebruijn/+archive/gcm-release I've just submitted the builds, so they'll be ready in an hour or six... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Mon Dec 7 15:08:18 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 7 Dec 2009 18:08:18 +0300 Subject: metadata Message-ID: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Hi, It looks like there are problems writing unicode text into icc metadata or reading from it. http://img39.imagefra.me/img/img39/1/12/7/prokoudine/f_0m_948cb9f.png Exhibit 1 is description of the profile that is obviously RU translation of "Custom" Exhibit 2 is copyright field that seems to be suffering from more or less same thing. Any ideas? Alexandre From pmjdebruijn at pcode.nl Mon Dec 7 15:14:42 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 16:14:42 +0100 Subject: metadata In-Reply-To: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Message-ID: <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> On Mon, Dec 7, 2009 at 4:08 PM, Alexandre Prokoudine wrote: > Hi, > > It looks like there are problems writing unicode text into icc > metadata or reading from it. > > http://img39.imagefra.me/img/img39/1/12/7/prokoudine/f_0m_948cb9f.png > > Exhibit 1 is description of the profile that is obviously RU > translation of "Custom" > > Exhibit 2 is copyright field that seems to be suffering from more or > less same thing. This actually begs a bigger question: Should we even try to internationalize metadata contents? I'd prefer not... It would make exchanging ICC profiles internationally harder... Say, one day, we make a big repository of ICC profiles which are ready to go for people who don't have a colorimeter. I'd be pretty crappy for someone in Germany to see Russian in the metadata? I'd prefer to stick to English... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Mon Dec 7 15:47:16 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 7 Dec 2009 18:47:16 +0300 Subject: metadata In-Reply-To: <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> Message-ID: <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> On Mon, Dec 7, 2009 at 6:14 PM, Pascal de Bruijn wrote: > This actually begs a bigger question: Should we even try to > internationalize metadata contents? > > I'd prefer not... It would make exchanging ICC profiles > internationally harder... Quite so :) > Say, one day, we make a big repository of ICC profiles which are ready > to go for people who don't have a colorimeter. Are we talking about profiles for DSLRs? :) Alexandre From hughsient at gmail.com Mon Dec 7 17:29:00 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 7 Dec 2009 17:29:00 +0000 Subject: gnome-color-manager 2.29.1 In-Reply-To: <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> Message-ID: <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> 2009/12/7 Pascal de Bruijn : > I've just submitted the builds, so they'll be ready in an hour or six... I've just started an official build for Fedora 12: http://koji.fedoraproject.org/koji/taskinfo?taskID=1860405 Richard. From pmjdebruijn at pcode.nl Mon Dec 7 19:21:36 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 20:21:36 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> Message-ID: <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> On Mon, Dec 7, 2009 at 6:29 PM, Richard Hughes wrote: > 2009/12/7 Pascal de Bruijn : >> I've just submitted the builds, so they'll be ready in an hour or six... > > I've just started an official build for Fedora 12: > http://koji.fedoraproject.org/koji/taskinfo?taskID=1860405 Richard, my mail to the mailing list got rejected: "Message rejected by filter rule match" Regards, Pascal de Bruijn From hughsient at gmail.com Tue Dec 8 10:15:58 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 10:15:58 +0000 Subject: gnome-color-manager 2.29.1 In-Reply-To: <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> Message-ID: <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> 2009/12/7 Pascal de Bruijn : > Richard, my mail to the mailing list got rejected: I think gnome-announce-list at gnome.org bounced the mail (correctly) -- dropping the cc should make things work again. Richard. From hughsient at gmail.com Tue Dec 8 11:02:50 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 11:02:50 +0000 Subject: metadata In-Reply-To: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Message-ID: <15e53e180912080302x2f413450v3b99d884d709ed57@mail.gmail.com> 2009/12/7 Alexandre Prokoudine : > Exhibit 1 is description of the profile that is obviously RU > translation of "Custom" Right, I made the Custom localized without thinking that it would involve non-ascii chars. I've fixed that in cfebe91aef33254709bc22ca547563044f1fbdd8. > Exhibit 2 is copyright field that seems to be suffering from more or > less same thing. > > Any ideas? Well, for textType, the profile spec mandates 7 bit ASCII, so we're wrong to translate Custom. For the unicodeTextType we should probably decode UTF-16 properly, although I've not seen a profile that uses this type in the wild yet. Note: if you've got a profile that breaks GCM in one way or another (and it's freely redistributable, which excludes the ICC and Adobe profiles..) then we should probably check it into GCM and wire it into the self test code. If nothing else you can use gcm-dump-profile -v thefilename.icc and send me the output. Richard. From hughsient at gmail.com Tue Dec 8 11:05:57 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 11:05:57 +0000 Subject: metadata In-Reply-To: <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> Message-ID: <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> 2009/12/7 Alexandre Prokoudine : >> Say, one day, we make a big repository of ICC profiles which are ready >> to go for people who don't have a colorimeter. > > Are we talking about profiles for DSLRs? :) Just in case, I've got a repo waiting: http://github.com/hughsie/gnome-color-profiles I'm going to be checking in the Adobe and ICC profiles first, but alas, they can't be installed by default in distros like Fedora as they are non free [1]. Richard. [1] https://www.redhat.com/archives/fedora-legal-list/2009-December/msg00009.html and https://www.redhat.com/archives/fedora-legal-list/2009-December/msg00010.html From pmjdebruijn at pcode.nl Tue Dec 8 11:16:35 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 12:16:35 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> Message-ID: <4c2c37f90912080316w42894a61m7ad18ee019eba3dd@mail.gmail.com> On Tue, Dec 8, 2009 at 11:15 AM, Richard Hughes wrote: > 2009/12/7 Pascal de Bruijn : >> Richard, my mail to the mailing list got rejected: > > I think ?gnome-announce-list at gnome.org bounced the mail (correctly) -- > dropping the cc should make things work again. Oops, silly me... Yeah, the gnome-color-manager list accepted my message just fine... Regards, Pascal de Bruijn From graeme2 at argyllcms.com Tue Dec 8 12:38:17 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Tue, 08 Dec 2009 23:38:17 +1100 Subject: metadata In-Reply-To: <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> Message-ID: <4B1E48B9.1080408@argyllcms.com> Richard Hughes wrote: > I'm going to be checking in the Adobe and ICC profiles first, but > alas, they can't be installed by default in distros like Fedora as > they are non free [1]. Note that Argyll V1.1.0 includes a public domain implementation of an sRGB profiles as well as an AdobeRGB1998 equivalent. Graeme Gill. From hughsient at gmail.com Tue Dec 8 12:51:45 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 12:51:45 +0000 Subject: metadata In-Reply-To: <4B1E48B9.1080408@argyllcms.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> <4B1E48B9.1080408@argyllcms.com> Message-ID: <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> 2009/12/8 Graeme Gill : > Richard Hughes wrote: > >> I'm going to be checking in the Adobe and ICC profiles first, but >> alas, they can't be installed by default in distros like Fedora as >> they are non free [1]. > > Note that Argyll V1.1.0 includes a public domain implementation > of an sRGB profiles as well as an AdobeRGB1998 equivalent. That's really good to know, and I might pinch those. What license are those available as? Really public domain? (if so, that's fantastic). Can they be installed alongside the real adobe profiles? Richard. From graeme2 at argyllcms.com Tue Dec 8 13:11:07 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Wed, 09 Dec 2009 00:11:07 +1100 Subject: metadata In-Reply-To: <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> <4B1E48B9.1080408@argyllcms.com> <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> Message-ID: <4B1E506B.4080709@argyllcms.com> Richard Hughes wrote: > 2009/12/8 Graeme Gill : >> Note that Argyll V1.1.0 includes a public domain implementation >> of an sRGB profiles as well as an AdobeRGB1998 equivalent. > > That's really good to know, and I might pinch those. What license are > those available as? Really public domain? (if so, that's fantastic). Yes. There didn't seem any sensible reason to retain copyright on them. > Can they be installed alongside the real adobe profiles? I'm not sure what you mean by that. Since "Adobe" is a registered company name, I've given the Adobe compatible profile a different name ("ClayRGB1998.icm", the best I could come up with at the time). The sRGB profile is called "sRGB.icm", just like the original one from HP/Microsoft that is no longer available, and to which it is functionally identical. Graeme Gill. From pmjdebruijn at pcode.nl Tue Dec 8 17:56:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 18:56:33 +0100 Subject: Icon Sizes In-Reply-To: <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> References: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> Message-ID: <4c2c37f90912080956y6dcf9264padf8217b2c3a5ab4@mail.gmail.com> On Sun, Dec 6, 2009 at 8:46 PM, Richard Hughes wrote: > 2009/12/6 Pascal de Bruijn : >> Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? > > commit 29b66375ed40e10bfffba601eeca9f25bd8a46b7 > Author: Richard Hughes > Date: ? Sun Dec 6 19:45:42 2009 +0000 > > ? ?Add a 64x64 pixel application icon Today, I had a chance to test this on my colleagues HP netbook again... And this seems to fix the "small icon" issue on Ubuntu Karmic Netbook Edition. Thanks, Pascal de Bruijn From pmjdebruijn at pcode.nl Tue Dec 8 17:58:02 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 18:58:02 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <4c2c37f90912080958s1dac9704qdc99372d9c1a3401@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays I had a chance to test this on a colleagues HP netbook today, and it seems to work fine. The CIE widget looks a bit silly at 50x50, but again, this is a netbook's owner choice... Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 9 12:47:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 12:47:28 +0000 Subject: Yelp isn't happy In-Reply-To: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> References: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> Message-ID: <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI > ?ghelp:gnome-color-manager?preferences? is incorrect". commit ef84a7c5dca50e40bfaf38b6c884c8c86983f334 Author: Richard Hughes Date: Wed Dec 9 12:46:33 2009 +0000 Fix the help file installation so that yelp recognises our help file Richard. From alexandre.prokoudine at gmail.com Wed Dec 9 13:18:19 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Wed, 9 Dec 2009 16:18:19 +0300 Subject: Yelp isn't happy In-Reply-To: <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> References: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> Message-ID: <733f2c730912090518w1ed31992g2921881109c65f9b@mail.gmail.com> On Wed, Dec 9, 2009 at 3:47 PM, Richard Hughes wrote: > 2009/12/6 Alexandre Prokoudine : >> When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI >> ?ghelp:gnome-color-manager?preferences? is incorrect". > > commit ef84a7c5dca50e40bfaf38b6c884c8c86983f334 > Author: Richard Hughes > Date: ? Wed Dec 9 12:46:33 2009 +0000 > > ? ?Fix the help file installation so that yelp recognises our help file Many thanks :) Alexandre From hughsient at gmail.com Wed Dec 9 13:18:35 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 13:18:35 +0000 Subject: Profile description for laptop displays In-Reply-To: <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> References: <4c2c37f90911290453x143c6820nb27f3a6b6188e3c7@mail.gmail.com> <15e53e180911291342o2b8755aavad083da3dcd849d2@mail.gmail.com> <4c2c37f90911291527v71d2cf9cy6bb41add8c7b3683@mail.gmail.com> <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> Message-ID: <15e53e180912090518h5b6e9deas414d6c5e1f1ec594@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > I wish we had some more EDID samples to compare... Maybe a call for > sample EDID data on Planet GNOME would get us some more samples? Could you send me your EDID please. Thanks. Richard. From hughsient at gmail.com Wed Dec 9 13:19:56 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 13:19:56 +0000 Subject: More verbose output in calibration? In-Reply-To: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> References: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> Message-ID: <15e53e180912090519v6481e569q867923aa08984dd3@mail.gmail.com> 2009/12/4 Lars Tore Gustavsen : > May I suggest that when I calibrate my monitor the details view are > more verbose than now (like dispcal -v). I guess a beginner will think > this is not working at all, when it measures all the black patches and > almost noting change on the display. ?I think more experienced users > also like to see the progress in the calibration ?and the deltaE > report at the end. ?In a longer time frame some progress bar (1) or > progress window (2) would be very nice eye candy in the calibration > process. You can already get the deltaE values in the dropdown widget. I think it would be very hard to do a progress bar from the data Argyll gives us. Richard. From hughsient at gmail.com Wed Dec 9 17:04:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 17:04:24 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B15B29C.2090105@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> Message-ID: <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> 2009/12/2 Graeme Gill : > For a cLut based profile, you would have to (as Lars Tore Gustavsen > suggested) run the primary values through the profile (or > do the equivalent of looking up what the cLUT values are at > the primaries). I'm sort-of stuck here. I'm confused about which cLUT I should be using (TRC curves?), as different profiles are doing very different things. Could you please elaborate about how to run the primaries through the table? Thanks. Richard. From hughsient at gmail.com Wed Dec 9 19:51:53 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 19:51:53 +0000 Subject: shared-color-profiles 0.1.0 release Message-ID: <15e53e180912091151y1f2303acw48cedde3eb0231f@mail.gmail.com> Version 0.1.0 ~~~~~~~~~~~~~ Released: 2009-12-09 * New Features: - First release! - Add the free Argyll profiles that are in the public domain (Richard Hughes) * Bugfix: - Don't install README and LICENSE in /usr/share/color (Richard Hughes) - Require color-filesystem on Fedora (Richard Hughes) The next release of gnome-color-manager will depend on this package. See http://blogs.gnome.org/hughsie/2009/12/08/shared-color-profiles/ for more details. Richard. From alexandre.prokoudine at gmail.com Wed Dec 9 23:02:22 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 02:02:22 +0300 Subject: sadly, segfault Message-ID: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> Hi, On commit f04af5c3072983e9b5113628f7782dc346c75b18: http://pastebin.com/m18c9d8c9 Alexandre From graeme2 at argyllcms.com Wed Dec 9 23:04:54 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Thu, 10 Dec 2009 10:04:54 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> Message-ID: <4B202D16.1000901@argyllcms.com> Richard Hughes wrote: > I'm sort-of stuck here. I'm confused about which cLUT I should be > using (TRC curves?), as different profiles are doing very different > things. Could you please elaborate about how to run the primaries > through the table? Thanks. Well I can only talk specifically about my set of tools: Using icclib command line tools it would be something like this: icclu -ff -ia -px sRGB.icm 1 0 0 1.000000 0.000000 0.000000 [RGB] -> MatrixFwd -> 0.412385 0.212596 0.019306 [XYZ] 0 1 0 0.000000 1.000000 0.000000 [RGB] -> MatrixFwd -> 0.357573 0.715199 0.119206 [XYZ] 0 0 1 0.000000 0.000000 1.000000 [RGB] -> MatrixFwd -> 0.180500 0.072205 0.950546 [XYZ] Progromatically, one would do the equivalent, e.g. using icclib open the profile, create a lookup object to do forward absolute colorimetric to XYZ space, and then feed colorant only combinations through it. icclu.c will illustrate the details. I'm sure something similar is possible using Marti's lcms. I wouldn't recommend trying to re-create the wheel and do it from ICC tags as there are a number of details to get right. Graeme Gill. From hughsient at gmail.com Thu Dec 10 08:07:04 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 08:07:04 +0000 Subject: sadly, segfault In-Reply-To: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> References: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> Message-ID: <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> 2009/12/9 Alexandre Prokoudine : > On commit f04af5c3072983e9b5113628f7782dc346c75b18: > http://pastebin.com/m18c9d8c9 Thanks, commit 64037aacc1ba6fa7bfa53deabc00dca57cdd658d Author: Richard Hughes Date: Thu Dec 10 08:05:25 2009 +0000 trivial: fix up an error in the last commit, so that if filename is not set we skip the load There are some more changes happening today, so please yell if anything breaks. Thanks. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 14:54:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 17:54:09 +0300 Subject: sadly, segfault In-Reply-To: <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> References: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> Message-ID: <733f2c730912100654g6eafc1e9v5c51b2783e119437@mail.gmail.com> On 12/10/09, Richard Hughes wrote: > There are some more changes happening today, so please yell if > anything breaks. Thanks. Well, apart from the profiles I sent you before whose CIE diagram doesn't render nicely, so far so good :) Alexandre From hughsient at gmail.com Thu Dec 10 15:26:35 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 15:26:35 +0000 Subject: LCMS dependency Message-ID: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> How would everyone feel if I added lcms as a hard dependency to gnome-color-manager? Now how would you all feel if I added lcms2 as a hard dep? Comments please. I think using lcms allows us to nuke a lot of the custom (fragile) profile code, but only lcms2 has the functionality I need to decode the vcgt tags. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 15:30:18 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 18:30:18 +0300 Subject: LCMS dependency In-Reply-To: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> Message-ID: <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> On 12/10/09, Richard Hughes wrote: > How would everyone feel if I added lcms as a hard dependency to > gnome-color-manager? Fine with me. Most graphics apps I use already have it as requirement. > Now how would you all feel if I added lcms2 as a hard dep? Marti does amazing job with lcms2, and since now it has a separate header for v2, this really might work :) Alexandre From pmjdebruijn at pcode.nl Thu Dec 10 15:47:06 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 16:47:06 +0100 Subject: LCMS dependency In-Reply-To: <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> Message-ID: <4c2c37f90912100747i1a0a507au47db84306b64dd42@mail.gmail.com> On Thu, Dec 10, 2009 at 4:30 PM, Alexandre Prokoudine wrote: > On 12/10/09, Richard Hughes wrote: >> How would everyone feel if I added lcms as a hard dependency to >> gnome-color-manager? > > Fine with me. Most graphics apps I use already have it as requirement. Indeed... >> Now how would you all feel if I added lcms2 as a hard dep? > > Marti does amazing job with lcms2, and since now it has a separate > header for v2, this really might work :) Adding lcms2 as a hard dependancy seems less then desirable (at this moment), since lcms2 hasn't been released yet. And it requires additional packaging... Adding it as an optional dependancy seems ok though. Regards, Pascal de Bruijn From graeme2 at argyllcms.com Thu Dec 10 16:05:34 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 11 Dec 2009 03:05:34 +1100 Subject: LCMS dependency In-Reply-To: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> Message-ID: <4B211C4E.9060502@argyllcms.com> Richard Hughes wrote: > I think using lcms allows us to nuke a lot of the custom (fragile) > profile code, but only lcms2 has the functionality I need to decode > the vcgt tags. Well of course icclib has support for accessing the vcgt tag. Graeme Gill. From hughsient at gmail.com Thu Dec 10 16:18:36 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 16:18:36 +0000 Subject: LCMS dependency In-Reply-To: <4B211C4E.9060502@argyllcms.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> Message-ID: <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> 2009/12/10 Graeme Gill : > Well of course icclib has support for accessing the vcgt tag. Sure, that's also a possibility. Richard. From pmjdebruijn at pcode.nl Thu Dec 10 17:00:03 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:00:03 +0100 Subject: LCMS dependency In-Reply-To: <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> Message-ID: <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> On Thu, Dec 10, 2009 at 5:18 PM, Richard Hughes wrote: > 2009/12/10 Graeme Gill : >> Well of course icclib has support for accessing the vcgt tag. > > Sure, that's also a possibility. Well, lcms is used by most OSS graphical apps for color management, so lcms is a natural choice when we don't want to introduce new dependancies into most systems in general. But as long as lcms2 hasn't been released and packaged into most distro's, it makes testing gcm a bit harder, which means feedback on git versions might degrade. Another slightly nasty option would be, to use lcms2, but just (for now) include the code into the gcm repo until it has been released, and just statically compile it into the gcm binaries. And yes, I'll wash my mouth with some soap :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 10 17:11:56 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:11:56 +0100 Subject: The new TRC graphs Message-ID: <4c2c37f90912100911o6f9c4b23y4648e95b2390d82a@mail.gmail.com> Hi, Just thought I should mention this... The new TRC graphs genuinely kick-ass! Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 10 17:14:31 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:14:31 +0000 Subject: LCMS dependency In-Reply-To: <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> Message-ID: <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > Well, lcms is used by most OSS graphical apps for color management, so > lcms is a natural choice when we don't want to introduce new > dependancies into most systems in general. Right. > But as long as lcms2 hasn't been released and packaged into most > distro's, it makes testing gcm a bit harder, which means feedback on > git versions might degrade. Agreed. > Another slightly nasty option would be, to use lcms2, but just (for > now) include the code into the gcm repo until it has been released, > and just statically compile it into the gcm binaries. Ick, ick. > And yes, I'll wash my mouth with some soap :) Okay, I think after what everyone has said I'll add an lcms dep, but not a lcms2 dep. We can convert when lcms2 has had it's first public release. Richard. From pmjdebruijn at pcode.nl Thu Dec 10 17:15:23 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:15:23 +0100 Subject: Bug: All profiles are 256 bytes Message-ID: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> Hi, I just built git again, and I noticed that GCM (on the profiles tab), tells me all my ICC profiles are 256 bytes... There is definitely something fishy about this :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 10 17:19:10 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:19:10 +0100 Subject: CIE widget drawing odd graphs In-Reply-To: <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> Message-ID: <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> On Sun, Dec 6, 2009 at 9:19 PM, Richard Hughes wrote: > 2009/12/6 Pascal de Bruijn : >> I might have uncovered some minor issues with the CIE widget drawing code. > > They look pretty major to me! Right :) >> I've attached an archive containing several non-standard profiles, >> which are drawn quite oddly. > > After the release I would like to look at the widget drawing code and > the profile extraction code (there are still quite a lot of FIXME's > and unimplemented stuff) and add a boat load more self tests and > internal validations. I fear I might not have time in the next couple > of weeks (lots of other work to do, which I've been putting off for > ages) so if anyone is familiar with XYZ co-ordinates and ICC profiles > I would appreciate any code review and possible patches. I just built git again, and I noticed XYZ drawing has improved... But I didn't see any actual improvements to the CIE code in git... I assume the problem wasn't in the CIE XYZ widget, but in the data that was fed to it? Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 10 17:42:38 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:42:38 +0000 Subject: Bug: All profiles are 256 bytes In-Reply-To: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> References: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> Message-ID: <15e53e180912100942t39465d40rf8875000038e9378@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > I just built git again, and I noticed that GCM (on the profiles tab), > tells me all my ICC profiles are 256 bytes... There is definitely > something fishy about this :) commit 99e9f26fe78202515187dc14b93917664ad01dc0 Author: Richard Hughes Date: Thu Dec 10 17:41:49 2009 +0000 trivial: show the file size in the prefs dialog, not the clut size Richard. From hughsient at gmail.com Thu Dec 10 17:43:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:43:40 +0000 Subject: CIE widget drawing odd graphs In-Reply-To: <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> Message-ID: <15e53e180912100943o413617a4j3d75cc58a096540b@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > I just built git again, and I noticed XYZ drawing has improved... But > I didn't see any actual improvements to the CIE code in git... The profile code had a subtle signing bug that wasn't tested for. The change from gchar to guint8 fixed this bug (and possibly a few more). Richard. From web.kiddo at free.fr Thu Dec 10 20:56:13 2009 From: web.kiddo at free.fr (Jeff) Date: Thu, 10 Dec 2009 15:56:13 -0500 Subject: Incorrect parsing of .icm files? Message-ID: <1260478573.20024.33.camel@kusanagi> Hello, I wanted to give gnome-color-manager a try just out of curiosity. I own a Dell Inspiron 910 (a.k.a. Mini 9) netbook, which (to my untrained eye) seems to have good colors out of the box. I downloaded the Windows drivers for this netbook's monitor here: http://support.us.dell.com/support/downloads/download.aspx?releaseid=R229102&fileid=329762 It seems like the monitor is called ST2010, there's an ST2010B and ST2010F version of the driver... both contain the exact same ".icm" file (verified with md5sum), which is attached to this message. Now, here comes the actual question: loading this .icm file in GNOME Color Manager and using it creates obviously incorrect colors (as if someone punched contrast and saturation way up), compared to the default colors (no color profile loaded; colors appear neutral). With the ICC profile loaded, for example, gradients in the Clearlooks GTK theme lose definition: the slight gray gradients on the combobox widgets disappears completely into white. I'm just wondering if - this is a bug in GNOME Color Manager - this is a buggy .icm file - it's a PEBKAC. I wish I could take a screenshot, but screenshots don't work in this case (and photo cameras have trouble photographing LCD screens)... -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ST2010.icm Type: image/x-tga Size: 552 bytes Desc: URL: From hughsient at gmail.com Thu Dec 10 21:24:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 21:24:28 +0000 Subject: Incorrect parsing of .icm files? In-Reply-To: <1260478573.20024.33.camel@kusanagi> References: <1260478573.20024.33.camel@kusanagi> Message-ID: <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> 2009/12/10 Jeff : > Now, here comes the actual question: loading this .icm file in GNOME Color > Manager and using it creates obviously incorrect colors (as if someone > punched contrast and saturation way up), compared to the default colors (no > color profile loaded; colors appear neutral). With the ICC profile loaded, > for example, gradients in the Clearlooks GTK theme lose definition: the > slight gray gradients on the combobox widgets disappears completely into > white. > > I'm just wondering if > - this is a bug in GNOME Color Manager > - this is a buggy .icm file Both, I think. If you try gnome-power-manager in git master, it won't "do" anything, as the profile does not have a vcgt or mlut tag, and even using the linear tables doesn't do anything. Even trying using dispwin doesn't appear to do anything. If you use the 2.29.1 release there's a bug where we're using the xTRC values as a gamma table (eek). Richard. From hughsient at gmail.com Thu Dec 10 21:28:59 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 21:28:59 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B202D16.1000901@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> Message-ID: <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> 2009/12/9 Graeme Gill : > Progromatically, one would do the equivalent, e.g. using icclib > open the profile, create a lookup object to do forward absolute colorimetric > to XYZ space, and then feed colorant only combinations through it. > icclu.c will illustrate the details. I'm sure something similar is > possible using Marti's lcms. I've just added code to git master to do this, and it indeed works to get the gamut range for a few scanner and print (RGB) profiles I've got here. It doesn't seem to work for CYMK profiles, although I'm sure that's something crazy I'm doing (or not doing..). Anyway, I'll give it another go to get CYMK profiles working correctly. Richard. From graeme2 at argyllcms.com Thu Dec 10 21:51:35 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 11 Dec 2009 08:51:35 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> Message-ID: <4B216D67.2080307@argyllcms.com> Richard Hughes wrote: > I've just added code to git master to do this, and it indeed works to > get the gamut range for a few scanner and print (RGB) profiles I've > got here. It doesn't seem to work for CYMK profiles, although I'm sure > that's something crazy I'm doing (or not doing..). Subtractive media profiles usually have less easily defined gamut extents than additive (RGB) profiles. Unlike additive, you can't expect it to be a nice polygon in XYZ/xy/u'v' space. Something roughly accurate for a chromaticity diagram would involve sampling along the color girdle (ie. going through combinations between CMY: 1 0 0, 1 1 0, 0 1 0, 0 1 1, 0 0 1, 1 0 1, e.g. 1 0 0, 1 .1 0, 1 .2 0 .. 1 .9 0, 1 1 0, .9 1 0, .8 1 0, ... etc.) Graeme Gill. From hughsient at gmail.com Thu Dec 10 22:08:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 22:08:49 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B216D67.2080307@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> <4B216D67.2080307@argyllcms.com> Message-ID: <15e53e180912101408w4b7dca45o4dac5869574a39bd@mail.gmail.com> 2009/12/10 Graeme Gill : > 1 1 0, 0 1 0, 0 1 1, 0 0 1, 1 0 1, > e.g. 1 0 0, 1 .1 0, 1 .2 0 .. 1 .9 0, 1 1 0, .9 1 0, .8 1 0, ... etc.) Excellent, this makes sense. I'll give it a go tomorrow. Thanks again. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 22:35:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 01:35:09 +0300 Subject: Incorrect parsing of .icm files? In-Reply-To: <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> References: <1260478573.20024.33.camel@kusanagi> <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> Message-ID: <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> On 12/11/09, Richard Hughes wrote: > Both, I think. If you try gnome-power-manager in git master, it won't > "do" anything, as the profile does not have a vcgt or mlut tag, and > even using the linear tables doesn't do anything. Even trying using > dispwin doesn't appear to do anything. If you use the 2.29.1 release > there's a bug where we're using the xTRC values as a gamma table > (eek). BTW, what's new regarding restoration of applied profile after dehibernation and leaving a screensaver? Alexandre From hughsient at gmail.com Fri Dec 11 08:55:43 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 08:55:43 +0000 Subject: Incorrect parsing of .icm files? In-Reply-To: <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> References: <1260478573.20024.33.camel@kusanagi> <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> Message-ID: <15e53e180912110055s2a1bc938ob58b1ef4508ef83a@mail.gmail.com> 2009/12/10 Alexandre Prokoudine : > BTW, what's new regarding restoration of applied profile after > dehibernation and leaving a screensaver? If you're using a KMS driver (the future) then the "thaw" should re-apply the per-output gamma tables. With the screensaver I want to work on a patch that either: * saves the gamma table before the table is destroyed * calls gcm-apply after the screensaver comes back from black I don't use a screensaver myself, but if nobody does the gnome-screensaver work I guess I'll have to add that to my list. Richard. From knizek.confy at volny.cz Fri Dec 11 09:46:18 2009 From: knizek.confy at volny.cz (Milan =?UTF-8?Q?Kn=C3=AD=C5=BEek?=) Date: Fri, 11 Dec 2009 10:46:18 +0100 Subject: shared-color-profiles 0.1.0 release In-Reply-To: <15e53e180912101003s2a013080qf839379565a40964@mail.gmail.com> References: <15e53e180912091151y1f2303acw48cedde3eb0231f@mail.gmail.com> <1260467822.3147.14.camel@athlon> <15e53e180912101003s2a013080qf839379565a40964@mail.gmail.com> Message-ID: <1260524778.2203.3.camel@athlon> (I am forwarding your response to the emailing list, since my original intention was not to email just you but the list and I did not notice the "reply" address is just yours.) Possibly someone knows about other pre-packaged profiles? regards, Milan Richard Hughes p??e v ?t 10. 12. 2009 v 18:03 +0000: > 2009/12/10 Milan Kn??ek : > > Will the project make "icc-profiles" and "oyranos-icc" obsolete or > > rather complementary? > > I hope we can standardize on one -- I must admit I had not heard of > icc-profiles until you pointed me at them. > > In the future, we'll be distributing shared-color-targets as well (8.7 > targets) to make all this stuff just work out of the box. > > Richard. > Milan Knizek knizek (dot) confy (at) volny (dot) cz http://www.milan-knizek.net - About linux and photography (Czech language only) From lars.tore at mulebakken.net Fri Dec 11 15:11:33 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 11 Dec 2009 16:11:33 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912110711n7d6a6e0ub9a3bf5b9f817b2f@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: > the x and y coordinates of the white point from the device ICC > profile. > The white point cross is nice, but I wonder if it would be nice to add a more human readable text string with the whitepoint somewhere: The wtpt program included in liblcms-utils outputs the correlated temperature or the locus temperature, so I guess it's easy to this with lcms. Here are some sample output: profile 1: White point near 4937K profile 2: WhitePoint : D65 (daylight) And I think this is more readable to most people than xy values. Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 11 15:51:36 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 11 Dec 2009 16:51:36 +0100 Subject: LCMS dependency In-Reply-To: <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> Message-ID: <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> On Thu, Dec 10, 2009 at 6:14 PM, Richard Hughes wrote: > 2009/12/10 Pascal de Bruijn : >> Well, lcms is used by most OSS graphical apps for color management, so >> lcms is a natural choice when we don't want to introduce new >> dependancies into most systems in general. > > Right. I see a lot of commits regarding lcms in git... Has all the ICC parsing code already been removed from gcm? I'm just asking this out of curiosity... Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Fri Dec 11 17:47:18 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 11 Dec 2009 18:47:18 +0100 Subject: Regression: the TRC graphs disappear on working space/input profiles Message-ID: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> Hi, With today's git checkout the TRC graphs disappear with working space profiles and input profiles. Is this intentional? Or just transitory during the lcms migration? Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Fri Dec 11 17:47:30 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 20:47:30 +0300 Subject: apply/session Message-ID: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> Hi, So, to get things right, what exactly is the difference between gcm-session and gcm-apply? Alexandre From hughsient at gmail.com Fri Dec 11 17:50:06 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:50:06 +0000 Subject: LCMS dependency In-Reply-To: <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> Message-ID: <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> 2009/12/11 Pascal de Bruijn : > I see a lot of commits regarding lcms in git... > Has all the ICC parsing code already been removed from gcm? About half. To remove the other half we need lcms2, which may be some months away. Richard. From hughsient at gmail.com Fri Dec 11 17:51:47 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:51:47 +0000 Subject: apply/session In-Reply-To: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> Message-ID: <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> 2009/12/11 Alexandre Prokoudine : > So, to get things right, what exactly is the difference between > gcm-session and gcm-apply? gcm-session is the dbus-activated session service, that provides an interface applications like gimp can use. gcm-apply is just a quick hack that sets (or resets) all the CLUT values to what they should be, in case something destroyed them (e.g. hibernate, screensaver dim, etc). Richard From hughsient at gmail.com Fri Dec 11 17:52:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:52:49 +0000 Subject: Regression: the TRC graphs disappear on working space/input profiles In-Reply-To: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> References: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> Message-ID: <15e53e180912110952s3f10d3dmc6305f350a17548@mail.gmail.com> 2009/12/11 Pascal de Bruijn : > With today's git checkout the TRC graphs disappear with working space > profiles and input profiles. > Is this intentional? Or just transitory during the lcms migration? Intentional, but they'll be coming back hopefully on Monday when I merge proper TRC support. Richard From alexandre.prokoudine at gmail.com Fri Dec 11 18:02:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 21:02:09 +0300 Subject: apply/session In-Reply-To: <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> Message-ID: <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> On Fri, Dec 11, 2009 at 8:51 PM, Richard Hughes wrote: > 2009/12/11 Alexandre Prokoudine : >> So, to get things right, what exactly is the difference between >> gcm-session and gcm-apply? > > gcm-session is the dbus-activated session service, that provides an > interface applications like gimp can use. And it's not currently used by anything, if I get it right? > gcm-apply is just a quick > hack that sets (or resets) all the CLUT values to what they should be, > in case something destroyed them (e.g. hibernate, screensaver dim, > etc). So vcgt part and setting the atom is done entirely by gcm-prefs? Alexandre From alexandre.prokoudine at gmail.com Fri Dec 11 18:06:41 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 21:06:41 +0300 Subject: LCMS dependency In-Reply-To: <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> Message-ID: <733f2c730912111006s640088c6g6b11a1266c54e4c8@mail.gmail.com> On Fri, Dec 11, 2009 at 8:50 PM, Richard Hughes wrote: >> Has all the ICC parsing code already been removed from gcm? > > About half. To remove the other half we need lcms2, which may be some > months away. Is this why TRC curve has (temp.) gone? :) Alexandre From claude at 2xlibre.net Fri Dec 11 22:36:40 2009 From: claude at 2xlibre.net (Claude Paroz) Date: Fri, 11 Dec 2009 23:36:40 +0100 Subject: Why not using strftime? Message-ID: <1260571000.4145.121.camel@shuttle> Hi, As I didn't find any bug tracker now, I'm reporting this here. I wonder why you're using a complicated machinery with gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of simply using strftime call in gcm-profile.c Something like (not tested code): ret = cmsTakeCreationDateTime (&created, priv->lcms_profile); if (ret) - priv->datetime = gcm_utils_format_date_time (1900 +created.tm_year, created.tm_mon+1, created + /* Translators: this is the profile creation date strftime format */ + strftime (priv->datetime, sizeof(priv->datetime), _("%d %B %Y, %H:%M:%S"), created); Months names should not be translated again, this is a waste of time for translators. Claude P.S. Please cc me on answer From hughsient at gmail.com Sat Dec 12 09:01:19 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 12 Dec 2009 09:01:19 +0000 Subject: apply/session In-Reply-To: <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> Message-ID: <15e53e180912120101n74bc1e92i31b4359ccaab961e@mail.gmail.com> 2009/12/11 Alexandre Prokoudine : >> gcm-session is the dbus-activated session service, that provides an >> interface applications like gimp can use. > > And it's not currently used ?by anything, if I get it right? Correct. It's the egg part of the chicken-and-egg question. >> gcm-apply is just a quick >> hack that sets (or resets) all the CLUT values to what they should be, >> in case something destroyed them (e.g. hibernate, screensaver dim, >> etc). > > So vcgt part and setting the atom is done entirely by gcm-prefs? No, when gcm "set the CLUT" it actually: * Applies the CLUT to each adaptor * Sets the _ICC_PROFILE atom on each output * Sets the _ICC_PROFILE atom on the screen for the primary output To apply the clut we extract the VCGT and then push it into the video card. Settign the CLUT is done by gcm-apply and gcm-prefs. You could argue the latter is redundant, unless you're changing the profile, and I would agree. Richard. From hughsient at gmail.com Sat Dec 12 09:21:57 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 12 Dec 2009 09:21:57 +0000 Subject: Why not using strftime? In-Reply-To: <1260571000.4145.121.camel@shuttle> References: <1260571000.4145.121.camel@shuttle> Message-ID: <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> 2009/12/11 Claude Paroz : > As I didn't find any bug tracker now, I'm reporting this here. I wonder > why you're using a complicated machinery with > gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of > simply using strftime call in gcm-profile.c Originally because we got some complicated date structure to decode, now we're using lcms we just get a nice sane tm struct. I've fixed things up in the following commit: commit 6607487433a1bf074638d6833bd7c493ef1d75ff Author: Richard Hughes Date: Sat Dec 12 09:19:57 2009 +0000 Use strftime rather than our own hand-rolled function :100644 100644 a6abf96... 0abfb35... M src/gcm-profile.c :100644 100644 203466c... 76e5b24... M src/gcm-utils.c :100644 100644 bc3ad9d... a570d6b... M src/gcm-utils.h Thanks Richard. From claude at 2xlibre.net Sat Dec 12 09:48:28 2009 From: claude at 2xlibre.net (Claude Paroz) Date: Sat, 12 Dec 2009 10:48:28 +0100 Subject: Why not using strftime? In-Reply-To: <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> References: <1260571000.4145.121.camel@shuttle> <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> Message-ID: <1260611308.3833.38.camel@shuttle> Le samedi 12 d?cembre 2009 ? 09:21 +0000, Richard Hughes a ?crit : > 2009/12/11 Claude Paroz : > > As I didn't find any bug tracker now, I'm reporting this here. I wonder > > why you're using a complicated machinery with > > gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of > > simply using strftime call in gcm-profile.c > > Originally because we got some complicated date structure to decode, > now we're using lcms we just get a nice sane tm struct. I've fixed > things up in the following commit: > > commit 6607487433a1bf074638d6833bd7c493ef1d75ff > Author: Richard Hughes > Date: Sat Dec 12 09:19:57 2009 +0000 > > Use strftime rather than our own hand-rolled function > > :100644 100644 a6abf96... 