Re: [Tracker] [PATCH] Re: Extracting Embedded Licenses
- From: jamie <jamiemcc blueyonder co uk>
- To: Jason Kivlighn <jkivlighn gmail com>
- Cc: CC Developer Mailing List <cc-devel lists ibiblio org>, tracker-list gnome org
- Subject: Re: [Tracker] [PATCH] Re: Extracting Embedded Licenses
- Date: Mon, 02 Jul 2007 00:56:12 +0100
On Sun, 2007-07-01 at 14:35 -0700, Jason Kivlighn wrote:
On Sat, 2007-06-30 at 20:19 -0700, Jason Kivlighn wrote:
these look ok - will apply as soon as the other stuff is tidied up (need
tracker_read_xmp before I can apply these)
Here's a patch that just adds tracker_read_xmp. It reads licenses and
dublin core metadata from XMP, and could easily be extended to other
schemas (like exif).
its not correct because DC is abstract in tracker and must not be used
to store metadata - you must use a subclass of the DC type instead
EG
for Docs
Doc:Title is a subclass of DC:Title
For Images
Image:Title is a subclass of DC:Title
what you will need to do is use the service type to determine which
subclass to use
For now you can just set it for Images if you like
Here's a patch that just uses Image:* for now and fixes the style issues
(I hope I got them all). Hopefully this can make it into 0.6, along
with the extractor updates.
Regarding using the service type, that's going to take more extensive
updates to Tracker. All extraction of XMP is done from tracker-extract,
which doesn't know the service type. Maybe extend
tracker_metadata_get_embedded(...) to somehow post-process the extracted
fields. Or maybe also pass along the service type to tracker-extract.
I'm not sure the most elegant solution here.
Cheers,
Jason
there are a few more styling issues still in there but I have committed
anyhow (I can tidy them up later)
I will do the extractor patches tomorrow (bedtime now)
I guess we might have to pass the service name to tracker-extract in the
future
thanks
jamie
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]