[Syncevolution-issues] [Bug 6378] automatic sync
http://bugzilla.moblin.org/show_bug.cgi?id=6378
--- Comment #34 from pohly <patrick.ohly(a)intel.com> 2010-02-27 01:46:19 PST ---
(In reply to comment #31)
>+ /** the 'syncURL' used by synchronization */
>+ string m_url;
>Wouldn't it be easier to use the vector<string> returned by getSyncURL()?
Then
>you don't have to do your own parsing. Or is this not a copy of the complete
>"syncURL" property, as the comment implies?
we have to parse, because we want to know whether a syncURL is enabled to auto
sync according to 'autoSync' property.
Yes, but where is the parsing done? Is this m_url still the comma-separated
value from "syncURL" or is it already a single URL after parsing?
(In reply to comment #30)
> Another unclear aspect is when the callbacks are allowed to
happen and
> what triggers them. We have to be aware that currently, only backends which
> integrate into the glib main loop have a chance to do change notifications.
Yes, you are right. Is it possible to add a mechanism to give opportunites of
backends to handle their changes?
Yes, at some point we'll have to do this. Probably together with out-of-process
backend implementations (which we will need at some point).
(In reply to comment #33)
> > I'm not sure whether we need this level of detail. It
makes implementing change
> > notification harder in the backends. What would the "bool raw"
parameter mean
> > anyway?
The same as the 'raw' parameter in insertItem, readItem,meaning item in
internal format instead of engine format.
But what does that mean in the context of change notification? When is the
parameter supposed to be true, when false?
(In reply to comment #30)
> I'm not sure whether we need this level of detail. It makes
implementing change
> notification harder in the backends. What would the "bool raw" parameter
mean
> anyway?
currently we don't need this kind of detail(only know it does change). but I
don't know whether there are other requirements for this.
An API can be both too limited and too complex. When it is too complex,
implementing it is unnecessarily hard. In this case I think the API is too
complex. My philosophy is that we shouldn't try to solve problems that don't
exist yet.
--
Configure bugmail: http://bugzilla.moblin.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching someone on the CC list of the bug.