0abfb35... M src/gcm-profile.c > :100644 100644 203466c... 76e5b24... M src/gcm-utils.c > :100644 100644 bc3ad9d... a570d6b... M src/gcm-utils.h > > Thanks Great, thanks ! Claude From pmjdebruijn at pcode.nl Mon Dec 14 14:44:19 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 14 Dec 2009 15:44:19 +0100 Subject: Patch: Debian binary detection fix In-Reply-To: <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> Message-ID: <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> On Thu, Nov 26, 2009 at 10:07 PM, Richard Hughes wrote: > 2009/11/26 Pascal de Bruijn : >> Therefore the detect should _first_ detect if the Debian renamed >> binaries exist, and if not, fallback to the original binary names. > > Agreed. I've attached the patch I've committed. Thanks. Maybe it's a thought to (last) search /usr/local/bin as well for the standard binary names? Regards, Pascal de Bruijn From hughsient at gmail.com Mon Dec 14 15:13:51 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 14 Dec 2009 15:13:51 +0000 Subject: Patch: Debian binary detection fix In-Reply-To: <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> Message-ID: <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> 2009/12/14 Pascal de Bruijn : > Maybe it's a thought to (last) search /usr/local/bin as well for the > standard binary names? Sure. I'll do a deal: You code it, and I'll apply it. If you need instructions, I can explain how. Richard. From hughsient at gmail.com Wed Dec 16 11:57:19 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 16 Dec 2009 11:57:19 +0000 Subject: PackageKit integration Message-ID: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> gnome-color-manager in git master now integrates with PackageKit in the following ways: * Offers to install argyllcms if the user tries to calibrate without argyllcms installed * Offers to install shared-color-targets if they are not installed, and the user tries to calibrate a non-display device This means gnome-color-manager can be shipped on live-cd's without bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) shared-color-targets, just in case the user wants to calibrate a device. If people want to turn off the PackageKit bits, then I can add a GConf key or #ifdef as required. Richard. From pmjdebruijn at pcode.nl Wed Dec 16 12:23:52 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 16 Dec 2009 13:23:52 +0100 Subject: PackageKit integration In-Reply-To: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> Message-ID: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> On Wed, Dec 16, 2009 at 12:57 PM, Richard Hughes wrote: > gnome-color-manager in git master now integrates with PackageKit in > the following ways: > > * Offers to install argyllcms if the user tries to calibrate without > argyllcms installed > * Offers to install shared-color-targets if they are not installed, > and the user tries to calibrate a non-display device > > This means gnome-color-manager can be shipped on live-cd's without > bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) > shared-color-targets, just in case the user wants to calibrate a > device. > > If people want to turn off the PackageKit bits, then I can add a GConf > key or #ifdef as required. Since not all distro's don't use PackageKit that would be very welcome. A ./configure option would probably be best? Since this is a distro specific setting, it needs to be set in the packaging process. And GConf is very user specific, so it would seem odd to use GConf for this particular purpose. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 16 12:32:06 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 16 Dec 2009 13:32:06 +0100 Subject: PackageKit integration In-Reply-To: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> Message-ID: <4c2c37f90912160432r5cc68f8as29bc4d7419eeecd8@mail.gmail.com> On Wed, Dec 16, 2009 at 1:23 PM, Pascal de Bruijn wrote: > On Wed, Dec 16, 2009 at 12:57 PM, Richard Hughes wrote: >> gnome-color-manager in git master now integrates with PackageKit in >> the following ways: >> >> * Offers to install argyllcms if the user tries to calibrate without >> argyllcms installed >> * Offers to install shared-color-targets if they are not installed, >> and the user tries to calibrate a non-display device >> >> This means gnome-color-manager can be shipped on live-cd's without >> bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) >> shared-color-targets, just in case the user wants to calibrate a >> device. >> >> If people want to turn off the PackageKit bits, then I can add a GConf >> key or #ifdef as required. > > Since not all distro's don't use PackageKit that would be very welcome. > > A ./configure option would probably be best? Since this is a distro > specific setting, it needs to be set in the packaging process. > > And GConf is very user specific, so it would seem odd to use GConf for > this particular purpose. Hmm doesn't totem do similar things with codecs and the gstreamer packages? GCM should probably follow a similar approach... Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 16 12:42:07 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 16 Dec 2009 12:42:07 +0000 Subject: PackageKit integration In-Reply-To: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> Message-ID: <15e53e180912160442v2f95d4rb4b670389fb556d9@mail.gmail.com> 2009/12/16 Pascal de Bruijn : > Since not all distro's don't use PackageKit that would be very welcome. commit 494ebc7e84d21d2626e06aa7e009cc90625e0fd8 Author: Richard Hughes Date: Wed Dec 16 12:41:02 2009 +0000 Add a configure option --enable-packagekit (default yes) to choose if auto-install functionality should be included Richard. From joedalton2 at yahoo.dk Wed Dec 16 22:16:18 2009 From: joedalton2 at yahoo.dk (Joe Dalton) Date: Wed, 16 Dec 2009 14:16:18 -0800 (PST) Subject: bugreport gnome-color-manager translation Message-ID: <977326.61727.qm@web24408.mail.ird.yahoo.com> "Report-Msgid-Bugs-To: http://bugzilla.gnome.org/enter_bug.cgi?product=gnome-" "color-manager\n" but https://bugzilla.gnome.org/enter_bug.cgi?product=gnome-color-manager gives this message Sorry, either the product gnome-color-manager does not exist or you aren't authorized to enter a bug into it. Please choose a valid product. What does intent mean here. Is the sentence broken in some way/missing something? #: ../data/gnome-color-manager.schemas.in.h:7 msgid "The output intent to use, e.g. 'perceptual'." Hi, a google did not help me, and i have only found one danish translation, which maybe is correct mayby not. Please give some more info. #. TRANSLATORS: rendering intent: you probably want to google this #: ../src/gcm-prefs.c:2029 msgid "Perceptual" bye Joe Danish translation ___________________________________________________________ Skal du k?be ny bil? Sammenlign priser p? brugte biler med Kelkoo og find et godt tilbud! - Se mere her http://dk.yahoo.com/r/pat/mmb From hughsient at gmail.com Thu Dec 17 09:31:22 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 17 Dec 2009 09:31:22 +0000 Subject: bugreport gnome-color-manager translation In-Reply-To: <977326.61727.qm@web24408.mail.ird.yahoo.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> Message-ID: <15e53e180912170131p5c35354dv6e3df58be1ee6782@mail.gmail.com> 2009/12/16 Joe Dalton : > Sorry, either the product gnome-color-manager ?does not exist or you aren't authorized to enter a bug into it. Please choose a valid product. I've opened a bug to have the bugzilla product created: https://bugzilla.gnome.org/show_bug.cgi?id=604811 > What does intent mean here. Is the sentence broken in some way/missing something? > > #: ../data/gnome-color-manager.schemas.in.h:7 > msgid "The output intent to use, e.g. 'perceptual'." The output intent is the way that one range of colors is mapped to another range, for instance, do we clip the extremes of color, or do we try to scale it so they are all represented. Intent in English means "what you would do if possible" although in this technical speak it means more of "what method do you want me to use when transforming the color" > Hi, a google did not help me, and i have only found one danish translation, which maybe is correct mayby not. Please give some more info. > #. TRANSLATORS: rendering intent: you probably want to google this > #: ../src/gcm-prefs.c:2029 > msgid "Perceptual" This is how I understand perceptual rendering intent: '''This rendering intent maps color "smoothly", preserving relationships between similar colors. This prevents "gamut clipping" with its potential loss of detail and "tonal banding" problems. Gamut clipping occurs when colors that are different in the input image appear the same when printed. Perceptual rendering intent makes small compromises throughout the entire color space in order to preserve color relationships. It sacrifices some precision of in-gamut colors in order to ensure pleasing results.''' Does that make things clearer? Feel free to change the translator comments to something more sensible -- you know what you are doing. :-) Richard. From alexandre.prokoudine at gmail.com Thu Dec 17 12:29:54 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 17 Dec 2009 15:29:54 +0300 Subject: bugreport gnome-color-manager translation In-Reply-To: <977326.61727.qm@web24408.mail.ird.yahoo.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> Message-ID: <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> On 12/17/09, Joe Dalton wrote: > What does intent mean here. Is the sentence broken in some way/missing > something? > > #: ../data/gnome-color-manager.schemas.in.h:7 > msgid "The output intent to use, e.g. 'perceptual'." > > > Hi, a google did not help me, and i have only found one danish translation, > which maybe is correct mayby not. I'm pretty sure that both GIMP, Inkscape and Scribus have Danish translations. Check them re. "Color management" tab in Preferences of all these applications. I also highly recommend "Real world color management" book. It was translated into many languages, maybe Danish was among them. You will understand a lot more about color management after reading it, trust me on this :) Alexandre From anders at brander.dk Thu Dec 17 12:35:12 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 17 Dec 2009 13:35:12 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> Message-ID: <1261053312.7873.1.camel@smutogtab> Hi, On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: > I also highly recommend "Real world color management" book. It was > translated into many languages, maybe Danish was among them. You will > understand a lot more about color management after reading it, trust > me on this :) I second that, go read "Real world color management" now, if you haven't already. It doesn't have a danish translation by the way. Regards, Anders Brander From pmjdebruijn at pcode.nl Thu Dec 17 12:38:42 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 17 Dec 2009 13:38:42 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <1261053312.7873.1.camel@smutogtab> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> Message-ID: <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> On Thu, Dec 17, 2009 at 1:35 PM, Anders Brander wrote: > Hi, > > On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: >> I also highly recommend "Real world color management" book. It was >> translated into many languages, maybe Danish was among them. You will >> understand a lot more about color management after reading it, trust >> me on this :) > > I second that, go read "Real world color management" now, if you haven't > already. It doesn't have a danish translation by the way. I third that, the first 75% is written completely generic and os/application independent. It's an excellent book... Generally for most photographical work you'll want Perceptual... Relative Colorimetric is usually most useful when you have very high end equipment, thus limiting clipping... Regards, Pascal de Bruijn From lars.tore at mulebakken.net Thu Dec 17 15:11:12 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 17 Dec 2009 16:11:12 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> Message-ID: <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> On Thu, Dec 17, 2009 at 1:38 PM, Pascal de Bruijn wrote: > On Thu, Dec 17, 2009 at 1:35 PM, Anders Brander wrote: >> Hi, >> >> On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: >>> I also highly recommend "Real world color management" book. It was >>> translated into many languages, maybe Danish was among them. You will >>> understand a lot more about color management after reading it, trust >>> me on this :) >> >> I second that, go read "Real world color management" now, if you haven't >> already. It doesn't have a danish translation by the way. > > I third that, the first 75% is written completely generic and > os/application independent. It's an excellent book... > > Generally for most photographical work you'll want Perceptual... > > Relative Colorimetric is usually most useful when you have very high > end equipment, thus limiting clipping... > > Regards, > Pascal de Bruijn > _______________________________________________ > gnome-color-manager-list mailing list > gnome-color-manager-list at gnome.org > http://mail.gnome.org/mailman/listinfo/gnome-color-manager-list > I think the book Understanding color managment by Abhay Sharma is a better book, and I have both :-) About the translation, in Norwegian (it might be close to danish) the correct word is "Perseptuell". But I think we don't have to translate all that specific terms. Since most of the written literature are in english, I guess most people who use color managment have no problems with the english terms on the rendering intets. Actually I think at least in my language that translated word will create a lot of confusion since the words are very seldom used, and everyone use the English terms. I guess we can blame photoshop for this. Lars Tore Gustavsen From anders at brander.dk Thu Dec 17 15:20:42 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 17 Dec 2009 16:20:42 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> Message-ID: <1261063242.15170.1.camel@smutogtab> Hi, On Thu, 2009-12-17 at 16:11 +0100, Lars Tore Gustavsen wrote: > About the translation, in Norwegian (it might be close to danish) the > correct word is "Perseptuell". I just looked it up in the danish Rawstudio translation, it's translated as "Opfattelse" - but I don't think that's correct (and I'm danish). Regards, Anders Brander From hughsient at gmail.com Sat Dec 19 17:32:08 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 19 Dec 2009 17:32:08 +0000 Subject: Bugzilla account created Message-ID: <15e53e180912190932x47e70876xcb147ce2718e3bff@mail.gmail.com> I've arranged for a bugzilla component to be created. If you want to file a bug, use https://bugzilla.gnome.org/enter_bug.cgi?product=gnome-color-manager but be sure to check for duplicates first, using https://bugzilla.gnome.org/buglist.cgi?product=gnome-color-manager At the moment, there are no bugs, which means GCM must be feature complete and perfect in every way. I can't imagine this situation will last long :-) Richard. From robert.ancell at gmail.com Sun Dec 20 22:34:24 2009 From: robert.ancell at gmail.com (Robert Ancell) Date: Mon, 21 Dec 2009 09:34:24 +1100 Subject: Color management and Simple Scan Message-ID: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Hi all, I've been developing a basic scanning application called Simple Scan ( http://launchpad.net/simple-scan) and one request is to add color management to it: https://bugs.edge.launchpad.net/simple-scan/+bug/498029 I don't know much about color management so my question is: - What do I need to do to make Simple Scan work with GNOME Color Manager? - Is there a GPL project I can lift some code from? Thanks, --Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughsient at gmail.com Mon Dec 21 08:28:27 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 21 Dec 2009 08:28:27 +0000 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Message-ID: <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> 2009/12/20 Robert Ancell : > I don't know much about color management so my question is: > - What do I need to do to make Simple Scan work with GNOME Color Manager? > - Is there a GPL project I can lift some code from? Well, first you need to ask gnome-color-manager for a list of suitable profiles using the DBus interface: http://git.gnome.org/browse/gnome-color-manager/plain/src/org.gnome.ColorManager.xml If you know the device you're using (you probably should) then you can call GetProfilesForDevice() with the sysfs path, and a list of profiles will be returned. If more than one profile is returned you probably need to show it in a combo-box or something, but it would be quite unusual for more than one profile to be returned. If you get no results, then it's probably because the user has never assigned a profile for that device (i.e. there is no a-priori mapping). In this case you can either default to not doing color conversion yourself, or instead call the GetProfilesForType() dbus method, using the type=scanner. This will return a list of profiles that are suitable for use with scanners. Now, when you have a filename of the profile you want to use it's an easy (!) case of acquiring the image like usual, and then you can either: * Just save the image with an embedded color profile (see bug https://bugzilla.gnome.org/show_bug.cgi?id=604610 -- I'm working as fast as I can on the infrastructure bits) * Convert the image from device-mapping to something like sRGB using lcms (there are examples in the lcms tarball) I think it's probably best you do the first one, unless you know you're going to be using the images in a non-color-aware application. Anyway, I hope that helps, Richard. From alexandre.prokoudine at gmail.com Mon Dec 21 16:01:01 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 21 Dec 2009 19:01:01 +0300 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Message-ID: <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> On Mon, Dec 21, 2009 at 1:34 AM, Robert Ancell wrote: > - Is there a GPL project I can lift some code from? XSane sounds like an obvious source :) Alexandre From hughsient at gmail.com Mon Dec 21 16:08:05 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 21 Dec 2009 16:08:05 +0000 Subject: Color management and Simple Scan In-Reply-To: <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> Message-ID: <15e53e180912210808y5fe054c7p1f2d41f4fdb97ee1@mail.gmail.com> 2009/12/21 Alexandre Prokoudine : > XSane sounds like an obvious source :) Well, depends on how you like your sanity. Richard. From robert.ancell at gmail.com Mon Dec 21 22:57:54 2009 From: robert.ancell at gmail.com (Robert Ancell) Date: Tue, 22 Dec 2009 09:57:54 +1100 Subject: Color management and Simple Scan In-Reply-To: <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> Message-ID: <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> Thanks very much Richard! Very interested in the GDK patch - I need to make a patch there too to embed resolution information: https://bugs.edge.launchpad.net/simple-scan/+bug/484616 I think I need to do the conversion regardless so the preview is displayed correctly. Thanks for the lcms link. --Robert 2009/12/21 Richard Hughes > 2009/12/20 Robert Ancell : > > I don't know much about color management so my question is: > > - What do I need to do to make Simple Scan work with GNOME Color Manager? > > - Is there a GPL project I can lift some code from? > > Well, first you need to ask gnome-color-manager for a list of suitable > profiles using the DBus interface: > > http://git.gnome.org/browse/gnome-color-manager/plain/src/org.gnome.ColorManager.xml > > If you know the device you're using (you probably should) then you can > call GetProfilesForDevice() with the sysfs path, and a list of > profiles will be returned. If more than one profile is returned you > probably need to show it in a combo-box or something, but it would be > quite unusual for more than one profile to be returned. If you get no > results, then it's probably because the user has never assigned a > profile for that device (i.e. there is no a-priori mapping). In this > case you can either default to not doing color conversion yourself, or > instead call the GetProfilesForType() dbus method, using the > type=scanner. This will return a list of profiles that are suitable > for use with scanners. > > Now, when you have a filename of the profile you want to use it's an > easy (!) case of acquiring the image like usual, and then you can > either: > > * Just save the image with an embedded color profile (see bug > https://bugzilla.gnome.org/show_bug.cgi?id=604610 -- I'm working as > fast as I can on the infrastructure bits) > * Convert the image from device-mapping to something like sRGB using > lcms (there are examples in the lcms tarball) > > I think it's probably best you do the first one, unless you know > you're going to be using the images in a non-color-aware application. > > Anyway, I hope that helps, > > Richard. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughsient at gmail.com Tue Dec 22 09:13:09 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 22 Dec 2009 09:13:09 +0000 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> Message-ID: <15e53e180912220113n329e8f81j982c21203734945f@mail.gmail.com> 2009/12/21 Robert Ancell : > Very interested in the GDK patch - I need to make a patch there too to embed > resolution information: > https://bugs.edge.launchpad.net/simple-scan/+bug/484616 Sure, looks like you just need to add a new pixbuf-option. It's not that hard, as long as you're comfortable just implementing a spec in otherwise fairly hairy code. :-) If you need a hand, give me a shout on or off list. Richard. From bcowan at fastmail.co.uk Thu Dec 24 20:06:29 2009 From: bcowan at fastmail.co.uk (Bruce Cowan) Date: Thu, 24 Dec 2009 20:06:29 +0000 Subject: String fixes Message-ID: <1261685189.7547.4.camel@localhost.localdomain> I've found a few string errors. Here's a patch to fix them. I thought I better post it here rather than just commit straight to the repository, because there is a potential point of disagreement. I've changed the strformat format to the en_US one, which could be a bit contentious. -- Bruce Cowan -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Minor-string-fixes.patch Type: text/x-patch Size: 2223 bytes Desc: not available URL: From hughsient at gmail.com Fri Dec 25 11:26:32 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 25 Dec 2009 11:26:32 +0000 Subject: Happy Christmas Message-ID: <15e53e180912250326n7d35ca86tc59fb282fd940a1c@mail.gmail.com> Hey guys. I just wanted to send to you all a Happy Christmas, and my thanks for the hard work you've put in this year. It's appreciated. Richard. From hughsient at gmail.com Fri Dec 25 11:28:10 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 25 Dec 2009 11:28:10 +0000 Subject: String fixes In-Reply-To: <1261685189.7547.4.camel@localhost.localdomain> References: <1261685189.7547.4.camel@localhost.localdomain> Message-ID: <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> 2009/12/24 Bruce Cowan : > I've found a few string errors. Here's a patch to fix them. I thought I > better post it here rather than just commit straight to the repository, > because there is a potential point of disagreement. The first two are obvious, so please commit those. > I've changed the strformat format to the en_US one, which could be a bit > contentious. I agree it's probably the correct thing to do, but it does however want me to rip my eyes out every time I see that format. :-) So, please commit the entire patch. Thanks. Richard. From pmjdebruijn at pcode.nl Fri Dec 25 12:28:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 25 Dec 2009 13:28:33 +0100 Subject: String fixes In-Reply-To: <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> References: <1261685189.7547.4.camel@localhost.localdomain> <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> Message-ID: <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> On Fri, Dec 25, 2009 at 12:28 PM, Richard Hughes wrote: > 2009/12/24 Bruce Cowan : >> I've found a few string errors. Here's a patch to fix them. I thought I >> better post it here rather than just commit straight to the repository, >> because there is a potential point of disagreement. > > The first two are obvious, so please commit those. > >> I've changed the strformat format to the en_US one, which could be a bit >> contentious. > > I agree it's probably the correct thing to do, but it does however > want me to rip my eyes out every time I see that format. :-) /me Shudders... We're introducing AM/PM time... oh well seems correct for the en_US locale... However, if one we're to switch to the en_UK locale, would AM/PM time be gone again? Regards, Pascal de Bruijn From bcowan at fastmail.co.uk Fri Dec 25 14:37:16 2009 From: bcowan at fastmail.co.uk (Bruce Cowan) Date: Fri, 25 Dec 2009 14:37:16 +0000 Subject: String fixes In-Reply-To: <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> References: <1261685189.7547.4.camel@localhost.localdomain> <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> Message-ID: <1261751836.1631.0.camel@localhost.localdomain> On Fri, 2009-12-25 at 13:28 +0100, Pascal de Bruijn wrote: > On Fri, Dec 25, 2009 at 12:28 PM, Richard Hughes wrote: > > 2009/12/24 Bruce Cowan : > >> I've found a few string errors. Here's a patch to fix them. I thought I > >> better post it here rather than just commit straight to the repository, > >> because there is a potential point of disagreement. > > > > The first two are obvious, so please commit those. > > > >> I've changed the strformat format to the en_US one, which could be a bit > >> contentious. > > > > I agree it's probably the correct thing to do, but it does however > > want me to rip my eyes out every time I see that format. :-) > > /me Shudders... > > We're introducing AM/PM time... oh well seems correct for the en_US locale... > > However, if one we're to switch to the en_UK locale, would AM/PM time > be gone again? Yup, I'll update the en_GB translation later as well, meaning we'll see 24 hour time still. -- Bruce Cowan From joedalton2 at yahoo.dk Fri Dec 25 17:59:33 2009 From: joedalton2 at yahoo.dk (Joe Dalton) Date: Fri, 25 Dec 2009 17:59:33 +0000 (GMT) Subject: bugreport gnome-color-manager translation In-Reply-To: <15e53e180912170131p5c35354dv6e3df58be1ee6782@mail.gmail.com> Message-ID: <381861.61449.qm@web24403.mail.ird.yahoo.com> thanx --- Den tors 17/12/09 skrev Richard Hughes : > Fra: Richard Hughes > Emne: Re: bugreport gnome-color-manager translation > Til: "Joe Dalton" > Cc: gnome-color-manager-list at gnome.org > Dato: torsdag 17. december 2009 10.31 > 2009/12/16 Joe Dalton : > > Sorry, either the product gnome-color-manager ?does > not exist or you aren't authorized to enter a bug into it. > Please choose a valid product. > > I've opened a bug to have the bugzilla product created: > https://bugzilla.gnome.org/show_bug.cgi?id=604811 > > > What does intent mean here. Is the sentence broken in > some way/missing something? > > > > #: ../data/gnome-color-manager.schemas.in.h:7 > > msgid "The output intent to use, e.g. 'perceptual'." > > The output intent is the way that one range of colors is > mapped to > another range, for instance, do we clip the extremes of > color, or do > we try to scale it so they are all represented. Intent in > English > means "what you would do if possible" although in this > technical speak > it means more of "what method do you want me to use when > transforming > the color" > > > Hi, a google did not help me, and i have only found > one danish translation, which maybe is correct mayby not. > Please give some more info. > > #. TRANSLATORS: rendering intent: you probably want to > google this > > #: ../src/gcm-prefs.c:2029 > > msgid "Perceptual" > > This is how I understand perceptual rendering intent: > > '''This rendering intent maps color "smoothly", preserving > relationships between similar colors. This prevents "gamut > clipping" > with its potential loss of detail and "tonal banding" > problems. Gamut > clipping occurs when colors that are different in the input > image > appear the same when printed. Perceptual rendering intent > makes small > compromises throughout the entire color space in order to > preserve > color relationships. It sacrifices some precision of > in-gamut colors > in order to ensure pleasing results.''' > > Does that make things clearer? Feel free to change the > translator > comments to something more sensible -- you know what you > are doing. > :-) > > Richard. > Find din nye laptop p? kelkoo.dk. Se de gode tilbud her - http://dk.yahoo.com/r/pat/mm From pmjdebruijn at pcode.nl Sun Dec 27 18:33:43 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 27 Dec 2009 19:33:43 +0100 Subject: Patch: Debian binary detection fix In-Reply-To: <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> Message-ID: <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> On Mon, Dec 14, 2009 at 4:13 PM, Richard Hughes wrote: > 2009/12/14 Pascal de Bruijn : >> Maybe it's a thought to (last) search /usr/local/bin as well for the >> standard binary names? > > Sure. I'll do a deal: You code it, and I'll apply it. If you need > instructions, I can explain how. I've submitted it has an enhancement on GNOME Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=605552 Regards, Pascal de Bruijn From hughsient at gmail.com Mon Dec 28 10:26:15 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 28 Dec 2009 10:26:15 +0000 Subject: Patch: Debian binary detection fix In-Reply-To: <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> Message-ID: <15e53e180912280226r2b1d3265x103a12fd65174192@mail.gmail.com> 2009/12/27 Pascal de Bruijn : > I've submitted it has an enhancement on GNOME Bugzilla: Committed, thanks. Richard. From pmjdebruijn at pcode.nl Mon Dec 28 18:20:54 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 28 Dec 2009 19:20:54 +0100 Subject: Documentation Images: Display OSD Message-ID: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Hi, I took a few shots of my display's OSD, mainly for my own purposes, but maybe they'll come in handy for GNOME Color Manager's documentation. These shots aren't perfect (at all), but it's a start. I can put these under any license you like... I'm fine with GPL and CC-BY-AC. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: lowres-osd-jpegs.zip Type: application/zip Size: 213749 bytes Desc: not available URL: From niklas.park at gmail.com Tue Dec 29 10:44:50 2009 From: niklas.park at gmail.com (Niklas Park) Date: Tue, 29 Dec 2009 11:44:50 +0100 Subject: Documentation Images: Display OSD In-Reply-To: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> References: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Message-ID: A little warning seems to be in place here. I can't see how you could put these screenshots under CC, FDL or any other license, since the copyright clearly belongs to the manufacturer of the device, or the person or entity the manufacturer licenses the OSD from. I can't really see that a screengrab could constitute original art under any jurisdiction. The best one could hope for is that using these images constitutes fair use, of course with correct attribution to the copyright holder, but IANAL. Kindly /Niklas Park 2009/12/28 Pascal de Bruijn > Hi, > > I took a few shots of my display's OSD, mainly for my own purposes, > but maybe they'll come in handy for GNOME Color Manager's > documentation. > > These shots aren't perfect (at all), but it's a start. > > I can put these under any license you like... I'm fine with GPL and > CC-BY-AC. > > Regards, > Pascal de Bruijn > > _______________________________________________ > gnome-color-manager-list mailing list > gnome-color-manager-list at gnome.org > http://mail.gnome.org/mailman/listinfo/gnome-color-manager-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pmjdebruijn at pcode.nl Tue Dec 29 14:41:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 29 Dec 2009 15:41:33 +0100 Subject: Documentation Images: Display OSD In-Reply-To: References: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Message-ID: <4c2c37f90912290641w8d6dff1y675a3a07844da84d@mail.gmail.com> On Tue, Dec 29, 2009 at 11:44 AM, Niklas Park wrote: > A little warning seems to be in place here. > I can't see how you could put these screenshots under CC, FDL or any other > license, since the copyright clearly belongs to the manufacturer of the > device, or the person or entity the manufacturer licenses the OSD from. I > can't really see that a screengrab could constitute original art under any > jurisdiction. Well, you've got a fair point there... In this case that would be HP... > The best one could hope for is that using these images constitutes fair use, > of course with correct attribution to the copyright holder, but IANAL. Since it's not a 1:1 (pixel-perfect) copy, and doesn't contain any brand logo's, it probably does... Regards, Pascal de Bruijn From hughsient at gmail.com Tue Dec 1 14:58:41 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 1 Dec 2009 14:58:41 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file Message-ID: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Attached is the new widget I've just added to gnome-color-manager git master. It allows us to show the user a preview of the gamut range of the profile in a graphical way, much like OSX does. You should be able to see the lightly shaded section. At the moment I've hardcoded the CIE REC 709 x-y values, but of course it's easy to change when put into GcmPrefs. So, the point of this email: I wanted to get the CIE x and y coordinates (or XYZ of course) of the three primary illuminants and the x and y coordinates of the white point from the device ICC profile. Is there an easy way to extract these end-point values from the ICC file? I guess mediaWhitePointTag is the latter half of the problem, but I'm not sure how to deal with the LUT values. Ideas welcome. Thanks. Richard. -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot.png Type: image/png Size: 49528 bytes Desc: not available URL: From lars.tore at mulebakken.net Tue Dec 1 16:43:31 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Tue, 1 Dec 2009 17:43:31 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912010843re5f1d81mef40fd5d57e50310@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: > Attached is the new widget I've just added to gnome-color-manager git > master. It allows us to show the user a preview of the gamut range of > the profile in a graphical way, much like OSX does. > > You should be able to see the lightly shaded section. At the moment > I've hardcoded the CIE REC 709 x-y values, but of course it's easy to > change when put into GcmPrefs. > > So, the point of this email: I wanted to get the CIE x and y > coordinates (or XYZ of course) of the three primary illuminants and > the x and y coordinates of the white point from the device ICC > profile. > > Is there an easy way to extract these end-point values from the ICC > file? I guess mediaWhitePointTag is the latter half of the problem, > but I'm not sure how to deal with the LUT values. Ideas welcome. > Thanks. > > Richard. > What about iccdump -v3 -t wtpt -t bkpt -t rXYZ -t gXYZ -t bXYZ myprofile.icc Lars Tore Gustavsen From lars.tore at mulebakken.net Tue Dec 1 18:12:38 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Tue, 1 Dec 2009 19:12:38 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912011012q42a44f4cvd385a8b97ca5f5eb@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: I'm not sure how to deal with the LUT values. Ideas welcome. > Thanks. > > Richard. > > _ I think I missed the last part. Can something like this be used? Just thinking loud here: echo 1 0 0|icclu -ff lutprofile.icc echo 0 1 0|icclu -ff lutprofile.icc echo 0 0 1|icclu -ff lutprofile.icc Lars Tore Gustavsen From graeme2 at argyllcms.com Wed Dec 2 00:19:40 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Wed, 02 Dec 2009 11:19:40 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <4B15B29C.2090105@argyllcms.com> Richard Hughes wrote: > Attached is the new widget I've just added to gnome-color-manager git > master. It allows us to show the user a preview of the gamut range of > the profile in a graphical way, much like OSX does. Professor Hunt would be jumping up and down in the first row at this point saying "Why are you using an XYZ chromaticity diagram ?" :-) :-) [Those who have been to the Color Imaging Conference will understand this.] Actually if you're wanting to plot Yxy values then it's appropriate, but I'll note that it's also very misleading in indicating gamut, since an xy diagram is highly visually non-uniform. A u' v' plot is a better choice to illustrate gamut, while still being linearly additive. > So, the point of this email: I wanted to get the CIE x and y > coordinates (or XYZ of course) of the three primary illuminants and > the x and y coordinates of the white point from the device ICC > profile. > > Is there an easy way to extract these end-point values from the ICC > file? I guess mediaWhitePointTag is the latter half of the problem, > but I'm not sure how to deal with the LUT values. Ideas welcome. The white point is in the white point tag (but see my final note). For a matrix profile the D50 adapted primaries are in the red/green/blue ColorantTags. For a cLut based profile, you would have to (as Lars Tore Gustavsen suggested) run the primary values through the profile (or do the equivalent of looking up what the cLUT values are at the primaries). Note a complication - you probably want the absolute colorant values, not the D50 adapted white ones. You need to undo the chromatic adaptation applied to the colorant tags, and/or do the lookup in absolute colorimetric intent mode. Note also that while absolute colorimetric interpretation of display profiles is perfectly consistent using Argyll/icclib based tools (and is consistent with the original sRGB and current AdobeRGB profiles), it probably will not be so for all other libaries and tools and profiles. This is due to a disagreement about the interpretation and intent of the ICC specification in this regard. Some display profiles have a white point tag of D50 even though this is not the white point of the display, and encode the absolute<->relative transformation in the ChromaticAdapation tag, meaning that the absolute white point and colorant values are not obtained using a normal absolute colorimetric lookup on these profiles. This latter interpretation is the one that has been adopted in the changes made for ICC V4 profiles, even though it is out of sync with much original ICC V2 practice. Graeme Gill. From hughsient at gmail.com Wed Dec 2 08:37:43 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 08:37:43 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B15B29C.2090105@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> Message-ID: <15e53e180912020037o6168de9dm4a20f429e7d665c@mail.gmail.com> 2009/12/2 Graeme Gill : > Actually if you're wanting to plot Yxy values then it's appropriate, > but I'll note that it's also very misleading in indicating gamut, > since an xy diagram is highly visually non-uniform. A u' v' plot > is a better choice to illustrate gamut, while still being linearly > additive. Sure, converting the code to plot u' v' should be pretty easy. The graph is really just for a passing glance, not for any serious comparison. > This is due to a disagreement about the interpretation and intent > of the ICC specification in this regard. Some display profiles > have a white point tag of D50 even though this is not the white > point of the display, and encode the absolute<->relative > transformation in the ChromaticAdapation tag, meaning that > the absolute white point and colorant values are not obtained > using a normal absolute colorimetric lookup on these profiles. This is exactly the information I needed, thanks very much for such a comprehensive answer. Richard. From hughsient at gmail.com Wed Dec 2 15:12:39 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 15:12:39 +0000 Subject: Uncaught bounce notification In-Reply-To: References: Message-ID: <15e53e180912020712l4506d8e1ja991ca132c63ae85@mail.gmail.com> 2009/12/2 : > i'v followed g-c-m using Pascal de Bruijn ppa (thanks to him), and for > these tests I'v built myself the last git (for CIE widget). For some reason your mail got bounced, apologies. I've forwarded it to the list. > So : > 1/ The window for listing screens are too big looking at the rest > (left/right panel) That should be fixed in git master now. > 2/ It's ok that this tool must be kept as simple as possible, but I > would love to have the possibility to set white K? That's on the assumption you're not using a hardware calibration device then? > 3/ if you play with icc profils, size of the right panel change and the > square of CIE widget may be crushed > 4/ I would love to see more (bigger/more details) the CIE widget when > clicking on it Sure, I'm not completely happy with the location or size of this widget. It seems to me I'm trying to cram the "I want to compare different profiles" and "assign a profile with a device" functionality into the single tab. maybe we just need to break out another tab called "Profiles" and list all the profiles along with some more information about them. In this way we can get rid of some of the clutter in the first page. Comments welcome. Richard. From ubuntu at kagou.fr Wed Dec 2 16:20:03 2009 From: ubuntu at kagou.fr (Patrice Vetsel) Date: Wed, 02 Dec 2009 17:20:03 +0100 Subject: Some tests Message-ID: <4B1693B3.8000809@kagou.fr> Hi, i'v followed g-c-m using Pascal de Bruijn ppa (thanks to him), and for these tests I'v built myself the last git (for CIE widget). So : 1/ The window for listing screens are too big looking at the rest (left/right panel) 2/ It's ok that this tool must be kept as simple as possible, but I would love to have the possibility to set white K? 3/ if you play with icc profils, size of the right panel change and the square of CIE widget may be crushed 4/ I would love to see more (bigger/more details) the CIE widget when clicking on it Thanks for this tool Regards Patrice Vetsel Aka Kagou From hughsient at gmail.com Wed Dec 2 16:47:21 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 16:47:21 +0000 Subject: Some tests In-Reply-To: <4B1693B3.8000809@kagou.fr> References: <4B1693B3.8000809@kagou.fr> Message-ID: <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> 2009/12/2 Patrice Vetsel : > 4/ I would love to see more (bigger/more details) the CIE widget when > clicking on it What do you all think of the attached mockups? Richard. -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-1.png Type: image/png Size: 42154 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-2.png Type: image/png Size: 86945 bytes Desc: not available URL: From pmjdebruijn at pcode.nl Wed Dec 2 17:09:30 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:09:30 +0100 Subject: Some tests In-Reply-To: <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> Message-ID: <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> On Wed, Dec 2, 2009 at 5:47 PM, Richard Hughes wrote: > 2009/12/2 Patrice Vetsel : >> 4/ I would love to see more (bigger/more details) the CIE widget when >> clicking on it > > What do you all think of the attached mockups? @Richard: Looks pretty good... The CIE widget is very nice. Though the profile whitepoint isn't being displayed yet, I'm assuming this is still on the TODO list. I also agree it would be nice (but not critical) if double clicking on the CIE widget would create a dialog with just the profile name and the CIE widget, for detailed viewing. And dare I ask, possibly with a right-click "Save (CIE diagram) as PNG/JPEG/PDF" option? Also, I notice the CIE widget has two "squares" of "space" to the left of the CIE diagram. Does this have a purpose? Otherwise couldn't this be decreased to a single white square. I also think either centering the CIE widget, or placing the buttons to the right would look a bit "better" esthetically... Or maybe move the button onder the list view? Giving the CIE diagram a bit more room to "grow"? @Patrice: I rebuild my PPA between 1 to 4 times per two days, depending on how interesting Richard's changes are... All in all it's pretty hard to keep up, and he's doing all the hard work :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 2 17:15:31 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:15:31 +0100 Subject: Idea: GTK/Cario CIE widget as a seperate library Message-ID: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> Hi, The following is just a thought, and is by no means very important... The GTK/Cairo CIE widget written for GCM is beginning to kick ass... However I guess it's GCM specific. Maybe for the long term it's a thought to separate it into a small library, so other apps can use it as well? For example whenever GIMP/UFraw open a dialog to import a new profile it would be very cool if the GNOME File Chooser would "preview" the profile by showing a CIE diagram. Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 2 17:35:04 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 17:35:04 +0000 Subject: Idea: GTK/Cario CIE widget as a seperate library In-Reply-To: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> References: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> Message-ID: <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > The GTK/Cairo CIE widget written for GCM is beginning to kick ass... > However I guess it's GCM specific. Not really, it's a standard GObject so it could be reused. > Maybe for the long term it's a thought to separate it into a small > library, so other apps can use it as well? Sure, there is loads of other common code that might be useful, but I don't think the project is at any level of internal ABI stability. Maybe we can revisit this idea in a few months time? Richard. From pmjdebruijn at pcode.nl Wed Dec 2 17:36:50 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:36:50 +0100 Subject: Profile description for laptop displays In-Reply-To: <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> References: <4c2c37f90911290453x143c6820nb27f3a6b6188e3c7@mail.gmail.com> <15e53e180911291342o2b8755aavad083da3dcd849d2@mail.gmail.com> <4c2c37f90911291527v71d2cf9cy6bb41add8c7b3683@mail.gmail.com> <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> Message-ID: <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> On Mon, Nov 30, 2009 at 11:10 AM, Richard Hughes wrote: > 2009/11/29 Pascal de Bruijn : >> I noticed you included the product_version as well, at least on my >> laptop it's actually the BIOS revision, I'm not sure what this is on >> other laptops. It's probably not that relevant... > > Hmm, it seems the manufacturers were in a rush when they designed your > laptop! The bios version, logically, belongs in bios_version :-) > > Could you attach the output of "cat /sys/class/dmi/id/*" and we'll try > and work around bios issues like this. Thanks. It's attached. >> I noticed you're currently filling the ICC Make with the username, it >> seems rather redundant with the copyright. Besides the point that the >> field wasn't ment to be used like that. I'd stick the EDID info there >> (even with laptops). For the Model field as well, EDID is most likely >> best and most accurate here. > > I deliberately changed it, as the vendor is supposed to be the person > or company that designed or created the profile. In this sense, it > seemed wrong to say that the profile was created "by Lenovo" when it > was created by me, for my Lenovo display. The ICCv2 specification disagrees: 6.4.15 deviceMfgDescTag Tag Type: textDescriptionType Tag Signature: ?dmnd? (646D6E64h) Structure containing invariant and localizable versions of the device manufacturer for display. The content of this structure is described in 6.5.17. 