So, to handle ical files, /usr/share/applications/defaults.list should point at
evolution-webcal . Instead, it points at evolution-2.0, which doesn't even
exist, and wouldn't work if it did. :) Probably this bug has to depend on the
existence of an evolution-webcal.desktop, but I'm not sure. I defer to your
excellence and wisdom, Mr. Nocera.
Sigh. I misfiled the version, and made an error in the bug as well; I'm not sure
that pointing it at evolution-webcal works either, since that doesn't work for
local files. Either way, evolution-2.0 is definitely not the solution that
you're looking for :)
Removed the following (broken) mappings in shared-mime-info for rawhide.
* Tue Aug 21 2007 - Bastien Nocera <firstname.lastname@example.org> - 0.22-2
- Don't say that webcal files are handled by evolution-2.0, it can't
- Disable vCard mapping as well, as evolution doesn't handle it
Reassigning to evolution-webcal. It should install a .desktop file advertising
the webcal mime-types, and only handle URLs, so that clicking on a link in a
browser does the right thing.
I can click on this link  in my browser and Evolution asks me if I want to
subscribe to the calendar and then proceeds to add it to my Calendar component.
Maybe I'm misunderstanding the problem but what aspect of this is not working?
I guess Bugzilla doesn't recognize webcal URLs, but I can still go to  and
click the "GNOME" link to subscribe. I'm using Fedora 7 with the Rawhide
Matthew, that works for "webcal://" type URLs, but doesn't for "http://" (when
we get a mime-type, etc.)
If you use webcal://, it will use the default URI handler for webcal from GConf,
and not do anything based on the mime-type.
Or download it locally and double-click.
Okay I see, thanks for clarifying Bastien.
Sounds like this should be moved upstream.
I did find this two-year-old upstream bug, but it kinda looks like dobey
stonewalled it. Thoughts?
Sorry for lag, my account was disabled.
I guess there are two bugs here:
(1) evo-as-packaged-in-fedora/RHEL doesn't even try to load, since it points at
evolution-2.0, which doesn't exist. That should be fixed to point at something
else; probably evolution-webcal.
(2) evolution-webcal should probably work, but doesn't; that is upstream 320865.
Okay so it looks like Bastien already handled part (1) and even pinged the
upstream bug. Any objection to me moving the remainder upstream?
Oh, I missed that.
(1) bad bastien- retitle the bug if you think it is now a different bug ;)
(2) all things considered, yeah, I guess the bug should be closed. Shame it
can't actually be fixed without smacking upstream about the head.
If Bastien or someone can suggest a reasonable workaround, I'm up for
maintaining it in Fedora/RHEL until upstream is, well, smacked. I myself am not
wise in these matters.
In the meantime, closing this as UPSTREAM.