I have spent some time thinking how to integrate both Tracker and Zeitgeist in Solang [1] and I am repeatedly hitting the wall when trying to figure out a way to abstract out accesses to two different databases. I am going to use Solang to describe my problem but I suspect this would be true to some extent for some other applications as well.
In Solang whenever we try to search (based on free text or URIs, tags, Exif meta-data) we put the criteria in a search manager which generates the SPARQL query and retrieves a list of photos from Tracker. You can find a bit more in the FOSDEM slides [2]. Now it would be only too natural to allow the user to search for the most recent photos and such in the same way she would initiate a search based on the other criteria. But this information, as far as I understand, is present in Zeitgeist. So the current approach of just iterating over all the criteria in the search manager and creating a SPARQL query won't really work if all the criteria are divided into Tracker and Zeitgeist. Ofcourse the UI can be divided into "search Tracker" and "search Zeitgeist" but might be a bad thing to do. After all the user does not really care. :-) So any suggestions? I am out of ideas at the moment. Thanks, Debarshi [1] http://projects.gnome.org/solang/ [2] http://rishi.fedorapeople.org/solang-fosdem.pdf -- One reason that life is complex is that it has a real part and an imaginary part. -- Andrew Koenig _______________________________________________ tracker-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/tracker-list