6.4.16 deviceModelDescTag Tag Type: textDescriptionType Tag Signature: ?dmdd? (646D6464h) Structure containing invariant and localizable versions of the device model for display. The content of this structure is described in 6.5.17. >> For example: >> >> Make: SEC (don't lookup in pnp.ids for storage in the ICC profile, >> only lookup in the gcm-prefs GUI). >> Model: 154AT07-H01 >> >> Storing the EDID info as verbatim as possible in the ICC make/models >> fields, will make it easier to auto match displays and profiles in the >> future. > > Hmm, I think the fields have to be displayable. We need to look at > adding private fields if we want to add any machine parsable data into > the ICC file. > >> Storing the EDID verbatim without looking up in pnp.ids prevents false >> data from conflicting lookups (like my Samsung/Seiko issue) from >> ending up in the profile itself. > > I think the lookup is valuable as it works most of the time. Your > panel should have encoded SAM into the PNP field, not SEC, unless of > course the panel really is a Seiko panel that's been rebadged as a > Samsung. For what it's worth, my nice expensive LG panel is actually a > generic Goldstar chipset with LG additions and plastic wrapping. > Hence, it shows up as "Goldstar" as the vendor. :-( I'm not saying my laptop's EDID info is not semi-broken... But translating SEC into Seiko for embedding in the profile, is moving from semi-wrong to completely-wrong... And I'm afraid my laptop isn't the only one having this or similar issues. I do wonder how Samsung's own brand sold displays identify themselves. I wish we had some more EDID samples to compare... Maybe a call for sample EDID data on Planet GNOME would get us some more samples? Regards, Pascal de Bruijn -------------- next part -------------- /sys/class/dmi/id/bios_date 03/12/2009 /sys/class/dmi/id/bios_vendor Hewlett-Packard /sys/class/dmi/id/bios_version 68PZU Ver. F.0C /sys/class/dmi/id/board_asset_tag /sys/class/dmi/id/board_name 30E8 /sys/class/dmi/id/board_serial /sys/class/dmi/id/board_vendor Hewlett-Packard /sys/class/dmi/id/board_version KBC Version 94.1C /sys/class/dmi/id/chassis_asset_tag CNU9100JZH /sys/class/dmi/id/chassis_serial CNU9100JZH /sys/class/dmi/id/chassis_type 10 /sys/class/dmi/id/chassis_vendor Hewlett-Packard /sys/class/dmi/id/chassis_version /sys/class/dmi/id/modalias dmi:bvnHewlett-Packard:bvr68PZUVer.F.0C:bd03/12/2009:svnHewlett-Packard:pnHPCompaq6730s:pvrF.0C:rvnHewlett-Packard:rn30E8:rvrKBCVersion94.1C:cvnHewlett-Packard:ct10:cvr: /sys/class/dmi/id/power cat: /sys/class/dmi/id/power: Is a directory /sys/class/dmi/id/product_name HP Compaq 6730s /sys/class/dmi/id/product_serial CNU9100JZH /sys/class/dmi/id/product_uuid 4377676E-D605-DE11-B150-80487F19A84A /sys/class/dmi/id/product_version F.0C /sys/class/dmi/id/subsystem cat: /sys/class/dmi/id/subsystem: Is a directory /sys/class/dmi/id/sys_vendor Hewlett-Packard /sys/class/dmi/id/uevent MODALIAS=dmi:bvnHewlett-Packard:bvr68PZUVer.F.0C:bd03/12/2009:svnHewlett-Packard:pnHPCompaq6730s:pvrF.0C:rvnHewlett-Packard:rn30E8:rvrKBCVersion94.1C:cvnHewlett-Packard:ct10:cvr: From pmjdebruijn at pcode.nl Wed Dec 2 17:46:29 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:46:29 +0100 Subject: Idea: GTK/Cario CIE widget as a seperate library In-Reply-To: <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> References: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> Message-ID: <4c2c37f90912020946r1a55d1afi1f7752bae2d80f58@mail.gmail.com> On Wed, Dec 2, 2009 at 6:35 PM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> The GTK/Cairo CIE widget written for GCM is beginning to kick ass... >> However I guess it's GCM specific. > > Not really, it's a standard GObject so it could be reused. Nice. >> Maybe for the long term it's a thought to separate it into a small >> library, so other apps can use it as well? > > Sure, there is loads of other common code that might be useful, but I > don't think the project is at any level of internal ABI stability. > Maybe we can revisit this idea in a few months time? Of course, like I said, it's just a thought. Rushing it won't make it better :) Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 2 19:47:00 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 19:47:00 +0000 Subject: Some tests In-Reply-To: <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> Message-ID: <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > The CIE widget is very nice. Though the profile whitepoint isn't being > displayed yet, I'm assuming this is still on the TODO list. I've just added this to git master. > I also agree it would be nice (but not critical) if double clicking on > the CIE widget would create a dialog with just the profile name and > the CIE widget, for detailed viewing. And dare I ask, possibly with a > right-click "Save (CIE diagram) as PNG/JPEG/PDF" option? Sure, we could do that. I'm tempted to leave that feature until the CIE widget displays more useful data. > Also, I notice the CIE widget has two "squares" of "space" to the left > of the CIE diagram. Does this have a purpose? Otherwise couldn't this > be decreased to a single white square. Sure, we can scale this later. > @Patrice: I rebuild my PPA between 1 to 4 times per two days, > depending on how interesting Richard's changes are... All in all it's > pretty hard to keep up, and he's doing all the hard work :) I'll take that as a compliment. After Monday development will slow quite a bit, as I've got a ton of other stuff to do. Richard. From lars.tore at mulebakken.net Wed Dec 2 20:29:04 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Wed, 2 Dec 2009 21:29:04 +0100 Subject: Some tests In-Reply-To: <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> Message-ID: <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> On Wed, Dec 2, 2009 at 8:47 PM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> The CIE widget is very nice. Though the profile whitepoint isn't being >> displayed yet, I'm assuming this is still on the TODO list. > > I've just added this to git master. > It works for matrix profiles. Lut profiles are already displayed black so it is not possibly to see the tiny cross. See attachment. Lars Tore Gustavsen -------------- next part -------------- A non-text attachment was scrubbed... Name: whitepoint-lut.png Type: image/png Size: 40793 bytes Desc: not available URL: From hughsient at gmail.com Wed Dec 2 20:57:23 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 20:57:23 +0000 Subject: Some tests In-Reply-To: <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> Message-ID: <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> 2009/12/2 Lars Tore Gustavsen : > It works for matrix profiles. Lut profiles are already displayed black > so it is not possibly to see the tiny cross. Sure, we need to add support for LUT profiles as described by Graeme. If you're feeling handy I would love a patch, although I can give this a go next week if nobody starts it before me. Richard. From pmjdebruijn at pcode.nl Wed Dec 2 22:06:57 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 23:06:57 +0100 Subject: Some tests In-Reply-To: <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> Message-ID: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> On Wed, Dec 2, 2009 at 9:57 PM, Richard Hughes wrote: > 2009/12/2 Lars Tore Gustavsen : >> It works for matrix profiles. Lut profiles are already displayed black >> so it is not possibly to see the tiny cross. > > Sure, we need to add support for LUT profiles as described by Graeme. > If you're feeling handy I would love a patch, although I can give this > a go next week if nobody starts it before me. LUT profiles have a normal white point defined, I think Lars ment, it would be nice to invert the color of the whitepoint cross, so it's visible on the black CIE diagram. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 2 22:32:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 23:32:01 +0100 Subject: GCM git fails to build Message-ID: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> Hi, I just tried to build GCM from git on Ubuntu Karmic, and it failed: cc1: warnings being treated as errors gcm-profile.c: In function ?gcm_profile_parse?: gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? from incompatible pointer type /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? but argument is of type ?guint *? make[4]: *** [libgcmshared_a-gcm-profile.o] Error 1 make[4]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager/src' make[3]: *** [all] Error 2 make[3]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager/src' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager' make[1]: *** [all] Error 2 make[1]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager' make: *** [debian/stamp-makefile-build] Error 2 dpkg-buildpackage: error: debian/rules build gave error exit status 2 Regards, Pascal de Bruijn From lars.tore at mulebakken.net Thu Dec 3 06:19:58 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 07:19:58 +0100 Subject: Some tests In-Reply-To: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> Message-ID: <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> On Wed, Dec 2, 2009 at 11:06 PM, Pascal de Bruijn wrote: > LUT profiles have a normal white point defined, I think Lars ment, it > would be nice to invert the color of the whitepoint cross, so it's > visible on the black CIE diagram. > Yep, I meant so. Anyway I like the new tabbed interface. Since we now have the profile size displayed in kb, I think it would be of great benefit to also show the profiles filename. Regards Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 08:03:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 08:03:40 +0000 Subject: GCM git fails to build In-Reply-To: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> Message-ID: <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > cc1: warnings being treated as errors > gcm-profile.c: In function ?gcm_profile_parse?: > gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? > from incompatible pointer type > /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? > but argument is of type ?guint *? Should be fixed in git, thanks. Richard. From hughsient at gmail.com Thu Dec 3 08:04:37 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 08:04:37 +0000 Subject: Some tests In-Reply-To: <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> Message-ID: <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> 2009/12/3 Lars Tore Gustavsen : > Anyway I like the new tabbed interface. Since we now have the profile > size displayed in kb, I think it would be of great benefit to also > show the profiles filename. Filename or filename with path? The latter is quite long. Richard. From lars.tore at mulebakken.net Thu Dec 3 09:12:46 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 10:12:46 +0100 Subject: Some tests In-Reply-To: <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> Message-ID: <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> On Thu, Dec 3, 2009 at 9:04 AM, Richard Hughes wrote: > 2009/12/3 Lars Tore Gustavsen : >> Anyway I like the new tabbed interface. Since we now have the profile >> size displayed in kb, I think it would be of great benefit to also >> show the profiles filename. > > Filename or filename with path? The latter is quite long. > > Richard. > Filename would be nice. I also thinks that full path will be to long for the GUI. Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 10:07:01 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 10:07:01 +0000 Subject: Some tests In-Reply-To: <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> Message-ID: <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> 2009/12/3 Lars Tore Gustavsen : > Filename would be nice. I also thinks that full path will be to long > for the GUI. commit 813c39747c4a0c01a5ee97842e32ab38ea5d71fc Author: Richard Hughes Date: Thu Dec 3 10:06:08 2009 +0000 Add the filename basename to the profile view Richard. From lars.tore at mulebakken.net Thu Dec 3 10:15:16 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 11:15:16 +0100 Subject: Some tests In-Reply-To: <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> Message-ID: <58497f010912030215n7a5e9ed6i8234bce4b3d54a6b@mail.gmail.com> On Thu, Dec 3, 2009 at 11:07 AM, Richard Hughes wrote: > 2009/12/3 Lars Tore Gustavsen : >> Filename would be nice. I also thinks that full path will be to long >> for the GUI. > > commit 813c39747c4a0c01a5ee97842e32ab38ea5d71fc > Author: Richard Hughes > Date: ? Thu Dec 3 10:06:08 2009 +0000 > > ? ?Add the filename basename to the profile view > > Richard. > Excellent, and it works. Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 11:42:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 11:42:49 +0000 Subject: Some tests In-Reply-To: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> Message-ID: <15e53e180912030342q2bde3d13ncb616f24c7a3617f@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > LUT profiles have a normal white point defined, I think Lars ment, it > would be nice to invert the color of the whitepoint cross, so it's > visible on the black CIE diagram. commit 6c77c29086ab851bec34c5667d0963bb32ec9217 Author: Richard Hughes Date: Thu Dec 3 11:41:21 2009 +0000 Show the white point cross in white if the CIE chart is all black Richard. From pmjdebruijn at pcode.nl Thu Dec 3 18:31:29 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 19:31:29 +0100 Subject: GCM git fails to build In-Reply-To: <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> Message-ID: <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> On Thu, Dec 3, 2009 at 9:03 AM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> cc1: warnings being treated as errors >> gcm-profile.c: In function ?gcm_profile_parse?: >> gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? >> from incompatible pointer type >> /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? >> but argument is of type ?guint *? > > Should be fixed in git, thanks. Indeed, it's fixed. Some minor trivia: it seems the above only broke on amd64, not on i386... Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 3 18:34:48 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 19:34:48 +0100 Subject: Camera input profiling make/model handling Message-ID: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> Hi, I haven't had time yet to properly check this, but what is used by GCM for make/model info when profiling cameras/scanners? At least for camera's it seems wise to use Make/Model from EXIF. This way, software like UFRaw and DarkTable (and others) can use the EXIF make/model to match which profiles should be valid for a certain RAW file... Luckily EXIF is much better human readable (and generally less ambiguous) than EDID... Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 3 19:02:31 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:02:31 +0000 Subject: GCM git fails to build In-Reply-To: <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> Message-ID: <15e53e180912031102w2c2e16dcmadc15da0983fb24c@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > Some minor trivia: it seems the above only broke on amd64, not on i386... Right, gsize is a different memory size to guint on 64bit architectures. Richard. From hughsient at gmail.com Thu Dec 3 19:09:22 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:09:22 +0000 Subject: Camera input profiling make/model handling In-Reply-To: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> References: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> Message-ID: <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > I haven't had time yet to properly check this, but what is used by GCM > for make/model info when profiling cameras/scanners? Information from sysfs. > At least for camera's it seems wise to use Make/Model from EXIF. This > way, software like UFRaw and DarkTable (and others) can use the EXIF > make/model to match which profiles should be valid for a certain RAW > file... GCM doesn't treat the device like a storage device, it treats it as a USB endpoint device, which means we don't look at the files contained within. > Luckily EXIF is much better human readable (and generally less > ambiguous) than EDID... I think this is the wrong layer. What happens if the EXIF data for one file on the "device" has different colorspaces or even two different device models? The device can carry all number of different EXIF files, but the device can only have one profile. If you inserted a pendrive of photos, would you expect to assign a per-pendrive ICC profile? I think it's sane for the end application to open the file with EXIF metadata and then decided what to do. If it wants to just use the embedded profile, that's fine, or it can ask gcm for the default device profile from the volume (which it can get from udev). Richard. From info at justinsseasonaltreasures.com Thu Dec 3 18:20:37 2009 From: info at justinsseasonaltreasures.com (Unknown) Date: Thu, 03 Dec 2009 10:20:37 -0800 Subject: Support for B9100 Message-ID: <1259864437.2325.1.camel@localhost> Will there be support for color management of the B9100? Thanks, Jazbo From hughsient at gmail.com Thu Dec 3 19:30:15 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:30:15 +0000 Subject: Support for B9100 In-Reply-To: <1259864437.2325.1.camel@localhost> References: <1259864437.2325.1.camel@localhost> Message-ID: <15e53e180912031130k38517da1j65cc586154c34625@mail.gmail.com> 2009/12/3 Unknown : > Will there be support for color management of the B9100? I assume you mean the HP printer. At the moment ArgyllCMS has some support for generating print profiles, but CUPS does not have a way to set the ICC profile for the device. This needs much more work (for the session CUPS process to talk to GCM and get the current profile for the device) before it will be a case of plug and play. Apple did a lot of the CUPS work, but we need to flesh that out for Linux. For using hplip it's a bit different, and we probably need to support that too. Either way, more help is really welcome. If you just want to assign a profile to a device, and then select that in an application, that should work now. Richard. From pmjdebruijn at pcode.nl Thu Dec 3 20:54:23 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 21:54:23 +0100 Subject: Camera input profiling make/model handling In-Reply-To: <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> References: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> Message-ID: <4c2c37f90912031254i1dc63397g70e35c11a902f5d2@mail.gmail.com> On Thu, Dec 3, 2009 at 8:09 PM, Richard Hughes wrote: > 2009/12/3 Pascal de Bruijn : >> I haven't had time yet to properly check this, but what is used by GCM >> for make/model info when profiling cameras/scanners? > > Information from sysfs. Maybe I'll test this over the weekend... >> At least for camera's it seems wise to use Make/Model from EXIF. This >> way, software like UFRaw and DarkTable (and others) can use the EXIF >> make/model to match which profiles should be valid for a certain RAW >> file... > > GCM doesn't treat the device like a storage device, it treats it as a > USB endpoint device, which means we don't look at the files contained > within. > >> Luckily EXIF is much better human readable (and generally less >> ambiguous) than EDID... > > I think this is the wrong layer. What happens if the EXIF data for one > file on the "device" has different colorspaces or even two different > device models? The device can carry all number of different EXIF > files, but the device can only have one profile. If you inserted a > pendrive of photos, would you expect to assign a per-pendrive ICC > profile? Huh? A "Device" does not really have different color spaces... a camera's sensor (RAW file) has a native color space characterized by the profiling of the IT8 target (or another target, which is not really relevant, it's a means to an end). Any camera that produces sRGB or AdobeRGB, is likely to be not fit to be really profiled. Since a camera most likely does not statically convert sensor data to sRGB/AdobeRGB, it's likely a dynamic (image dependant) process. Anyway I filed a feature request with UFRaw, for "special" profiling mode, to convert a camera RAW file an image with linear gamma in the native camera color space: http://sourceforge.net/tracker/?func=detail&aid=2841222&group_id=127649&atid=709089 If one would have a pen drive of photos, the pen drive would be completely and utterly irrelevant from a profiling/profile assignment perspective. Each and every file on the drive could require a different profile (at least for proper results). And normally EXIF could be reasonably reliably used to "match" that. > I think it's sane for the end application to open the file with EXIF > metadata and then decided what to do. If it wants to just use the > embedded profile, that's fine, or it can ask gcm for the default > device profile from the volume (which it can get from udev). Again, with images, only the image color space is truely relevant (embedded or not). However, RAW files do not have assigned (embedded) profiles. UFRaw/DCRaw "solves" this by ripping generic profiles (Color Matrix) from Adobe DNG SDK, and applying them on a by EXIF make/model basis. Please do note, the above applies to camera's. I'm not really sure about scanners. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 3 22:23:48 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 23:23:48 +0100 Subject: ICC operating system identifiers Message-ID: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Hi, Currently there is no registered operating system identifier for Linux in general. Some examples: Microsoft 'msft' Apple 'appl' Sun Microsystems 'sunw' In the past I tried to mail the Linux Foundation about this, since they seem to be the best (least unfit) umbrella organisation to register such an identifier with the ICC. For example: Linux Foundation 'lnxf' However, it seems the mail was ignored. Richard, could you possibly attempt this, since you can mail from a gnome.org e-mail adres, the request might be more seriously considered. Or do you have other thoughts on how to orchestrate this? Regards, Pascal de Bruijn From anders at brander.dk Thu Dec 3 22:32:11 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 03 Dec 2009 23:32:11 +0100 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <1259879531.25715.4.camel@video64> Hi, On Thu, 2009-12-03 at 23:23 +0100, Pascal de Bruijn wrote: > Currently there is no registered operating system identifier for Linux > in general. Is this relevant in this day and age? Is there any CMM/CMS that does anything useful with this information? Regards, Anders Brander From pmjdebruijn at pcode.nl Thu Dec 3 22:46:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 23:46:01 +0100 Subject: ICC operating system identifiers In-Reply-To: <1259879531.25715.4.camel@video64> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> <1259879531.25715.4.camel@video64> Message-ID: <4c2c37f90912031446p58a57498i6cf869fb254962a1@mail.gmail.com> On Thu, Dec 3, 2009 at 11:32 PM, Anders Brander wrote: > Hi, > > On Thu, 2009-12-03 at 23:23 +0100, Pascal de Bruijn wrote: >> Currently there is no registered operating system identifier for Linux >> in general. > > Is this relevant in this day and age? Is there any CMM/CMS that does > anything useful with this information? I think this remains relevant until the ICC officially deprecates the fields... Regards, Pascal de Bruijn From graeme2 at argyllcms.com Fri Dec 4 00:03:34 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 04 Dec 2009 11:03:34 +1100 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <4B1851D6.8080801@argyllcms.com> Pascal de Bruijn wrote: > For example: > Linux Foundation 'lnxf' Hi, note that Argyll uses '*nix' as an unofficial platform signature for Linux and other Unix like platforms. See also this thread on the OpenICC list: Graeme Gill. From hughsient at gmail.com Fri Dec 4 09:22:59 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 09:22:59 +0000 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > Richard, could you possibly attempt this, since you can mail from a > gnome.org e-mail adres, the request might be more seriously > considered. I would suspect my @redhat email account might have more gravitas, although I'm not sure what it would achieve. Do programs actually do anything differently depending on the OSI information, or even display it at any point? If you think it may be useful, then I guess "*nix" makes most sense as Graeme has been using this for ages. If anything, Graeme's email would convey more weight than mine ever would. Richard. From graeme2 at argyllcms.com Fri Dec 4 11:20:47 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 04 Dec 2009 22:20:47 +1100 Subject: ICC operating system identifiers In-Reply-To: <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> Message-ID: <4B18F08F.3080306@argyllcms.com> Richard Hughes wrote: > If you think it may be useful, then I guess "*nix" makes most sense as > Graeme has been using this for ages. If anything, Graeme's email would > convey more weight than mine ever would. The issues are the following: 1) The platform signature is not of great importance, and it's difficult to know what the original intention of it was. ICC profiles are typically not platform specific. It's lack of importance means that it might be difficult to excite anybodies interest in this issue. The only reason I added '*nix' to my code was an interest in not being misleading. My other alternatives were to use one of the other two Unix based platform signatures (SGI or Sun), or to set it to Unknown (0). 2) Most of the other ICC signatures are in registries, so it's not that bigger deal to ask someone to add a signature to a registry - a single person will have the authority to just do it (hence 'argl' added as a CMM signature). The platform signatures only appear in the ICC spec. though, and they also seem to correlate with the founding members (Although I notice that Taligent has been dropped between V2 an V4). So to add a new signature means someone within the ICC has to propose a spec. change and it then has to be approved by the members at some point ! 3) I'm not sure how much consensus there is about a signature that would cover Linux. My thought process was that the existing signatures were both too vague and too specific. They are really operating system vendors, not specific operating system platforms, and so are fairly loose (there is a noticeable difference between Win3.1 and WinNT, between OS9 and OS X). They are also too specific, in that Sun and SGI were both Unix based systems, although SGI has switch to Windows, so what does it mean ? If the signatures were just a registry, there wouldn't be any problem being very specific (Linux, FreeBSD, OpenBSD, RedHatLinux, etc. etc.) So I went for something that would cover cover Linux, BSD etc. and would be enough to distinguish it from Apple and Microsoft. As I suggested on the OpenICC list, if there is sufficient consensus then I'm happy to mail a couple of people and request that a new platform signature be put forward as a modification to the ICC spec., but on the other hand I won't be surprised if action on it is a little slow :-) cheers, Graeme Gill. From lars.tore at mulebakken.net Fri Dec 4 11:53:06 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 4 Dec 2009 12:53:06 +0100 Subject: More verbose output in calibration? Message-ID: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> May I suggest that when I calibrate my monitor the details view are more verbose than now (like dispcal -v). I guess a beginner will think this is not working at all, when it measures all the black patches and almost noting change on the display. I think more experienced users also like to see the progress in the calibration and the deltaE report at the end. In a longer time frame some progress bar (1) or progress window (2) would be very nice eye candy in the calibration process. 1 http://developer.gnome.org/projects/gup/hig/2.0/controls-progress-bars.html 2 http://developer.gnome.org/projects/gup/hig/2.0/windows-progress.html Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 4 14:51:47 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 15:51:47 +0100 Subject: Bug: Profiles tab is empty as long as no profiles have been assigned. Message-ID: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> Hi, I think I found a small bug. When starting gcm-prefs after a clean install, it does list all the available profiles in the profiles dropdown box on the Devices tab. But the list on the Profiles tab is completely empty. When I restart gcm-prefs the list is still empty. However, when I assign a profile (for example blueish) to the screen, it seems to change nothing at first glance. However after restarting gcm-prefs again, suddenly the dialog is bigger, and the list on the Profiles tabs has been filled. I haven't been able to test this, but when a new profile is created, does this get properly added the the list on the Profiles tab as well? Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Fri Dec 4 14:58:25 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 15:58:25 +0100 Subject: gcm-prefs on netbook Message-ID: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> Hi, I managed to test-drive gcm-prefs on a friends netbook (which runs Ubuntu Karmic Netbook Edition). And when the list on the Profiles tab is filled, the dialog quickly grows to big. I've attached a screenshot. I personally don't particularly like netbooks, but they are popular, and thus I think it might be nice for gcm-prefs to work well on netbooks too. I'm quite aware netbooks aren't particularly well suited for "graphics-work", but that's not really the point here. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: gcm-prefs-netbook.png Type: image/png Size: 83037 bytes Desc: not available URL: From hughsient at gmail.com Fri Dec 4 15:00:38 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:00:38 +0000 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> Message-ID: <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > I personally don't particularly like netbooks, but they are popular, > and thus I think it might be nice for gcm-prefs to work well on > netbooks too. What about if the window size < 1024x768 then make the cie widget smaller? Richard. From hughsient at gmail.com Fri Dec 4 15:18:18 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:18:18 +0000 Subject: Bug: Profiles tab is empty as long as no profiles have been assigned. In-Reply-To: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> References: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> Message-ID: <15e53e180912040718m678446b7s3dd3797494ac05e9@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > I think I found a small bug. When starting gcm-prefs after a clean > install, it does list all the available profiles in the profiles > dropdown box on the Devices tab. But the list on the Profiles tab is > completely empty. Ahh, thanks. commit 902fbf2f4271d6be79985d40338fcf4b35bbc2a1 Author: Richard Hughes Date: Fri Dec 4 15:16:37 2009 +0000 Don't skip phase2 startup if we can't load the config file :100644 100644 2726f42... 792b729... M src/gcm-prefs.c Richard. From hughsient at gmail.com Fri Dec 4 15:18:39 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:18:39 +0000 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> Message-ID: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> 2009/12/4 Richard Hughes : > What about if the window size < 1024x768 then make the cie widget smaller? commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 Author: Richard Hughes Date: Fri Dec 4 15:11:19 2009 +0000 Reduce the size of the ICC widget on small displays :100644 100644 4a1551d... 2726f42... M src/gcm-prefs.c Richard. From pmjdebruijn at pcode.nl Fri Dec 4 15:49:27 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 16:49:27 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <4c2c37f90912040749m6c8cdea1vb5c4aa14b43555@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays > > :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c Excellent. The CIE widget isn't the most import thing ever, however nice it may be. The user chose to have a small (low resolution) screen, so having the buttons visible is much more important. Regards, Pascal de Bruijn From lars.tore at mulebakken.net Fri Dec 4 17:16:10 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 4 Dec 2009 18:16:10 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays > > :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c > > Richard. On my 1280x1024 monitor the CIE diagram is quite small now. I can live with it, but I'm not sure if 1280x1024 was meant to be a small display? Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 4 17:18:10 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 18:18:10 +0100 Subject: gcm-prefs on netbook In-Reply-To: <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> Message-ID: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> On Fri, Dec 4, 2009 at 6:16 PM, Lars Tore Gustavsen wrote: > On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: >> 2009/12/4 Richard Hughes : >>> What about if the window size < 1024x768 then make the cie widget smaller? >> >> commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 >> Author: Richard Hughes >> Date: ? Fri Dec 4 15:11:19 2009 +0000 >> >> ? ?Reduce the size of the ICC widget on small displays >> >> :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c >> >> Richard. > > On my 1280x1024 monitor the CIE diagram is quite small now. I can live > with it, but I'm not sure if 1280x1024 was meant to be a small > display? That wasn't intentional: + if (gdk_screen_get_width (screen) < 10240 || That's a typo :) Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Fri Dec 4 17:24:17 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 4 Dec 2009 20:24:17 +0300 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> Message-ID: <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> On Fri, Dec 4, 2009 at 8:18 PM, Pascal de Bruijn wrote: >> On my 1280x1024 monitor the CIE diagram is quite small now. I can live >> with it, but I'm not sure if 1280x1024 was meant to be a small >> display? > > That wasn't intentional: So rectangular shape is? :) http://img37.imagefra.me/img/img37/1/12/4/prokoudine/f_1m_444c758.png Alexandre From hughsient at gmail.com Fri Dec 4 17:44:41 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 17:44:41 +0000 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> Message-ID: <15e53e180912040944k21f00207qa486932397da75eb@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > + if (gdk_screen_get_width (screen) < 10240 || > > That's a typo :) Fixed, thanks. Richard. From hughsient at gmail.com Fri Dec 4 17:49:13 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 17:49:13 +0000 Subject: gcm-prefs on netbook In-Reply-To: <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> Message-ID: <15e53e180912040949p144a83b8gce479add4b1a829@mail.gmail.com> 2009/12/4 Alexandre Prokoudine : > So rectangular shape is? :) Wrong! commit e818364b44830c6178a2fcbd81a255f47ad01a6d Author: Richard Hughes Date: Fri Dec 4 17:47:31 2009 +0000 Ensure the CIE widget is square Richard. From pmjdebruijn at pcode.nl Sat Dec 5 12:29:07 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 13:29:07 +0100 Subject: Devices list sorting Message-ID: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> Hi, Since I'm _guessing_ GCM will most commonly be used to calibrate displays, it might be a thought to sort on device "class" first, and then alphabetically? Possibly displaying displays at the top of the list? I'm guessing this order makes most sense (to me): - Displays - Printers - Scanners - Camera's - Video Camera's Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Sat Dec 5 12:30:56 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 13:30:56 +0100 Subject: Feature: Nautilus Thumbnailer for profiles Message-ID: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> Hi, Another idea for the long term: Wouldn't it severely kick ass, if Nautilus were to thumbnail ICC profiles using the CIE widget? Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Sat Dec 5 13:11:25 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sat, 5 Dec 2009 16:11:25 +0300 Subject: Feature: Nautilus Thumbnailer for profiles In-Reply-To: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> References: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> Message-ID: <733f2c730912050511s6634231qf84b50fb000b148b@mail.gmail.com> On 12/5/09, Pascal de Bruijn wrote: > Another idea for the long term: > > Wouldn't it severely kick ass, if Nautilus were to thumbnail ICC > profiles using the CIE widget? +2^32 :) Alexandre From hughsient at gmail.com Sat Dec 5 15:52:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 5 Dec 2009 15:52:24 +0000 Subject: Devices list sorting In-Reply-To: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> Message-ID: <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> 2009/12/5 Pascal de Bruijn : > Since I'm _guessing_ GCM will most commonly be used to calibrate > displays, it might be a thought to sort on device "class" first, and > then alphabetically? The profile title or filename as the secondary sort? > Possibly displaying displays at the top of the list? I'm guessing this > order makes most sense (to me): > > - Displays > - Printers > - Scanners > - Camera's > - Video Camera's I'm not sure it matters, but sure, it would be better if they were sorted. Richard. From pmjdebruijn at pcode.nl Sat Dec 5 15:54:18 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 16:54:18 +0100 Subject: Devices list sorting In-Reply-To: <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> Message-ID: <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> On Sat, Dec 5, 2009 at 4:52 PM, Richard Hughes wrote: > 2009/12/5 Pascal de Bruijn : >> Since I'm _guessing_ GCM will most commonly be used to calibrate >> displays, it might be a thought to sort on device "class" first, and >> then alphabetically? > > The profile title or filename as the secondary sort? I'd use the profile description as secondary sort... >> Possibly displaying displays at the top of the list? I'm guessing this >> order makes most sense (to me): >> >> - Displays >> - Printers >> - Scanners >> - Camera's >> - Video Camera's > > I'm not sure it matters, but sure, it would be better if they were sorted. It's a convenience thing... For example, since video4linux support has been commited, on my laptop I get my Chicony webcam as the first entry, and below that my display... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Sun Dec 6 09:31:43 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:31:43 +0300 Subject: gcm-inspect segfaults Message-ID: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> (gdb) run -d Starting program: /usr/bin/gcm-inspect -d [Thread debugging using libthread_db enabled] Output profile 'VGA1': not set Output profile 'HDMI1': not set Output profile 'DP1': not set Output profile 'HDMI2': not set Output profile 'DP2': not set Program received signal SIGSEGV, Segmentation fault. 0x00007ffff3ffba52 in g_atomic_int_exchange_and_add () from /lib/libglib-2.0.so.0 (gdb) bt #0 0x00007ffff3ffba52 in g_atomic_int_exchange_and_add () from /lib/libglib-2.0.so.0 #1 0x00007ffff401162e in g_hash_table_unref () from /lib/libglib-2.0.so.0 #2 0x0000000000404ee7 in gcm_inspect_get_properties (argc=1, argv=0x7fffffffe3d8) at gcm-inspect.c:371 #3 main (argc=1, argv=0x7fffffffe3d8) at gcm-inspect.c:438 Should I be concerned about that sort of thing? :) Alexandre From alexandre.prokoudine at gmail.com Sun Dec 6 09:45:02 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:45:02 +0300 Subject: Yelp isn't happy Message-ID: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> Hi, When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI ?ghelp:gnome-color-manager?preferences? is incorrect". Alexandre From alexandre.prokoudine at gmail.com Sun Dec 6 09:51:56 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:51:56 +0300 Subject: yet another PO comment bug Message-ID: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> Hi, A comment for "Setting up device" says "TRANSLATORS: title, a profile is a ICC file" Alexandre From hughsient at gmail.com Sun Dec 6 10:09:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 10:09:49 +0000 Subject: gcm-inspect segfaults In-Reply-To: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> References: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> Message-ID: <15e53e180912060209i151c479gd985e4d60e4a829e@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > #1 ?0x00007ffff401162e in g_hash_table_unref () from /lib/libglib-2.0.so.0 Fixed in git, thanks. Richard. From hughsient at gmail.com Sun Dec 6 10:16:29 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 10:16:29 +0000 Subject: yet another PO comment bug In-Reply-To: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> References: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> Message-ID: <15e53e180912060216y4b0c3daeh2c850b6b15b8ce35@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > A comment for "Setting up device" says "TRANSLATORS: title, a profile > is a ICC file" Fixed, thanks. Richard. From pmjdebruijn at pcode.nl Sun Dec 6 11:39:55 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 6 Dec 2009 12:39:55 +0100 Subject: CIE widget drawing odd graphs Message-ID: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> Hi, I might have uncovered some minor issues with the CIE widget drawing code. I've attached an archive containing several non-standard profiles, which are drawn quite oddly. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: weirdciewidget.zip Type: application/zip Size: 31374 bytes Desc: not available URL: From pmjdebruijn at pcode.nl Sun Dec 6 12:33:52 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 6 Dec 2009 13:33:52 +0100 Subject: Icon Sizes Message-ID: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> Hi, I managed to test GCM before on a netbook with Ubuntu Karmic Netbook edition installed. I noticed somehow the GCM icon was very small in comparison to all the other application icons... While I'm not entirely sure about this, I think the Ubuntu Netbook edition uses the 64x64 icons (and does not fall back to scalable to save CPU cycles on low end CPUs), since GCM does not supply a 64x64 icon, it falls back to a smaller one. Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? Regards, Pascal de Bruijn From lars.tore at mulebakken.net Sun Dec 6 16:04:46 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Sun, 6 Dec 2009 17:04:46 +0100 Subject: Bluish.icc In-Reply-To: <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> References: <58497f010911110351o3d7df74dt8ac35ef7c8cb5e4@mail.gmail.com> <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> Message-ID: <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> On Sun, Dec 6, 2009 at 3:32 PM, Pascal de Bruijn wrote: > On Wed, Nov 11, 2009 at 12:51 PM, Lars Tore Gustavsen > wrote: >> The icc profile include with gnome-color-manager is very nice to test >> if everything works. However it advertise itself as "sRGB >> IEC61966-2-1" >> >> I have edited the attached profile and it now reads "bluish test" in gcm-prefs. > > Would it be possible for you to rename it to "Blueish Test" instead of > "blueish test"? > > Not extremely important, but it does look a bit better :) > > Regards, > Pascal de Bruijn > Here we go: It's done with Andrew Shepherd's ICC Profile Toolkit http://www.tlbtlb.com/links/ Regards Lars Tore Gustavsen -------------- next part -------------- A non-text attachment was scrubbed... Name: bluish.icc Type: application/x-icc Size: 3966 bytes Desc: not available URL: From hughsient at gmail.com Sun Dec 6 19:37:09 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 19:37:09 +0000 Subject: Bluish.icc In-Reply-To: <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> References: <58497f010911110351o3d7df74dt8ac35ef7c8cb5e4@mail.gmail.com> <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> Message-ID: <15e53e180912061137h57d3f969m2a57af3568e547e7@mail.gmail.com> 2009/12/6 Lars Tore Gustavsen : > Here we go: Applied, thanks. Richard From hughsient at gmail.com Sun Dec 6 19:46:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 19:46:40 +0000 Subject: Icon Sizes In-Reply-To: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> References: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> Message-ID: <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> 2009/12/6 Pascal de Bruijn : > Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? commit 29b66375ed40e10bfffba601eeca9f25bd8a46b7 Author: Richard Hughes Date: Sun Dec 6 19:45:42 2009 +0000 Add a 64x64 pixel application icon Richard. From hughsient at gmail.com Sun Dec 6 20:06:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 20:06:24 +0000 Subject: Devices list sorting In-Reply-To: <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> Message-ID: <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> 2009/12/5 Pascal de Bruijn : > For example, since video4linux support has been commited, on my laptop > I get my Chicony webcam as the first entry, and below that my > display... commit dfe5fc80cef25cac2deddb5687719ab298e786e0 Author: Richard Hughes Date: Sun Dec 6 20:05:18 2009 +0000 Make the list orders predictable by setting a sort string Richard. From hughsient at gmail.com Sun Dec 6 20:19:37 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 20:19:37 +0000 Subject: CIE widget drawing odd graphs In-Reply-To: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> Message-ID: <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> 2009/12/6 Pascal de Bruijn : > I might have uncovered some minor issues with the CIE widget drawing code. They look pretty major to me! > I've attached an archive containing several non-standard profiles, > which are drawn quite oddly. After the release I would like to look at the widget drawing code and the profile extraction code (there are still quite a lot of FIXME's and unimplemented stuff) and add a boat load more self tests and internal validations. I fear I might not have time in the next couple of weeks (lots of other work to do, which I've been putting off for ages) so if anyone is familiar with XYZ co-ordinates and ICC profiles I would appreciate any code review and possible patches. Thanks, Richard. From pmjdebruijn at pcode.nl Sun Dec 6 23:35:43 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 00:35:43 +0100 Subject: Devices list sorting In-Reply-To: <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> Message-ID: <4c2c37f90912061535n22c07324s60f6a5edbc7e5147@mail.gmail.com> On Sun, Dec 6, 2009 at 9:06 PM, Richard Hughes wrote: > 2009/12/5 Pascal de Bruijn : >> For example, since video4linux support has been commited, on my laptop >> I get my Chicony webcam as the first entry, and below that my >> display... > > commit dfe5fc80cef25cac2deddb5687719ab298e786e0 > Author: Richard Hughes > Date: ? Sun Dec 6 20:05:18 2009 +0000 > > ? ?Make the list orders predictable by setting a sort string Seems to work excellently here... Thanks, Pascal de Bruijn From hughsient at gmail.com Mon Dec 7 12:03:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 7 Dec 2009 12:03:28 +0000 Subject: gnome-color-manager 2.29.1 Message-ID: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> gnome-color-manager is a session program that makes it easy to manage, install and generate color profiles in the GNOME desktop. Version 2.29.1 ~~~~~~~~~~~~~~ Released: 2009-12-07 * Translations - Add Russian translation (Alexandre Prokoudine) - Add British English translation (Bruce Cowan) - Add Indonesian translation (Andika Triwidada) - Add French translation (Claude Paroz) - Add Swedish translation (Daniel Nylander) - Add Brazilian Portuguese translation (Flamarion Jorge) - Add Lithuanian translation (Gintautas Miliauskas) - Add German translation (Hendrik Brandt) - Add Danish translation (Joe Hansen) - Add Spanish translation (Jorge Gonz?lez) - Add Norwegian bokm?l translation (Kjartan Maraas) - Add Thai translation (Phondanai Khanti) - Add Polish translation (Piotr Dr?g) - Add Estonian translation (Priit Laes, Mattias P?ldaru) - Add Tamil translation (vasudeven) - Add Russian user guide translation (Alexandre Prokoudine) - Add Indonesian user guide translation (Andika Triwidada) * New Features: - Add gcm-import, a helper to allow double clicking on ICC profiles to import them (Richard Hughes) - Add gcm-prefs, a utility to assign profiles to devices, examine profiles, and set session-wide defaults (Richard Hughes) - Add gcm-session, a dbus-activated session daemon for applications to get the profiles for a device, or device class and to get session-wide defaults. It exits when no longer used to save resources. (Richard Hughes) - Add gcm-self-test, a self test framework that tests GCM functionality (Richard Hughes) - Add gcm-apply, a simple utility to just set (or reset) display profiles (Richard Hughes) - Add gcm-inspect, a debugging utility to inspect the profiles set in the session (Richard Hughes) - Add gcm-dump-edid, a utility to dump the EDID to disk for debugging (Richard Hughes) - Add gcm-dump-profile, a utility to dump the ICC profile to the screen (Richard Hughes) - Add some simple man pages and help document (Richard Hughes) - Add ArgyllCMS support to generate device profiles (Richard Hughes) - Add color calibration hardware auto-detection (Richard Hughes) - Add code to set the _ICC_PROFILE atom per-output and also per-screen (Richard Hughes) - Add some pre-calibration steps for external displays (Pascal de Bruijn) - Add hardware support for gphoto supported cameras (Richard Hughes) - Add hardware support for SANE suppoerted scanners (Richard Hughes) - Add hardware support for video4linux supported video devices (Richard Hughes) - Add hardware support for XRandR supported displays (Richard Hughes) - Add hardware support for hplip supported printers (Richard Hughes) - Add CIE widget to display visual data about different profiles (Richard Hughes) - Use the system DMI data to better itentify internal LCD panels (Richard Hughes) - Parse the EDID to get a better device description for displays (Richard Hughes) - Make the list orders predictable by setting a sort string (Richard Hughes) * Bugfix: - Use XDG directory to store data (Baptiste Mille-Mathias) - Remove markup from GTKBuilder translatable strings (Claude Paroz) - Update bluish.icc title (Lars Tore Gustavsen, Pascal de Bruijn) - Enable adding xrandr devices with no EDID (Martin Szulecki) - Avoid reporting a (false) failure on first import (Stephane Delcroix) - Fix the message-received cb signature (Stephane Delcroix) - Fix up numerous small bugs prior to first release (Richard Hughes) - Look for the debian-named argyllcms binaries first (Richard Hughes) - Set the brightness to 100% on internal LCD panels before we generate a output profile (Richard Hughes) Richard. From pmjdebruijn at pcode.nl Mon Dec 7 12:37:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 13:37:01 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> Message-ID: <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> On Mon, Dec 7, 2009 at 1:03 PM, Richard Hughes wrote: > gnome-color-manager is a session program that makes it easy to manage, install > and generate color profiles in the GNOME desktop. > > Version 2.29.1 > ~~~~~~~~~~~~~~ > Released: 2009-12-07 Excellent :) Packages are available from my PPA as usual. However, since I'll be tracking git again in my main repo, I've decided to start a gcm-release PPA: https://launchpad.net/~pmjdebruijn/+archive/gcm-release I've just submitted the builds, so they'll be ready in an hour or six... