В сообщении от Среда 03 февраля 2010 01:08:34 автор Jamboarder написал: > > From: Evgeny Egorochkin > > > > > > So in theory, libstreamanalyzer can be the best, most detailed, precise > > and semantically-aware data extractor but in practice someone needs to > > write the code :) > > ... > > I have a Xine-based fallback analyzer here right now and it even works. I > > need to test it and polish a bit. I hope this will motivate you guys to > > finally get rid of duplicate file indexing services :) > > Count me in as motivated. :-)
Good :) > As long as it's not too difficult to write the plugins. I'm definitely not > a file format junkie since I just use whatever existing/available libs can > provide. But if it takes a little extra work to extend the file indexer > capabilities at install time, then I'm down to help out by working on > plugins for Bangarang 2.0 cycle. Till then I'll address the potential > duplication of data in the nepomuk store as a bugfix by giving Strigi an > opportunity to index any opened file before Bangarang does. I don't think you'll have to write any plugins as much as help me test them, figure out what needs improvement etc > peace and much respect, > Andrew Lake > > P.S. Just a minor note that the duplication of functionality in this case > is incidental since Bangarang, like any media player, will still need to > be able to read metadata directly from files But if you can instruct the file indexing service to do this, why do you even need this? > and will still need to be > able to write directly to the nepomuk datastore for the in-context and > non-extractable use cases. Not surprising :) > I'll just need to add a branch the > nepomukwriter code path to call the file indexer, hopefully extended via > plugins, for technically extractable data. -- Evgeny _______________________________________________ Nepomuk mailing list [email protected] https://mail.kde.org/mailman/listinfo/nepomuk
