Re: TM/JSON first draft
Benjamin Bock <[email protected]>
| Newsgroups | gmane.text.xml.xtm.general |
|---|---|
| Message-ID | <[email protected]> |
Sorry, I didn't want to spread FUD. Actually, I haven't seen any results of CXTM-tests for CouchTM, so I have no clue if it relates to the Associations. I was just thinking of Lars Heuer's email on the German TMlist to which I was pointing Hans-Henning to before writing the mail here: > Der aktuelle TMAPIX Snapshot beinhaltet nun einen CXTM-Writer, der für > viele TMAPI-Implementierungen funktionieren sollte (es ist tinyTiMs > CXTM-Writer mit allen tinyTiM-spezfischen Referenzen durch > TMAPI-Referenzen ersetzt) > > Wahrscheinlich funktioniert er noch nicht für Implementierungen, die > type-instance Beziehungen als Association modellieren, Wer seine > Implementierung testen möchte, kann optional ein internes Interface > des CXTMWriters implementieren, das vor dem Schreiben etwaige > Duplikate entfernt (siehe [1]). > > Der aktuelle Snapshot von TMAPIX liegt im Maven-Repository: > > <http://repository.semagia.com/snapshots/org/tmapix/tmapix-io/0.4.0-SNAPSHOT/> > > Rückmeldungen sind willkommen auch oder inbesondere über Probleme mit > Engines, die type-instance Beziehungen als Association verwalten. > > [1] <https://code.google.com/p/tmapix/source/browse/tmapix-io/trunk/src/main/java/org/tmapix/io/CXTMTopicMapWriter.java?spec=svn244&r=244#723> Short translation: There's a CXTM writer in TMAPIX now which probably won't work for implementations using Associations for type-instance. Additionally, Hans-Henning wrote to me that the RubyTopicMaps specs using the CouchTM backend pass for almost everything except some of the type index related stuff. I asked him to look into that. -- Benjamin On Tue, Aug 24, 2010 at 7:14 PM, Lars Marius Garshol <[email protected]> wrote: > > * Benjamin Bock >> >> It currently passes the TMAPI 2.0 test suite but AFAIK not yet the >> CXTM tests (because it uses type-instance Associations, maybe other >> reasons as well). > > That's not the reason. Type-instance is alwasy represented using associations in CXTM. There's no other way to do it, in fact. > > (Of course, I have no idea what the reason really is, but it can't be this.) > > --Lars M. > http://www.garshol.priv.no/tmphoto/ > http://www.garshol.priv.no/blog/ > _______________________________________________ > topicmapmail mailing list > [email protected] > http://www.infoloom.com/mailman/listinfo/topicmapmail > _______________________________________________ topicmapmail mailing list [email protected] http://www.infoloom.com/mailman/listinfo/topicmapmail