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Mon Dec 7 15:08:18 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 7 Dec 2009 18:08:18 +0300 Subject: metadata Message-ID: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Hi, It looks like there are problems writing unicode text into icc metadata or reading from it. http://img39.imagefra.me/img/img39/1/12/7/prokoudine/f_0m_948cb9f.png Exhibit 1 is description of the profile that is obviously RU translation of "Custom" Exhibit 2 is copyright field that seems to be suffering from more or less same thing. Any ideas? Alexandre From pmjdebruijn at pcode.nl Mon Dec 7 15:14:42 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 16:14:42 +0100 Subject: metadata In-Reply-To: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Message-ID: <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> On Mon, Dec 7, 2009 at 4:08 PM, Alexandre Prokoudine wrote: > Hi, > > It looks like there are problems writing unicode text into icc > metadata or reading from it. > > http://img39.imagefra.me/img/img39/1/12/7/prokoudine/f_0m_948cb9f.png > > Exhibit 1 is description of the profile that is obviously RU > translation of "Custom" > > Exhibit 2 is copyright field that seems to be suffering from more or > less same thing. This actually begs a bigger question: Should we even try to internationalize metadata contents? I'd prefer not... It would make exchanging ICC profiles internationally harder... Say, one day, we make a big repository of ICC profiles which are ready to go for people who don't have a colorimeter. I'd be pretty crappy for someone in Germany to see Russian in the metadata? I'd prefer to stick to English... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Mon Dec 7 15:47:16 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 7 Dec 2009 18:47:16 +0300 Subject: metadata In-Reply-To: <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> Message-ID: <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> On Mon, Dec 7, 2009 at 6:14 PM, Pascal de Bruijn wrote: > This actually begs a bigger question: Should we even try to > internationalize metadata contents? > > I'd prefer not... It would make exchanging ICC profiles > internationally harder... Quite so :) > Say, one day, we make a big repository of ICC profiles which are ready > to go for people who don't have a colorimeter. Are we talking about profiles for DSLRs? :) Alexandre From hughsient at gmail.com Mon Dec 7 17:29:00 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 7 Dec 2009 17:29:00 +0000 Subject: gnome-color-manager 2.29.1 In-Reply-To: <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> Message-ID: <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> 2009/12/7 Pascal de Bruijn : > I've just submitted the builds, so they'll be ready in an hour or six... I've just started an official build for Fedora 12: http://koji.fedoraproject.org/koji/taskinfo?taskID=1860405 Richard. From pmjdebruijn at pcode.nl Mon Dec 7 19:21:36 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 20:21:36 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> Message-ID: <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> On Mon, Dec 7, 2009 at 6:29 PM, Richard Hughes wrote: > 2009/12/7 Pascal de Bruijn : >> I've just submitted the builds, so they'll be ready in an hour or six... > > I've just started an official build for Fedora 12: > http://koji.fedoraproject.org/koji/taskinfo?taskID=1860405 Richard, my mail to the mailing list got rejected: "Message rejected by filter rule match" Regards, Pascal de Bruijn From hughsient at gmail.com Tue Dec 8 10:15:58 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 10:15:58 +0000 Subject: gnome-color-manager 2.29.1 In-Reply-To: <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> Message-ID: <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> 2009/12/7 Pascal de Bruijn : > Richard, my mail to the mailing list got rejected: I think gnome-announce-list at gnome.org bounced the mail (correctly) -- dropping the cc should make things work again. Richard. From hughsient at gmail.com Tue Dec 8 11:02:50 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 11:02:50 +0000 Subject: metadata In-Reply-To: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Message-ID: <15e53e180912080302x2f413450v3b99d884d709ed57@mail.gmail.com> 2009/12/7 Alexandre Prokoudine : > Exhibit 1 is description of the profile that is obviously RU > translation of "Custom" Right, I made the Custom localized without thinking that it would involve non-ascii chars. I've fixed that in cfebe91aef33254709bc22ca547563044f1fbdd8. > Exhibit 2 is copyright field that seems to be suffering from more or > less same thing. > > Any ideas? Well, for textType, the profile spec mandates 7 bit ASCII, so we're wrong to translate Custom. For the unicodeTextType we should probably decode UTF-16 properly, although I've not seen a profile that uses this type in the wild yet. Note: if you've got a profile that breaks GCM in one way or another (and it's freely redistributable, which excludes the ICC and Adobe profiles..) then we should probably check it into GCM and wire it into the self test code. If nothing else you can use gcm-dump-profile -v thefilename.icc and send me the output. Richard. From hughsient at gmail.com Tue Dec 8 11:05:57 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 11:05:57 +0000 Subject: metadata In-Reply-To: <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> Message-ID: <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> 2009/12/7 Alexandre Prokoudine : >> Say, one day, we make a big repository of ICC profiles which are ready >> to go for people who don't have a colorimeter. > > Are we talking about profiles for DSLRs? :) Just in case, I've got a repo waiting: http://github.com/hughsie/gnome-color-profiles I'm going to be checking in the Adobe and ICC profiles first, but alas, they can't be installed by default in distros like Fedora as they are non free [1]. Richard. [1] https://www.redhat.com/archives/fedora-legal-list/2009-December/msg00009.html and https://www.redhat.com/archives/fedora-legal-list/2009-December/msg00010.html From pmjdebruijn at pcode.nl Tue Dec 8 11:16:35 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 12:16:35 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> Message-ID: <4c2c37f90912080316w42894a61m7ad18ee019eba3dd@mail.gmail.com> On Tue, Dec 8, 2009 at 11:15 AM, Richard Hughes wrote: > 2009/12/7 Pascal de Bruijn : >> Richard, my mail to the mailing list got rejected: > > I think ?gnome-announce-list at gnome.org bounced the mail (correctly) -- > dropping the cc should make things work again. Oops, silly me... Yeah, the gnome-color-manager list accepted my message just fine... Regards, Pascal de Bruijn From graeme2 at argyllcms.com Tue Dec 8 12:38:17 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Tue, 08 Dec 2009 23:38:17 +1100 Subject: metadata In-Reply-To: <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> Message-ID: <4B1E48B9.1080408@argyllcms.com> Richard Hughes wrote: > I'm going to be checking in the Adobe and ICC profiles first, but > alas, they can't be installed by default in distros like Fedora as > they are non free [1]. Note that Argyll V1.1.0 includes a public domain implementation of an sRGB profiles as well as an AdobeRGB1998 equivalent. Graeme Gill. From hughsient at gmail.com Tue Dec 8 12:51:45 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 12:51:45 +0000 Subject: metadata In-Reply-To: <4B1E48B9.1080408@argyllcms.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> <4B1E48B9.1080408@argyllcms.com> Message-ID: <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> 2009/12/8 Graeme Gill : > Richard Hughes wrote: > >> I'm going to be checking in the Adobe and ICC profiles first, but >> alas, they can't be installed by default in distros like Fedora as >> they are non free [1]. > > Note that Argyll V1.1.0 includes a public domain implementation > of an sRGB profiles as well as an AdobeRGB1998 equivalent. That's really good to know, and I might pinch those. What license are those available as? Really public domain? (if so, that's fantastic). Can they be installed alongside the real adobe profiles? Richard. From graeme2 at argyllcms.com Tue Dec 8 13:11:07 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Wed, 09 Dec 2009 00:11:07 +1100 Subject: metadata In-Reply-To: <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> <4B1E48B9.1080408@argyllcms.com> <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> Message-ID: <4B1E506B.4080709@argyllcms.com> Richard Hughes wrote: > 2009/12/8 Graeme Gill : >> Note that Argyll V1.1.0 includes a public domain implementation >> of an sRGB profiles as well as an AdobeRGB1998 equivalent. > > That's really good to know, and I might pinch those. What license are > those available as? Really public domain? (if so, that's fantastic). Yes. There didn't seem any sensible reason to retain copyright on them. > Can they be installed alongside the real adobe profiles? I'm not sure what you mean by that. Since "Adobe" is a registered company name, I've given the Adobe compatible profile a different name ("ClayRGB1998.icm", the best I could come up with at the time). The sRGB profile is called "sRGB.icm", just like the original one from HP/Microsoft that is no longer available, and to which it is functionally identical. Graeme Gill. From pmjdebruijn at pcode.nl Tue Dec 8 17:56:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 18:56:33 +0100 Subject: Icon Sizes In-Reply-To: <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> References: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> Message-ID: <4c2c37f90912080956y6dcf9264padf8217b2c3a5ab4@mail.gmail.com> On Sun, Dec 6, 2009 at 8:46 PM, Richard Hughes wrote: > 2009/12/6 Pascal de Bruijn : >> Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? > > commit 29b66375ed40e10bfffba601eeca9f25bd8a46b7 > Author: Richard Hughes > Date: ? Sun Dec 6 19:45:42 2009 +0000 > > ? ?Add a 64x64 pixel application icon Today, I had a chance to test this on my colleagues HP netbook again... And this seems to fix the "small icon" issue on Ubuntu Karmic Netbook Edition. Thanks, Pascal de Bruijn From pmjdebruijn at pcode.nl Tue Dec 8 17:58:02 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 18:58:02 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <4c2c37f90912080958s1dac9704qdc99372d9c1a3401@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays I had a chance to test this on a colleagues HP netbook today, and it seems to work fine. The CIE widget looks a bit silly at 50x50, but again, this is a netbook's owner choice... Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 9 12:47:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 12:47:28 +0000 Subject: Yelp isn't happy In-Reply-To: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> References: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> Message-ID: <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI > ?ghelp:gnome-color-manager?preferences? is incorrect". commit ef84a7c5dca50e40bfaf38b6c884c8c86983f334 Author: Richard Hughes Date: Wed Dec 9 12:46:33 2009 +0000 Fix the help file installation so that yelp recognises our help file Richard. From alexandre.prokoudine at gmail.com Wed Dec 9 13:18:19 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Wed, 9 Dec 2009 16:18:19 +0300 Subject: Yelp isn't happy In-Reply-To: <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> References: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> Message-ID: <733f2c730912090518w1ed31992g2921881109c65f9b@mail.gmail.com> On Wed, Dec 9, 2009 at 3:47 PM, Richard Hughes wrote: > 2009/12/6 Alexandre Prokoudine : >> When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI >> ?ghelp:gnome-color-manager?preferences? is incorrect". > > commit ef84a7c5dca50e40bfaf38b6c884c8c86983f334 > Author: Richard Hughes > Date: ? Wed Dec 9 12:46:33 2009 +0000 > > ? ?Fix the help file installation so that yelp recognises our help file Many thanks :) Alexandre From hughsient at gmail.com Wed Dec 9 13:18:35 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 13:18:35 +0000 Subject: Profile description for laptop displays In-Reply-To: <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> References: <4c2c37f90911290453x143c6820nb27f3a6b6188e3c7@mail.gmail.com> <15e53e180911291342o2b8755aavad083da3dcd849d2@mail.gmail.com> <4c2c37f90911291527v71d2cf9cy6bb41add8c7b3683@mail.gmail.com> <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> Message-ID: <15e53e180912090518h5b6e9deas414d6c5e1f1ec594@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > I wish we had some more EDID samples to compare... Maybe a call for > sample EDID data on Planet GNOME would get us some more samples? Could you send me your EDID please. Thanks. Richard. From hughsient at gmail.com Wed Dec 9 13:19:56 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 13:19:56 +0000 Subject: More verbose output in calibration? In-Reply-To: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> References: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> Message-ID: <15e53e180912090519v6481e569q867923aa08984dd3@mail.gmail.com> 2009/12/4 Lars Tore Gustavsen : > May I suggest that when I calibrate my monitor the details view are > more verbose than now (like dispcal -v). I guess a beginner will think > this is not working at all, when it measures all the black patches and > almost noting change on the display. ?I think more experienced users > also like to see the progress in the calibration ?and the deltaE > report at the end. ?In a longer time frame some progress bar (1) or > progress window (2) would be very nice eye candy in the calibration > process. You can already get the deltaE values in the dropdown widget. I think it would be very hard to do a progress bar from the data Argyll gives us. Richard. From hughsient at gmail.com Wed Dec 9 17:04:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 17:04:24 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B15B29C.2090105@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> Message-ID: <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> 2009/12/2 Graeme Gill : > For a cLut based profile, you would have to (as Lars Tore Gustavsen > suggested) run the primary values through the profile (or > do the equivalent of looking up what the cLUT values are at > the primaries). I'm sort-of stuck here. I'm confused about which cLUT I should be using (TRC curves?), as different profiles are doing very different things. Could you please elaborate about how to run the primaries through the table? Thanks. Richard. From hughsient at gmail.com Wed Dec 9 19:51:53 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 19:51:53 +0000 Subject: shared-color-profiles 0.1.0 release Message-ID: <15e53e180912091151y1f2303acw48cedde3eb0231f@mail.gmail.com> Version 0.1.0 ~~~~~~~~~~~~~ Released: 2009-12-09 * New Features: - First release! - Add the free Argyll profiles that are in the public domain (Richard Hughes) * Bugfix: - Don't install README and LICENSE in /usr/share/color (Richard Hughes) - Require color-filesystem on Fedora (Richard Hughes) The next release of gnome-color-manager will depend on this package. See http://blogs.gnome.org/hughsie/2009/12/08/shared-color-profiles/ for more details. Richard. From alexandre.prokoudine at gmail.com Wed Dec 9 23:02:22 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 02:02:22 +0300 Subject: sadly, segfault Message-ID: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> Hi, On commit f04af5c3072983e9b5113628f7782dc346c75b18: http://pastebin.com/m18c9d8c9 Alexandre From graeme2 at argyllcms.com Wed Dec 9 23:04:54 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Thu, 10 Dec 2009 10:04:54 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> Message-ID: <4B202D16.1000901@argyllcms.com> Richard Hughes wrote: > I'm sort-of stuck here. I'm confused about which cLUT I should be > using (TRC curves?), as different profiles are doing very different > things. Could you please elaborate about how to run the primaries > through the table? Thanks. Well I can only talk specifically about my set of tools: Using icclib command line tools it would be something like this: icclu -ff -ia -px sRGB.icm 1 0 0 1.000000 0.000000 0.000000 [RGB] -> MatrixFwd -> 0.412385 0.212596 0.019306 [XYZ] 0 1 0 0.000000 1.000000 0.000000 [RGB] -> MatrixFwd -> 0.357573 0.715199 0.119206 [XYZ] 0 0 1 0.000000 0.000000 1.000000 [RGB] -> MatrixFwd -> 0.180500 0.072205 0.950546 [XYZ] Progromatically, one would do the equivalent, e.g. using icclib open the profile, create a lookup object to do forward absolute colorimetric to XYZ space, and then feed colorant only combinations through it. icclu.c will illustrate the details. I'm sure something similar is possible using Marti's lcms. I wouldn't recommend trying to re-create the wheel and do it from ICC tags as there are a number of details to get right. Graeme Gill. From hughsient at gmail.com Thu Dec 10 08:07:04 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 08:07:04 +0000 Subject: sadly, segfault In-Reply-To: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> References: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> Message-ID: <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> 2009/12/9 Alexandre Prokoudine : > On commit f04af5c3072983e9b5113628f7782dc346c75b18: > http://pastebin.com/m18c9d8c9 Thanks, commit 64037aacc1ba6fa7bfa53deabc00dca57cdd658d Author: Richard Hughes Date: Thu Dec 10 08:05:25 2009 +0000 trivial: fix up an error in the last commit, so that if filename is not set we skip the load There are some more changes happening today, so please yell if anything breaks. Thanks. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 14:54:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 17:54:09 +0300 Subject: sadly, segfault In-Reply-To: <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> References: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> Message-ID: <733f2c730912100654g6eafc1e9v5c51b2783e119437@mail.gmail.com> On 12/10/09, Richard Hughes wrote: > There are some more changes happening today, so please yell if > anything breaks. Thanks. Well, apart from the profiles I sent you before whose CIE diagram doesn't render nicely, so far so good :) Alexandre From hughsient at gmail.com Thu Dec 10 15:26:35 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 15:26:35 +0000 Subject: LCMS dependency Message-ID: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> How would everyone feel if I added lcms as a hard dependency to gnome-color-manager? Now how would you all feel if I added lcms2 as a hard dep? Comments please. I think using lcms allows us to nuke a lot of the custom (fragile) profile code, but only lcms2 has the functionality I need to decode the vcgt tags. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 15:30:18 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 18:30:18 +0300 Subject: LCMS dependency In-Reply-To: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> Message-ID: <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> On 12/10/09, Richard Hughes wrote: > How would everyone feel if I added lcms as a hard dependency to > gnome-color-manager? Fine with me. Most graphics apps I use already have it as requirement. > Now how would you all feel if I added lcms2 as a hard dep? Marti does amazing job with lcms2, and since now it has a separate header for v2, this really might work :) Alexandre From pmjdebruijn at pcode.nl Thu Dec 10 15:47:06 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 16:47:06 +0100 Subject: LCMS dependency In-Reply-To: <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> Message-ID: <4c2c37f90912100747i1a0a507au47db84306b64dd42@mail.gmail.com> On Thu, Dec 10, 2009 at 4:30 PM, Alexandre Prokoudine wrote: > On 12/10/09, Richard Hughes wrote: >> How would everyone feel if I added lcms as a hard dependency to >> gnome-color-manager? > > Fine with me. Most graphics apps I use already have it as requirement. Indeed... >> Now how would you all feel if I added lcms2 as a hard dep? > > Marti does amazing job with lcms2, and since now it has a separate > header for v2, this really might work :) Adding lcms2 as a hard dependancy seems less then desirable (at this moment), since lcms2 hasn't been released yet. And it requires additional packaging... Adding it as an optional dependancy seems ok though. Regards, Pascal de Bruijn From graeme2 at argyllcms.com Thu Dec 10 16:05:34 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 11 Dec 2009 03:05:34 +1100 Subject: LCMS dependency In-Reply-To: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> Message-ID: <4B211C4E.9060502@argyllcms.com> Richard Hughes wrote: > I think using lcms allows us to nuke a lot of the custom (fragile) > profile code, but only lcms2 has the functionality I need to decode > the vcgt tags. Well of course icclib has support for accessing the vcgt tag. Graeme Gill. From hughsient at gmail.com Thu Dec 10 16:18:36 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 16:18:36 +0000 Subject: LCMS dependency In-Reply-To: <4B211C4E.9060502@argyllcms.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> Message-ID: <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> 2009/12/10 Graeme Gill : > Well of course icclib has support for accessing the vcgt tag. Sure, that's also a possibility. Richard. From pmjdebruijn at pcode.nl Thu Dec 10 17:00:03 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:00:03 +0100 Subject: LCMS dependency In-Reply-To: <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> Message-ID: <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> On Thu, Dec 10, 2009 at 5:18 PM, Richard Hughes wrote: > 2009/12/10 Graeme Gill : >> Well of course icclib has support for accessing the vcgt tag. > > Sure, that's also a possibility. Well, lcms is used by most OSS graphical apps for color management, so lcms is a natural choice when we don't want to introduce new dependancies into most systems in general. But as long as lcms2 hasn't been released and packaged into most distro's, it makes testing gcm a bit harder, which means feedback on git versions might degrade. Another slightly nasty option would be, to use lcms2, but just (for now) include the code into the gcm repo until it has been released, and just statically compile it into the gcm binaries. And yes, I'll wash my mouth with some soap :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 10 17:11:56 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:11:56 +0100 Subject: The new TRC graphs Message-ID: <4c2c37f90912100911o6f9c4b23y4648e95b2390d82a@mail.gmail.com> Hi, Just thought I should mention this... The new TRC graphs genuinely kick-ass! Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 10 17:14:31 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:14:31 +0000 Subject: LCMS dependency In-Reply-To: <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> Message-ID: <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > Well, lcms is used by most OSS graphical apps for color management, so > lcms is a natural choice when we don't want to introduce new > dependancies into most systems in general. Right. > But as long as lcms2 hasn't been released and packaged into most > distro's, it makes testing gcm a bit harder, which means feedback on > git versions might degrade. Agreed. > Another slightly nasty option would be, to use lcms2, but just (for > now) include the code into the gcm repo until it has been released, > and just statically compile it into the gcm binaries. Ick, ick. > And yes, I'll wash my mouth with some soap :) Okay, I think after what everyone has said I'll add an lcms dep, but not a lcms2 dep. We can convert when lcms2 has had it's first public release. Richard. From pmjdebruijn at pcode.nl Thu Dec 10 17:15:23 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:15:23 +0100 Subject: Bug: All profiles are 256 bytes Message-ID: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> Hi, I just built git again, and I noticed that GCM (on the profiles tab), tells me all my ICC profiles are 256 bytes... There is definitely something fishy about this :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 10 17:19:10 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:19:10 +0100 Subject: CIE widget drawing odd graphs In-Reply-To: <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> Message-ID: <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> On Sun, Dec 6, 2009 at 9:19 PM, Richard Hughes wrote: > 2009/12/6 Pascal de Bruijn : >> I might have uncovered some minor issues with the CIE widget drawing code. > > They look pretty major to me! Right :) >> I've attached an archive containing several non-standard profiles, >> which are drawn quite oddly. > > After the release I would like to look at the widget drawing code and > the profile extraction code (there are still quite a lot of FIXME's > and unimplemented stuff) and add a boat load more self tests and > internal validations. I fear I might not have time in the next couple > of weeks (lots of other work to do, which I've been putting off for > ages) so if anyone is familiar with XYZ co-ordinates and ICC profiles > I would appreciate any code review and possible patches. I just built git again, and I noticed XYZ drawing has improved... But I didn't see any actual improvements to the CIE code in git... I assume the problem wasn't in the CIE XYZ widget, but in the data that was fed to it? Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 10 17:42:38 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:42:38 +0000 Subject: Bug: All profiles are 256 bytes In-Reply-To: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> References: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> Message-ID: <15e53e180912100942t39465d40rf8875000038e9378@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > I just built git again, and I noticed that GCM (on the profiles tab), > tells me all my ICC profiles are 256 bytes... There is definitely > something fishy about this :) commit 99e9f26fe78202515187dc14b93917664ad01dc0 Author: Richard Hughes Date: Thu Dec 10 17:41:49 2009 +0000 trivial: show the file size in the prefs dialog, not the clut size Richard. From hughsient at gmail.com Thu Dec 10 17:43:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:43:40 +0000 Subject: CIE widget drawing odd graphs In-Reply-To: <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> Message-ID: <15e53e180912100943o413617a4j3d75cc58a096540b@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > I just built git again, and I noticed XYZ drawing has improved... But > I didn't see any actual improvements to the CIE code in git... The profile code had a subtle signing bug that wasn't tested for. The change from gchar to guint8 fixed this bug (and possibly a few more). Richard. From web.kiddo at free.fr Thu Dec 10 20:56:13 2009 From: web.kiddo at free.fr (Jeff) Date: Thu, 10 Dec 2009 15:56:13 -0500 Subject: Incorrect parsing of .icm files? Message-ID: <1260478573.20024.33.camel@kusanagi> Hello, I wanted to give gnome-color-manager a try just out of curiosity. I own a Dell Inspiron 910 (a.k.a. Mini 9) netbook, which (to my untrained eye) seems to have good colors out of the box. I downloaded the Windows drivers for this netbook's monitor here: http://support.us.dell.com/support/downloads/download.aspx?releaseid=R229102&fileid=329762 It seems like the monitor is called ST2010, there's an ST2010B and ST2010F version of the driver... both contain the exact same ".icm" file (verified with md5sum), which is attached to this message. Now, here comes the actual question: loading this .icm file in GNOME Color Manager and using it creates obviously incorrect colors (as if someone punched contrast and saturation way up), compared to the default colors (no color profile loaded; colors appear neutral). With the ICC profile loaded, for example, gradients in the Clearlooks GTK theme lose definition: the slight gray gradients on the combobox widgets disappears completely into white. I'm just wondering if - this is a bug in GNOME Color Manager - this is a buggy .icm file - it's a PEBKAC. I wish I could take a screenshot, but screenshots don't work in this case (and photo cameras have trouble photographing LCD screens)... -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ST2010.icm Type: image/x-tga Size: 552 bytes Desc: URL: From hughsient at gmail.com Thu Dec 10 21:24:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 21:24:28 +0000 Subject: Incorrect parsing of .icm files? In-Reply-To: <1260478573.20024.33.camel@kusanagi> References: <1260478573.20024.33.camel@kusanagi> Message-ID: <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> 2009/12/10 Jeff : > Now, here comes the actual question: loading this .icm file in GNOME Color > Manager and using it creates obviously incorrect colors (as if someone > punched contrast and saturation way up), compared to the default colors (no > color profile loaded; colors appear neutral). With the ICC profile loaded, > for example, gradients in the Clearlooks GTK theme lose definition: the > slight gray gradients on the combobox widgets disappears completely into > white. > > I'm just wondering if > - this is a bug in GNOME Color Manager > - this is a buggy .icm file Both, I think. If you try gnome-power-manager in git master, it won't "do" anything, as the profile does not have a vcgt or mlut tag, and even using the linear tables doesn't do anything. Even trying using dispwin doesn't appear to do anything. If you use the 2.29.1 release there's a bug where we're using the xTRC values as a gamma table (eek). Richard. From hughsient at gmail.com Thu Dec 10 21:28:59 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 21:28:59 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B202D16.1000901@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> Message-ID: <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> 2009/12/9 Graeme Gill : > Progromatically, one would do the equivalent, e.g. using icclib > open the profile, create a lookup object to do forward absolute colorimetric > to XYZ space, and then feed colorant only combinations through it. > icclu.c will illustrate the details. I'm sure something similar is > possible using Marti's lcms. I've just added code to git master to do this, and it indeed works to get the gamut range for a few scanner and print (RGB) profiles I've got here. It doesn't seem to work for CYMK profiles, although I'm sure that's something crazy I'm doing (or not doing..). Anyway, I'll give it another go to get CYMK profiles working correctly. Richard. From graeme2 at argyllcms.com Thu Dec 10 21:51:35 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 11 Dec 2009 08:51:35 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> Message-ID: <4B216D67.2080307@argyllcms.com> Richard Hughes wrote: > I've just added code to git master to do this, and it indeed works to > get the gamut range for a few scanner and print (RGB) profiles I've > got here. It doesn't seem to work for CYMK profiles, although I'm sure > that's something crazy I'm doing (or not doing..). Subtractive media profiles usually have less easily defined gamut extents than additive (RGB) profiles. Unlike additive, you can't expect it to be a nice polygon in XYZ/xy/u'v' space. Something roughly accurate for a chromaticity diagram would involve sampling along the color girdle (ie. going through combinations between CMY: 1 0 0, 1 1 0, 0 1 0, 0 1 1, 0 0 1, 1 0 1, e.g. 1 0 0, 1 .1 0, 1 .2 0 .. 1 .9 0, 1 1 0, .9 1 0, .8 1 0, ... etc.) Graeme Gill. From hughsient at gmail.com Thu Dec 10 22:08:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 22:08:49 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B216D67.2080307@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> <4B216D67.2080307@argyllcms.com> Message-ID: <15e53e180912101408w4b7dca45o4dac5869574a39bd@mail.gmail.com> 2009/12/10 Graeme Gill : > 1 1 0, 0 1 0, 0 1 1, 0 0 1, 1 0 1, > e.g. 1 0 0, 1 .1 0, 1 .2 0 .. 1 .9 0, 1 1 0, .9 1 0, .8 1 0, ... etc.) Excellent, this makes sense. I'll give it a go tomorrow. Thanks again. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 22:35:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 01:35:09 +0300 Subject: Incorrect parsing of .icm files? In-Reply-To: <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> References: <1260478573.20024.33.camel@kusanagi> <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> Message-ID: <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> On 12/11/09, Richard Hughes wrote: > Both, I think. If you try gnome-power-manager in git master, it won't > "do" anything, as the profile does not have a vcgt or mlut tag, and > even using the linear tables doesn't do anything. Even trying using > dispwin doesn't appear to do anything. If you use the 2.29.1 release > there's a bug where we're using the xTRC values as a gamma table > (eek). BTW, what's new regarding restoration of applied profile after dehibernation and leaving a screensaver? Alexandre From hughsient at gmail.com Fri Dec 11 08:55:43 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 08:55:43 +0000 Subject: Incorrect parsing of .icm files? In-Reply-To: <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> References: <1260478573.20024.33.camel@kusanagi> <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> Message-ID: <15e53e180912110055s2a1bc938ob58b1ef4508ef83a@mail.gmail.com> 2009/12/10 Alexandre Prokoudine : > BTW, what's new regarding restoration of applied profile after > dehibernation and leaving a screensaver? If you're using a KMS driver (the future) then the "thaw" should re-apply the per-output gamma tables. With the screensaver I want to work on a patch that either: * saves the gamma table before the table is destroyed * calls gcm-apply after the screensaver comes back from black I don't use a screensaver myself, but if nobody does the gnome-screensaver work I guess I'll have to add that to my list. Richard. From knizek.confy at volny.cz Fri Dec 11 09:46:18 2009 From: knizek.confy at volny.cz (Milan =?UTF-8?Q?Kn=C3=AD=C5=BEek?=) Date: Fri, 11 Dec 2009 10:46:18 +0100 Subject: shared-color-profiles 0.1.0 release In-Reply-To: <15e53e180912101003s2a013080qf839379565a40964@mail.gmail.com> References: <15e53e180912091151y1f2303acw48cedde3eb0231f@mail.gmail.com> <1260467822.3147.14.camel@athlon> <15e53e180912101003s2a013080qf839379565a40964@mail.gmail.com> Message-ID: <1260524778.2203.3.camel@athlon> (I am forwarding your response to the emailing list, since my original intention was not to email just you but the list and I did not notice the "reply" address is just yours.) Possibly someone knows about other pre-packaged profiles? regards, Milan Richard Hughes p??e v ?t 10. 12. 2009 v 18:03 +0000: > 2009/12/10 Milan Kn??ek : > > Will the project make "icc-profiles" and "oyranos-icc" obsolete or > > rather complementary? > > I hope we can standardize on one -- I must admit I had not heard of > icc-profiles until you pointed me at them. > > In the future, we'll be distributing shared-color-targets as well (8.7 > targets) to make all this stuff just work out of the box. > > Richard. > Milan Knizek knizek (dot) confy (at) volny (dot) cz http://www.milan-knizek.net - About linux and photography (Czech language only) From lars.tore at mulebakken.net Fri Dec 11 15:11:33 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 11 Dec 2009 16:11:33 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912110711n7d6a6e0ub9a3bf5b9f817b2f@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: > the x and y coordinates of the white point from the device ICC > profile. > The white point cross is nice, but I wonder if it would be nice to add a more human readable text string with the whitepoint somewhere: The wtpt program included in liblcms-utils outputs the correlated temperature or the locus temperature, so I guess it's easy to this with lcms. Here are some sample output: profile 1: White point near 4937K profile 2: WhitePoint : D65 (daylight) And I think this is more readable to most people than xy values. Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 11 15:51:36 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 11 Dec 2009 16:51:36 +0100 Subject: LCMS dependency In-Reply-To: <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> Message-ID: <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> On Thu, Dec 10, 2009 at 6:14 PM, Richard Hughes wrote: > 2009/12/10 Pascal de Bruijn : >> Well, lcms is used by most OSS graphical apps for color management, so >> lcms is a natural choice when we don't want to introduce new >> dependancies into most systems in general. > > Right. I see a lot of commits regarding lcms in git... Has all the ICC parsing code already been removed from gcm? I'm just asking this out of curiosity... Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Fri Dec 11 17:47:18 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 11 Dec 2009 18:47:18 +0100 Subject: Regression: the TRC graphs disappear on working space/input profiles Message-ID: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> Hi, With today's git checkout the TRC graphs disappear with working space profiles and input profiles. Is this intentional? Or just transitory during the lcms migration? Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Fri Dec 11 17:47:30 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 20:47:30 +0300 Subject: apply/session Message-ID: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> Hi, So, to get things right, what exactly is the difference between gcm-session and gcm-apply? Alexandre From hughsient at gmail.com Fri Dec 11 17:50:06 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:50:06 +0000 Subject: LCMS dependency In-Reply-To: <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> Message-ID: <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> 2009/12/11 Pascal de Bruijn : > I see a lot of commits regarding lcms in git... > Has all the ICC parsing code already been removed from gcm? About half. To remove the other half we need lcms2, which may be some months away. Richard. From hughsient at gmail.com Fri Dec 11 17:51:47 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:51:47 +0000 Subject: apply/session In-Reply-To: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> Message-ID: <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> 2009/12/11 Alexandre Prokoudine : > So, to get things right, what exactly is the difference between > gcm-session and gcm-apply? gcm-session is the dbus-activated session service, that provides an interface applications like gimp can use. gcm-apply is just a quick hack that sets (or resets) all the CLUT values to what they should be, in case something destroyed them (e.g. hibernate, screensaver dim, etc). Richard From hughsient at gmail.com Fri Dec 11 17:52:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:52:49 +0000 Subject: Regression: the TRC graphs disappear on working space/input profiles In-Reply-To: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> References: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> Message-ID: <15e53e180912110952s3f10d3dmc6305f350a17548@mail.gmail.com> 2009/12/11 Pascal de Bruijn : > With today's git checkout the TRC graphs disappear with working space > profiles and input profiles. > Is this intentional? Or just transitory during the lcms migration? Intentional, but they'll be coming back hopefully on Monday when I merge proper TRC support. Richard From alexandre.prokoudine at gmail.com Fri Dec 11 18:02:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 21:02:09 +0300 Subject: apply/session In-Reply-To: <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> Message-ID: <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> On Fri, Dec 11, 2009 at 8:51 PM, Richard Hughes wrote: > 2009/12/11 Alexandre Prokoudine : >> So, to get things right, what exactly is the difference between >> gcm-session and gcm-apply? > > gcm-session is the dbus-activated session service, that provides an > interface applications like gimp can use. And it's not currently used by anything, if I get it right? > gcm-apply is just a quick > hack that sets (or resets) all the CLUT values to what they should be, > in case something destroyed them (e.g. hibernate, screensaver dim, > etc). So vcgt part and setting the atom is done entirely by gcm-prefs? Alexandre From alexandre.prokoudine at gmail.com Fri Dec 11 18:06:41 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 21:06:41 +0300 Subject: LCMS dependency In-Reply-To: <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> Message-ID: <733f2c730912111006s640088c6g6b11a1266c54e4c8@mail.gmail.com> On Fri, Dec 11, 2009 at 8:50 PM, Richard Hughes wrote: >> Has all the ICC parsing code already been removed from gcm? > > About half. To remove the other half we need lcms2, which may be some > months away. Is this why TRC curve has (temp.) gone? :) Alexandre From claude at 2xlibre.net Fri Dec 11 22:36:40 2009 From: claude at 2xlibre.net (Claude Paroz) Date: Fri, 11 Dec 2009 23:36:40 +0100 Subject: Why not using strftime? Message-ID: <1260571000.4145.121.camel@shuttle> Hi, As I didn't find any bug tracker now, I'm reporting this here. I wonder why you're using a complicated machinery with gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of simply using strftime call in gcm-profile.c Something like (not tested code): ret = cmsTakeCreationDateTime (&created, priv->lcms_profile); if (ret) - priv->datetime = gcm_utils_format_date_time (1900 +created.tm_year, created.tm_mon+1, created + /* Translators: this is the profile creation date strftime format */ + strftime (priv->datetime, sizeof(priv->datetime), _("%d %B %Y, %H:%M:%S"), created); Months names should not be translated again, this is a waste of time for translators. Claude P.S. Please cc me on answer From hughsient at gmail.com Sat Dec 12 09:01:19 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 12 Dec 2009 09:01:19 +0000 Subject: apply/session In-Reply-To: <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> Message-ID: <15e53e180912120101n74bc1e92i31b4359ccaab961e@mail.gmail.com> 2009/12/11 Alexandre Prokoudine : >> gcm-session is the dbus-activated session service, that provides an >> interface applications like gimp can use. > > And it's not currently used ?by anything, if I get it right? Correct. It's the egg part of the chicken-and-egg question. >> gcm-apply is just a quick >> hack that sets (or resets) all the CLUT values to what they should be, >> in case something destroyed them (e.g. hibernate, screensaver dim, >> etc). > > So vcgt part and setting the atom is done entirely by gcm-prefs? No, when gcm "set the CLUT" it actually: * Applies the CLUT to each adaptor * Sets the _ICC_PROFILE atom on each output * Sets the _ICC_PROFILE atom on the screen for the primary output To apply the clut we extract the VCGT and then push it into the video card. Settign the CLUT is done by gcm-apply and gcm-prefs. You could argue the latter is redundant, unless you're changing the profile, and I would agree. Richard. From hughsient at gmail.com Sat Dec 12 09:21:57 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 12 Dec 2009 09:21:57 +0000 Subject: Why not using strftime? In-Reply-To: <1260571000.4145.121.camel@shuttle> References: <1260571000.4145.121.camel@shuttle> Message-ID: <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> 2009/12/11 Claude Paroz : > As I didn't find any bug tracker now, I'm reporting this here. I wonder > why you're using a complicated machinery with > gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of > simply using strftime call in gcm-profile.c Originally because we got some complicated date structure to decode, now we're using lcms we just get a nice sane tm struct. I've fixed things up in the following commit: commit 6607487433a1bf074638d6833bd7c493ef1d75ff Author: Richard Hughes Date: Sat Dec 12 09:19:57 2009 +0000 Use strftime rather than our own hand-rolled function :100644 100644 a6abf96... 