Re: [patch] gqview 2.1 and FUJI raw files (.raf)

/ 2005-05-14 18:26:27 -0400
\ John Ellis:
> > quickfix for this particular problem:
> > --- gqview-2.1.0/src/thumb_standard.c   2005-03-03 05:09:25.000000000 +0100
> > +++ gqview-2.1.0.lge/src/thumb_standard.c       2005-04-26 09:59:05.970224344 +0200
> > @@ -709,14 +709,14 @@
> >                 {
> >                 gint found;
> > 
> > -               if (thumb_loader_std_fail_check(tl)) return FALSE;
> > -
> >                 tl->thumb_path = thumb_loader_std_cache_path(tl, FALSE, NULL, FALSE);
> >                 tl->thumb_path_local = FALSE;
> > 
> >                 found = isfile(tl->thumb_path);
> >                 if (found && thumb_loader_std_setup(tl, tl->thumb_path)) return TRUE;
> > 
> > +               if (thumb_loader_std_fail_check(tl)) return FALSE;
> > +
> >                 return thumb_loader_std_next_source(tl, found);
> >                 }
> 
> The fail check is first because it tests if the image had previously failed so
> that GQview does not attempt to generate a thumbnail again. Instead of the
> proposed patch, an option could be added to the maintenance dialog to reset the
> fail markers (remove contents of ~/.thumbnails/fail/gqview-1.0).

current gqview checks its fail markers before it looks for existing
thumbnails (created by other applications like gimp), and I think that
is wrong.

it should first check whether there is a usable thumbnail,
regardless of whom created it.
if there is no thumbnail yet, then it should check if _gqview_ failed to
create one before.  if so, do not attempt to create one, but fail again.
if there is no gqview specific fail marker, try to create the thumbnail.

think about it again, that is exactly the reasoning for application
specific fail markers, but generic thumbnails. and thats what the
patchlet above does.
see, only if found == 0 AND thumb_loader_std_fail_check(tl) == FALSE
thumb_loader_std_next_source is called.

> > and src/raw_format_offsets.c, the later containing some stubs and the
> > implementation for the fuji .raf format.
> 
> I prefer to keep all image formats to stay with using a gdk-pixbuf loader.

right. but.
I did not want to mess with a huge project like gtk.
and it would not help for the exif data, either,
so you'd have to special-case it in gqview anyways, right?
unless you use some of the exif sources which already handle raw
formats, libexif knows about some since december 2004.

> <wish>
> If only all the raw image format hackers could get together and create some sort
> of gdk-pixbuf-extras package for file formats not supported by GTK+ proper. The
> tedious part would be staying on top of bugs and potential security issues as
> they would effect all GTK+ apps (probably why no one has started such a project
> up to this point).
> </wish>

I know what you mean.

hm. I think someone even started such a project,
but I don't recall where I came across it :-(
I sort of ignored it, since it did not support .raf
I'll dig through my history and try to find it again...

cheers,

	Lars Ellenberg


-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7393&alloc_id=16281&op=click