2008/10/16 Sebastian Trüg <[EMAIL PROTECTED]> > On Monday 06 October 2008 22:58:06 Mikkel Kamstrup Erlandsen wrote: > > Hi, > > > > A few of us made a more or less complete draft of a Xesam Metadata > > API. I just brushed it up a bit. You can now find it here: > > http://xesam.org/main/XesamMetadataAPI > > > > Please comment. > > Hm, the way I see it this is basically an RDF API. Thus, I am coming back > to > the "Model" part of the Soprano API [1] which I already proposed a while > back. > It allows to add Statements which are triples (or quadruples actually but > you > could forget about that). > So please have a look and tell me if there are real differences because I > don't think so. Except for details such as the Node class which only makes > things more flexible. > We don't have to use exactly that but why not reuse it and only change it > slightly to adjust it to our needs? After all, this is an API which is > already in use. > > [1] > > http://api.kde.org/kdesupport-api/kdesupport-apidocs/soprano/html/soprano_server_dbus.html >
I think I need some more documentation (or a little help) to grok that API... - It is unclear to me how this Model abstraction is going to be used - What does all the ((isss)(isss)(isss)(isss)) arguments mean? And still, as we discussed in the past[2], I don't think exposing RDF in the API that way is the right way to go about this. The Soprano API does contain some nice things that are not in the proposed Xesam Metadata API though. Namely iterators and signals for changes. It does however also have some shortcomings one prominent one being that it is not batch oriented. [2]: http://lists.freedesktop.org/archives/xesam/2008-January/000048.html -- Cheers, Mikkel
_______________________________________________ Xesam mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/xesam
