[sync] Inconsistent "ReccurenceType" for yearly events

[lst_hoe02 at kwsoft.de]

Zitat von Jan Schneider <jan at horde.org>:

> Zitat von lst_hoe02 at kwsoft.de:
>
>> Zitat von lst_hoe02 at kwsoft.de:
>>
>>> Zitat von lst_hoe02 at kwsoft.de:
>>>
>>>> Hello
>>>>
>>>> we have the problem that events that have yearly reccurence  
>>>> change reccurence after being modified on the server and synced  
>>>> to the device. This is with Outlook Funambol PlugIN with all  
>>>> latest versions. What we have found is that the event is created  
>>>> with ReccurenceType=5 but sent to the client with  
>>>> RecurrenceType=6 after modified at the server.
>>>>
>>>> To reproduce :
>>>>
>>>> Create event with yearly schedule in Outlook and sync to the  
>>>> server. In Kronolith the event shows up fine.
>>>> Change something other than schedule (eg. Starttime) with  
>>>> Kronolith and sync it back to the client.
>>>>
>>>> The event shows up in Outlook with yearly xx weekday in <month>
>>>
>>> I have tracked down the problem to  
>>> horde/lib/SyncML/Device/Sync4i.php. It seams the more komplex  
>>> recurrence pattern are not (well) supported until now in  
>>> Horde/SyncML despite the fact that Kronolith (iCal) is able to do  
>>> so. I will try to patch Sync4i.php to get a better match for  
>>> recurrence between SIF-E and iCal.
>>
>> I have created a patch for "Sync4i.php" but there are other cases  
>> where Sync4i.php create wrong SIF-E or iCal values. Strange thing  
>> is that some of them are corrected as soon as they show up in  
>> kronolith. Is there some other place where the files from /tmp/sync  
>> are altered again? I was under the impression that the files from  
>> /tmp/sync get imported to kronolith directly??
>
> They do. But some corrections might be applied when converting the  
> iCalendar objects to native Kronolith events.
>
>> Someone has a flow-chart at hand for SyncML??
>
> http://cvs.horde.org/framework/SyncML/docs/
> But those haven't been updated in a while.
>
> Jan.

Thanks, i will have a look at it...
BTW : I have created patches for #6847 and the mentioned #7790.

Regards

Andreas