Content types

[Bastien Nocera]

On Fri, 2008-07-18 at 12:52 +0100, Bastien Nocera wrote:
> On Thu, 2008-07-17 at 15:27 +0100, Bastien Nocera wrote:
> > On Fri, 2008-07-11 at 10:25 -0400, Matthias Clasen wrote:
> <snip>
> > That looks good to me, apart from ignore-case. I think we should default
> > to ignoring the case, as I can't think of any content types for which we
> > shouldn't (a pure 8.3 CD will show AUTORUN.INF, probably Autorun.inf for
> > Joliet, and autorun.inf with Rockridge/Unix support).
> > 
> > Could you send me your latest patches privately?
> > 
> > I'll take care of creating the test suite for the types we handle right
> > now, and probably update totem-pl-parser's totem-disc to use the
> > functionality.
> 
> Committed in shared-mime-info HEAD, along with a test program.
> 
> Still need to move the matching data from nautilus to the
> shared-mime-info database.

Data moved, and added test trees for those as well.

In the docs for g_content_type_guess_for_tree() you mention:
The types returned all have the form x-content/foo, e.g.
x-content/audio-cdda (for audio CDs) or x-content/image-dcf

But we can't detect x-content/audio-cdda with the current code... Is
this something that will be reimplemented in gvfs as well, so it can
aggregate the data from HAL and other sources?

Cheers