0abfb35... M src/gcm-profile.c :100644 100644 203466c... 76e5b24... M src/gcm-utils.c :100644 100644 bc3ad9d... a570d6b... M src/gcm-utils.h Thanks Richard. From claude at 2xlibre.net Sat Dec 12 09:48:28 2009 From: claude at 2xlibre.net (Claude Paroz) Date: Sat, 12 Dec 2009 10:48:28 +0100 Subject: Why not using strftime? In-Reply-To: <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> References: <1260571000.4145.121.camel@shuttle> <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> Message-ID: <1260611308.3833.38.camel@shuttle> Le samedi 12 d?cembre 2009 ? 09:21 +0000, Richard Hughes a ?crit : > 2009/12/11 Claude Paroz : > > As I didn't find any bug tracker now, I'm reporting this here. I wonder > > why you're using a complicated machinery with > > gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of > > simply using strftime call in gcm-profile.c > > Originally because we got some complicated date structure to decode, > now we're using lcms we just get a nice sane tm struct. I've fixed > things up in the following commit: > > commit 6607487433a1bf074638d6833bd7c493ef1d75ff > Author: Richard Hughes > Date: Sat Dec 12 09:19:57 2009 +0000 > > Use strftime rather than our own hand-rolled function > > :100644 100644 a6abf96... 0abfb35... M src/gcm-profile.c > :100644 100644 203466c... 76e5b24... M src/gcm-utils.c > :100644 100644 bc3ad9d... a570d6b... M src/gcm-utils.h > > Thanks Great, thanks ! Claude From pmjdebruijn at pcode.nl Mon Dec 14 14:44:19 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 14 Dec 2009 15:44:19 +0100 Subject: Patch: Debian binary detection fix In-Reply-To: <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> Message-ID: <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> On Thu, Nov 26, 2009 at 10:07 PM, Richard Hughes wrote: > 2009/11/26 Pascal de Bruijn : >> Therefore the detect should _first_ detect if the Debian renamed >> binaries exist, and if not, fallback to the original binary names. > > Agreed. I've attached the patch I've committed. Thanks. Maybe it's a thought to (last) search /usr/local/bin as well for the standard binary names? Regards, Pascal de Bruijn From hughsient at gmail.com Mon Dec 14 15:13:51 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 14 Dec 2009 15:13:51 +0000 Subject: Patch: Debian binary detection fix In-Reply-To: <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> Message-ID: <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> 2009/12/14 Pascal de Bruijn : > Maybe it's a thought to (last) search /usr/local/bin as well for the > standard binary names? Sure. I'll do a deal: You code it, and I'll apply it. If you need instructions, I can explain how. Richard. From hughsient at gmail.com Wed Dec 16 11:57:19 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 16 Dec 2009 11:57:19 +0000 Subject: PackageKit integration Message-ID: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> gnome-color-manager in git master now integrates with PackageKit in the following ways: * Offers to install argyllcms if the user tries to calibrate without argyllcms installed * Offers to install shared-color-targets if they are not installed, and the user tries to calibrate a non-display device This means gnome-color-manager can be shipped on live-cd's without bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) shared-color-targets, just in case the user wants to calibrate a device. If people want to turn off the PackageKit bits, then I can add a GConf key or #ifdef as required. Richard. From pmjdebruijn at pcode.nl Wed Dec 16 12:23:52 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 16 Dec 2009 13:23:52 +0100 Subject: PackageKit integration In-Reply-To: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> Message-ID: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> On Wed, Dec 16, 2009 at 12:57 PM, Richard Hughes wrote: > gnome-color-manager in git master now integrates with PackageKit in > the following ways: > > * Offers to install argyllcms if the user tries to calibrate without > argyllcms installed > * Offers to install shared-color-targets if they are not installed, > and the user tries to calibrate a non-display device > > This means gnome-color-manager can be shipped on live-cd's without > bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) > shared-color-targets, just in case the user wants to calibrate a > device. > > If people want to turn off the PackageKit bits, then I can add a GConf > key or #ifdef as required. Since not all distro's don't use PackageKit that would be very welcome. A ./configure option would probably be best? Since this is a distro specific setting, it needs to be set in the packaging process. And GConf is very user specific, so it would seem odd to use GConf for this particular purpose. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 16 12:32:06 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 16 Dec 2009 13:32:06 +0100 Subject: PackageKit integration In-Reply-To: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> Message-ID: <4c2c37f90912160432r5cc68f8as29bc4d7419eeecd8@mail.gmail.com> On Wed, Dec 16, 2009 at 1:23 PM, Pascal de Bruijn wrote: > On Wed, Dec 16, 2009 at 12:57 PM, Richard Hughes wrote: >> gnome-color-manager in git master now integrates with PackageKit in >> the following ways: >> >> * Offers to install argyllcms if the user tries to calibrate without >> argyllcms installed >> * Offers to install shared-color-targets if they are not installed, >> and the user tries to calibrate a non-display device >> >> This means gnome-color-manager can be shipped on live-cd's without >> bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) >> shared-color-targets, just in case the user wants to calibrate a >> device. >> >> If people want to turn off the PackageKit bits, then I can add a GConf >> key or #ifdef as required. > > Since not all distro's don't use PackageKit that would be very welcome. > > A ./configure option would probably be best? Since this is a distro > specific setting, it needs to be set in the packaging process. > > And GConf is very user specific, so it would seem odd to use GConf for > this particular purpose. Hmm doesn't totem do similar things with codecs and the gstreamer packages? GCM should probably follow a similar approach... Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 16 12:42:07 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 16 Dec 2009 12:42:07 +0000 Subject: PackageKit integration In-Reply-To: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> Message-ID: <15e53e180912160442v2f95d4rb4b670389fb556d9@mail.gmail.com> 2009/12/16 Pascal de Bruijn : > Since not all distro's don't use PackageKit that would be very welcome. commit 494ebc7e84d21d2626e06aa7e009cc90625e0fd8 Author: Richard Hughes Date: Wed Dec 16 12:41:02 2009 +0000 Add a configure option --enable-packagekit (default yes) to choose if auto-install functionality should be included Richard. From joedalton2 at yahoo.dk Wed Dec 16 22:16:18 2009 From: joedalton2 at yahoo.dk (Joe Dalton) Date: Wed, 16 Dec 2009 14:16:18 -0800 (PST) Subject: bugreport gnome-color-manager translation Message-ID: <977326.61727.qm@web24408.mail.ird.yahoo.com> "Report-Msgid-Bugs-To: http://bugzilla.gnome.org/enter_bug.cgi?product=gnome-" "color-manager\n" but https://bugzilla.gnome.org/enter_bug.cgi?product=gnome-color-manager gives this message Sorry, either the product gnome-color-manager does not exist or you aren't authorized to enter a bug into it. Please choose a valid product. What does intent mean here. Is the sentence broken in some way/missing something? #: ../data/gnome-color-manager.schemas.in.h:7 msgid "The output intent to use, e.g. 'perceptual'." Hi, a google did not help me, and i have only found one danish translation, which maybe is correct mayby not. Please give some more info. #. TRANSLATORS: rendering intent: you probably want to google this #: ../src/gcm-prefs.c:2029 msgid "Perceptual" bye Joe Danish translation ___________________________________________________________ Skal du k?be ny bil? Sammenlign priser p? brugte biler med Kelkoo og find et godt tilbud! - Se mere her http://dk.yahoo.com/r/pat/mmb From hughsient at gmail.com Thu Dec 17 09:31:22 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 17 Dec 2009 09:31:22 +0000 Subject: bugreport gnome-color-manager translation In-Reply-To: <977326.61727.qm@web24408.mail.ird.yahoo.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> Message-ID: <15e53e180912170131p5c35354dv6e3df58be1ee6782@mail.gmail.com> 2009/12/16 Joe Dalton : > Sorry, either the product gnome-color-manager ?does not exist or you aren't authorized to enter a bug into it. Please choose a valid product. I've opened a bug to have the bugzilla product created: https://bugzilla.gnome.org/show_bug.cgi?id=604811 > What does intent mean here. Is the sentence broken in some way/missing something? > > #: ../data/gnome-color-manager.schemas.in.h:7 > msgid "The output intent to use, e.g. 'perceptual'." The output intent is the way that one range of colors is mapped to another range, for instance, do we clip the extremes of color, or do we try to scale it so they are all represented. Intent in English means "what you would do if possible" although in this technical speak it means more of "what method do you want me to use when transforming the color" > Hi, a google did not help me, and i have only found one danish translation, which maybe is correct mayby not. Please give some more info. > #. TRANSLATORS: rendering intent: you probably want to google this > #: ../src/gcm-prefs.c:2029 > msgid "Perceptual" This is how I understand perceptual rendering intent: '''This rendering intent maps color "smoothly", preserving relationships between similar colors. This prevents "gamut clipping" with its potential loss of detail and "tonal banding" problems. Gamut clipping occurs when colors that are different in the input image appear the same when printed. Perceptual rendering intent makes small compromises throughout the entire color space in order to preserve color relationships. It sacrifices some precision of in-gamut colors in order to ensure pleasing results.''' Does that make things clearer? Feel free to change the translator comments to something more sensible -- you know what you are doing. :-) Richard. From alexandre.prokoudine at gmail.com Thu Dec 17 12:29:54 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 17 Dec 2009 15:29:54 +0300 Subject: bugreport gnome-color-manager translation In-Reply-To: <977326.61727.qm@web24408.mail.ird.yahoo.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> Message-ID: <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> On 12/17/09, Joe Dalton wrote: > What does intent mean here. Is the sentence broken in some way/missing > something? > > #: ../data/gnome-color-manager.schemas.in.h:7 > msgid "The output intent to use, e.g. 'perceptual'." > > > Hi, a google did not help me, and i have only found one danish translation, > which maybe is correct mayby not. I'm pretty sure that both GIMP, Inkscape and Scribus have Danish translations. Check them re. "Color management" tab in Preferences of all these applications. I also highly recommend "Real world color management" book. It was translated into many languages, maybe Danish was among them. You will understand a lot more about color management after reading it, trust me on this :) Alexandre From anders at brander.dk Thu Dec 17 12:35:12 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 17 Dec 2009 13:35:12 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> Message-ID: <1261053312.7873.1.camel@smutogtab> Hi, On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: > I also highly recommend "Real world color management" book. It was > translated into many languages, maybe Danish was among them. You will > understand a lot more about color management after reading it, trust > me on this :) I second that, go read "Real world color management" now, if you haven't already. It doesn't have a danish translation by the way. Regards, Anders Brander From pmjdebruijn at pcode.nl Thu Dec 17 12:38:42 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 17 Dec 2009 13:38:42 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <1261053312.7873.1.camel@smutogtab> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> Message-ID: <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> On Thu, Dec 17, 2009 at 1:35 PM, Anders Brander wrote: > Hi, > > On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: >> I also highly recommend "Real world color management" book. It was >> translated into many languages, maybe Danish was among them. You will >> understand a lot more about color management after reading it, trust >> me on this :) > > I second that, go read "Real world color management" now, if you haven't > already. It doesn't have a danish translation by the way. I third that, the first 75% is written completely generic and os/application independent. It's an excellent book... Generally for most photographical work you'll want Perceptual... Relative Colorimetric is usually most useful when you have very high end equipment, thus limiting clipping... Regards, Pascal de Bruijn From lars.tore at mulebakken.net Thu Dec 17 15:11:12 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 17 Dec 2009 16:11:12 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> Message-ID: <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> On Thu, Dec 17, 2009 at 1:38 PM, Pascal de Bruijn wrote: > On Thu, Dec 17, 2009 at 1:35 PM, Anders Brander wrote: >> Hi, >> >> On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: >>> I also highly recommend "Real world color management" book. It was >>> translated into many languages, maybe Danish was among them. You will >>> understand a lot more about color management after reading it, trust >>> me on this :) >> >> I second that, go read "Real world color management" now, if you haven't >> already. It doesn't have a danish translation by the way. > > I third that, the first 75% is written completely generic and > os/application independent. It's an excellent book... > > Generally for most photographical work you'll want Perceptual... > > Relative Colorimetric is usually most useful when you have very high > end equipment, thus limiting clipping... > > Regards, > Pascal de Bruijn > _______________________________________________ > gnome-color-manager-list mailing list > gnome-color-manager-list at gnome.org > http://mail.gnome.org/mailman/listinfo/gnome-color-manager-list > I think the book Understanding color managment by Abhay Sharma is a better book, and I have both :-) About the translation, in Norwegian (it might be close to danish) the correct word is "Perseptuell". But I think we don't have to translate all that specific terms. Since most of the written literature are in english, I guess most people who use color managment have no problems with the english terms on the rendering intets. Actually I think at least in my language that translated word will create a lot of confusion since the words are very seldom used, and everyone use the English terms. I guess we can blame photoshop for this. Lars Tore Gustavsen From anders at brander.dk Thu Dec 17 15:20:42 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 17 Dec 2009 16:20:42 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> Message-ID: <1261063242.15170.1.camel@smutogtab> Hi, On Thu, 2009-12-17 at 16:11 +0100, Lars Tore Gustavsen wrote: > About the translation, in Norwegian (it might be close to danish) the > correct word is "Perseptuell". I just looked it up in the danish Rawstudio translation, it's translated as "Opfattelse" - but I don't think that's correct (and I'm danish). Regards, Anders Brander From hughsient at gmail.com Sat Dec 19 17:32:08 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 19 Dec 2009 17:32:08 +0000 Subject: Bugzilla account created Message-ID: <15e53e180912190932x47e70876xcb147ce2718e3bff@mail.gmail.com> I've arranged for a bugzilla component to be created. If you want to file a bug, use https://bugzilla.gnome.org/enter_bug.cgi?product=gnome-color-manager but be sure to check for duplicates first, using https://bugzilla.gnome.org/buglist.cgi?product=gnome-color-manager At the moment, there are no bugs, which means GCM must be feature complete and perfect in every way. I can't imagine this situation will last long :-) Richard. From robert.ancell at gmail.com Sun Dec 20 22:34:24 2009 From: robert.ancell at gmail.com (Robert Ancell) Date: Mon, 21 Dec 2009 09:34:24 +1100 Subject: Color management and Simple Scan Message-ID: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Hi all, I've been developing a basic scanning application called Simple Scan ( http://launchpad.net/simple-scan) and one request is to add color management to it: https://bugs.edge.launchpad.net/simple-scan/+bug/498029 I don't know much about color management so my question is: - What do I need to do to make Simple Scan work with GNOME Color Manager? - Is there a GPL project I can lift some code from? Thanks, --Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughsient at gmail.com Mon Dec 21 08:28:27 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 21 Dec 2009 08:28:27 +0000 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Message-ID: <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> 2009/12/20 Robert Ancell : > I don't know much about color management so my question is: > - What do I need to do to make Simple Scan work with GNOME Color Manager? > - Is there a GPL project I can lift some code from? Well, first you need to ask gnome-color-manager for a list of suitable profiles using the DBus interface: http://git.gnome.org/browse/gnome-color-manager/plain/src/org.gnome.ColorManager.xml If you know the device you're using (you probably should) then you can call GetProfilesForDevice() with the sysfs path, and a list of profiles will be returned. If more than one profile is returned you probably need to show it in a combo-box or something, but it would be quite unusual for more than one profile to be returned. If you get no results, then it's probably because the user has never assigned a profile for that device (i.e. there is no a-priori mapping). In this case you can either default to not doing color conversion yourself, or instead call the GetProfilesForType() dbus method, using the type=scanner. This will return a list of profiles that are suitable for use with scanners. Now, when you have a filename of the profile you want to use it's an easy (!) case of acquiring the image like usual, and then you can either: * Just save the image with an embedded color profile (see bug https://bugzilla.gnome.org/show_bug.cgi?id=604610 -- I'm working as fast as I can on the infrastructure bits) * Convert the image from device-mapping to something like sRGB using lcms (there are examples in the lcms tarball) I think it's probably best you do the first one, unless you know you're going to be using the images in a non-color-aware application. Anyway, I hope that helps, Richard. From alexandre.prokoudine at gmail.com Mon Dec 21 16:01:01 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 21 Dec 2009 19:01:01 +0300 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Message-ID: <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> On Mon, Dec 21, 2009 at 1:34 AM, Robert Ancell wrote: > - Is there a GPL project I can lift some code from? XSane sounds like an obvious source :) Alexandre From hughsient at gmail.com Mon Dec 21 16:08:05 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 21 Dec 2009 16:08:05 +0000 Subject: Color management and Simple Scan In-Reply-To: <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> Message-ID: <15e53e180912210808y5fe054c7p1f2d41f4fdb97ee1@mail.gmail.com> 2009/12/21 Alexandre Prokoudine : > XSane sounds like an obvious source :) Well, depends on how you like your sanity. Richard. From robert.ancell at gmail.com Mon Dec 21 22:57:54 2009 From: robert.ancell at gmail.com (Robert Ancell) Date: Tue, 22 Dec 2009 09:57:54 +1100 Subject: Color management and Simple Scan In-Reply-To: <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> Message-ID: <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> Thanks very much Richard! Very interested in the GDK patch - I need to make a patch there too to embed resolution information: https://bugs.edge.launchpad.net/simple-scan/+bug/484616 I think I need to do the conversion regardless so the preview is displayed correctly. Thanks for the lcms link. --Robert 2009/12/21 Richard Hughes > 2009/12/20 Robert Ancell : > > I don't know much about color management so my question is: > > - What do I need to do to make Simple Scan work with GNOME Color Manager? > > - Is there a GPL project I can lift some code from? > > Well, first you need to ask gnome-color-manager for a list of suitable > profiles using the DBus interface: > > http://git.gnome.org/browse/gnome-color-manager/plain/src/org.gnome.ColorManager.xml > > If you know the device you're using (you probably should) then you can > call GetProfilesForDevice() with the sysfs path, and a list of > profiles will be returned. If more than one profile is returned you > probably need to show it in a combo-box or something, but it would be > quite unusual for more than one profile to be returned. If you get no > results, then it's probably because the user has never assigned a > profile for that device (i.e. there is no a-priori mapping). In this > case you can either default to not doing color conversion yourself, or > instead call the GetProfilesForType() dbus method, using the > type=scanner. This will return a list of profiles that are suitable > for use with scanners. > > Now, when you have a filename of the profile you want to use it's an > easy (!) case of acquiring the image like usual, and then you can > either: > > * Just save the image with an embedded color profile (see bug > https://bugzilla.gnome.org/show_bug.cgi?id=604610 -- I'm working as > fast as I can on the infrastructure bits) > * Convert the image from device-mapping to something like sRGB using > lcms (there are examples in the lcms tarball) > > I think it's probably best you do the first one, unless you know > you're going to be using the images in a non-color-aware application. > > Anyway, I hope that helps, > > Richard. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughsient at gmail.com Tue Dec 22 09:13:09 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 22 Dec 2009 09:13:09 +0000 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> Message-ID: <15e53e180912220113n329e8f81j982c21203734945f@mail.gmail.com> 2009/12/21 Robert Ancell : > Very interested in the GDK patch - I need to make a patch there too to embed > resolution information: > https://bugs.edge.launchpad.net/simple-scan/+bug/484616 Sure, looks like you just need to add a new pixbuf-option. It's not that hard, as long as you're comfortable just implementing a spec in otherwise fairly hairy code. :-) If you need a hand, give me a shout on or off list. Richard. From bcowan at fastmail.co.uk Thu Dec 24 20:06:29 2009 From: bcowan at fastmail.co.uk (Bruce Cowan) Date: Thu, 24 Dec 2009 20:06:29 +0000 Subject: String fixes Message-ID: <1261685189.7547.4.camel@localhost.localdomain> I've found a few string errors. Here's a patch to fix them. I thought I better post it here rather than just commit straight to the repository, because there is a potential point of disagreement. I've changed the strformat format to the en_US one, which could be a bit contentious. -- Bruce Cowan -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Minor-string-fixes.patch Type: text/x-patch Size: 2223 bytes Desc: not available URL: From hughsient at gmail.com Fri Dec 25 11:26:32 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 25 Dec 2009 11:26:32 +0000 Subject: Happy Christmas Message-ID: <15e53e180912250326n7d35ca86tc59fb282fd940a1c@mail.gmail.com> Hey guys. I just wanted to send to you all a Happy Christmas, and my thanks for the hard work you've put in this year. It's appreciated. Richard. From hughsient at gmail.com Fri Dec 25 11:28:10 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 25 Dec 2009 11:28:10 +0000 Subject: String fixes In-Reply-To: <1261685189.7547.4.camel@localhost.localdomain> References: <1261685189.7547.4.camel@localhost.localdomain> Message-ID: <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> 2009/12/24 Bruce Cowan : > I've found a few string errors. Here's a patch to fix them. I thought I > better post it here rather than just commit straight to the repository, > because there is a potential point of disagreement. The first two are obvious, so please commit those. > I've changed the strformat format to the en_US one, which could be a bit > contentious. I agree it's probably the correct thing to do, but it does however want me to rip my eyes out every time I see that format. :-) So, please commit the entire patch. Thanks. Richard. From pmjdebruijn at pcode.nl Fri Dec 25 12:28:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 25 Dec 2009 13:28:33 +0100 Subject: String fixes In-Reply-To: <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> References: <1261685189.7547.4.camel@localhost.localdomain> <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> Message-ID: <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> On Fri, Dec 25, 2009 at 12:28 PM, Richard Hughes wrote: > 2009/12/24 Bruce Cowan : >> I've found a few string errors. Here's a patch to fix them. I thought I >> better post it here rather than just commit straight to the repository, >> because there is a potential point of disagreement. > > The first two are obvious, so please commit those. > >> I've changed the strformat format to the en_US one, which could be a bit >> contentious. > > I agree it's probably the correct thing to do, but it does however > want me to rip my eyes out every time I see that format. :-) /me Shudders... We're introducing AM/PM time... oh well seems correct for the en_US locale... However, if one we're to switch to the en_UK locale, would AM/PM time be gone again? Regards, Pascal de Bruijn From bcowan at fastmail.co.uk Fri Dec 25 14:37:16 2009 From: bcowan at fastmail.co.uk (Bruce Cowan) Date: Fri, 25 Dec 2009 14:37:16 +0000 Subject: String fixes In-Reply-To: <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> References: <1261685189.7547.4.camel@localhost.localdomain> <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> Message-ID: <1261751836.1631.0.camel@localhost.localdomain> On Fri, 2009-12-25 at 13:28 +0100, Pascal de Bruijn wrote: > On Fri, Dec 25, 2009 at 12:28 PM, Richard Hughes wrote: > > 2009/12/24 Bruce Cowan : > >> I've found a few string errors. Here's a patch to fix them. I thought I > >> better post it here rather than just commit straight to the repository, > >> because there is a potential point of disagreement. > > > > The first two are obvious, so please commit those. > > > >> I've changed the strformat format to the en_US one, which could be a bit > >> contentious. > > > > I agree it's probably the correct thing to do, but it does however > > want me to rip my eyes out every time I see that format. :-) > > /me Shudders... > > We're introducing AM/PM time... oh well seems correct for the en_US locale... > > However, if one we're to switch to the en_UK locale, would AM/PM time > be gone again? Yup, I'll update the en_GB translation later as well, meaning we'll see 24 hour time still. -- Bruce Cowan From joedalton2 at yahoo.dk Fri Dec 25 17:59:33 2009 From: joedalton2 at yahoo.dk (Joe Dalton) Date: Fri, 25 Dec 2009 17:59:33 +0000 (GMT) Subject: bugreport gnome-color-manager translation In-Reply-To: <15e53e180912170131p5c35354dv6e3df58be1ee6782@mail.gmail.com> Message-ID: <381861.61449.qm@web24403.mail.ird.yahoo.com> thanx --- Den tors 17/12/09 skrev Richard Hughes : > Fra: Richard Hughes > Emne: Re: bugreport gnome-color-manager translation > Til: "Joe Dalton" > Cc: gnome-color-manager-list at gnome.org > Dato: torsdag 17. december 2009 10.31 > 2009/12/16 Joe Dalton : > > Sorry, either the product gnome-color-manager ?does > not exist or you aren't authorized to enter a bug into it. > Please choose a valid product. > > I've opened a bug to have the bugzilla product created: > https://bugzilla.gnome.org/show_bug.cgi?id=604811 > > > What does intent mean here. Is the sentence broken in > some way/missing something? > > > > #: ../data/gnome-color-manager.schemas.in.h:7 > > msgid "The output intent to use, e.g. 'perceptual'." > > The output intent is the way that one range of colors is > mapped to > another range, for instance, do we clip the extremes of > color, or do > we try to scale it so they are all represented. Intent in > English > means "what you would do if possible" although in this > technical speak > it means more of "what method do you want me to use when > transforming > the color" > > > Hi, a google did not help me, and i have only found > one danish translation, which maybe is correct mayby not. > Please give some more info. > > #. TRANSLATORS: rendering intent: you probably want to > google this > > #: ../src/gcm-prefs.c:2029 > > msgid "Perceptual" > > This is how I understand perceptual rendering intent: > > '''This rendering intent maps color "smoothly", preserving > relationships between similar colors. This prevents "gamut > clipping" > with its potential loss of detail and "tonal banding" > problems. Gamut > clipping occurs when colors that are different in the input > image > appear the same when printed. Perceptual rendering intent > makes small > compromises throughout the entire color space in order to > preserve > color relationships. It sacrifices some precision of > in-gamut colors > in order to ensure pleasing results.''' > > Does that make things clearer? Feel free to change the > translator > comments to something more sensible -- you know what you > are doing. > :-) > > Richard. > Find din nye laptop p? kelkoo.dk. Se de gode tilbud her - http://dk.yahoo.com/r/pat/mm From pmjdebruijn at pcode.nl Sun Dec 27 18:33:43 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 27 Dec 2009 19:33:43 +0100 Subject: Patch: Debian binary detection fix In-Reply-To: <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> Message-ID: <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> On Mon, Dec 14, 2009 at 4:13 PM, Richard Hughes wrote: > 2009/12/14 Pascal de Bruijn : >> Maybe it's a thought to (last) search /usr/local/bin as well for the >> standard binary names? > > Sure. I'll do a deal: You code it, and I'll apply it. If you need > instructions, I can explain how. I've submitted it has an enhancement on GNOME Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=605552 Regards, Pascal de Bruijn From hughsient at gmail.com Mon Dec 28 10:26:15 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 28 Dec 2009 10:26:15 +0000 Subject: Patch: Debian binary detection fix In-Reply-To: <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> Message-ID: <15e53e180912280226r2b1d3265x103a12fd65174192@mail.gmail.com> 2009/12/27 Pascal de Bruijn : > I've submitted it has an enhancement on GNOME Bugzilla: Committed, thanks. Richard. From pmjdebruijn at pcode.nl Mon Dec 28 18:20:54 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 28 Dec 2009 19:20:54 +0100 Subject: Documentation Images: Display OSD Message-ID: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Hi, I took a few shots of my display's OSD, mainly for my own purposes, but maybe they'll come in handy for GNOME Color Manager's documentation. These shots aren't perfect (at all), but it's a start. I can put these under any license you like... I'm fine with GPL and CC-BY-AC. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: lowres-osd-jpegs.zip Type: application/zip Size: 213749 bytes Desc: not available URL: From niklas.park at gmail.com Tue Dec 29 10:44:50 2009 From: niklas.park at gmail.com (Niklas Park) Date: Tue, 29 Dec 2009 11:44:50 +0100 Subject: Documentation Images: Display OSD In-Reply-To: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> References: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Message-ID: A little warning seems to be in place here. I can't see how you could put these screenshots under CC, FDL or any other license, since the copyright clearly belongs to the manufacturer of the device, or the person or entity the manufacturer licenses the OSD from. I can't really see that a screengrab could constitute original art under any jurisdiction. The best one could hope for is that using these images constitutes fair use, of course with correct attribution to the copyright holder, but IANAL. Kindly /Niklas Park 2009/12/28 Pascal de Bruijn > Hi, > > I took a few shots of my display's OSD, mainly for my own purposes, > but maybe they'll come in handy for GNOME Color Manager's > documentation. > > These shots aren't perfect (at all), but it's a start. > > I can put these under any license you like... I'm fine with GPL and > CC-BY-AC. > > Regards, > Pascal de Bruijn > > _______________________________________________ > gnome-color-manager-list mailing list > gnome-color-manager-list at gnome.org > http://mail.gnome.org/mailman/listinfo/gnome-color-manager-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pmjdebruijn at pcode.nl Tue Dec 29 14:41:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 29 Dec 2009 15:41:33 +0100 Subject: Documentation Images: Display OSD In-Reply-To: References: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Message-ID: <4c2c37f90912290641w8d6dff1y675a3a07844da84d@mail.gmail.com> On Tue, Dec 29, 2009 at 11:44 AM, Niklas Park wrote: > A little warning seems to be in place here. > I can't see how you could put these screenshots under CC, FDL or any other > license, since the copyright clearly belongs to the manufacturer of the > device, or the person or entity the manufacturer licenses the OSD from. I > can't really see that a screengrab could constitute original art under any > jurisdiction. Well, you've got a fair point there... In this case that would be HP... > The best one could hope for is that using these images constitutes fair use, > of course with correct attribution to the copyright holder, but IANAL. Since it's not a 1:1 (pixel-perfect) copy, and doesn't contain any brand logo's, it probably does... Regards, Pascal de Bruijn From hughsient at gmail.com Tue Dec 1 14:58:41 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 1 Dec 2009 14:58:41 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file Message-ID: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Attached is the new widget I've just added to gnome-color-manager git master. It allows us to show the user a preview of the gamut range of the profile in a graphical way, much like OSX does. You should be able to see the lightly shaded section. At the moment I've hardcoded the CIE REC 709 x-y values, but of course it's easy to change when put into GcmPrefs. So, the point of this email: I wanted to get the CIE x and y coordinates (or XYZ of course) of the three primary illuminants and the x and y coordinates of the white point from the device ICC profile. Is there an easy way to extract these end-point values from the ICC file? I guess mediaWhitePointTag is the latter half of the problem, but I'm not sure how to deal with the LUT values. Ideas welcome. Thanks. Richard. -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot.png Type: image/png Size: 49528 bytes Desc: not available URL: From lars.tore at mulebakken.net Tue Dec 1 16:43:31 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Tue, 1 Dec 2009 17:43:31 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912010843re5f1d81mef40fd5d57e50310@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: > Attached is the new widget I've just added to gnome-color-manager git > master. It allows us to show the user a preview of the gamut range of > the profile in a graphical way, much like OSX does. > > You should be able to see the lightly shaded section. At the moment > I've hardcoded the CIE REC 709 x-y values, but of course it's easy to > change when put into GcmPrefs. > > So, the point of this email: I wanted to get the CIE x and y > coordinates (or XYZ of course) of the three primary illuminants and > the x and y coordinates of the white point from the device ICC > profile. > > Is there an easy way to extract these end-point values from the ICC > file? I guess mediaWhitePointTag is the latter half of the problem, > but I'm not sure how to deal with the LUT values. Ideas welcome. > Thanks. > > Richard. > What about iccdump -v3 -t wtpt -t bkpt -t rXYZ -t gXYZ -t bXYZ myprofile.icc Lars Tore Gustavsen From lars.tore at mulebakken.net Tue Dec 1 18:12:38 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Tue, 1 Dec 2009 19:12:38 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912011012q42a44f4cvd385a8b97ca5f5eb@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: I'm not sure how to deal with the LUT values. Ideas welcome. > Thanks. > > Richard. > > _ I think I missed the last part. Can something like this be used? Just thinking loud here: echo 1 0 0|icclu -ff lutprofile.icc echo 0 1 0|icclu -ff lutprofile.icc echo 0 0 1|icclu -ff lutprofile.icc Lars Tore Gustavsen From graeme2 at argyllcms.com Wed Dec 2 00:19:40 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Wed, 02 Dec 2009 11:19:40 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <4B15B29C.2090105@argyllcms.com> Richard Hughes wrote: > Attached is the new widget I've just added to gnome-color-manager git > master. It allows us to show the user a preview of the gamut range of > the profile in a graphical way, much like OSX does. Professor Hunt would be jumping up and down in the first row at this point saying "Why are you using an XYZ chromaticity diagram ?" :-) :-) [Those who have been to the Color Imaging Conference will understand this.] Actually if you're wanting to plot Yxy values then it's appropriate, but I'll note that it's also very misleading in indicating gamut, since an xy diagram is highly visually non-uniform. A u' v' plot is a better choice to illustrate gamut, while still being linearly additive. > So, the point of this email: I wanted to get the CIE x and y > coordinates (or XYZ of course) of the three primary illuminants and > the x and y coordinates of the white point from the device ICC > profile. > > Is there an easy way to extract these end-point values from the ICC > file? I guess mediaWhitePointTag is the latter half of the problem, > but I'm not sure how to deal with the LUT values. Ideas welcome. The white point is in the white point tag (but see my final note). For a matrix profile the D50 adapted primaries are in the red/green/blue ColorantTags. For a cLut based profile, you would have to (as Lars Tore Gustavsen suggested) run the primary values through the profile (or do the equivalent of looking up what the cLUT values are at the primaries). Note a complication - you probably want the absolute colorant values, not the D50 adapted white ones. You need to undo the chromatic adaptation applied to the colorant tags, and/or do the lookup in absolute colorimetric intent mode. Note also that while absolute colorimetric interpretation of display profiles is perfectly consistent using Argyll/icclib based tools (and is consistent with the original sRGB and current AdobeRGB profiles), it probably will not be so for all other libaries and tools and profiles. This is due to a disagreement about the interpretation and intent of the ICC specification in this regard. Some display profiles have a white point tag of D50 even though this is not the white point of the display, and encode the absolute<->relative transformation in the ChromaticAdapation tag, meaning that the absolute white point and colorant values are not obtained using a normal absolute colorimetric lookup on these profiles. This latter interpretation is the one that has been adopted in the changes made for ICC V4 profiles, even though it is out of sync with much original ICC V2 practice. Graeme Gill. From hughsient at gmail.com Wed Dec 2 08:37:43 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 08:37:43 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B15B29C.2090105@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> Message-ID: <15e53e180912020037o6168de9dm4a20f429e7d665c@mail.gmail.com> 2009/12/2 Graeme Gill : > Actually if you're wanting to plot Yxy values then it's appropriate, > but I'll note that it's also very misleading in indicating gamut, > since an xy diagram is highly visually non-uniform. A u' v' plot > is a better choice to illustrate gamut, while still being linearly > additive. Sure, converting the code to plot u' v' should be pretty easy. The graph is really just for a passing glance, not for any serious comparison. > This is due to a disagreement about the interpretation and intent > of the ICC specification in this regard. Some display profiles > have a white point tag of D50 even though this is not the white > point of the display, and encode the absolute<->relative > transformation in the ChromaticAdapation tag, meaning that > the absolute white point and colorant values are not obtained > using a normal absolute colorimetric lookup on these profiles. This is exactly the information I needed, thanks very much for such a comprehensive answer. Richard. From hughsient at gmail.com Wed Dec 2 15:12:39 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 15:12:39 +0000 Subject: Uncaught bounce notification In-Reply-To: References: Message-ID: <15e53e180912020712l4506d8e1ja991ca132c63ae85@mail.gmail.com> 2009/12/2 : > i'v followed g-c-m using Pascal de Bruijn ppa (thanks to him), and for > these tests I'v built myself the last git (for CIE widget). For some reason your mail got bounced, apologies. I've forwarded it to the list. > So : > 1/ The window for listing screens are too big looking at the rest > (left/right panel) That should be fixed in git master now. > 2/ It's ok that this tool must be kept as simple as possible, but I > would love to have the possibility to set white K? That's on the assumption you're not using a hardware calibration device then? > 3/ if you play with icc profils, size of the right panel change and the > square of CIE widget may be crushed > 4/ I would love to see more (bigger/more details) the CIE widget when > clicking on it Sure, I'm not completely happy with the location or size of this widget. It seems to me I'm trying to cram the "I want to compare different profiles" and "assign a profile with a device" functionality into the single tab. maybe we just need to break out another tab called "Profiles" and list all the profiles along with some more information about them. In this way we can get rid of some of the clutter in the first page. Comments welcome. Richard. From ubuntu at kagou.fr Wed Dec 2 16:20:03 2009 From: ubuntu at kagou.fr (Patrice Vetsel) Date: Wed, 02 Dec 2009 17:20:03 +0100 Subject: Some tests Message-ID: <4B1693B3.8000809@kagou.fr> Hi, i'v followed g-c-m using Pascal de Bruijn ppa (thanks to him), and for these tests I'v built myself the last git (for CIE widget). So : 1/ The window for listing screens are too big looking at the rest (left/right panel) 2/ It's ok that this tool must be kept as simple as possible, but I would love to have the possibility to set white K? 3/ if you play with icc profils, size of the right panel change and the square of CIE widget may be crushed 4/ I would love to see more (bigger/more details) the CIE widget when clicking on it Thanks for this tool Regards Patrice Vetsel Aka Kagou From hughsient at gmail.com Wed Dec 2 16:47:21 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 16:47:21 +0000 Subject: Some tests In-Reply-To: <4B1693B3.8000809@kagou.fr> References: <4B1693B3.8000809@kagou.fr> Message-ID: <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> 2009/12/2 Patrice Vetsel : > 4/ I would love to see more (bigger/more details) the CIE widget when > clicking on it What do you all think of the attached mockups? Richard. -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-1.png Type: image/png Size: 42154 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot-2.png Type: image/png Size: 86945 bytes Desc: not available URL: From pmjdebruijn at pcode.nl Wed Dec 2 17:09:30 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:09:30 +0100 Subject: Some tests In-Reply-To: <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> Message-ID: <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> On Wed, Dec 2, 2009 at 5:47 PM, Richard Hughes wrote: > 2009/12/2 Patrice Vetsel : >> 4/ I would love to see more (bigger/more details) the CIE widget when >> clicking on it > > What do you all think of the attached mockups? @Richard: Looks pretty good... The CIE widget is very nice. Though the profile whitepoint isn't being displayed yet, I'm assuming this is still on the TODO list. I also agree it would be nice (but not critical) if double clicking on the CIE widget would create a dialog with just the profile name and the CIE widget, for detailed viewing. And dare I ask, possibly with a right-click "Save (CIE diagram) as PNG/JPEG/PDF" option? Also, I notice the CIE widget has two "squares" of "space" to the left of the CIE diagram. Does this have a purpose? Otherwise couldn't this be decreased to a single white square. I also think either centering the CIE widget, or placing the buttons to the right would look a bit "better" esthetically... Or maybe move the button onder the list view? Giving the CIE diagram a bit more room to "grow"? @Patrice: I rebuild my PPA between 1 to 4 times per two days, depending on how interesting Richard's changes are... All in all it's pretty hard to keep up, and he's doing all the hard work :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 2 17:15:31 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:15:31 +0100 Subject: Idea: GTK/Cario CIE widget as a seperate library Message-ID: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> Hi, The following is just a thought, and is by no means very important... The GTK/Cairo CIE widget written for GCM is beginning to kick ass... However I guess it's GCM specific. Maybe for the long term it's a thought to separate it into a small library, so other apps can use it as well? For example whenever GIMP/UFraw open a dialog to import a new profile it would be very cool if the GNOME File Chooser would "preview" the profile by showing a CIE diagram. Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 2 17:35:04 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 17:35:04 +0000 Subject: Idea: GTK/Cario CIE widget as a seperate library In-Reply-To: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> References: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> Message-ID: <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > The GTK/Cairo CIE widget written for GCM is beginning to kick ass... > However I guess it's GCM specific. Not really, it's a standard GObject so it could be reused. > Maybe for the long term it's a thought to separate it into a small > library, so other apps can use it as well? Sure, there is loads of other common code that might be useful, but I don't think the project is at any level of internal ABI stability. Maybe we can revisit this idea in a few months time? Richard. From pmjdebruijn at pcode.nl Wed Dec 2 17:36:50 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:36:50 +0100 Subject: Profile description for laptop displays In-Reply-To: <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> References: <4c2c37f90911290453x143c6820nb27f3a6b6188e3c7@mail.gmail.com> <15e53e180911291342o2b8755aavad083da3dcd849d2@mail.gmail.com> <4c2c37f90911291527v71d2cf9cy6bb41add8c7b3683@mail.gmail.com> <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> Message-ID: <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> On Mon, Nov 30, 2009 at 11:10 AM, Richard Hughes wrote: > 2009/11/29 Pascal de Bruijn : >> I noticed you included the product_version as well, at least on my >> laptop it's actually the BIOS revision, I'm not sure what this is on >> other laptops. It's probably not that relevant... > > Hmm, it seems the manufacturers were in a rush when they designed your > laptop! The bios version, logically, belongs in bios_version :-) > > Could you attach the output of "cat /sys/class/dmi/id/*" and we'll try > and work around bios issues like this. Thanks. It's attached. >> I noticed you're currently filling the ICC Make with the username, it >> seems rather redundant with the copyright. Besides the point that the >> field wasn't ment to be used like that. I'd stick the EDID info there >> (even with laptops). For the Model field as well, EDID is most likely >> best and most accurate here. > > I deliberately changed it, as the vendor is supposed to be the person > or company that designed or created the profile. In this sense, it > seemed wrong to say that the profile was created "by Lenovo" when it > was created by me, for my Lenovo display. The ICCv2 specification disagrees: 6.4.15 deviceMfgDescTag Tag Type: textDescriptionType Tag Signature: ?dmnd? (646D6E64h) Structure containing invariant and localizable versions of the device manufacturer for display. The content of this structure is described in 6.5.17. 6.4.16 deviceModelDescTag Tag Type: textDescriptionType Tag Signature: ?dmdd? (646D6464h) Structure containing invariant and localizable versions of the device model for display. The content of this structure is described in 6.5.17. >> For example: >> >> Make: SEC (don't lookup in pnp.ids for storage in the ICC profile, >> only lookup in the gcm-prefs GUI). >> Model: 154AT07-H01 >> >> Storing the EDID info as verbatim as possible in the ICC make/models >> fields, will make it easier to auto match displays and profiles in the >> future. > > Hmm, I think the fields have to be displayable. We need to look at > adding private fields if we want to add any machine parsable data into > the ICC file. > >> Storing the EDID verbatim without looking up in pnp.ids prevents false >> data from conflicting lookups (like my Samsung/Seiko issue) from >> ending up in the profile itself. > > I think the lookup is valuable as it works most of the time. Your > panel should have encoded SAM into the PNP field, not SEC, unless of > course the panel really is a Seiko panel that's been rebadged as a > Samsung. For what it's worth, my nice expensive LG panel is actually a > generic Goldstar chipset with LG additions and plastic wrapping. > Hence, it shows up as "Goldstar" as the vendor. :-( I'm not saying my laptop's EDID info is not semi-broken... But translating SEC into Seiko for embedding in the profile, is moving from semi-wrong to completely-wrong... And I'm afraid my laptop isn't the only one having this or similar issues. I do wonder how Samsung's own brand sold displays identify themselves. I wish we had some more EDID samples to compare... Maybe a call for sample EDID data on Planet GNOME would get us some more samples? Regards, Pascal de Bruijn -------------- next part -------------- /sys/class/dmi/id/bios_date 03/12/2009 /sys/class/dmi/id/bios_vendor Hewlett-Packard /sys/class/dmi/id/bios_version 68PZU Ver. F.0C /sys/class/dmi/id/board_asset_tag /sys/class/dmi/id/board_name 30E8 /sys/class/dmi/id/board_serial /sys/class/dmi/id/board_vendor Hewlett-Packard /sys/class/dmi/id/board_version KBC Version 94.1C /sys/class/dmi/id/chassis_asset_tag CNU9100JZH /sys/class/dmi/id/chassis_serial CNU9100JZH /sys/class/dmi/id/chassis_type 10 /sys/class/dmi/id/chassis_vendor Hewlett-Packard /sys/class/dmi/id/chassis_version /sys/class/dmi/id/modalias dmi:bvnHewlett-Packard:bvr68PZUVer.F.0C:bd03/12/2009:svnHewlett-Packard:pnHPCompaq6730s:pvrF.0C:rvnHewlett-Packard:rn30E8:rvrKBCVersion94.1C:cvnHewlett-Packard:ct10:cvr: /sys/class/dmi/id/power cat: /sys/class/dmi/id/power: Is a directory /sys/class/dmi/id/product_name HP Compaq 6730s /sys/class/dmi/id/product_serial CNU9100JZH /sys/class/dmi/id/product_uuid 4377676E-D605-DE11-B150-80487F19A84A /sys/class/dmi/id/product_version F.0C /sys/class/dmi/id/subsystem cat: /sys/class/dmi/id/subsystem: Is a directory /sys/class/dmi/id/sys_vendor Hewlett-Packard /sys/class/dmi/id/uevent MODALIAS=dmi:bvnHewlett-Packard:bvr68PZUVer.F.0C:bd03/12/2009:svnHewlett-Packard:pnHPCompaq6730s:pvrF.0C:rvnHewlett-Packard:rn30E8:rvrKBCVersion94.1C:cvnHewlett-Packard:ct10:cvr: From pmjdebruijn at pcode.nl Wed Dec 2 17:46:29 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 18:46:29 +0100 Subject: Idea: GTK/Cario CIE widget as a seperate library In-Reply-To: <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> References: <4c2c37f90912020915p5c0cfb8atcc1f91bfd646016@mail.gmail.com> <15e53e180912020935p5854522bj61ccb699879f479a@mail.gmail.com> Message-ID: <4c2c37f90912020946r1a55d1afi1f7752bae2d80f58@mail.gmail.com> On Wed, Dec 2, 2009 at 6:35 PM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> The GTK/Cairo CIE widget written for GCM is beginning to kick ass... >> However I guess it's GCM specific. > > Not really, it's a standard GObject so it could be reused. Nice. >> Maybe for the long term it's a thought to separate it into a small >> library, so other apps can use it as well? > > Sure, there is loads of other common code that might be useful, but I > don't think the project is at any level of internal ABI stability. > Maybe we can revisit this idea in a few months time? Of course, like I said, it's just a thought. Rushing it won't make it better :) Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 2 19:47:00 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 19:47:00 +0000 Subject: Some tests In-Reply-To: <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> Message-ID: <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > The CIE widget is very nice. Though the profile whitepoint isn't being > displayed yet, I'm assuming this is still on the TODO list. I've just added this to git master. > I also agree it would be nice (but not critical) if double clicking on > the CIE widget would create a dialog with just the profile name and > the CIE widget, for detailed viewing. And dare I ask, possibly with a > right-click "Save (CIE diagram) as PNG/JPEG/PDF" option? Sure, we could do that. I'm tempted to leave that feature until the CIE widget displays more useful data. > Also, I notice the CIE widget has two "squares" of "space" to the left > of the CIE diagram. Does this have a purpose? Otherwise couldn't this > be decreased to a single white square. Sure, we can scale this later. > @Patrice: I rebuild my PPA between 1 to 4 times per two days, > depending on how interesting Richard's changes are... All in all it's > pretty hard to keep up, and he's doing all the hard work :) I'll take that as a compliment. After Monday development will slow quite a bit, as I've got a ton of other stuff to do. Richard. From lars.tore at mulebakken.net Wed Dec 2 20:29:04 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Wed, 2 Dec 2009 21:29:04 +0100 Subject: Some tests In-Reply-To: <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> Message-ID: <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> On Wed, Dec 2, 2009 at 8:47 PM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> The CIE widget is very nice. Though the profile whitepoint isn't being >> displayed yet, I'm assuming this is still on the TODO list. > > I've just added this to git master. > It works for matrix profiles. Lut profiles are already displayed black so it is not possibly to see the tiny cross. See attachment. Lars Tore Gustavsen -------------- next part -------------- A non-text attachment was scrubbed... Name: whitepoint-lut.png Type: image/png Size: 40793 bytes Desc: not available URL: From hughsient at gmail.com Wed Dec 2 20:57:23 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 2 Dec 2009 20:57:23 +0000 Subject: Some tests In-Reply-To: <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> Message-ID: <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> 2009/12/2 Lars Tore Gustavsen : > It works for matrix profiles. Lut profiles are already displayed black > so it is not possibly to see the tiny cross. Sure, we need to add support for LUT profiles as described by Graeme. If you're feeling handy I would love a patch, although I can give this a go next week if nobody starts it before me. Richard. From pmjdebruijn at pcode.nl Wed Dec 2 22:06:57 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 23:06:57 +0100 Subject: Some tests In-Reply-To: <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> Message-ID: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> On Wed, Dec 2, 2009 at 9:57 PM, Richard Hughes wrote: > 2009/12/2 Lars Tore Gustavsen : >> It works for matrix profiles. Lut profiles are already displayed black >> so it is not possibly to see the tiny cross. > > Sure, we need to add support for LUT profiles as described by Graeme. > If you're feeling handy I would love a patch, although I can give this > a go next week if nobody starts it before me. LUT profiles have a normal white point defined, I think Lars ment, it would be nice to invert the color of the whitepoint cross, so it's visible on the black CIE diagram. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 2 22:32:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 2 Dec 2009 23:32:01 +0100 Subject: GCM git fails to build Message-ID: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> Hi, I just tried to build GCM from git on Ubuntu Karmic, and it failed: cc1: warnings being treated as errors gcm-profile.c: In function ?gcm_profile_parse?: gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? from incompatible pointer type /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? but argument is of type ?guint *? make[4]: *** [libgcmshared_a-gcm-profile.o] Error 1 make[4]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager/src' make[3]: *** [all] Error 2 make[3]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager/src' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager' make[1]: *** [all] Error 2 make[1]: Leaving directory `/home/pmjdebruijn/Packages/Karmic/gcm/build/gnome-color-manager' make: *** [debian/stamp-makefile-build] Error 2 dpkg-buildpackage: error: debian/rules build gave error exit status 2 Regards, Pascal de Bruijn From lars.tore at mulebakken.net Thu Dec 3 06:19:58 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 07:19:58 +0100 Subject: Some tests In-Reply-To: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> Message-ID: <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> On Wed, Dec 2, 2009 at 11:06 PM, Pascal de Bruijn wrote: > LUT profiles have a normal white point defined, I think Lars ment, it > would be nice to invert the color of the whitepoint cross, so it's > visible on the black CIE diagram. > Yep, I meant so. Anyway I like the new tabbed interface. Since we now have the profile size displayed in kb, I think it would be of great benefit to also show the profiles filename. Regards Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 08:03:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 08:03:40 +0000 Subject: GCM git fails to build In-Reply-To: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> Message-ID: <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > cc1: warnings being treated as errors > gcm-profile.c: In function ?gcm_profile_parse?: > gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? > from incompatible pointer type > /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? > but argument is of type ?guint *? Should be fixed in git, thanks. Richard. From hughsient at gmail.com Thu Dec 3 08:04:37 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 08:04:37 +0000 Subject: Some tests In-Reply-To: <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> Message-ID: <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> 2009/12/3 Lars Tore Gustavsen : > Anyway I like the new tabbed interface. Since we now have the profile > size displayed in kb, I think it would be of great benefit to also > show the profiles filename. Filename or filename with path? The latter is quite long. Richard. From lars.tore at mulebakken.net Thu Dec 3 09:12:46 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 10:12:46 +0100 Subject: Some tests In-Reply-To: <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> Message-ID: <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> On Thu, Dec 3, 2009 at 9:04 AM, Richard Hughes wrote: > 2009/12/3 Lars Tore Gustavsen : >> Anyway I like the new tabbed interface. Since we now have the profile >> size displayed in kb, I think it would be of great benefit to also >> show the profiles filename. > > Filename or filename with path? The latter is quite long. > > Richard. > Filename would be nice. I also thinks that full path will be to long for the GUI. Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 10:07:01 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 10:07:01 +0000 Subject: Some tests In-Reply-To: <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> Message-ID: <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> 2009/12/3 Lars Tore Gustavsen : > Filename would be nice. I also thinks that full path will be to long > for the GUI. commit 813c39747c4a0c01a5ee97842e32ab38ea5d71fc Author: Richard Hughes Date: Thu Dec 3 10:06:08 2009 +0000 Add the filename basename to the profile view Richard. From lars.tore at mulebakken.net Thu Dec 3 10:15:16 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 3 Dec 2009 11:15:16 +0100 Subject: Some tests In-Reply-To: <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> <58497f010912022219t321739b8l4c44f91d2f46616a@mail.gmail.com> <15e53e180912030004q62923173qaefa1b0ff631070d@mail.gmail.com> <58497f010912030112m75be866bld94325ac5344ef5e@mail.gmail.com> <15e53e180912030207pcd54902xc766b05d8d8b95a@mail.gmail.com> Message-ID: <58497f010912030215n7a5e9ed6i8234bce4b3d54a6b@mail.gmail.com> On Thu, Dec 3, 2009 at 11:07 AM, Richard Hughes wrote: > 2009/12/3 Lars Tore Gustavsen : >> Filename would be nice. I also thinks that full path will be to long >> for the GUI. > > commit 813c39747c4a0c01a5ee97842e32ab38ea5d71fc > Author: Richard Hughes > Date: ? Thu Dec 3 10:06:08 2009 +0000 > > ? ?Add the filename basename to the profile view > > Richard. > Excellent, and it works. Lars Tore Gustavsen From hughsient at gmail.com Thu Dec 3 11:42:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 11:42:49 +0000 Subject: Some tests In-Reply-To: <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> References: <4B1693B3.8000809@kagou.fr> <15e53e180912020847h1a9b327bm41f611b4fb56beb6@mail.gmail.com> <4c2c37f90912020909p4a67185apa2e44301b45e49e9@mail.gmail.com> <15e53e180912021147qc4c140ct863f91e041d67643@mail.gmail.com> <58497f010912021229y48044902w84daf79a1961a8a1@mail.gmail.com> <15e53e180912021257j6a7e9195o99999ce57280cfc5@mail.gmail.com> <4c2c37f90912021406u4d958c9bx1ad7935ae8cc7ca7@mail.gmail.com> Message-ID: <15e53e180912030342q2bde3d13ncb616f24c7a3617f@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > LUT profiles have a normal white point defined, I think Lars ment, it > would be nice to invert the color of the whitepoint cross, so it's > visible on the black CIE diagram. commit 6c77c29086ab851bec34c5667d0963bb32ec9217 Author: Richard Hughes Date: Thu Dec 3 11:41:21 2009 +0000 Show the white point cross in white if the CIE chart is all black Richard. From pmjdebruijn at pcode.nl Thu Dec 3 18:31:29 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 19:31:29 +0100 Subject: GCM git fails to build In-Reply-To: <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> Message-ID: <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> On Thu, Dec 3, 2009 at 9:03 AM, Richard Hughes wrote: > 2009/12/2 Pascal de Bruijn : >> cc1: warnings being treated as errors >> gcm-profile.c: In function ?gcm_profile_parse?: >> gcm-profile.c:868: error: passing argument 3 of ?g_file_get_contents? >> from incompatible pointer type >> /usr/include/glib-2.0/glib/gfileutils.h:89: note: expected ?gsize *? >> but argument is of type ?guint *? > > Should be fixed in git, thanks. Indeed, it's fixed. Some minor trivia: it seems the above only broke on amd64, not on i386... Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 3 18:34:48 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 19:34:48 +0100 Subject: Camera input profiling make/model handling Message-ID: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> Hi, I haven't had time yet to properly check this, but what is used by GCM for make/model info when profiling cameras/scanners? At least for camera's it seems wise to use Make/Model from EXIF. This way, software like UFRaw and DarkTable (and others) can use the EXIF make/model to match which profiles should be valid for a certain RAW file... Luckily EXIF is much better human readable (and generally less ambiguous) than EDID... Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 3 19:02:31 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:02:31 +0000 Subject: GCM git fails to build In-Reply-To: <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> References: <4c2c37f90912021432p4a36681fxca3dee7457888e0f@mail.gmail.com> <15e53e180912030003o7014f147kad993ae3ea9d3a35@mail.gmail.com> <4c2c37f90912031031m7a4b75efu1858c5876307a3df@mail.gmail.com> Message-ID: <15e53e180912031102w2c2e16dcmadc15da0983fb24c@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > Some minor trivia: it seems the above only broke on amd64, not on i386... Right, gsize is a different memory size to guint on 64bit architectures. Richard. From hughsient at gmail.com Thu Dec 3 19:09:22 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:09:22 +0000 Subject: Camera input profiling make/model handling In-Reply-To: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> References: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> Message-ID: <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > I haven't had time yet to properly check this, but what is used by GCM > for make/model info when profiling cameras/scanners? Information from sysfs. > At least for camera's it seems wise to use Make/Model from EXIF. This > way, software like UFRaw and DarkTable (and others) can use the EXIF > make/model to match which profiles should be valid for a certain RAW > file... GCM doesn't treat the device like a storage device, it treats it as a USB endpoint device, which means we don't look at the files contained within. > Luckily EXIF is much better human readable (and generally less > ambiguous) than EDID... I think this is the wrong layer. What happens if the EXIF data for one file on the "device" has different colorspaces or even two different device models? The device can carry all number of different EXIF files, but the device can only have one profile. If you inserted a pendrive of photos, would you expect to assign a per-pendrive ICC profile? I think it's sane for the end application to open the file with EXIF metadata and then decided what to do. If it wants to just use the embedded profile, that's fine, or it can ask gcm for the default device profile from the volume (which it can get from udev). Richard. From info at justinsseasonaltreasures.com Thu Dec 3 18:20:37 2009 From: info at justinsseasonaltreasures.com (Unknown) Date: Thu, 03 Dec 2009 10:20:37 -0800 Subject: Support for B9100 Message-ID: <1259864437.2325.1.camel@localhost> Will there be support for color management of the B9100? Thanks, Jazbo From hughsient at gmail.com Thu Dec 3 19:30:15 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 3 Dec 2009 19:30:15 +0000 Subject: Support for B9100 In-Reply-To: <1259864437.2325.1.camel@localhost> References: <1259864437.2325.1.camel@localhost> Message-ID: <15e53e180912031130k38517da1j65cc586154c34625@mail.gmail.com> 2009/12/3 Unknown : > Will there be support for color management of the B9100? I assume you mean the HP printer. At the moment ArgyllCMS has some support for generating print profiles, but CUPS does not have a way to set the ICC profile for the device. This needs much more work (for the session CUPS process to talk to GCM and get the current profile for the device) before it will be a case of plug and play. Apple did a lot of the CUPS work, but we need to flesh that out for Linux. For using hplip it's a bit different, and we probably need to support that too. Either way, more help is really welcome. If you just want to assign a profile to a device, and then select that in an application, that should work now. Richard. From pmjdebruijn at pcode.nl Thu Dec 3 20:54:23 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 21:54:23 +0100 Subject: Camera input profiling make/model handling In-Reply-To: <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> References: <4c2c37f90912031034o2f67ef24had89227e68ef0201@mail.gmail.com> <15e53e180912031109x64d72a6cs498ded3a9fafa419@mail.gmail.com> Message-ID: <4c2c37f90912031254i1dc63397g70e35c11a902f5d2@mail.gmail.com> On Thu, Dec 3, 2009 at 8:09 PM, Richard Hughes wrote: > 2009/12/3 Pascal de Bruijn : >> I haven't had time yet to properly check this, but what is used by GCM >> for make/model info when profiling cameras/scanners? > > Information from sysfs. Maybe I'll test this over the weekend... >> At least for camera's it seems wise to use Make/Model from EXIF. This >> way, software like UFRaw and DarkTable (and others) can use the EXIF >> make/model to match which profiles should be valid for a certain RAW >> file... > > GCM doesn't treat the device like a storage device, it treats it as a > USB endpoint device, which means we don't look at the files contained > within. > >> Luckily EXIF is much better human readable (and generally less >> ambiguous) than EDID... > > I think this is the wrong layer. What happens if the EXIF data for one > file on the "device" has different colorspaces or even two different > device models? The device can carry all number of different EXIF > files, but the device can only have one profile. If you inserted a > pendrive of photos, would you expect to assign a per-pendrive ICC > profile? Huh? A "Device" does not really have different color spaces... a camera's sensor (RAW file) has a native color space characterized by the profiling of the IT8 target (or another target, which is not really relevant, it's a means to an end). Any camera that produces sRGB or AdobeRGB, is likely to be not fit to be really profiled. Since a camera most likely does not statically convert sensor data to sRGB/AdobeRGB, it's likely a dynamic (image dependant) process. Anyway I filed a feature request with UFRaw, for "special" profiling mode, to convert a camera RAW file an image with linear gamma in the native camera color space: http://sourceforge.net/tracker/?func=detail&aid=2841222&group_id=127649&atid=709089 If one would have a pen drive of photos, the pen drive would be completely and utterly irrelevant from a profiling/profile assignment perspective. Each and every file on the drive could require a different profile (at least for proper results). And normally EXIF could be reasonably reliably used to "match" that. > I think it's sane for the end application to open the file with EXIF > metadata and then decided what to do. If it wants to just use the > embedded profile, that's fine, or it can ask gcm for the default > device profile from the volume (which it can get from udev). Again, with images, only the image color space is truely relevant (embedded or not). However, RAW files do not have assigned (embedded) profiles. UFRaw/DCRaw "solves" this by ripping generic profiles (Color Matrix) from Adobe DNG SDK, and applying them on a by EXIF make/model basis. Please do note, the above applies to camera's. I'm not really sure about scanners. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 3 22:23:48 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 23:23:48 +0100 Subject: ICC operating system identifiers Message-ID: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Hi, Currently there is no registered operating system identifier for Linux in general. Some examples: Microsoft 'msft' Apple 'appl' Sun Microsystems 'sunw' In the past I tried to mail the Linux Foundation about this, since they seem to be the best (least unfit) umbrella organisation to register such an identifier with the ICC. For example: Linux Foundation 'lnxf' However, it seems the mail was ignored. Richard, could you possibly attempt this, since you can mail from a gnome.org e-mail adres, the request might be more seriously considered. Or do you have other thoughts on how to orchestrate this? Regards, Pascal de Bruijn From anders at brander.dk Thu Dec 3 22:32:11 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 03 Dec 2009 23:32:11 +0100 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <1259879531.25715.4.camel@video64> Hi, On Thu, 2009-12-03 at 23:23 +0100, Pascal de Bruijn wrote: > Currently there is no registered operating system identifier for Linux > in general. Is this relevant in this day and age? Is there any CMM/CMS that does anything useful with this information? Regards, Anders Brander From pmjdebruijn at pcode.nl Thu Dec 3 22:46:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 3 Dec 2009 23:46:01 +0100 Subject: ICC operating system identifiers In-Reply-To: <1259879531.25715.4.camel@video64> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> <1259879531.25715.4.camel@video64> Message-ID: <4c2c37f90912031446p58a57498i6cf869fb254962a1@mail.gmail.com> On Thu, Dec 3, 2009 at 11:32 PM, Anders Brander wrote: > Hi, > > On Thu, 2009-12-03 at 23:23 +0100, Pascal de Bruijn wrote: >> Currently there is no registered operating system identifier for Linux >> in general. > > Is this relevant in this day and age? Is there any CMM/CMS that does > anything useful with this information? I think this remains relevant until the ICC officially deprecates the fields... Regards, Pascal de Bruijn From graeme2 at argyllcms.com Fri Dec 4 00:03:34 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 04 Dec 2009 11:03:34 +1100 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <4B1851D6.8080801@argyllcms.com> Pascal de Bruijn wrote: > For example: > Linux Foundation 'lnxf' Hi, note that Argyll uses '*nix' as an unofficial platform signature for Linux and other Unix like platforms. See also this thread on the OpenICC list: Graeme Gill. From hughsient at gmail.com Fri Dec 4 09:22:59 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 09:22:59 +0000 Subject: ICC operating system identifiers In-Reply-To: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> Message-ID: <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> 2009/12/3 Pascal de Bruijn : > Richard, could you possibly attempt this, since you can mail from a > gnome.org e-mail adres, the request might be more seriously > considered. I would suspect my @redhat email account might have more gravitas, although I'm not sure what it would achieve. Do programs actually do anything differently depending on the OSI information, or even display it at any point? If you think it may be useful, then I guess "*nix" makes most sense as Graeme has been using this for ages. If anything, Graeme's email would convey more weight than mine ever would. Richard. From graeme2 at argyllcms.com Fri Dec 4 11:20:47 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 04 Dec 2009 22:20:47 +1100 Subject: ICC operating system identifiers In-Reply-To: <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> References: <4c2c37f90912031423w7c969a5fu2ff7aae1f046f314@mail.gmail.com> <15e53e180912040122w7128849fu846b802004a99aab@mail.gmail.com> Message-ID: <4B18F08F.3080306@argyllcms.com> Richard Hughes wrote: > If you think it may be useful, then I guess "*nix" makes most sense as > Graeme has been using this for ages. If anything, Graeme's email would > convey more weight than mine ever would. The issues are the following: 1) The platform signature is not of great importance, and it's difficult to know what the original intention of it was. ICC profiles are typically not platform specific. It's lack of importance means that it might be difficult to excite anybodies interest in this issue. The only reason I added '*nix' to my code was an interest in not being misleading. My other alternatives were to use one of the other two Unix based platform signatures (SGI or Sun), or to set it to Unknown (0). 2) Most of the other ICC signatures are in registries, so it's not that bigger deal to ask someone to add a signature to a registry - a single person will have the authority to just do it (hence 'argl' added as a CMM signature). The platform signatures only appear in the ICC spec. though, and they also seem to correlate with the founding members (Although I notice that Taligent has been dropped between V2 an V4). So to add a new signature means someone within the ICC has to propose a spec. change and it then has to be approved by the members at some point ! 3) I'm not sure how much consensus there is about a signature that would cover Linux. My thought process was that the existing signatures were both too vague and too specific. They are really operating system vendors, not specific operating system platforms, and so are fairly loose (there is a noticeable difference between Win3.1 and WinNT, between OS9 and OS X). They are also too specific, in that Sun and SGI were both Unix based systems, although SGI has switch to Windows, so what does it mean ? If the signatures were just a registry, there wouldn't be any problem being very specific (Linux, FreeBSD, OpenBSD, RedHatLinux, etc. etc.) So I went for something that would cover cover Linux, BSD etc. and would be enough to distinguish it from Apple and Microsoft. As I suggested on the OpenICC list, if there is sufficient consensus then I'm happy to mail a couple of people and request that a new platform signature be put forward as a modification to the ICC spec., but on the other hand I won't be surprised if action on it is a little slow :-) cheers, Graeme Gill. From lars.tore at mulebakken.net Fri Dec 4 11:53:06 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 4 Dec 2009 12:53:06 +0100 Subject: More verbose output in calibration? Message-ID: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> May I suggest that when I calibrate my monitor the details view are more verbose than now (like dispcal -v). I guess a beginner will think this is not working at all, when it measures all the black patches and almost noting change on the display. I think more experienced users also like to see the progress in the calibration and the deltaE report at the end. In a longer time frame some progress bar (1) or progress window (2) would be very nice eye candy in the calibration process. 1 http://developer.gnome.org/projects/gup/hig/2.0/controls-progress-bars.html 2 http://developer.gnome.org/projects/gup/hig/2.0/windows-progress.html Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 4 14:51:47 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 15:51:47 +0100 Subject: Bug: Profiles tab is empty as long as no profiles have been assigned. Message-ID: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> Hi, I think I found a small bug. When starting gcm-prefs after a clean install, it does list all the available profiles in the profiles dropdown box on the Devices tab. But the list on the Profiles tab is completely empty. When I restart gcm-prefs the list is still empty. However, when I assign a profile (for example blueish) to the screen, it seems to change nothing at first glance. However after restarting gcm-prefs again, suddenly the dialog is bigger, and the list on the Profiles tabs has been filled. I haven't been able to test this, but when a new profile is created, does this get properly added the the list on the Profiles tab as well? Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Fri Dec 4 14:58:25 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 15:58:25 +0100 Subject: gcm-prefs on netbook Message-ID: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> Hi, I managed to test-drive gcm-prefs on a friends netbook (which runs Ubuntu Karmic Netbook Edition). And when the list on the Profiles tab is filled, the dialog quickly grows to big. I've attached a screenshot. I personally don't particularly like netbooks, but they are popular, and thus I think it might be nice for gcm-prefs to work well on netbooks too. I'm quite aware netbooks aren't particularly well suited for "graphics-work", but that's not really the point here. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: gcm-prefs-netbook.png Type: image/png Size: 83037 bytes Desc: not available URL: From hughsient at gmail.com Fri Dec 4 15:00:38 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:00:38 +0000 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> Message-ID: <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > I personally don't particularly like netbooks, but they are popular, > and thus I think it might be nice for gcm-prefs to work well on > netbooks too. What about if the window size < 1024x768 then make the cie widget smaller? Richard. From hughsient at gmail.com Fri Dec 4 15:18:18 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:18:18 +0000 Subject: Bug: Profiles tab is empty as long as no profiles have been assigned. In-Reply-To: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> References: <4c2c37f90912040651g11e1adfbq39b1fb2f5ac296b6@mail.gmail.com> Message-ID: <15e53e180912040718m678446b7s3dd3797494ac05e9@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > I think I found a small bug. When starting gcm-prefs after a clean > install, it does list all the available profiles in the profiles > dropdown box on the Devices tab. But the list on the Profiles tab is > completely empty. Ahh, thanks. commit 902fbf2f4271d6be79985d40338fcf4b35bbc2a1 Author: Richard Hughes Date: Fri Dec 4 15:16:37 2009 +0000 Don't skip phase2 startup if we can't load the config file :100644 100644 2726f42... 792b729... M src/gcm-prefs.c Richard. From hughsient at gmail.com Fri Dec 4 15:18:39 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 15:18:39 +0000 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> Message-ID: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> 2009/12/4 Richard Hughes : > What about if the window size < 1024x768 then make the cie widget smaller? commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 Author: Richard Hughes Date: Fri Dec 4 15:11:19 2009 +0000 Reduce the size of the ICC widget on small displays :100644 100644 4a1551d... 2726f42... M src/gcm-prefs.c Richard. From pmjdebruijn at pcode.nl Fri Dec 4 15:49:27 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 16:49:27 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <4c2c37f90912040749m6c8cdea1vb5c4aa14b43555@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays > > :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c Excellent. The CIE widget isn't the most import thing ever, however nice it may be. The user chose to have a small (low resolution) screen, so having the buttons visible is much more important. Regards, Pascal de Bruijn From lars.tore at mulebakken.net Fri Dec 4 17:16:10 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 4 Dec 2009 18:16:10 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays > > :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c > > Richard. On my 1280x1024 monitor the CIE diagram is quite small now. I can live with it, but I'm not sure if 1280x1024 was meant to be a small display? Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 4 17:18:10 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 4 Dec 2009 18:18:10 +0100 Subject: gcm-prefs on netbook In-Reply-To: <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> Message-ID: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> On Fri, Dec 4, 2009 at 6:16 PM, Lars Tore Gustavsen wrote: > On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: >> 2009/12/4 Richard Hughes : >>> What about if the window size < 1024x768 then make the cie widget smaller? >> >> commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 >> Author: Richard Hughes >> Date: ? Fri Dec 4 15:11:19 2009 +0000 >> >> ? ?Reduce the size of the ICC widget on small displays >> >> :100644 100644 4a1551d... 2726f42... M ?src/gcm-prefs.c >> >> Richard. > > On my 1280x1024 monitor the CIE diagram is quite small now. I can live > with it, but I'm not sure if 1280x1024 was meant to be a small > display? That wasn't intentional: + if (gdk_screen_get_width (screen) < 10240 || That's a typo :) Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Fri Dec 4 17:24:17 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 4 Dec 2009 20:24:17 +0300 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> Message-ID: <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> On Fri, Dec 4, 2009 at 8:18 PM, Pascal de Bruijn wrote: >> On my 1280x1024 monitor the CIE diagram is quite small now. I can live >> with it, but I'm not sure if 1280x1024 was meant to be a small >> display? > > That wasn't intentional: So rectangular shape is? :) http://img37.imagefra.me/img/img37/1/12/4/prokoudine/f_1m_444c758.png Alexandre From hughsient at gmail.com Fri Dec 4 17:44:41 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 17:44:41 +0000 Subject: gcm-prefs on netbook In-Reply-To: <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> Message-ID: <15e53e180912040944k21f00207qa486932397da75eb@mail.gmail.com> 2009/12/4 Pascal de Bruijn : > + if (gdk_screen_get_width (screen) < 10240 || > > That's a typo :) Fixed, thanks. Richard. From hughsient at gmail.com Fri Dec 4 17:49:13 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 4 Dec 2009 17:49:13 +0000 Subject: gcm-prefs on netbook In-Reply-To: <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> <58497f010912040916s2a398f16i5d272900da653cfb@mail.gmail.com> <4c2c37f90912040918y6e25908crdbd32641bb64349e@mail.gmail.com> <733f2c730912040924q669cb8f5ve305c71b64a960c8@mail.gmail.com> Message-ID: <15e53e180912040949p144a83b8gce479add4b1a829@mail.gmail.com> 2009/12/4 Alexandre Prokoudine : > So rectangular shape is? :) Wrong! commit e818364b44830c6178a2fcbd81a255f47ad01a6d Author: Richard Hughes Date: Fri Dec 4 17:47:31 2009 +0000 Ensure the CIE widget is square Richard. From pmjdebruijn at pcode.nl Sat Dec 5 12:29:07 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 13:29:07 +0100 Subject: Devices list sorting Message-ID: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> Hi, Since I'm _guessing_ GCM will most commonly be used to calibrate displays, it might be a thought to sort on device "class" first, and then alphabetically? Possibly displaying displays at the top of the list? I'm guessing this order makes most sense (to me): - Displays - Printers - Scanners - Camera's - Video Camera's Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Sat Dec 5 12:30:56 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 13:30:56 +0100 Subject: Feature: Nautilus Thumbnailer for profiles Message-ID: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> Hi, Another idea for the long term: Wouldn't it severely kick ass, if Nautilus were to thumbnail ICC profiles using the CIE widget? Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Sat Dec 5 13:11:25 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sat, 5 Dec 2009 16:11:25 +0300 Subject: Feature: Nautilus Thumbnailer for profiles In-Reply-To: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> References: <4c2c37f90912050430r23271f5p575ad481e8cbb7c3@mail.gmail.com> Message-ID: <733f2c730912050511s6634231qf84b50fb000b148b@mail.gmail.com> On 12/5/09, Pascal de Bruijn wrote: > Another idea for the long term: > > Wouldn't it severely kick ass, if Nautilus were to thumbnail ICC > profiles using the CIE widget? +2^32 :) Alexandre From hughsient at gmail.com Sat Dec 5 15:52:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 5 Dec 2009 15:52:24 +0000 Subject: Devices list sorting In-Reply-To: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> Message-ID: <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> 2009/12/5 Pascal de Bruijn : > Since I'm _guessing_ GCM will most commonly be used to calibrate > displays, it might be a thought to sort on device "class" first, and > then alphabetically? The profile title or filename as the secondary sort? > Possibly displaying displays at the top of the list? I'm guessing this > order makes most sense (to me): > > - Displays > - Printers > - Scanners > - Camera's > - Video Camera's I'm not sure it matters, but sure, it would be better if they were sorted. Richard. From pmjdebruijn at pcode.nl Sat Dec 5 15:54:18 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sat, 5 Dec 2009 16:54:18 +0100 Subject: Devices list sorting In-Reply-To: <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> Message-ID: <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> On Sat, Dec 5, 2009 at 4:52 PM, Richard Hughes wrote: > 2009/12/5 Pascal de Bruijn : >> Since I'm _guessing_ GCM will most commonly be used to calibrate >> displays, it might be a thought to sort on device "class" first, and >> then alphabetically? > > The profile title or filename as the secondary sort? I'd use the profile description as secondary sort... >> Possibly displaying displays at the top of the list? I'm guessing this >> order makes most sense (to me): >> >> - Displays >> - Printers >> - Scanners >> - Camera's >> - Video Camera's > > I'm not sure it matters, but sure, it would be better if they were sorted. It's a convenience thing... For example, since video4linux support has been commited, on my laptop I get my Chicony webcam as the first entry, and below that my display... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Sun Dec 6 09:31:43 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:31:43 +0300 Subject: gcm-inspect segfaults Message-ID: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> (gdb) run -d Starting program: /usr/bin/gcm-inspect -d [Thread debugging using libthread_db enabled] Output profile 'VGA1': not set Output profile 'HDMI1': not set Output profile 'DP1': not set Output profile 'HDMI2': not set Output profile 'DP2': not set Program received signal SIGSEGV, Segmentation fault. 0x00007ffff3ffba52 in g_atomic_int_exchange_and_add () from /lib/libglib-2.0.so.0 (gdb) bt #0 0x00007ffff3ffba52 in g_atomic_int_exchange_and_add () from /lib/libglib-2.0.so.0 #1 0x00007ffff401162e in g_hash_table_unref () from /lib/libglib-2.0.so.0 #2 0x0000000000404ee7 in gcm_inspect_get_properties (argc=1, argv=0x7fffffffe3d8) at gcm-inspect.c:371 #3 main (argc=1, argv=0x7fffffffe3d8) at gcm-inspect.c:438 Should I be concerned about that sort of thing? :) Alexandre From alexandre.prokoudine at gmail.com Sun Dec 6 09:45:02 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:45:02 +0300 Subject: Yelp isn't happy Message-ID: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> Hi, When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI ?ghelp:gnome-color-manager?preferences? is incorrect". Alexandre From alexandre.prokoudine at gmail.com Sun Dec 6 09:51:56 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Sun, 6 Dec 2009 12:51:56 +0300 Subject: yet another PO comment bug Message-ID: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> Hi, A comment for "Setting up device" says "TRANSLATORS: title, a profile is a ICC file" Alexandre From hughsient at gmail.com Sun Dec 6 10:09:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 10:09:49 +0000 Subject: gcm-inspect segfaults In-Reply-To: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> References: <733f2c730912060131t66aa6d09i5a1b120c1ea87ced@mail.gmail.com> Message-ID: <15e53e180912060209i151c479gd985e4d60e4a829e@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > #1 ?0x00007ffff401162e in g_hash_table_unref () from /lib/libglib-2.0.so.0 Fixed in git, thanks. Richard. From hughsient at gmail.com Sun Dec 6 10:16:29 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 10:16:29 +0000 Subject: yet another PO comment bug In-Reply-To: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> References: <733f2c730912060151y6835c62fxf815211d1e35b7c5@mail.gmail.com> Message-ID: <15e53e180912060216y4b0c3daeh2c850b6b15b8ce35@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > A comment for "Setting up device" says "TRANSLATORS: title, a profile > is a ICC file" Fixed, thanks. Richard. From pmjdebruijn at pcode.nl Sun Dec 6 11:39:55 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 6 Dec 2009 12:39:55 +0100 Subject: CIE widget drawing odd graphs Message-ID: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> Hi, I might have uncovered some minor issues with the CIE widget drawing code. I've attached an archive containing several non-standard profiles, which are drawn quite oddly. