[GroupDAV] Working on charset problem

Ole Dalgaard groupdav@opengroupware.org
Fri, 11 Aug 2006 21:24:36 +0200


Hi

I am currently working on a new patch for the charset problem. It seems 
that calendar events from my phone can look like this (before the 
vcal1.0 -> vcal2.0 conversion):

BEGIN:VCALENDAR
VERSION:1.0
BEGIN:VEVENT
DTSTART:20051216T090000
DTEND:20051216T090000
CATEGORIES:REMINDER
SUMMARY;CHARSET=ISO-8859-1;ENCODING=QUOTED-PRINTABLE:Juleafslutning i b=F8r=
nehaven
END:VEVENT
END:VCALENDAR

Just ignoring the line containing  CHARSET leaves out  a very important 
part of the calendar event . :-) The problem is that the charset 
parameter is not part of vcal2.0, it assumes utf8, therefore the lines 
containing charset needs to be converted.
I am preparing a patch that reads the charset and encoding parameters 
and removes them, then assembling all lines for that line (the "nehaven" 
above is part of the SUMMARY) and decoding it, using commons-codec, 
which I found already being part of the funambol package. I need to test 
it a little bit more before sending it.

During the tests I just deleted the status file that the groupdav code 
makes. This made the number of events in zidestore increase for every 
attempted sync, when I reached about 100 events, my phone timed out 
during the sync, the groupdav connectors communication with zidestore 
seemed to take too long... :-(

Is there any calendar bulk delete function in opengroupware?

/Ole