On Wed, 2005-03-09 at 00:09 +0100, Armin Bauer wrote: > > Anyhow, I think whats really happening is that Evolution wants to use > > the CATEGORIES as a whole single string and breaks it up in the UI > > itself. The \, means we are treating the categories as a single block > > rather a delimited list, that is valid icalendar, if a bit unusual. > > > > So, judging from your answer, that you are not going to change this?
Not in the short term, is there a specific issue this is causing you? > > > >> Type example: > >> > >> CATEGORIES:TRAVEL AGENT > >> > >> CATEGORIES:INTERNET,IETF,INDUSTRY,INFORMATION TECHNOLOGY > >> > >>so this would be valid: sometext\,,somemoretext According to the spec, yes it would. > >>>>The second one is that when i add several tasks/events, request them via > >>>>e_cal_get_changes, delete all of them and request again using > >>>>e_cal_get_changes the returned GList will contain more items than delete > >>>>d and the additional items look like random memory locations. I attached > >>>>the code that produces this result. I took a look at the bug database > >>>>and there are quite some bugs which reporting crashing etc when deleting > >>>>tasks/events, so maybe this is related. > >>> > >>> > >>>Probably unrelated, but it might help against some pilot bugs. What is > >>>the exactly version you are coding against? > >> > >>ii evolution 2.0.4-0.1 The groupware suite > >>ii evolution-data 1.0.4-0.1 evolution database backend server > >> > >>entry [EMAIL PROTECTED] > >>entry [EMAIL PROTECTED] > >>entry [EMAIL PROTECTED] > >>entry [EMAIL PROTECTED] > >>entry [EMAIL PROTECTED] > >>entry [EMAIL PROTECTED] > >> > >>Note that i deleted 3 tasks (the ones with *1000-1*) > > > > > > Have you looked at the .xml file on disk to see if thats being updated > > properly? > > Yes. i attached the ics and .db files after the first sync (after i > added 3 tasks) and after the second one (with the deletion). Nothing > suspicious... Well, next step would be to debug the code in the file backend (e-d-s/calendar/backends/file). It will be several days before i could get to this, but you can ping me with any questions if you want to take a crack. > But there seems to be another bug: take a look at before.ics. there are > errors in there which i guess is not correct. (but these errors are not > the root of the problem, i checked this) Yah, libical is sticking those in there. -JP -- JP Rosevear <[EMAIL PROTECTED]> Novell, Inc. _______________________________________________ evolution-hackers maillist - [email protected] http://lists.ximian.com/mailman/listinfo/evolution-hackers
