Re: 0.8 Breaks CR2?



On 10-12-11 08:33 PM, Tim Howard wrote:
My mistake, I thought you could import them it would just take forever.
The XMP sidecar feature as I understand it operates in two modes: ALWAYS
and OFF. The ALWAYS mode is obvious, in every case metadata will be
written to a .XMP file instead of the image. If you are not using the
ALWAYS preference then sidecar files won't be written unless we
determine we cannot safely write to the image. If there's something we
don't understand about the image layout, existing metadata, etc. then
you'll get a sidecar regardless of the preference setting. It's just the
safest thing to do.

I need to figure out why those are not being created in this case.

Tim,

I retried the experiment. f-spot crashes if it has to generate the thumbnail for the cr2 file in the import window, so it was crashing when I was trying to import just the three files.

When importing the whole directory, I managed to get it all, though it took very long.

NONE of the cr2 files (even working ones) produced XMP files.

If I did the experiment with both RAW and JPEGS in the directory, I got XMP files. But I also got a lot of this on the terminal:

[Error 21:14:06.233] Error syncing metadata to file
System.NullReferenceException: Object reference not set to an instance of an object at FSpot.Utils.SidecarXmpExtensions.ParseXmpSidecar (TagLib.Image.File file, IFileAbstraction resource) [0x00000] in <filename unknown>:0 at FSpot.Utils.Metadata.Parse (Hyena.SafeUri uri) [0x00000] in <filename unknown>:0 at FSpot.Jobs.SyncMetadataJob.WriteMetadataToImage (FSpot.Photo photo) [0x00000] in <filename unknown>:0 at FSpot.Jobs.SyncMetadataJob.Execute () [0x00000] in <filename unknown>:0

If I repeat the experiment with ONLY JPEGs in the directory, the XMP files get produced correctly.

Not sure what to make of any of this. Seeing as XMPs don't get generated even from working CR2 files, I think you can go ahead and check in your patch. I personally don't use XMP anyway, and would consider that a different issue altogether. It doesn't seem to be your patch.

--Pat


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]