I just read the Data conversion and testing page on the wiki
(
http://syncevolution.org/development/data-conversion-testing)
A number of scenarios are covered, but the one which
currently interests me is the preservation of unique fields
across clients...the last test scenario covered on the wiki page.
Apparently the Synthesis library is quite capable in this regard and
so Syncevolution should be robust as well.. provided that 1) (as
mentioned in the second scenario on the page) that the XML that defines
mapping from native to internal synthesis server format is complete and
2) that the server accurately advertise their capabilities
I'm still reviewing the source code and getting my head around the
architecture... but I think I'd like to look at the XML and/or
the code segments which advertise capabilities....
The syncevolution scenario of interest to me is: N900 with gui front-end
and HTTP server setup of syncevolution on the "server" ie: desktop side.
Actually, if someone has reviewed the N900 with HTTP server use case
already, then I might be more assured - perhaps I'm worrying too much.
Thank you for your thoughts...
--Jeff