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: weirdciewidget.zip Type: application/zip Size: 31374 bytes Desc: not available URL: From pmjdebruijn at pcode.nl Sun Dec 6 12:33:52 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 6 Dec 2009 13:33:52 +0100 Subject: Icon Sizes Message-ID: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> Hi, I managed to test GCM before on a netbook with Ubuntu Karmic Netbook edition installed. I noticed somehow the GCM icon was very small in comparison to all the other application icons... While I'm not entirely sure about this, I think the Ubuntu Netbook edition uses the 64x64 icons (and does not fall back to scalable to save CPU cycles on low end CPUs), since GCM does not supply a 64x64 icon, it falls back to a smaller one. Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? Regards, Pascal de Bruijn From lars.tore at mulebakken.net Sun Dec 6 16:04:46 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Sun, 6 Dec 2009 17:04:46 +0100 Subject: Bluish.icc In-Reply-To: <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> References: <58497f010911110351o3d7df74dt8ac35ef7c8cb5e4@mail.gmail.com> <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> Message-ID: <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> On Sun, Dec 6, 2009 at 3:32 PM, Pascal de Bruijn wrote: > On Wed, Nov 11, 2009 at 12:51 PM, Lars Tore Gustavsen > wrote: >> The icc profile include with gnome-color-manager is very nice to test >> if everything works. However it advertise itself as "sRGB >> IEC61966-2-1" >> >> I have edited the attached profile and it now reads "bluish test" in gcm-prefs. > > Would it be possible for you to rename it to "Blueish Test" instead of > "blueish test"? > > Not extremely important, but it does look a bit better :) > > Regards, > Pascal de Bruijn > Here we go: It's done with Andrew Shepherd's ICC Profile Toolkit http://www.tlbtlb.com/links/ Regards Lars Tore Gustavsen -------------- next part -------------- A non-text attachment was scrubbed... Name: bluish.icc Type: application/x-icc Size: 3966 bytes Desc: not available URL: From hughsient at gmail.com Sun Dec 6 19:37:09 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 19:37:09 +0000 Subject: Bluish.icc In-Reply-To: <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> References: <58497f010911110351o3d7df74dt8ac35ef7c8cb5e4@mail.gmail.com> <4c2c37f90912060632k4ddb5a54u8f49d7392eefa449@mail.gmail.com> <58497f010912060804l53c1ae31k39cad89bfa83da68@mail.gmail.com> Message-ID: <15e53e180912061137h57d3f969m2a57af3568e547e7@mail.gmail.com> 2009/12/6 Lars Tore Gustavsen : > Here we go: Applied, thanks. Richard From hughsient at gmail.com Sun Dec 6 19:46:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 19:46:40 +0000 Subject: Icon Sizes In-Reply-To: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> References: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> Message-ID: <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> 2009/12/6 Pascal de Bruijn : > Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? commit 29b66375ed40e10bfffba601eeca9f25bd8a46b7 Author: Richard Hughes Date: Sun Dec 6 19:45:42 2009 +0000 Add a 64x64 pixel application icon Richard. From hughsient at gmail.com Sun Dec 6 20:06:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 20:06:24 +0000 Subject: Devices list sorting In-Reply-To: <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> Message-ID: <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> 2009/12/5 Pascal de Bruijn : > For example, since video4linux support has been commited, on my laptop > I get my Chicony webcam as the first entry, and below that my > display... commit dfe5fc80cef25cac2deddb5687719ab298e786e0 Author: Richard Hughes Date: Sun Dec 6 20:05:18 2009 +0000 Make the list orders predictable by setting a sort string Richard. From hughsient at gmail.com Sun Dec 6 20:19:37 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sun, 6 Dec 2009 20:19:37 +0000 Subject: CIE widget drawing odd graphs In-Reply-To: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> Message-ID: <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> 2009/12/6 Pascal de Bruijn : > I might have uncovered some minor issues with the CIE widget drawing code. They look pretty major to me! > I've attached an archive containing several non-standard profiles, > which are drawn quite oddly. After the release I would like to look at the widget drawing code and the profile extraction code (there are still quite a lot of FIXME's and unimplemented stuff) and add a boat load more self tests and internal validations. I fear I might not have time in the next couple of weeks (lots of other work to do, which I've been putting off for ages) so if anyone is familiar with XYZ co-ordinates and ICC profiles I would appreciate any code review and possible patches. Thanks, Richard. From pmjdebruijn at pcode.nl Sun Dec 6 23:35:43 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 00:35:43 +0100 Subject: Devices list sorting In-Reply-To: <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> References: <4c2c37f90912050429u5142e877wbe371cd423e3e983@mail.gmail.com> <15e53e180912050752q300c2a24p90ad7006e7e7ef06@mail.gmail.com> <4c2c37f90912050754j7403226fv534f8f634415a1bf@mail.gmail.com> <15e53e180912061206w100bde6cs29f3aaaf1932c1e6@mail.gmail.com> Message-ID: <4c2c37f90912061535n22c07324s60f6a5edbc7e5147@mail.gmail.com> On Sun, Dec 6, 2009 at 9:06 PM, Richard Hughes wrote: > 2009/12/5 Pascal de Bruijn : >> For example, since video4linux support has been commited, on my laptop >> I get my Chicony webcam as the first entry, and below that my >> display... > > commit dfe5fc80cef25cac2deddb5687719ab298e786e0 > Author: Richard Hughes > Date: ? Sun Dec 6 20:05:18 2009 +0000 > > ? ?Make the list orders predictable by setting a sort string Seems to work excellently here... Thanks, Pascal de Bruijn From hughsient at gmail.com Mon Dec 7 12:03:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 7 Dec 2009 12:03:28 +0000 Subject: gnome-color-manager 2.29.1 Message-ID: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> gnome-color-manager is a session program that makes it easy to manage, install and generate color profiles in the GNOME desktop. Version 2.29.1 ~~~~~~~~~~~~~~ Released: 2009-12-07 * Translations - Add Russian translation (Alexandre Prokoudine) - Add British English translation (Bruce Cowan) - Add Indonesian translation (Andika Triwidada) - Add French translation (Claude Paroz) - Add Swedish translation (Daniel Nylander) - Add Brazilian Portuguese translation (Flamarion Jorge) - Add Lithuanian translation (Gintautas Miliauskas) - Add German translation (Hendrik Brandt) - Add Danish translation (Joe Hansen) - Add Spanish translation (Jorge Gonz?lez) - Add Norwegian bokm?l translation (Kjartan Maraas) - Add Thai translation (Phondanai Khanti) - Add Polish translation (Piotr Dr?g) - Add Estonian translation (Priit Laes, Mattias P?ldaru) - Add Tamil translation (vasudeven) - Add Russian user guide translation (Alexandre Prokoudine) - Add Indonesian user guide translation (Andika Triwidada) * New Features: - Add gcm-import, a helper to allow double clicking on ICC profiles to import them (Richard Hughes) - Add gcm-prefs, a utility to assign profiles to devices, examine profiles, and set session-wide defaults (Richard Hughes) - Add gcm-session, a dbus-activated session daemon for applications to get the profiles for a device, or device class and to get session-wide defaults. It exits when no longer used to save resources. (Richard Hughes) - Add gcm-self-test, a self test framework that tests GCM functionality (Richard Hughes) - Add gcm-apply, a simple utility to just set (or reset) display profiles (Richard Hughes) - Add gcm-inspect, a debugging utility to inspect the profiles set in the session (Richard Hughes) - Add gcm-dump-edid, a utility to dump the EDID to disk for debugging (Richard Hughes) - Add gcm-dump-profile, a utility to dump the ICC profile to the screen (Richard Hughes) - Add some simple man pages and help document (Richard Hughes) - Add ArgyllCMS support to generate device profiles (Richard Hughes) - Add color calibration hardware auto-detection (Richard Hughes) - Add code to set the _ICC_PROFILE atom per-output and also per-screen (Richard Hughes) - Add some pre-calibration steps for external displays (Pascal de Bruijn) - Add hardware support for gphoto supported cameras (Richard Hughes) - Add hardware support for SANE suppoerted scanners (Richard Hughes) - Add hardware support for video4linux supported video devices (Richard Hughes) - Add hardware support for XRandR supported displays (Richard Hughes) - Add hardware support for hplip supported printers (Richard Hughes) - Add CIE widget to display visual data about different profiles (Richard Hughes) - Use the system DMI data to better itentify internal LCD panels (Richard Hughes) - Parse the EDID to get a better device description for displays (Richard Hughes) - Make the list orders predictable by setting a sort string (Richard Hughes) * Bugfix: - Use XDG directory to store data (Baptiste Mille-Mathias) - Remove markup from GTKBuilder translatable strings (Claude Paroz) - Update bluish.icc title (Lars Tore Gustavsen, Pascal de Bruijn) - Enable adding xrandr devices with no EDID (Martin Szulecki) - Avoid reporting a (false) failure on first import (Stephane Delcroix) - Fix the message-received cb signature (Stephane Delcroix) - Fix up numerous small bugs prior to first release (Richard Hughes) - Look for the debian-named argyllcms binaries first (Richard Hughes) - Set the brightness to 100% on internal LCD panels before we generate a output profile (Richard Hughes) Richard. From pmjdebruijn at pcode.nl Mon Dec 7 12:37:01 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 13:37:01 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> Message-ID: <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> On Mon, Dec 7, 2009 at 1:03 PM, Richard Hughes wrote: > gnome-color-manager is a session program that makes it easy to manage, install > and generate color profiles in the GNOME desktop. > > Version 2.29.1 > ~~~~~~~~~~~~~~ > Released: 2009-12-07 Excellent :) Packages are available from my PPA as usual. However, since I'll be tracking git again in my main repo, I've decided to start a gcm-release PPA: https://launchpad.net/~pmjdebruijn/+archive/gcm-release I've just submitted the builds, so they'll be ready in an hour or six... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Mon Dec 7 15:08:18 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 7 Dec 2009 18:08:18 +0300 Subject: metadata Message-ID: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Hi, It looks like there are problems writing unicode text into icc metadata or reading from it. http://img39.imagefra.me/img/img39/1/12/7/prokoudine/f_0m_948cb9f.png Exhibit 1 is description of the profile that is obviously RU translation of "Custom" Exhibit 2 is copyright field that seems to be suffering from more or less same thing. Any ideas? Alexandre From pmjdebruijn at pcode.nl Mon Dec 7 15:14:42 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 16:14:42 +0100 Subject: metadata In-Reply-To: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Message-ID: <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> On Mon, Dec 7, 2009 at 4:08 PM, Alexandre Prokoudine wrote: > Hi, > > It looks like there are problems writing unicode text into icc > metadata or reading from it. > > http://img39.imagefra.me/img/img39/1/12/7/prokoudine/f_0m_948cb9f.png > > Exhibit 1 is description of the profile that is obviously RU > translation of "Custom" > > Exhibit 2 is copyright field that seems to be suffering from more or > less same thing. This actually begs a bigger question: Should we even try to internationalize metadata contents? I'd prefer not... It would make exchanging ICC profiles internationally harder... Say, one day, we make a big repository of ICC profiles which are ready to go for people who don't have a colorimeter. I'd be pretty crappy for someone in Germany to see Russian in the metadata? I'd prefer to stick to English... Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Mon Dec 7 15:47:16 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 7 Dec 2009 18:47:16 +0300 Subject: metadata In-Reply-To: <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> Message-ID: <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> On Mon, Dec 7, 2009 at 6:14 PM, Pascal de Bruijn wrote: > This actually begs a bigger question: Should we even try to > internationalize metadata contents? > > I'd prefer not... It would make exchanging ICC profiles > internationally harder... Quite so :) > Say, one day, we make a big repository of ICC profiles which are ready > to go for people who don't have a colorimeter. Are we talking about profiles for DSLRs? :) Alexandre From hughsient at gmail.com Mon Dec 7 17:29:00 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 7 Dec 2009 17:29:00 +0000 Subject: gnome-color-manager 2.29.1 In-Reply-To: <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> Message-ID: <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> 2009/12/7 Pascal de Bruijn : > I've just submitted the builds, so they'll be ready in an hour or six... I've just started an official build for Fedora 12: http://koji.fedoraproject.org/koji/taskinfo?taskID=1860405 Richard. From pmjdebruijn at pcode.nl Mon Dec 7 19:21:36 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 7 Dec 2009 20:21:36 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> Message-ID: <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> On Mon, Dec 7, 2009 at 6:29 PM, Richard Hughes wrote: > 2009/12/7 Pascal de Bruijn : >> I've just submitted the builds, so they'll be ready in an hour or six... > > I've just started an official build for Fedora 12: > http://koji.fedoraproject.org/koji/taskinfo?taskID=1860405 Richard, my mail to the mailing list got rejected: "Message rejected by filter rule match" Regards, Pascal de Bruijn From hughsient at gmail.com Tue Dec 8 10:15:58 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 10:15:58 +0000 Subject: gnome-color-manager 2.29.1 In-Reply-To: <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> Message-ID: <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> 2009/12/7 Pascal de Bruijn : > Richard, my mail to the mailing list got rejected: I think gnome-announce-list at gnome.org bounced the mail (correctly) -- dropping the cc should make things work again. Richard. From hughsient at gmail.com Tue Dec 8 11:02:50 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 11:02:50 +0000 Subject: metadata In-Reply-To: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> Message-ID: <15e53e180912080302x2f413450v3b99d884d709ed57@mail.gmail.com> 2009/12/7 Alexandre Prokoudine : > Exhibit 1 is description of the profile that is obviously RU > translation of "Custom" Right, I made the Custom localized without thinking that it would involve non-ascii chars. I've fixed that in cfebe91aef33254709bc22ca547563044f1fbdd8. > Exhibit 2 is copyright field that seems to be suffering from more or > less same thing. > > Any ideas? Well, for textType, the profile spec mandates 7 bit ASCII, so we're wrong to translate Custom. For the unicodeTextType we should probably decode UTF-16 properly, although I've not seen a profile that uses this type in the wild yet. Note: if you've got a profile that breaks GCM in one way or another (and it's freely redistributable, which excludes the ICC and Adobe profiles..) then we should probably check it into GCM and wire it into the self test code. If nothing else you can use gcm-dump-profile -v thefilename.icc and send me the output. Richard. From hughsient at gmail.com Tue Dec 8 11:05:57 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 11:05:57 +0000 Subject: metadata In-Reply-To: <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> Message-ID: <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> 2009/12/7 Alexandre Prokoudine : >> Say, one day, we make a big repository of ICC profiles which are ready >> to go for people who don't have a colorimeter. > > Are we talking about profiles for DSLRs? :) Just in case, I've got a repo waiting: http://github.com/hughsie/gnome-color-profiles I'm going to be checking in the Adobe and ICC profiles first, but alas, they can't be installed by default in distros like Fedora as they are non free [1]. Richard. [1] https://www.redhat.com/archives/fedora-legal-list/2009-December/msg00009.html and https://www.redhat.com/archives/fedora-legal-list/2009-December/msg00010.html From pmjdebruijn at pcode.nl Tue Dec 8 11:16:35 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 12:16:35 +0100 Subject: gnome-color-manager 2.29.1 In-Reply-To: <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> References: <15e53e180912070403q5384b17bga676c7edb9627e8b@mail.gmail.com> <4c2c37f90912070437l55b9ea15r34290aac597fba4c@mail.gmail.com> <15e53e180912070929u366a5d94hdfb0d9fa667cf8b0@mail.gmail.com> <4c2c37f90912071121s12ba134aj2f7d4a621d064bfa@mail.gmail.com> <15e53e180912080215k324546f8kec9a9b64c6bdecf8@mail.gmail.com> Message-ID: <4c2c37f90912080316w42894a61m7ad18ee019eba3dd@mail.gmail.com> On Tue, Dec 8, 2009 at 11:15 AM, Richard Hughes wrote: > 2009/12/7 Pascal de Bruijn : >> Richard, my mail to the mailing list got rejected: > > I think ?gnome-announce-list at gnome.org bounced the mail (correctly) -- > dropping the cc should make things work again. Oops, silly me... Yeah, the gnome-color-manager list accepted my message just fine... Regards, Pascal de Bruijn From graeme2 at argyllcms.com Tue Dec 8 12:38:17 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Tue, 08 Dec 2009 23:38:17 +1100 Subject: metadata In-Reply-To: <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> Message-ID: <4B1E48B9.1080408@argyllcms.com> Richard Hughes wrote: > I'm going to be checking in the Adobe and ICC profiles first, but > alas, they can't be installed by default in distros like Fedora as > they are non free [1]. Note that Argyll V1.1.0 includes a public domain implementation of an sRGB profiles as well as an AdobeRGB1998 equivalent. Graeme Gill. From hughsient at gmail.com Tue Dec 8 12:51:45 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 8 Dec 2009 12:51:45 +0000 Subject: metadata In-Reply-To: <4B1E48B9.1080408@argyllcms.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> <4B1E48B9.1080408@argyllcms.com> Message-ID: <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> 2009/12/8 Graeme Gill : > Richard Hughes wrote: > >> I'm going to be checking in the Adobe and ICC profiles first, but >> alas, they can't be installed by default in distros like Fedora as >> they are non free [1]. > > Note that Argyll V1.1.0 includes a public domain implementation > of an sRGB profiles as well as an AdobeRGB1998 equivalent. That's really good to know, and I might pinch those. What license are those available as? Really public domain? (if so, that's fantastic). Can they be installed alongside the real adobe profiles? Richard. From graeme2 at argyllcms.com Tue Dec 8 13:11:07 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Wed, 09 Dec 2009 00:11:07 +1100 Subject: metadata In-Reply-To: <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> References: <733f2c730912070708x104779f7y1a95ea33b36fe569@mail.gmail.com> <4c2c37f90912070714y20dd1609h7202e680be7479a6@mail.gmail.com> <733f2c730912070747o4bcd8565l38ed26a1d96c9ad8@mail.gmail.com> <15e53e180912080305n5520ff47s6f045d1e67ea1885@mail.gmail.com> <4B1E48B9.1080408@argyllcms.com> <15e53e180912080451i3affe46bl554e7e0e7e706aa0@mail.gmail.com> Message-ID: <4B1E506B.4080709@argyllcms.com> Richard Hughes wrote: > 2009/12/8 Graeme Gill : >> Note that Argyll V1.1.0 includes a public domain implementation >> of an sRGB profiles as well as an AdobeRGB1998 equivalent. > > That's really good to know, and I might pinch those. What license are > those available as? Really public domain? (if so, that's fantastic). Yes. There didn't seem any sensible reason to retain copyright on them. > Can they be installed alongside the real adobe profiles? I'm not sure what you mean by that. Since "Adobe" is a registered company name, I've given the Adobe compatible profile a different name ("ClayRGB1998.icm", the best I could come up with at the time). The sRGB profile is called "sRGB.icm", just like the original one from HP/Microsoft that is no longer available, and to which it is functionally identical. Graeme Gill. From pmjdebruijn at pcode.nl Tue Dec 8 17:56:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 18:56:33 +0100 Subject: Icon Sizes In-Reply-To: <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> References: <4c2c37f90912060433o5e65fd97vc71934c48ddbf98e@mail.gmail.com> <15e53e180912061146x13c16feexa1d195240b1f860@mail.gmail.com> Message-ID: <4c2c37f90912080956y6dcf9264padf8217b2c3a5ab4@mail.gmail.com> On Sun, Dec 6, 2009 at 8:46 PM, Richard Hughes wrote: > 2009/12/6 Pascal de Bruijn : >> Maybe it's possible to include a 64x64 rendered PNG of the GCM icon? > > commit 29b66375ed40e10bfffba601eeca9f25bd8a46b7 > Author: Richard Hughes > Date: ? Sun Dec 6 19:45:42 2009 +0000 > > ? ?Add a 64x64 pixel application icon Today, I had a chance to test this on my colleagues HP netbook again... And this seems to fix the "small icon" issue on Ubuntu Karmic Netbook Edition. Thanks, Pascal de Bruijn From pmjdebruijn at pcode.nl Tue Dec 8 17:58:02 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 8 Dec 2009 18:58:02 +0100 Subject: gcm-prefs on netbook In-Reply-To: <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> References: <4c2c37f90912040658w6592e4a2v246dbf17c89cc592@mail.gmail.com> <15e53e180912040700q6884b4a5h176d7f9bdc1498e0@mail.gmail.com> <15e53e180912040718s66a221a9s65dff40e05758047@mail.gmail.com> Message-ID: <4c2c37f90912080958s1dac9704qdc99372d9c1a3401@mail.gmail.com> On Fri, Dec 4, 2009 at 4:18 PM, Richard Hughes wrote: > 2009/12/4 Richard Hughes : >> What about if the window size < 1024x768 then make the cie widget smaller? > > commit 820955239cffb0d9a9bc40c12e62240ea6a3a416 > Author: Richard Hughes > Date: ? Fri Dec 4 15:11:19 2009 +0000 > > ? ?Reduce the size of the ICC widget on small displays I had a chance to test this on a colleagues HP netbook today, and it seems to work fine. The CIE widget looks a bit silly at 50x50, but again, this is a netbook's owner choice... Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 9 12:47:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 12:47:28 +0000 Subject: Yelp isn't happy In-Reply-To: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> References: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> Message-ID: <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> 2009/12/6 Alexandre Prokoudine : > When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI > ?ghelp:gnome-color-manager?preferences? is incorrect". commit ef84a7c5dca50e40bfaf38b6c884c8c86983f334 Author: Richard Hughes Date: Wed Dec 9 12:46:33 2009 +0000 Fix the help file installation so that yelp recognises our help file Richard. From alexandre.prokoudine at gmail.com Wed Dec 9 13:18:19 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Wed, 9 Dec 2009 16:18:19 +0300 Subject: Yelp isn't happy In-Reply-To: <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> References: <733f2c730912060145u4422174s368bba65fa23adf@mail.gmail.com> <15e53e180912090447y7741610ai3411f9f118b7b047@mail.gmail.com> Message-ID: <733f2c730912090518w1ed31992g2921881109c65f9b@mail.gmail.com> On Wed, Dec 9, 2009 at 3:47 PM, Richard Hughes wrote: > 2009/12/6 Alexandre Prokoudine : >> When I click "Help" button in gcm-prefs, Yelp tells me "Requested URI >> ?ghelp:gnome-color-manager?preferences? is incorrect". > > commit ef84a7c5dca50e40bfaf38b6c884c8c86983f334 > Author: Richard Hughes > Date: ? Wed Dec 9 12:46:33 2009 +0000 > > ? ?Fix the help file installation so that yelp recognises our help file Many thanks :) Alexandre From hughsient at gmail.com Wed Dec 9 13:18:35 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 13:18:35 +0000 Subject: Profile description for laptop displays In-Reply-To: <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> References: <4c2c37f90911290453x143c6820nb27f3a6b6188e3c7@mail.gmail.com> <15e53e180911291342o2b8755aavad083da3dcd849d2@mail.gmail.com> <4c2c37f90911291527v71d2cf9cy6bb41add8c7b3683@mail.gmail.com> <15e53e180911300210tae5b9a0kaa94ead83633e413@mail.gmail.com> <4c2c37f90912020936k5096c2c2s537ad34784e86f2a@mail.gmail.com> Message-ID: <15e53e180912090518h5b6e9deas414d6c5e1f1ec594@mail.gmail.com> 2009/12/2 Pascal de Bruijn : > I wish we had some more EDID samples to compare... Maybe a call for > sample EDID data on Planet GNOME would get us some more samples? Could you send me your EDID please. Thanks. Richard. From hughsient at gmail.com Wed Dec 9 13:19:56 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 13:19:56 +0000 Subject: More verbose output in calibration? In-Reply-To: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> References: <58497f010912040353l67d737c7p8820923f0098f388@mail.gmail.com> Message-ID: <15e53e180912090519v6481e569q867923aa08984dd3@mail.gmail.com> 2009/12/4 Lars Tore Gustavsen : > May I suggest that when I calibrate my monitor the details view are > more verbose than now (like dispcal -v). I guess a beginner will think > this is not working at all, when it measures all the black patches and > almost noting change on the display. ?I think more experienced users > also like to see the progress in the calibration ?and the deltaE > report at the end. ?In a longer time frame some progress bar (1) or > progress window (2) would be very nice eye candy in the calibration > process. You can already get the deltaE values in the dropdown widget. I think it would be very hard to do a progress bar from the data Argyll gives us. Richard. From hughsient at gmail.com Wed Dec 9 17:04:24 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 17:04:24 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B15B29C.2090105@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> Message-ID: <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> 2009/12/2 Graeme Gill : > For a cLut based profile, you would have to (as Lars Tore Gustavsen > suggested) run the primary values through the profile (or > do the equivalent of looking up what the cLUT values are at > the primaries). I'm sort-of stuck here. I'm confused about which cLUT I should be using (TRC curves?), as different profiles are doing very different things. Could you please elaborate about how to run the primaries through the table? Thanks. Richard. From hughsient at gmail.com Wed Dec 9 19:51:53 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 9 Dec 2009 19:51:53 +0000 Subject: shared-color-profiles 0.1.0 release Message-ID: <15e53e180912091151y1f2303acw48cedde3eb0231f@mail.gmail.com> Version 0.1.0 ~~~~~~~~~~~~~ Released: 2009-12-09 * New Features: - First release! - Add the free Argyll profiles that are in the public domain (Richard Hughes) * Bugfix: - Don't install README and LICENSE in /usr/share/color (Richard Hughes) - Require color-filesystem on Fedora (Richard Hughes) The next release of gnome-color-manager will depend on this package. See http://blogs.gnome.org/hughsie/2009/12/08/shared-color-profiles/ for more details. Richard. From alexandre.prokoudine at gmail.com Wed Dec 9 23:02:22 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 02:02:22 +0300 Subject: sadly, segfault Message-ID: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> Hi, On commit f04af5c3072983e9b5113628f7782dc346c75b18: http://pastebin.com/m18c9d8c9 Alexandre From graeme2 at argyllcms.com Wed Dec 9 23:04:54 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Thu, 10 Dec 2009 10:04:54 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> Message-ID: <4B202D16.1000901@argyllcms.com> Richard Hughes wrote: > I'm sort-of stuck here. I'm confused about which cLUT I should be > using (TRC curves?), as different profiles are doing very different > things. Could you please elaborate about how to run the primaries > through the table? Thanks. Well I can only talk specifically about my set of tools: Using icclib command line tools it would be something like this: icclu -ff -ia -px sRGB.icm 1 0 0 1.000000 0.000000 0.000000 [RGB] -> MatrixFwd -> 0.412385 0.212596 0.019306 [XYZ] 0 1 0 0.000000 1.000000 0.000000 [RGB] -> MatrixFwd -> 0.357573 0.715199 0.119206 [XYZ] 0 0 1 0.000000 0.000000 1.000000 [RGB] -> MatrixFwd -> 0.180500 0.072205 0.950546 [XYZ] Progromatically, one would do the equivalent, e.g. using icclib open the profile, create a lookup object to do forward absolute colorimetric to XYZ space, and then feed colorant only combinations through it. icclu.c will illustrate the details. I'm sure something similar is possible using Marti's lcms. I wouldn't recommend trying to re-create the wheel and do it from ICC tags as there are a number of details to get right. Graeme Gill. From hughsient at gmail.com Thu Dec 10 08:07:04 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 08:07:04 +0000 Subject: sadly, segfault In-Reply-To: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> References: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> Message-ID: <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> 2009/12/9 Alexandre Prokoudine : > On commit f04af5c3072983e9b5113628f7782dc346c75b18: > http://pastebin.com/m18c9d8c9 Thanks, commit 64037aacc1ba6fa7bfa53deabc00dca57cdd658d Author: Richard Hughes Date: Thu Dec 10 08:05:25 2009 +0000 trivial: fix up an error in the last commit, so that if filename is not set we skip the load There are some more changes happening today, so please yell if anything breaks. Thanks. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 14:54:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 17:54:09 +0300 Subject: sadly, segfault In-Reply-To: <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> References: <733f2c730912091502k5d43fd03ka29b8f8d8bbf43a4@mail.gmail.com> <15e53e180912100007u4e552f10yaba6803a787ba8ca@mail.gmail.com> Message-ID: <733f2c730912100654g6eafc1e9v5c51b2783e119437@mail.gmail.com> On 12/10/09, Richard Hughes wrote: > There are some more changes happening today, so please yell if > anything breaks. Thanks. Well, apart from the profiles I sent you before whose CIE diagram doesn't render nicely, so far so good :) Alexandre From hughsient at gmail.com Thu Dec 10 15:26:35 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 15:26:35 +0000 Subject: LCMS dependency Message-ID: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> How would everyone feel if I added lcms as a hard dependency to gnome-color-manager? Now how would you all feel if I added lcms2 as a hard dep? Comments please. I think using lcms allows us to nuke a lot of the custom (fragile) profile code, but only lcms2 has the functionality I need to decode the vcgt tags. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 15:30:18 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 10 Dec 2009 18:30:18 +0300 Subject: LCMS dependency In-Reply-To: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> Message-ID: <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> On 12/10/09, Richard Hughes wrote: > How would everyone feel if I added lcms as a hard dependency to > gnome-color-manager? Fine with me. Most graphics apps I use already have it as requirement. > Now how would you all feel if I added lcms2 as a hard dep? Marti does amazing job with lcms2, and since now it has a separate header for v2, this really might work :) Alexandre From pmjdebruijn at pcode.nl Thu Dec 10 15:47:06 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 16:47:06 +0100 Subject: LCMS dependency In-Reply-To: <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <733f2c730912100730m4a0971f0x1b5d5055b168e026@mail.gmail.com> Message-ID: <4c2c37f90912100747i1a0a507au47db84306b64dd42@mail.gmail.com> On Thu, Dec 10, 2009 at 4:30 PM, Alexandre Prokoudine wrote: > On 12/10/09, Richard Hughes wrote: >> How would everyone feel if I added lcms as a hard dependency to >> gnome-color-manager? > > Fine with me. Most graphics apps I use already have it as requirement. Indeed... >> Now how would you all feel if I added lcms2 as a hard dep? > > Marti does amazing job with lcms2, and since now it has a separate > header for v2, this really might work :) Adding lcms2 as a hard dependancy seems less then desirable (at this moment), since lcms2 hasn't been released yet. And it requires additional packaging... Adding it as an optional dependancy seems ok though. Regards, Pascal de Bruijn From graeme2 at argyllcms.com Thu Dec 10 16:05:34 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 11 Dec 2009 03:05:34 +1100 Subject: LCMS dependency In-Reply-To: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> Message-ID: <4B211C4E.9060502@argyllcms.com> Richard Hughes wrote: > I think using lcms allows us to nuke a lot of the custom (fragile) > profile code, but only lcms2 has the functionality I need to decode > the vcgt tags. Well of course icclib has support for accessing the vcgt tag. Graeme Gill. From hughsient at gmail.com Thu Dec 10 16:18:36 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 16:18:36 +0000 Subject: LCMS dependency In-Reply-To: <4B211C4E.9060502@argyllcms.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> Message-ID: <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> 2009/12/10 Graeme Gill : > Well of course icclib has support for accessing the vcgt tag. Sure, that's also a possibility. Richard. From pmjdebruijn at pcode.nl Thu Dec 10 17:00:03 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:00:03 +0100 Subject: LCMS dependency In-Reply-To: <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> Message-ID: <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> On Thu, Dec 10, 2009 at 5:18 PM, Richard Hughes wrote: > 2009/12/10 Graeme Gill : >> Well of course icclib has support for accessing the vcgt tag. > > Sure, that's also a possibility. Well, lcms is used by most OSS graphical apps for color management, so lcms is a natural choice when we don't want to introduce new dependancies into most systems in general. But as long as lcms2 hasn't been released and packaged into most distro's, it makes testing gcm a bit harder, which means feedback on git versions might degrade. Another slightly nasty option would be, to use lcms2, but just (for now) include the code into the gcm repo until it has been released, and just statically compile it into the gcm binaries. And yes, I'll wash my mouth with some soap :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 10 17:11:56 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:11:56 +0100 Subject: The new TRC graphs Message-ID: <4c2c37f90912100911o6f9c4b23y4648e95b2390d82a@mail.gmail.com> Hi, Just thought I should mention this... The new TRC graphs genuinely kick-ass! Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 10 17:14:31 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:14:31 +0000 Subject: LCMS dependency In-Reply-To: <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> Message-ID: <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > Well, lcms is used by most OSS graphical apps for color management, so > lcms is a natural choice when we don't want to introduce new > dependancies into most systems in general. Right. > But as long as lcms2 hasn't been released and packaged into most > distro's, it makes testing gcm a bit harder, which means feedback on > git versions might degrade. Agreed. > Another slightly nasty option would be, to use lcms2, but just (for > now) include the code into the gcm repo until it has been released, > and just statically compile it into the gcm binaries. Ick, ick. > And yes, I'll wash my mouth with some soap :) Okay, I think after what everyone has said I'll add an lcms dep, but not a lcms2 dep. We can convert when lcms2 has had it's first public release. Richard. From pmjdebruijn at pcode.nl Thu Dec 10 17:15:23 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:15:23 +0100 Subject: Bug: All profiles are 256 bytes Message-ID: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> Hi, I just built git again, and I noticed that GCM (on the profiles tab), tells me all my ICC profiles are 256 bytes... There is definitely something fishy about this :) Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Thu Dec 10 17:19:10 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 10 Dec 2009 18:19:10 +0100 Subject: CIE widget drawing odd graphs In-Reply-To: <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> Message-ID: <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> On Sun, Dec 6, 2009 at 9:19 PM, Richard Hughes wrote: > 2009/12/6 Pascal de Bruijn : >> I might have uncovered some minor issues with the CIE widget drawing code. > > They look pretty major to me! Right :) >> I've attached an archive containing several non-standard profiles, >> which are drawn quite oddly. > > After the release I would like to look at the widget drawing code and > the profile extraction code (there are still quite a lot of FIXME's > and unimplemented stuff) and add a boat load more self tests and > internal validations. I fear I might not have time in the next couple > of weeks (lots of other work to do, which I've been putting off for > ages) so if anyone is familiar with XYZ co-ordinates and ICC profiles > I would appreciate any code review and possible patches. I just built git again, and I noticed XYZ drawing has improved... But I didn't see any actual improvements to the CIE code in git... I assume the problem wasn't in the CIE XYZ widget, but in the data that was fed to it? Regards, Pascal de Bruijn From hughsient at gmail.com Thu Dec 10 17:42:38 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:42:38 +0000 Subject: Bug: All profiles are 256 bytes In-Reply-To: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> References: <4c2c37f90912100915u7283710euf23b96e8b45f4b64@mail.gmail.com> Message-ID: <15e53e180912100942t39465d40rf8875000038e9378@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > I just built git again, and I noticed that GCM (on the profiles tab), > tells me all my ICC profiles are 256 bytes... There is definitely > something fishy about this :) commit 99e9f26fe78202515187dc14b93917664ad01dc0 Author: Richard Hughes Date: Thu Dec 10 17:41:49 2009 +0000 trivial: show the file size in the prefs dialog, not the clut size Richard. From hughsient at gmail.com Thu Dec 10 17:43:40 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 17:43:40 +0000 Subject: CIE widget drawing odd graphs In-Reply-To: <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> References: <4c2c37f90912060339i7ed57c89y17cb96d176637b9c@mail.gmail.com> <15e53e180912061219q13bf9adfu3443eaa03a22ffe6@mail.gmail.com> <4c2c37f90912100919n2756b62fnf897f9c083a9d169@mail.gmail.com> Message-ID: <15e53e180912100943o413617a4j3d75cc58a096540b@mail.gmail.com> 2009/12/10 Pascal de Bruijn : > I just built git again, and I noticed XYZ drawing has improved... But > I didn't see any actual improvements to the CIE code in git... The profile code had a subtle signing bug that wasn't tested for. The change from gchar to guint8 fixed this bug (and possibly a few more). Richard. From web.kiddo at free.fr Thu Dec 10 20:56:13 2009 From: web.kiddo at free.fr (Jeff) Date: Thu, 10 Dec 2009 15:56:13 -0500 Subject: Incorrect parsing of .icm files? Message-ID: <1260478573.20024.33.camel@kusanagi> Hello, I wanted to give gnome-color-manager a try just out of curiosity. I own a Dell Inspiron 910 (a.k.a. Mini 9) netbook, which (to my untrained eye) seems to have good colors out of the box. I downloaded the Windows drivers for this netbook's monitor here: http://support.us.dell.com/support/downloads/download.aspx?releaseid=R229102&fileid=329762 It seems like the monitor is called ST2010, there's an ST2010B and ST2010F version of the driver... both contain the exact same ".icm" file (verified with md5sum), which is attached to this message. Now, here comes the actual question: loading this .icm file in GNOME Color Manager and using it creates obviously incorrect colors (as if someone punched contrast and saturation way up), compared to the default colors (no color profile loaded; colors appear neutral). With the ICC profile loaded, for example, gradients in the Clearlooks GTK theme lose definition: the slight gray gradients on the combobox widgets disappears completely into white. I'm just wondering if - this is a bug in GNOME Color Manager - this is a buggy .icm file - it's a PEBKAC. I wish I could take a screenshot, but screenshots don't work in this case (and photo cameras have trouble photographing LCD screens)... -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ST2010.icm Type: image/x-tga Size: 552 bytes Desc: URL: From hughsient at gmail.com Thu Dec 10 21:24:28 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 21:24:28 +0000 Subject: Incorrect parsing of .icm files? In-Reply-To: <1260478573.20024.33.camel@kusanagi> References: <1260478573.20024.33.camel@kusanagi> Message-ID: <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> 2009/12/10 Jeff : > Now, here comes the actual question: loading this .icm file in GNOME Color > Manager and using it creates obviously incorrect colors (as if someone > punched contrast and saturation way up), compared to the default colors (no > color profile loaded; colors appear neutral). With the ICC profile loaded, > for example, gradients in the Clearlooks GTK theme lose definition: the > slight gray gradients on the combobox widgets disappears completely into > white. > > I'm just wondering if > - this is a bug in GNOME Color Manager > - this is a buggy .icm file Both, I think. If you try gnome-power-manager in git master, it won't "do" anything, as the profile does not have a vcgt or mlut tag, and even using the linear tables doesn't do anything. Even trying using dispwin doesn't appear to do anything. If you use the 2.29.1 release there's a bug where we're using the xTRC values as a gamma table (eek). Richard. From hughsient at gmail.com Thu Dec 10 21:28:59 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 21:28:59 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B202D16.1000901@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> Message-ID: <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> 2009/12/9 Graeme Gill : > Progromatically, one would do the equivalent, e.g. using icclib > open the profile, create a lookup object to do forward absolute colorimetric > to XYZ space, and then feed colorant only combinations through it. > icclu.c will illustrate the details. I'm sure something similar is > possible using Marti's lcms. I've just added code to git master to do this, and it indeed works to get the gamut range for a few scanner and print (RGB) profiles I've got here. It doesn't seem to work for CYMK profiles, although I'm sure that's something crazy I'm doing (or not doing..). Anyway, I'll give it another go to get CYMK profiles working correctly. Richard. From graeme2 at argyllcms.com Thu Dec 10 21:51:35 2009 From: graeme2 at argyllcms.com (Graeme Gill) Date: Fri, 11 Dec 2009 08:51:35 +1100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> Message-ID: <4B216D67.2080307@argyllcms.com> Richard Hughes wrote: > I've just added code to git master to do this, and it indeed works to > get the gamut range for a few scanner and print (RGB) profiles I've > got here. It doesn't seem to work for CYMK profiles, although I'm sure > that's something crazy I'm doing (or not doing..). Subtractive media profiles usually have less easily defined gamut extents than additive (RGB) profiles. Unlike additive, you can't expect it to be a nice polygon in XYZ/xy/u'v' space. Something roughly accurate for a chromaticity diagram would involve sampling along the color girdle (ie. going through combinations between CMY: 1 0 0, 1 1 0, 0 1 0, 0 1 1, 0 0 1, 1 0 1, e.g. 1 0 0, 1 .1 0, 1 .2 0 .. 1 .9 0, 1 1 0, .9 1 0, .8 1 0, ... etc.) Graeme Gill. From hughsient at gmail.com Thu Dec 10 22:08:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 10 Dec 2009 22:08:49 +0000 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <4B216D67.2080307@argyllcms.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> <4B15B29C.2090105@argyllcms.com> <15e53e180912090904o489081c4y8c959c0dbc9c6428@mail.gmail.com> <4B202D16.1000901@argyllcms.com> <15e53e180912101328r246ee990t1488a4c96ece7a75@mail.gmail.com> <4B216D67.2080307@argyllcms.com> Message-ID: <15e53e180912101408w4b7dca45o4dac5869574a39bd@mail.gmail.com> 2009/12/10 Graeme Gill : > 1 1 0, 0 1 0, 0 1 1, 0 0 1, 1 0 1, > e.g. 1 0 0, 1 .1 0, 1 .2 0 .. 1 .9 0, 1 1 0, .9 1 0, .8 1 0, ... etc.) Excellent, this makes sense. I'll give it a go tomorrow. Thanks again. Richard. From alexandre.prokoudine at gmail.com Thu Dec 10 22:35:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 01:35:09 +0300 Subject: Incorrect parsing of .icm files? In-Reply-To: <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> References: <1260478573.20024.33.camel@kusanagi> <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> Message-ID: <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> On 12/11/09, Richard Hughes wrote: > Both, I think. If you try gnome-power-manager in git master, it won't > "do" anything, as the profile does not have a vcgt or mlut tag, and > even using the linear tables doesn't do anything. Even trying using > dispwin doesn't appear to do anything. If you use the 2.29.1 release > there's a bug where we're using the xTRC values as a gamma table > (eek). BTW, what's new regarding restoration of applied profile after dehibernation and leaving a screensaver? Alexandre From hughsient at gmail.com Fri Dec 11 08:55:43 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 08:55:43 +0000 Subject: Incorrect parsing of .icm files? In-Reply-To: <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> References: <1260478573.20024.33.camel@kusanagi> <15e53e180912101324y610082f1tb1367f64c48c4e2d@mail.gmail.com> <733f2c730912101435g5939c476t1708b40492eeca98@mail.gmail.com> Message-ID: <15e53e180912110055s2a1bc938ob58b1ef4508ef83a@mail.gmail.com> 2009/12/10 Alexandre Prokoudine : > BTW, what's new regarding restoration of applied profile after > dehibernation and leaving a screensaver? If you're using a KMS driver (the future) then the "thaw" should re-apply the per-output gamma tables. With the screensaver I want to work on a patch that either: * saves the gamma table before the table is destroyed * calls gcm-apply after the screensaver comes back from black I don't use a screensaver myself, but if nobody does the gnome-screensaver work I guess I'll have to add that to my list. Richard. From knizek.confy at volny.cz Fri Dec 11 09:46:18 2009 From: knizek.confy at volny.cz (Milan =?UTF-8?Q?Kn=C3=AD=C5=BEek?