[sync] Stripping UIDs

[Jan Schneider]

Zitat von Karsten Fourmont <fourmont at gmx.de>:

> some phones seem to send the local ID of an entry as UID, see
> http://wiki.horde.org/SyncMLDevicesFileFormats
> The Nokia 9500 sends something like
> BEGIN VTODO
> UID: 170
>
> I doubt this is a globally unique ID :-(

Ah, too bad. And extremely silly.

> We can enhance device.php to include a flag "producesValidUIDs()"  
> and allow UIDs for devices where this is true.
> Or alternatively do some heuristic checks (length, contains "@")  
> before accepting a UID.

I'm actually going to move the complete import code into the  
applications' APIs, since SyncML is only a transportation protocol and  
doesn't know anything about the exchanged data. And maybe it might  
become handy to be able to import and export SIF data without using  
SyncML.
Thus we can't strip the UIDs in Device.php anymore. So we probably  
need to tell the import methods whether to ignore any existing UIDs.

> Quoting Jan Schneider <jan at horde.org>:
>
>> Does anybody know a good reason why we strip the UIDs from objects  
>> during synchronization, just to re-add them when synchronizing in  
>> the different direction? This doesn't make much sense to me.
>>
>> Jan.
>>
>> -- 
>> Do you need professional PHP or Horde consulting?
>> http://horde.org/consulting/
>>
>> -- 
>> sync mailing list - Join the hunt: http://horde.org/bounties/#sync
>> Frequently Asked Questions: http://horde.org/faq/
>> To unsubscribe, mail: sync-unsubscribe at lists.horde.org
>
>
>
> -- 
> sync mailing list - Join the hunt: http://horde.org/bounties/#sync
> Frequently Asked Questions: http://horde.org/faq/
> To unsubscribe, mail: sync-unsubscribe at lists.horde.org



Jan.

-- 
Do you need professional PHP or Horde consulting?
http://horde.org/consulting/