> As for finding the default app, wouldn't that be xdg-mime query default
xdg-mime postdates the creation of this code. Bug 296443 covers using it.
Again, we're not actually using mimeinfo.cache ourselves. It's your GNOME configuration that's using it.
> Besides, before upgrading from 9 to 10 I didn't have a problem with this behavior
That's quite odd. I don't think this code changed from 9 to 10. Would you be willing to use nightly builds to find when the behavior changed for you?
> 2. use the GNOME application order in the correct way (not via the cache file)
Again, we are not doing anything with the cache file ourselves. We're just calling the official "get the app for a MIME type" GNOME APIs. First g_app_info_get_default_for_type and then if GIO is not around gnome_vfs_mime_get_default_application. You can see the code for yourself right here:
I'd like to focus on figuring out why there was a behavior change from 9 to 10 on Dutch Kind's machine here, though. _Something_ weird is going on there.
> As for finding the default app, wouldn't that be xdg-mime query default
xdg-mime postdates the creation of this code. Bug 296443 covers using it.
Again, we're not actually using mimeinfo.cache ourselves. It's your GNOME configuration that's using it.
> Besides, before upgrading from 9 to 10 I didn't have a problem with this behavior
That's quite odd. I don't think this code changed from 9 to 10. Would you be willing to use nightly builds to find when the behavior changed for you?
> 2. use the GNOME application order in the correct way (not via the cache file)
Again, we are not doing anything with the cache file ourselves. We're just calling the official "get the app for a MIME type" GNOME APIs. First g_app_info_ get_default_ for_type and then if GIO is not around gnome_vfs_ mime_get_ default_ application. You can see the code for yourself right here:
http:// mxr.mozilla. org/mozilla- central/ source/ toolkit/ system/ gnome/nsGnomeVF SService. cpp#204 mxr.mozilla. org/mozilla- central/ source/ toolkit/ system/ gnome/nsGIOServ ice.cpp# 352 mxr.mozilla. org/mozilla- central/ source/ uriloader/ exthandler/ unix/nsGNOMEReg istry.cpp# 172
http://
http://
I'd like to focus on figuring out why there was a behavior change from 9 to 10 on Dutch Kind's machine here, though. _Something_ weird is going on there.