Version: 1.4.0 (using KDE 4.5.1) OS: Linux Quoted mail from Akunambol's LikeBack system: A developer responded on your LikeBack feedback message. This is the message you sent us: > This is what Akunambol did to a previously empty addressbook: > http://bitdump.msquadrat.de/2010/akunambol-vcfs.png (might be an Akonadi > bug?) A developer, ruphy, set the status to closed (Solved), and wrote this message to you: > It looks like it was indeed an akonadi bug. I just pushed a commit to > workaround this behiavour temporarily (on the 0.2.x branch): the problem > was that akunambol sets the mimetype "x-vcard", instead of > "text/directory". This behiavour should be correct, but kaddressbook > doesn't recognize it (and this is the bug), and falls back on showing a > default "id" instead of the name. > Now, some people in Akonadi are working on fixing it, so you can expect > to see your data again with newer kontact releases. I'll probably be > making a 0.2.1.1 release soon that includes this workaround, including a > little tutorial that teaches you how to "convert" your items to the > "text/directory" mimetypes, by performing a non-destructive fully new > sync. > > Can you please open a bug on bugs.kde.org too, against akonadi? This > would be really helpful to ensure that the bug is really fixed. > > Thanks! Reproducible: Always
Hej, contact objects stored inside Akonadi have to use the text/directory mime type by definition. If some other app doesn't follow the standard I can't see how this should be a bug in Akonadi... Ciao, Tobias
Hi, apart for the fact that this is an undocumented behiavour (besides Addressee::mimeType() which -now- returns text/directory), I still think that this is a bug, and that's why: - The problem is, as diagnosticated by dfaure, that Akonadi doesn't resolve mimetype aliases correctly, by not using KMimeType. - x-vcard is an alias for text/directory, add this to the fact that what Akonadi stores *are* vcards, I think would just mean sanity to support "text/x-vcard". I also remember seeing it advertised as supported by the collection, at some point. - This is a regression over 4.4. So I think that this bug is still to be considered valid, as confirmed in the #akonadi channel (not today, but when I asked for clues about it).
Tobias, or other KDEPIM developers, would you please share your thoughts on whether you consider valid the reasons that I brought to consider this a bug? I don't want to re-open it out of nothing, but please respond! :)
I rephrased the subject and set this to wishlist.
Perhaps related to bug 183026?
This I think has been fixed by now.