[Kde-pim] Nepomukfeeeder issues summary

[Christian Mollekopf]

Hey,

I've written down some of what I know about the indexer, for one to
share some knowledge, but also for me to make it a bit easier to
identify some of the problems we have currently. I hope that it will
also help us to prioritize our todos and gives a solid base for further
discussion, so we can get the feeders in better shape for 4.10. 

You can find my braindump here:
http://techbase.kde.org/User:Cmollekopf/Drafts/AkonadiNepmukFeeder

I especially hope we can improve the queries so we can move to recurring
checks if all items have been indexed, which would enable us to solve a
couple of problems already (more on the reliability side and not so much
on the performance side).

Vishesh, I know you had already some points written down from Akademy,
but couldn't find them anymore. Also I'd be interested whether you think
the mass querying would be feasible for nepomuk (aka. give me all
indexed items of that collection), in performance and also because of
the rather large result set.

Cheers,
Christian
_______________________________________________
KDE PIM mailing list kde-pim at kde.org
https://mail.kde.org/mailman/listinfo/kde-pim
KDE PIM home page at http://pim.kde.org/