=) Date: Fri, 11 Dec 2009 10:46:18 +0100 Subject: shared-color-profiles 0.1.0 release In-Reply-To: <15e53e180912101003s2a013080qf839379565a40964@mail.gmail.com> References: <15e53e180912091151y1f2303acw48cedde3eb0231f@mail.gmail.com> <1260467822.3147.14.camel@athlon> <15e53e180912101003s2a013080qf839379565a40964@mail.gmail.com> Message-ID: <1260524778.2203.3.camel@athlon> (I am forwarding your response to the emailing list, since my original intention was not to email just you but the list and I did not notice the "reply" address is just yours.) Possibly someone knows about other pre-packaged profiles? regards, Milan Richard Hughes p??e v ?t 10. 12. 2009 v 18:03 +0000: > 2009/12/10 Milan Kn??ek : > > Will the project make "icc-profiles" and "oyranos-icc" obsolete or > > rather complementary? > > I hope we can standardize on one -- I must admit I had not heard of > icc-profiles until you pointed me at them. > > In the future, we'll be distributing shared-color-targets as well (8.7 > targets) to make all this stuff just work out of the box. > > Richard. > Milan Knizek knizek (dot) confy (at) volny (dot) cz http://www.milan-knizek.net - About linux and photography (Czech language only) From lars.tore at mulebakken.net Fri Dec 11 15:11:33 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Fri, 11 Dec 2009 16:11:33 +0100 Subject: Help required: Getting primary illuminants and white point from ICC file In-Reply-To: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> References: <15e53e180912010658k109b0cb8y8b7d686cf3704073@mail.gmail.com> Message-ID: <58497f010912110711n7d6a6e0ub9a3bf5b9f817b2f@mail.gmail.com> On Tue, Dec 1, 2009 at 3:58 PM, Richard Hughes wrote: > the x and y coordinates of the white point from the device ICC > profile. > The white point cross is nice, but I wonder if it would be nice to add a more human readable text string with the whitepoint somewhere: The wtpt program included in liblcms-utils outputs the correlated temperature or the locus temperature, so I guess it's easy to this with lcms. Here are some sample output: profile 1: White point near 4937K profile 2: WhitePoint : D65 (daylight) And I think this is more readable to most people than xy values. Regards Lars Tore Gustavsen From pmjdebruijn at pcode.nl Fri Dec 11 15:51:36 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 11 Dec 2009 16:51:36 +0100 Subject: LCMS dependency In-Reply-To: <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> Message-ID: <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> On Thu, Dec 10, 2009 at 6:14 PM, Richard Hughes wrote: > 2009/12/10 Pascal de Bruijn : >> Well, lcms is used by most OSS graphical apps for color management, so >> lcms is a natural choice when we don't want to introduce new >> dependancies into most systems in general. > > Right. I see a lot of commits regarding lcms in git... Has all the ICC parsing code already been removed from gcm? I'm just asking this out of curiosity... Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Fri Dec 11 17:47:18 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 11 Dec 2009 18:47:18 +0100 Subject: Regression: the TRC graphs disappear on working space/input profiles Message-ID: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> Hi, With today's git checkout the TRC graphs disappear with working space profiles and input profiles. Is this intentional? Or just transitory during the lcms migration? Regards, Pascal de Bruijn From alexandre.prokoudine at gmail.com Fri Dec 11 17:47:30 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 20:47:30 +0300 Subject: apply/session Message-ID: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> Hi, So, to get things right, what exactly is the difference between gcm-session and gcm-apply? Alexandre From hughsient at gmail.com Fri Dec 11 17:50:06 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:50:06 +0000 Subject: LCMS dependency In-Reply-To: <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> Message-ID: <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> 2009/12/11 Pascal de Bruijn : > I see a lot of commits regarding lcms in git... > Has all the ICC parsing code already been removed from gcm? About half. To remove the other half we need lcms2, which may be some months away. Richard. From hughsient at gmail.com Fri Dec 11 17:51:47 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:51:47 +0000 Subject: apply/session In-Reply-To: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> Message-ID: <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> 2009/12/11 Alexandre Prokoudine : > So, to get things right, what exactly is the difference between > gcm-session and gcm-apply? gcm-session is the dbus-activated session service, that provides an interface applications like gimp can use. gcm-apply is just a quick hack that sets (or resets) all the CLUT values to what they should be, in case something destroyed them (e.g. hibernate, screensaver dim, etc). Richard From hughsient at gmail.com Fri Dec 11 17:52:49 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 11 Dec 2009 17:52:49 +0000 Subject: Regression: the TRC graphs disappear on working space/input profiles In-Reply-To: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> References: <4c2c37f90912110947v7f879c60ue781da65b9602348@mail.gmail.com> Message-ID: <15e53e180912110952s3f10d3dmc6305f350a17548@mail.gmail.com> 2009/12/11 Pascal de Bruijn : > With today's git checkout the TRC graphs disappear with working space > profiles and input profiles. > Is this intentional? Or just transitory during the lcms migration? Intentional, but they'll be coming back hopefully on Monday when I merge proper TRC support. Richard From alexandre.prokoudine at gmail.com Fri Dec 11 18:02:09 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 21:02:09 +0300 Subject: apply/session In-Reply-To: <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> Message-ID: <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> On Fri, Dec 11, 2009 at 8:51 PM, Richard Hughes wrote: > 2009/12/11 Alexandre Prokoudine : >> So, to get things right, what exactly is the difference between >> gcm-session and gcm-apply? > > gcm-session is the dbus-activated session service, that provides an > interface applications like gimp can use. And it's not currently used by anything, if I get it right? > gcm-apply is just a quick > hack that sets (or resets) all the CLUT values to what they should be, > in case something destroyed them (e.g. hibernate, screensaver dim, > etc). So vcgt part and setting the atom is done entirely by gcm-prefs? Alexandre From alexandre.prokoudine at gmail.com Fri Dec 11 18:06:41 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Fri, 11 Dec 2009 21:06:41 +0300 Subject: LCMS dependency In-Reply-To: <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> References: <15e53e180912100726u9e1b11v2a47db0225a8ac96@mail.gmail.com> <4B211C4E.9060502@argyllcms.com> <15e53e180912100818l7af5dec8jdaf230304fc0a1a6@mail.gmail.com> <4c2c37f90912100900o717a7c27o1c30c23208c05f92@mail.gmail.com> <15e53e180912100914pf12493fkc795ac55994b3049@mail.gmail.com> <4c2c37f90912110751r4344cbaem2b4b00e6e58f8c8@mail.gmail.com> <15e53e180912110950m1a174a2bi3e469a05e1a14e0e@mail.gmail.com> Message-ID: <733f2c730912111006s640088c6g6b11a1266c54e4c8@mail.gmail.com> On Fri, Dec 11, 2009 at 8:50 PM, Richard Hughes wrote: >> Has all the ICC parsing code already been removed from gcm? > > About half. To remove the other half we need lcms2, which may be some > months away. Is this why TRC curve has (temp.) gone? :) Alexandre From claude at 2xlibre.net Fri Dec 11 22:36:40 2009 From: claude at 2xlibre.net (Claude Paroz) Date: Fri, 11 Dec 2009 23:36:40 +0100 Subject: Why not using strftime? Message-ID: <1260571000.4145.121.camel@shuttle> Hi, As I didn't find any bug tracker now, I'm reporting this here. I wonder why you're using a complicated machinery with gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of simply using strftime call in gcm-profile.c Something like (not tested code): ret = cmsTakeCreationDateTime (&created, priv->lcms_profile); if (ret) - priv->datetime = gcm_utils_format_date_time (1900 +created.tm_year, created.tm_mon+1, created + /* Translators: this is the profile creation date strftime format */ + strftime (priv->datetime, sizeof(priv->datetime), _("%d %B %Y, %H:%M:%S"), created); Months names should not be translated again, this is a waste of time for translators. Claude P.S. Please cc me on answer From hughsient at gmail.com Sat Dec 12 09:01:19 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 12 Dec 2009 09:01:19 +0000 Subject: apply/session In-Reply-To: <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> References: <733f2c730912110947i387f7d4dkb3f8d9f26e1f3854@mail.gmail.com> <15e53e180912110951i73aefe24sd38c7bcfbd64e862@mail.gmail.com> <733f2c730912111002p6808bb4ah2a28b9a89f4f064c@mail.gmail.com> Message-ID: <15e53e180912120101n74bc1e92i31b4359ccaab961e@mail.gmail.com> 2009/12/11 Alexandre Prokoudine : >> gcm-session is the dbus-activated session service, that provides an >> interface applications like gimp can use. > > And it's not currently used ?by anything, if I get it right? Correct. It's the egg part of the chicken-and-egg question. >> gcm-apply is just a quick >> hack that sets (or resets) all the CLUT values to what they should be, >> in case something destroyed them (e.g. hibernate, screensaver dim, >> etc). > > So vcgt part and setting the atom is done entirely by gcm-prefs? No, when gcm "set the CLUT" it actually: * Applies the CLUT to each adaptor * Sets the _ICC_PROFILE atom on each output * Sets the _ICC_PROFILE atom on the screen for the primary output To apply the clut we extract the VCGT and then push it into the video card. Settign the CLUT is done by gcm-apply and gcm-prefs. You could argue the latter is redundant, unless you're changing the profile, and I would agree. Richard. From hughsient at gmail.com Sat Dec 12 09:21:57 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 12 Dec 2009 09:21:57 +0000 Subject: Why not using strftime? In-Reply-To: <1260571000.4145.121.camel@shuttle> References: <1260571000.4145.121.camel@shuttle> Message-ID: <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> 2009/12/11 Claude Paroz : > As I didn't find any bug tracker now, I'm reporting this here. I wonder > why you're using a complicated machinery with > gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of > simply using strftime call in gcm-profile.c Originally because we got some complicated date structure to decode, now we're using lcms we just get a nice sane tm struct. I've fixed things up in the following commit: commit 6607487433a1bf074638d6833bd7c493ef1d75ff Author: Richard Hughes Date: Sat Dec 12 09:19:57 2009 +0000 Use strftime rather than our own hand-rolled function :100644 100644 a6abf96... 0abfb35... M src/gcm-profile.c :100644 100644 203466c... 76e5b24... M src/gcm-utils.c :100644 100644 bc3ad9d... a570d6b... M src/gcm-utils.h Thanks Richard. From claude at 2xlibre.net Sat Dec 12 09:48:28 2009 From: claude at 2xlibre.net (Claude Paroz) Date: Sat, 12 Dec 2009 10:48:28 +0100 Subject: Why not using strftime? In-Reply-To: <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> References: <1260571000.4145.121.camel@shuttle> <15e53e180912120121y2dfde519y994dcea52ab76c26@mail.gmail.com> Message-ID: <1260611308.3833.38.camel@shuttle> Le samedi 12 d?cembre 2009 ? 09:21 +0000, Richard Hughes a ?crit : > 2009/12/11 Claude Paroz : > > As I didn't find any bug tracker now, I'm reporting this here. I wonder > > why you're using a complicated machinery with > > gcm_utils_format_date_time/gcm_utils_month_to_localized_text instead of > > simply using strftime call in gcm-profile.c > > Originally because we got some complicated date structure to decode, > now we're using lcms we just get a nice sane tm struct. I've fixed > things up in the following commit: > > commit 6607487433a1bf074638d6833bd7c493ef1d75ff > Author: Richard Hughes > Date: Sat Dec 12 09:19:57 2009 +0000 > > Use strftime rather than our own hand-rolled function > > :100644 100644 a6abf96... 0abfb35... M src/gcm-profile.c > :100644 100644 203466c... 76e5b24... M src/gcm-utils.c > :100644 100644 bc3ad9d... a570d6b... M src/gcm-utils.h > > Thanks Great, thanks ! Claude From pmjdebruijn at pcode.nl Mon Dec 14 14:44:19 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 14 Dec 2009 15:44:19 +0100 Subject: Patch: Debian binary detection fix In-Reply-To: <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> Message-ID: <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> On Thu, Nov 26, 2009 at 10:07 PM, Richard Hughes wrote: > 2009/11/26 Pascal de Bruijn : >> Therefore the detect should _first_ detect if the Debian renamed >> binaries exist, and if not, fallback to the original binary names. > > Agreed. I've attached the patch I've committed. Thanks. Maybe it's a thought to (last) search /usr/local/bin as well for the standard binary names? Regards, Pascal de Bruijn From hughsient at gmail.com Mon Dec 14 15:13:51 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 14 Dec 2009 15:13:51 +0000 Subject: Patch: Debian binary detection fix In-Reply-To: <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> Message-ID: <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> 2009/12/14 Pascal de Bruijn : > Maybe it's a thought to (last) search /usr/local/bin as well for the > standard binary names? Sure. I'll do a deal: You code it, and I'll apply it. If you need instructions, I can explain how. Richard. From hughsient at gmail.com Wed Dec 16 11:57:19 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 16 Dec 2009 11:57:19 +0000 Subject: PackageKit integration Message-ID: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> gnome-color-manager in git master now integrates with PackageKit in the following ways: * Offers to install argyllcms if the user tries to calibrate without argyllcms installed * Offers to install shared-color-targets if they are not installed, and the user tries to calibrate a non-display device This means gnome-color-manager can be shipped on live-cd's without bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) shared-color-targets, just in case the user wants to calibrate a device. If people want to turn off the PackageKit bits, then I can add a GConf key or #ifdef as required. Richard. From pmjdebruijn at pcode.nl Wed Dec 16 12:23:52 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 16 Dec 2009 13:23:52 +0100 Subject: PackageKit integration In-Reply-To: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> Message-ID: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> On Wed, Dec 16, 2009 at 12:57 PM, Richard Hughes wrote: > gnome-color-manager in git master now integrates with PackageKit in > the following ways: > > * Offers to install argyllcms if the user tries to calibrate without > argyllcms installed > * Offers to install shared-color-targets if they are not installed, > and the user tries to calibrate a non-display device > > This means gnome-color-manager can be shipped on live-cd's without > bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) > shared-color-targets, just in case the user wants to calibrate a > device. > > If people want to turn off the PackageKit bits, then I can add a GConf > key or #ifdef as required. Since not all distro's don't use PackageKit that would be very welcome. A ./configure option would probably be best? Since this is a distro specific setting, it needs to be set in the packaging process. And GConf is very user specific, so it would seem odd to use GConf for this particular purpose. Regards, Pascal de Bruijn From pmjdebruijn at pcode.nl Wed Dec 16 12:32:06 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Wed, 16 Dec 2009 13:32:06 +0100 Subject: PackageKit integration In-Reply-To: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> Message-ID: <4c2c37f90912160432r5cc68f8as29bc4d7419eeecd8@mail.gmail.com> On Wed, Dec 16, 2009 at 1:23 PM, Pascal de Bruijn wrote: > On Wed, Dec 16, 2009 at 12:57 PM, Richard Hughes wrote: >> gnome-color-manager in git master now integrates with PackageKit in >> the following ways: >> >> * Offers to install argyllcms if the user tries to calibrate without >> argyllcms installed >> * Offers to install shared-color-targets if they are not installed, >> and the user tries to calibrate a non-display device >> >> This means gnome-color-manager can be shipped on live-cd's without >> bringing in the 5.0Mb argyllcms, and the 1.4Mb (and rising) >> shared-color-targets, just in case the user wants to calibrate a >> device. >> >> If people want to turn off the PackageKit bits, then I can add a GConf >> key or #ifdef as required. > > Since not all distro's don't use PackageKit that would be very welcome. > > A ./configure option would probably be best? Since this is a distro > specific setting, it needs to be set in the packaging process. > > And GConf is very user specific, so it would seem odd to use GConf for > this particular purpose. Hmm doesn't totem do similar things with codecs and the gstreamer packages? GCM should probably follow a similar approach... Regards, Pascal de Bruijn From hughsient at gmail.com Wed Dec 16 12:42:07 2009 From: hughsient at gmail.com (Richard Hughes) Date: Wed, 16 Dec 2009 12:42:07 +0000 Subject: PackageKit integration In-Reply-To: <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> References: <15e53e180912160357r23a01207pdc15daf7b550c666@mail.gmail.com> <4c2c37f90912160423n7fa9f7far32a64bdd4017cd74@mail.gmail.com> Message-ID: <15e53e180912160442v2f95d4rb4b670389fb556d9@mail.gmail.com> 2009/12/16 Pascal de Bruijn : > Since not all distro's don't use PackageKit that would be very welcome. commit 494ebc7e84d21d2626e06aa7e009cc90625e0fd8 Author: Richard Hughes Date: Wed Dec 16 12:41:02 2009 +0000 Add a configure option --enable-packagekit (default yes) to choose if auto-install functionality should be included Richard. From joedalton2 at yahoo.dk Wed Dec 16 22:16:18 2009 From: joedalton2 at yahoo.dk (Joe Dalton) Date: Wed, 16 Dec 2009 14:16:18 -0800 (PST) Subject: bugreport gnome-color-manager translation Message-ID: <977326.61727.qm@web24408.mail.ird.yahoo.com> "Report-Msgid-Bugs-To: http://bugzilla.gnome.org/enter_bug.cgi?product=gnome-" "color-manager\n" but https://bugzilla.gnome.org/enter_bug.cgi?product=gnome-color-manager gives this message Sorry, either the product gnome-color-manager does not exist or you aren't authorized to enter a bug into it. Please choose a valid product. What does intent mean here. Is the sentence broken in some way/missing something? #: ../data/gnome-color-manager.schemas.in.h:7 msgid "The output intent to use, e.g. 'perceptual'." Hi, a google did not help me, and i have only found one danish translation, which maybe is correct mayby not. Please give some more info. #. TRANSLATORS: rendering intent: you probably want to google this #: ../src/gcm-prefs.c:2029 msgid "Perceptual" bye Joe Danish translation ___________________________________________________________ Skal du k?be ny bil? Sammenlign priser p? brugte biler med Kelkoo og find et godt tilbud! - Se mere her http://dk.yahoo.com/r/pat/mmb From hughsient at gmail.com Thu Dec 17 09:31:22 2009 From: hughsient at gmail.com (Richard Hughes) Date: Thu, 17 Dec 2009 09:31:22 +0000 Subject: bugreport gnome-color-manager translation In-Reply-To: <977326.61727.qm@web24408.mail.ird.yahoo.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> Message-ID: <15e53e180912170131p5c35354dv6e3df58be1ee6782@mail.gmail.com> 2009/12/16 Joe Dalton : > Sorry, either the product gnome-color-manager ?does not exist or you aren't authorized to enter a bug into it. Please choose a valid product. I've opened a bug to have the bugzilla product created: https://bugzilla.gnome.org/show_bug.cgi?id=604811 > What does intent mean here. Is the sentence broken in some way/missing something? > > #: ../data/gnome-color-manager.schemas.in.h:7 > msgid "The output intent to use, e.g. 'perceptual'." The output intent is the way that one range of colors is mapped to another range, for instance, do we clip the extremes of color, or do we try to scale it so they are all represented. Intent in English means "what you would do if possible" although in this technical speak it means more of "what method do you want me to use when transforming the color" > Hi, a google did not help me, and i have only found one danish translation, which maybe is correct mayby not. Please give some more info. > #. TRANSLATORS: rendering intent: you probably want to google this > #: ../src/gcm-prefs.c:2029 > msgid "Perceptual" This is how I understand perceptual rendering intent: '''This rendering intent maps color "smoothly", preserving relationships between similar colors. This prevents "gamut clipping" with its potential loss of detail and "tonal banding" problems. Gamut clipping occurs when colors that are different in the input image appear the same when printed. Perceptual rendering intent makes small compromises throughout the entire color space in order to preserve color relationships. It sacrifices some precision of in-gamut colors in order to ensure pleasing results.''' Does that make things clearer? Feel free to change the translator comments to something more sensible -- you know what you are doing. :-) Richard. From alexandre.prokoudine at gmail.com Thu Dec 17 12:29:54 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Thu, 17 Dec 2009 15:29:54 +0300 Subject: bugreport gnome-color-manager translation In-Reply-To: <977326.61727.qm@web24408.mail.ird.yahoo.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> Message-ID: <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> On 12/17/09, Joe Dalton wrote: > What does intent mean here. Is the sentence broken in some way/missing > something? > > #: ../data/gnome-color-manager.schemas.in.h:7 > msgid "The output intent to use, e.g. 'perceptual'." > > > Hi, a google did not help me, and i have only found one danish translation, > which maybe is correct mayby not. I'm pretty sure that both GIMP, Inkscape and Scribus have Danish translations. Check them re. "Color management" tab in Preferences of all these applications. I also highly recommend "Real world color management" book. It was translated into many languages, maybe Danish was among them. You will understand a lot more about color management after reading it, trust me on this :) Alexandre From anders at brander.dk Thu Dec 17 12:35:12 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 17 Dec 2009 13:35:12 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> Message-ID: <1261053312.7873.1.camel@smutogtab> Hi, On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: > I also highly recommend "Real world color management" book. It was > translated into many languages, maybe Danish was among them. You will > understand a lot more about color management after reading it, trust > me on this :) I second that, go read "Real world color management" now, if you haven't already. It doesn't have a danish translation by the way. Regards, Anders Brander From pmjdebruijn at pcode.nl Thu Dec 17 12:38:42 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Thu, 17 Dec 2009 13:38:42 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <1261053312.7873.1.camel@smutogtab> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> Message-ID: <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> On Thu, Dec 17, 2009 at 1:35 PM, Anders Brander wrote: > Hi, > > On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: >> I also highly recommend "Real world color management" book. It was >> translated into many languages, maybe Danish was among them. You will >> understand a lot more about color management after reading it, trust >> me on this :) > > I second that, go read "Real world color management" now, if you haven't > already. It doesn't have a danish translation by the way. I third that, the first 75% is written completely generic and os/application independent. It's an excellent book... Generally for most photographical work you'll want Perceptual... Relative Colorimetric is usually most useful when you have very high end equipment, thus limiting clipping... Regards, Pascal de Bruijn From lars.tore at mulebakken.net Thu Dec 17 15:11:12 2009 From: lars.tore at mulebakken.net (Lars Tore Gustavsen) Date: Thu, 17 Dec 2009 16:11:12 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> Message-ID: <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> On Thu, Dec 17, 2009 at 1:38 PM, Pascal de Bruijn wrote: > On Thu, Dec 17, 2009 at 1:35 PM, Anders Brander wrote: >> Hi, >> >> On Thu, 2009-12-17 at 15:29 +0300, Alexandre Prokoudine wrote: >>> I also highly recommend "Real world color management" book. It was >>> translated into many languages, maybe Danish was among them. You will >>> understand a lot more about color management after reading it, trust >>> me on this :) >> >> I second that, go read "Real world color management" now, if you haven't >> already. It doesn't have a danish translation by the way. > > I third that, the first 75% is written completely generic and > os/application independent. It's an excellent book... > > Generally for most photographical work you'll want Perceptual... > > Relative Colorimetric is usually most useful when you have very high > end equipment, thus limiting clipping... > > Regards, > Pascal de Bruijn > _______________________________________________ > gnome-color-manager-list mailing list > gnome-color-manager-list at gnome.org > http://mail.gnome.org/mailman/listinfo/gnome-color-manager-list > I think the book Understanding color managment by Abhay Sharma is a better book, and I have both :-) About the translation, in Norwegian (it might be close to danish) the correct word is "Perseptuell". But I think we don't have to translate all that specific terms. Since most of the written literature are in english, I guess most people who use color managment have no problems with the english terms on the rendering intets. Actually I think at least in my language that translated word will create a lot of confusion since the words are very seldom used, and everyone use the English terms. I guess we can blame photoshop for this. Lars Tore Gustavsen From anders at brander.dk Thu Dec 17 15:20:42 2009 From: anders at brander.dk (Anders Brander) Date: Thu, 17 Dec 2009 16:20:42 +0100 Subject: bugreport gnome-color-manager translation In-Reply-To: <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> References: <977326.61727.qm@web24408.mail.ird.yahoo.com> <733f2c730912170429i432019cfj5b80294b83ff33f2@mail.gmail.com> <1261053312.7873.1.camel@smutogtab> <4c2c37f90912170438p5cbc9b39ia8784bf8423203ac@mail.gmail.com> <58497f010912170711hd596c1dtd5b7a63f4ee9db31@mail.gmail.com> Message-ID: <1261063242.15170.1.camel@smutogtab> Hi, On Thu, 2009-12-17 at 16:11 +0100, Lars Tore Gustavsen wrote: > About the translation, in Norwegian (it might be close to danish) the > correct word is "Perseptuell". I just looked it up in the danish Rawstudio translation, it's translated as "Opfattelse" - but I don't think that's correct (and I'm danish). Regards, Anders Brander From hughsient at gmail.com Sat Dec 19 17:32:08 2009 From: hughsient at gmail.com (Richard Hughes) Date: Sat, 19 Dec 2009 17:32:08 +0000 Subject: Bugzilla account created Message-ID: <15e53e180912190932x47e70876xcb147ce2718e3bff@mail.gmail.com> I've arranged for a bugzilla component to be created. If you want to file a bug, use https://bugzilla.gnome.org/enter_bug.cgi?product=gnome-color-manager but be sure to check for duplicates first, using https://bugzilla.gnome.org/buglist.cgi?product=gnome-color-manager At the moment, there are no bugs, which means GCM must be feature complete and perfect in every way. I can't imagine this situation will last long :-) Richard. From robert.ancell at gmail.com Sun Dec 20 22:34:24 2009 From: robert.ancell at gmail.com (Robert Ancell) Date: Mon, 21 Dec 2009 09:34:24 +1100 Subject: Color management and Simple Scan Message-ID: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Hi all, I've been developing a basic scanning application called Simple Scan ( http://launchpad.net/simple-scan) and one request is to add color management to it: https://bugs.edge.launchpad.net/simple-scan/+bug/498029 I don't know much about color management so my question is: - What do I need to do to make Simple Scan work with GNOME Color Manager? - Is there a GPL project I can lift some code from? Thanks, --Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughsient at gmail.com Mon Dec 21 08:28:27 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 21 Dec 2009 08:28:27 +0000 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Message-ID: <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> 2009/12/20 Robert Ancell : > I don't know much about color management so my question is: > - What do I need to do to make Simple Scan work with GNOME Color Manager? > - Is there a GPL project I can lift some code from? Well, first you need to ask gnome-color-manager for a list of suitable profiles using the DBus interface: http://git.gnome.org/browse/gnome-color-manager/plain/src/org.gnome.ColorManager.xml If you know the device you're using (you probably should) then you can call GetProfilesForDevice() with the sysfs path, and a list of profiles will be returned. If more than one profile is returned you probably need to show it in a combo-box or something, but it would be quite unusual for more than one profile to be returned. If you get no results, then it's probably because the user has never assigned a profile for that device (i.e. there is no a-priori mapping). In this case you can either default to not doing color conversion yourself, or instead call the GetProfilesForType() dbus method, using the type=scanner. This will return a list of profiles that are suitable for use with scanners. Now, when you have a filename of the profile you want to use it's an easy (!) case of acquiring the image like usual, and then you can either: * Just save the image with an embedded color profile (see bug https://bugzilla.gnome.org/show_bug.cgi?id=604610 -- I'm working as fast as I can on the infrastructure bits) * Convert the image from device-mapping to something like sRGB using lcms (there are examples in the lcms tarball) I think it's probably best you do the first one, unless you know you're going to be using the images in a non-color-aware application. Anyway, I hope that helps, Richard. From alexandre.prokoudine at gmail.com Mon Dec 21 16:01:01 2009 From: alexandre.prokoudine at gmail.com (Alexandre Prokoudine) Date: Mon, 21 Dec 2009 19:01:01 +0300 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> Message-ID: <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> On Mon, Dec 21, 2009 at 1:34 AM, Robert Ancell wrote: > - Is there a GPL project I can lift some code from? XSane sounds like an obvious source :) Alexandre From hughsient at gmail.com Mon Dec 21 16:08:05 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 21 Dec 2009 16:08:05 +0000 Subject: Color management and Simple Scan In-Reply-To: <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <733f2c730912210801ld9cc0f9u478099eaf3bb234@mail.gmail.com> Message-ID: <15e53e180912210808y5fe054c7p1f2d41f4fdb97ee1@mail.gmail.com> 2009/12/21 Alexandre Prokoudine : > XSane sounds like an obvious source :) Well, depends on how you like your sanity. Richard. From robert.ancell at gmail.com Mon Dec 21 22:57:54 2009 From: robert.ancell at gmail.com (Robert Ancell) Date: Tue, 22 Dec 2009 09:57:54 +1100 Subject: Color management and Simple Scan In-Reply-To: <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> Message-ID: <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> Thanks very much Richard! Very interested in the GDK patch - I need to make a patch there too to embed resolution information: https://bugs.edge.launchpad.net/simple-scan/+bug/484616 I think I need to do the conversion regardless so the preview is displayed correctly. Thanks for the lcms link. --Robert 2009/12/21 Richard Hughes > 2009/12/20 Robert Ancell : > > I don't know much about color management so my question is: > > - What do I need to do to make Simple Scan work with GNOME Color Manager? > > - Is there a GPL project I can lift some code from? > > Well, first you need to ask gnome-color-manager for a list of suitable > profiles using the DBus interface: > > http://git.gnome.org/browse/gnome-color-manager/plain/src/org.gnome.ColorManager.xml > > If you know the device you're using (you probably should) then you can > call GetProfilesForDevice() with the sysfs path, and a list of > profiles will be returned. If more than one profile is returned you > probably need to show it in a combo-box or something, but it would be > quite unusual for more than one profile to be returned. If you get no > results, then it's probably because the user has never assigned a > profile for that device (i.e. there is no a-priori mapping). In this > case you can either default to not doing color conversion yourself, or > instead call the GetProfilesForType() dbus method, using the > type=scanner. This will return a list of profiles that are suitable > for use with scanners. > > Now, when you have a filename of the profile you want to use it's an > easy (!) case of acquiring the image like usual, and then you can > either: > > * Just save the image with an embedded color profile (see bug > https://bugzilla.gnome.org/show_bug.cgi?id=604610 -- I'm working as > fast as I can on the infrastructure bits) > * Convert the image from device-mapping to something like sRGB using > lcms (there are examples in the lcms tarball) > > I think it's probably best you do the first one, unless you know > you're going to be using the images in a non-color-aware application. > > Anyway, I hope that helps, > > Richard. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From hughsient at gmail.com Tue Dec 22 09:13:09 2009 From: hughsient at gmail.com (Richard Hughes) Date: Tue, 22 Dec 2009 09:13:09 +0000 Subject: Color management and Simple Scan In-Reply-To: <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> References: <9a0b92c0912201434w5bce10bewa1d8e1636219ed9d@mail.gmail.com> <15e53e180912210028l49c39b4auc02902223b8a58ad@mail.gmail.com> <9a0b92c0912211457s5590c31ejd9fe5a5934de8cdf@mail.gmail.com> Message-ID: <15e53e180912220113n329e8f81j982c21203734945f@mail.gmail.com> 2009/12/21 Robert Ancell : > Very interested in the GDK patch - I need to make a patch there too to embed > resolution information: > https://bugs.edge.launchpad.net/simple-scan/+bug/484616 Sure, looks like you just need to add a new pixbuf-option. It's not that hard, as long as you're comfortable just implementing a spec in otherwise fairly hairy code. :-) If you need a hand, give me a shout on or off list. Richard. From bcowan at fastmail.co.uk Thu Dec 24 20:06:29 2009 From: bcowan at fastmail.co.uk (Bruce Cowan) Date: Thu, 24 Dec 2009 20:06:29 +0000 Subject: String fixes Message-ID: <1261685189.7547.4.camel@localhost.localdomain> I've found a few string errors. Here's a patch to fix them. I thought I better post it here rather than just commit straight to the repository, because there is a potential point of disagreement. I've changed the strformat format to the en_US one, which could be a bit contentious. -- Bruce Cowan -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-Minor-string-fixes.patch Type: text/x-patch Size: 2223 bytes Desc: not available URL: From hughsient at gmail.com Fri Dec 25 11:26:32 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 25 Dec 2009 11:26:32 +0000 Subject: Happy Christmas Message-ID: <15e53e180912250326n7d35ca86tc59fb282fd940a1c@mail.gmail.com> Hey guys. I just wanted to send to you all a Happy Christmas, and my thanks for the hard work you've put in this year. It's appreciated. Richard. From hughsient at gmail.com Fri Dec 25 11:28:10 2009 From: hughsient at gmail.com (Richard Hughes) Date: Fri, 25 Dec 2009 11:28:10 +0000 Subject: String fixes In-Reply-To: <1261685189.7547.4.camel@localhost.localdomain> References: <1261685189.7547.4.camel@localhost.localdomain> Message-ID: <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> 2009/12/24 Bruce Cowan : > I've found a few string errors. Here's a patch to fix them. I thought I > better post it here rather than just commit straight to the repository, > because there is a potential point of disagreement. The first two are obvious, so please commit those. > I've changed the strformat format to the en_US one, which could be a bit > contentious. I agree it's probably the correct thing to do, but it does however want me to rip my eyes out every time I see that format. :-) So, please commit the entire patch. Thanks. Richard. From pmjdebruijn at pcode.nl Fri Dec 25 12:28:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Fri, 25 Dec 2009 13:28:33 +0100 Subject: String fixes In-Reply-To: <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> References: <1261685189.7547.4.camel@localhost.localdomain> <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> Message-ID: <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> On Fri, Dec 25, 2009 at 12:28 PM, Richard Hughes wrote: > 2009/12/24 Bruce Cowan : >> I've found a few string errors. Here's a patch to fix them. I thought I >> better post it here rather than just commit straight to the repository, >> because there is a potential point of disagreement. > > The first two are obvious, so please commit those. > >> I've changed the strformat format to the en_US one, which could be a bit >> contentious. > > I agree it's probably the correct thing to do, but it does however > want me to rip my eyes out every time I see that format. :-) /me Shudders... We're introducing AM/PM time... oh well seems correct for the en_US locale... However, if one we're to switch to the en_UK locale, would AM/PM time be gone again? Regards, Pascal de Bruijn From bcowan at fastmail.co.uk Fri Dec 25 14:37:16 2009 From: bcowan at fastmail.co.uk (Bruce Cowan) Date: Fri, 25 Dec 2009 14:37:16 +0000 Subject: String fixes In-Reply-To: <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> References: <1261685189.7547.4.camel@localhost.localdomain> <15e53e180912250328h513660c3v6d96cca72ce507af@mail.gmail.com> <4c2c37f90912250428j7e5dc582y24ac25b08f67610f@mail.gmail.com> Message-ID: <1261751836.1631.0.camel@localhost.localdomain> On Fri, 2009-12-25 at 13:28 +0100, Pascal de Bruijn wrote: > On Fri, Dec 25, 2009 at 12:28 PM, Richard Hughes wrote: > > 2009/12/24 Bruce Cowan : > >> I've found a few string errors. Here's a patch to fix them. I thought I > >> better post it here rather than just commit straight to the repository, > >> because there is a potential point of disagreement. > > > > The first two are obvious, so please commit those. > > > >> I've changed the strformat format to the en_US one, which could be a bit > >> contentious. > > > > I agree it's probably the correct thing to do, but it does however > > want me to rip my eyes out every time I see that format. :-) > > /me Shudders... > > We're introducing AM/PM time... oh well seems correct for the en_US locale... > > However, if one we're to switch to the en_UK locale, would AM/PM time > be gone again? Yup, I'll update the en_GB translation later as well, meaning we'll see 24 hour time still. -- Bruce Cowan From joedalton2 at yahoo.dk Fri Dec 25 17:59:33 2009 From: joedalton2 at yahoo.dk (Joe Dalton) Date: Fri, 25 Dec 2009 17:59:33 +0000 (GMT) Subject: bugreport gnome-color-manager translation In-Reply-To: <15e53e180912170131p5c35354dv6e3df58be1ee6782@mail.gmail.com> Message-ID: <381861.61449.qm@web24403.mail.ird.yahoo.com> thanx --- Den tors 17/12/09 skrev Richard Hughes : > Fra: Richard Hughes > Emne: Re: bugreport gnome-color-manager translation > Til: "Joe Dalton" > Cc: gnome-color-manager-list at gnome.org > Dato: torsdag 17. december 2009 10.31 > 2009/12/16 Joe Dalton : > > Sorry, either the product gnome-color-manager ?does > not exist or you aren't authorized to enter a bug into it. > Please choose a valid product. > > I've opened a bug to have the bugzilla product created: > https://bugzilla.gnome.org/show_bug.cgi?id=604811 > > > What does intent mean here. Is the sentence broken in > some way/missing something? > > > > #: ../data/gnome-color-manager.schemas.in.h:7 > > msgid "The output intent to use, e.g. 'perceptual'." > > The output intent is the way that one range of colors is > mapped to > another range, for instance, do we clip the extremes of > color, or do > we try to scale it so they are all represented. Intent in > English > means "what you would do if possible" although in this > technical speak > it means more of "what method do you want me to use when > transforming > the color" > > > Hi, a google did not help me, and i have only found > one danish translation, which maybe is correct mayby not. > Please give some more info. > > #. TRANSLATORS: rendering intent: you probably want to > google this > > #: ../src/gcm-prefs.c:2029 > > msgid "Perceptual" > > This is how I understand perceptual rendering intent: > > '''This rendering intent maps color "smoothly", preserving > relationships between similar colors. This prevents "gamut > clipping" > with its potential loss of detail and "tonal banding" > problems. Gamut > clipping occurs when colors that are different in the input > image > appear the same when printed. Perceptual rendering intent > makes small > compromises throughout the entire color space in order to > preserve > color relationships. It sacrifices some precision of > in-gamut colors > in order to ensure pleasing results.''' > > Does that make things clearer? Feel free to change the > translator > comments to something more sensible -- you know what you > are doing. > :-) > > Richard. > Find din nye laptop p? kelkoo.dk. Se de gode tilbud her - http://dk.yahoo.com/r/pat/mm From pmjdebruijn at pcode.nl Sun Dec 27 18:33:43 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Sun, 27 Dec 2009 19:33:43 +0100 Subject: Patch: Debian binary detection fix In-Reply-To: <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> Message-ID: <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> On Mon, Dec 14, 2009 at 4:13 PM, Richard Hughes wrote: > 2009/12/14 Pascal de Bruijn : >> Maybe it's a thought to (last) search /usr/local/bin as well for the >> standard binary names? > > Sure. I'll do a deal: You code it, and I'll apply it. If you need > instructions, I can explain how. I've submitted it has an enhancement on GNOME Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=605552 Regards, Pascal de Bruijn From hughsient at gmail.com Mon Dec 28 10:26:15 2009 From: hughsient at gmail.com (Richard Hughes) Date: Mon, 28 Dec 2009 10:26:15 +0000 Subject: Patch: Debian binary detection fix In-Reply-To: <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> References: <4c2c37f90911261254y49b35366t13b85e32f505c013@mail.gmail.com> <15e53e180911261307l6b8b4427g13bc6de54b014f00@mail.gmail.com> <4c2c37f90912140644o32a5ea81la5da841d100c5a5d@mail.gmail.com> <15e53e180912140713x5bd8f12jc4a1126d9de0cf88@mail.gmail.com> <4c2c37f90912271033q215ab22cye03f7c6eb912a3b5@mail.gmail.com> Message-ID: <15e53e180912280226r2b1d3265x103a12fd65174192@mail.gmail.com> 2009/12/27 Pascal de Bruijn : > I've submitted it has an enhancement on GNOME Bugzilla: Committed, thanks. Richard. From pmjdebruijn at pcode.nl Mon Dec 28 18:20:54 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Mon, 28 Dec 2009 19:20:54 +0100 Subject: Documentation Images: Display OSD Message-ID: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Hi, I took a few shots of my display's OSD, mainly for my own purposes, but maybe they'll come in handy for GNOME Color Manager's documentation. These shots aren't perfect (at all), but it's a start. I can put these under any license you like... I'm fine with GPL and CC-BY-AC. Regards, Pascal de Bruijn -------------- next part -------------- A non-text attachment was scrubbed... Name: lowres-osd-jpegs.zip Type: application/zip Size: 213749 bytes Desc: not available URL: From niklas.park at gmail.com Tue Dec 29 10:44:50 2009 From: niklas.park at gmail.com (Niklas Park) Date: Tue, 29 Dec 2009 11:44:50 +0100 Subject: Documentation Images: Display OSD In-Reply-To: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> References: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Message-ID: A little warning seems to be in place here. I can't see how you could put these screenshots under CC, FDL or any other license, since the copyright clearly belongs to the manufacturer of the device, or the person or entity the manufacturer licenses the OSD from. I can't really see that a screengrab could constitute original art under any jurisdiction. The best one could hope for is that using these images constitutes fair use, of course with correct attribution to the copyright holder, but IANAL. Kindly /Niklas Park 2009/12/28 Pascal de Bruijn > Hi, > > I took a few shots of my display's OSD, mainly for my own purposes, > but maybe they'll come in handy for GNOME Color Manager's > documentation. > > These shots aren't perfect (at all), but it's a start. > > I can put these under any license you like... I'm fine with GPL and > CC-BY-AC. > > Regards, > Pascal de Bruijn > > _______________________________________________ > gnome-color-manager-list mailing list > gnome-color-manager-list at gnome.org > http://mail.gnome.org/mailman/listinfo/gnome-color-manager-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pmjdebruijn at pcode.nl Tue Dec 29 14:41:33 2009 From: pmjdebruijn at pcode.nl (Pascal de Bruijn) Date: Tue, 29 Dec 2009 15:41:33 +0100 Subject: Documentation Images: Display OSD In-Reply-To: References: <4c2c37f90912281020r14f14a8csd02b125cdbd5f7b9@mail.gmail.com> Message-ID: <4c2c37f90912290641w8d6dff1y675a3a07844da84d@mail.gmail.com> On Tue, Dec 29, 2009 at 11:44 AM, Niklas Park wrote: > A little warning seems to be in place here. > I can't see how you could put these screenshots under CC, FDL or any other > license, since the copyright clearly belongs to the manufacturer of the > device, or the person or entity the manufacturer licenses the OSD from. I > can't really see that a screengrab could constitute original art under any > jurisdiction. Well, you've got a fair point there... In this case that would be HP... > The best one could hope for is that using these images constitutes fair use, > of course with correct attribution to the copyright holder, but IANAL. Since it's not a 1:1 (pixel-perfect) copy, and doesn't contain any brand logo's, it probably does... Regards, Pascal de Bruijn