Re: [Exist-open] XQuery Joins are slow: Optimizer does not work?

[Joe W. <jo...@gm...>]

Are the doubly nested predicates really necessary?  Would this work for you?

collection("/db/setA/")/content[.//link/@href =
collection("/db/setB/")/content/@id]/@id/string()

Do you have indexes set on @href and @id qnames?  If not, you should
create an index definition, and reindex the collection.  See "Prefer
simple index definitions" (http://exist-db.org/tuning.html#idxdefs).

With indexes created, you might Another way of reworking this query
according to the principle of "Use the ancestor or parent axis instead
of a top-down approach" (http://exist-db.org/tuning.html#N1034C):

collection("/db/setA/")//link[@href =
collection("/db/setB/")/content/@id]/ancestor::content/@id/string()

Joe


On Wed, Jun 9, 2010 at 8:05 AM, Markus <m2...@li...> wrote:
> Hi,
>
> perhaps my first post was a little bit too complex ...
>
> 1. I have two large set of xml documents which are linked with each other.
> 2. I want to get all documents of set A which have valid links to
> documents in set B.
>
> Example XQuery:
> collection("/db/setA/")/content[.//link[@href=collection("/db/setB/")/content/@id]]/@id/string()
>
> Question:
> Is it possible to write an xquery which will return fast results (e.g.
> by doing an intersection of two sets)?
> Perhaps the exist:join option is quite right for me. But has no effect.
> see http://article.gmane.org/gmane.text.xml.exist/17129/match=join
>
> Kind regards
>
> Markus
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit.  See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
> _______________________________________________
> Exist-open mailing list
> Exi...@li...
> https://lists.sourceforge.net/lists/listinfo/exist-open
>