On 25/06/12 22:47, Patrick Ohly wrote:
> On Mon, 2012-06-25 at 21:52 +1200, Jane Atkinson wrote:
>> On 25/06/12 20:29, Patrick Ohly wrote:
>>> On Sat, 2012-06-23 at 15:31 +1200, Jane Atkinson wrote:
>>>> I've noticed some odd behaviour with the memo sync. I have two 
>>>> installations of SyncEvolution, on different machines. One uses
>>>> calDAV and the other uses EDS. They interface with the same
>>>> Radicale memo file. Each installation is used with one phone
>>>> only.
>>>>
>>>> I have two memos which were created on the 6120c phone and
>>>> introduced into the system via calDAV sync. Now they seem to
>>>> update themselves on the E63 every time I sync the E63 via EDS.
>>>> No other memos are doing this.
>>>>
>>>> I'm happy to provide logs if needed.
>>>
>>> That sounds like the "last-modified" information given to
>>> SyncEvolution by EDS indicates a change each time SyncEvolution
>>> runs. Can you check the LAST-MODIFIED property in the
>>> "syncevolution --export - <E63 config name> memo" output? Does it
>>> stay the same when you repeat the command?
>>>
>>> I assume here that "memo" is the source name for the memos in your 
>>> config.
>>>
>>> If it stays the same, then yes, please provide further logs.
>>>
>>> If it changes, then the problem lies somewhere between EDS and
>>> Radicale. Does the memo have a LAST-MODIFIED field in Radicale's
>>> data file (somewhere in ~/.config/radicale/....)?
>>>
>>
>> I notice that the two memos in question don't have a LAST-MODIFIED
>> field at all.
> 
> SyncEvolution assumes that the CalDAV server maintains that field. Looks
> like Radicale doesn't do that, or (more likely) doesn't do it when
> storing a VJOURNAL which doesn't already have a LAST-MODIFIED field.
> 
> I'll test that.
> 

I've just noticed that these two memos have no dates at all. Here's the
entire contents of one of them:

BEGIN:VJOURNAL
SUMMARY:New note 6120c
DESCRIPTION:Start new line.
UID:0935ca07-8fd0-4286-bfd9-454ec410822a
X-RADICALE-NAME:0935ca07-8fd0-4286-bfd9-454ec410822a.ics
END:VJOURNAL

As I mentioned, new memos added via EDS (whether it's Evolution or the
E63) don't have this problem, and they have all the date fields.


I've started playing with Darwin Calendarserver, because a large
calendar in Radicale takes a long time to load. Assuming I can get memos
to work in DCS, might it be the better solution in the longer term?
Radicale seems to have a lot of quirks...

Jane Atkinson
_______________________________________________
SyncEvolution mailing list
[email protected]
http://lists.syncevolution.org/listinfo/syncevolution

Reply via email to