jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Brian Moseley <...@osafoundation.org>
Subject Re: handling indexing failure
Date Wed, 11 Jan 2006 15:52:10 GMT
On 1/11/06, Tobias Bocanegra <tobias.bocanegra@day.com> wrote:
> ...wait for jsr283 and the synchronous/vetoable observation [1] :-)

heh, unfortunately i would like to ship before then :)

> well, if the peformance does not suffer to much (i.e. if the indexing
> is fast), you could index the data twice. once, just for checking
> somewhere in the dav server part, and once in the filter.

yeah, it's not that fast, cos sometimes it has to do hard things like
expanding recurring icalendar events. in fact, we've clocked the
indexing of an infinitely recurring event at almost 5 seconds (though
that is due to a probable bug/inefficiency in in ical4j). so
duplicating indexing is definitely something i don't want to do.

> we also planed to introduce some sort of custom constraint filter,
> that can be somehow registered to the repository, in order to check
> the property values.

that would work, as long as it executes after the indexing filters.

View raw message