Re: [SyncEvolution] Getting the concepts clear (Was: Configuring a target)
On 10/04/2014 21:33:05, "Patrick Ohly" <patrick.ohly(a)intel.com> wrote:
> A
> peer can have multiple sync configs, and a peer loosely translates to
>a
> device or service or other syncml instance,
Correct.
> but is merely a notational
> convention to mentally cluster several sync configs. Correct?
I wouldn't describe it like that. A peer typically really exists in the
physical world, so it is more than just an abstract convention.
I understand that,
so it's a very useful convention, but it's not
represented anywhere in the datamodel as far as I can see.
> OK, I've moved them and simply renamed that entity to
SyncSource for
> now. The 'sync' property is currently located on the DataSource --
>that
> is still correct?
No. What you had is "mode" in "SyncMode" is the "sync"
property value.
OK, good. So, 'sync' is called a 'source' property if I go look in
--source-property ?, but it is actually hosted in what I've dubbed
'SyncSource' (the property previously known as Xmn), and as such can
have multiple values per source -- one value per existing sync config,
to be exact.
The "sync" property lives in the intersection of data source
and sync
source, so it *is* part of Xmn. I would even say that it is the most
important part of it, although there are others ("uri", "syncFormat"
and
"forceSyncFormat").
OK, I understand that. But then how should I
understand the earlier
"Nothing stops you from setting a "sync" property in one of the Xmn
boxes. But these values are not used and therefore don't make sense
there." when we were discussing how target-config fits into all of this?
Regards,
Emile