AFAIK aggregate UDFs only work with range indexes and other lexicon-type indexes. If they *did* work directly with fragments, I would expect them to perform about like other fragment-filtering operations.
-- Mike On Feb 11, 2013, at 10:22, bek <[email protected]> wrote: >> From what I can read, UDFs look pretty interesting. I'm intrigued by the >> idea of being able to use the map and reduce functions to summarize values >> from elements in my documents without having to add indices to everything. >> However, it seems that to do this, it would be the equivalent of running a >> filtered search so the elements can be evaluated, thus raising a performance >> penalty with large document counts. Did I miss something in the >> documentation? > > thanks > bek > _______________________________________________ > General mailing list > [email protected] > http://developer.marklogic.com/mailman/listinfo/general > _______________________________________________ General mailing list [email protected] http://developer.marklogic.com/mailman/listinfo/general
