Many thanks again for your answer Geert. I will take in account your answer.
Lanz On Fri, Nov 12, 2010 at 18:41, Geert Josten <geert.jos...@daidalos.nl> wrote: > Hi Lanz, > > You could try to use XPath, but you need some changes to make sure the > expression can be optimized. > > First, you need a searchable path expression. Searchable expression cannot > start with variables, so get the base-uri of $doc and start your path with > doc($uri) instead of $doc. > > Second, I think it is best to add Item as Fragment Root, and replace > /Items/Item with //Item. > > Third thing you should do is replace count() with xdmp:estimate(). The speed > of indexes is all about not loading data into memory you don't need. Count > forces the processor to load all Items into memory, slowing the count down. > Xdmp:estimate just grabs references to fragments from the index and counts > those, much quicker. > > There are some side notes to xdmp:estimate, but this should work I guess. > > http://developer.marklogic.com/pubs/4.2/apidocs/Ext-4.html#xdmp:estimate > > Kind regards, > Geert > >> > > > drs. G.P.H. (Geert) Josten > Consultant > > Daidalos BV > Hoekeindsehof 1-4 > 2665 JZ Bleiswijk > > T +31 (0)10 850 1200 > F +31 (0)10 850 1199 > > mailto:geert.jos...@daidalos.nl > http://www.daidalos.nl/ > > KvK 27164984 > > > De informatie - verzonden in of met dit e-mailbericht - is afkomstig van > Daidalos BV en is uitsluitend bestemd voor de geadresseerde. Indien u dit > bericht onbedoeld hebt ontvangen, verzoeken wij u het te verwijderen. Aan dit > bericht kunnen geen rechten worden ontleend. > >> From: general-boun...@developer.marklogic.com >> [mailto:general-boun...@developer.marklogic.com] On Behalf Of Lanz >> Sent: vrijdag 12 november 2010 16:19 >> To: general >> Subject: [MarkLogic Dev General] Optimize time spent for a >> query using Element range Index >> >> Hi All, >> >> I'm trying to take advantage of 'Element range Index' in a query but >> can't find how... >> >> Here is my context : >> <Items> >> <Item> >> <Productid>72764</Productid> >> <ItemType>Article</ItemType> >> <DOI>budget-10xxxyyy</DOI> >> <Element>C</Element> >> <VersionCode>E</VersionCode> >> <Theme> >> <ProductId>72764</ProductId> >> <ThemeCode>37</ThemeCode> >> </Theme> >> <Theme> >> <ProductId>72764</ProductId> >> <ThemeCode>39</ThemeCode> >> </Theme> >> </Item> >> <Item> >> <Productid>72692</Productid> >> <ItemType>Book</ItemType> >> <DOI>budget-10-yyyyyxxxx</DOI> >> <Element>C</Element> >> <VersionCode>E</VersionCode> >> <Theme> >> <ProductId>72692</ProductId> >> <ThemeCode>37</ThemeCode> >> </Theme> >> <Theme> >> <ProductId>72692</ProductId> >> <ThemeCode>39</ThemeCode> >> </Theme> >> </Item> >> .... >> </Items> >> >> I want to get statistics of how many Items I have for a given >> ThemeCode by ItemType (ie for the ThemeCode '37' : I have 53 items of >> ItemType 'Article' for the ThemeCode '37' and 89 items of ItemType >> 'Book' ). >> I have the right result using pure Xpath but it takes time : >> fn:count($doc/Items/Item[ItemType = $item/@name and Theme/ThemeCode = >> $themeId]) (where $doc is my searched document, $item/@name is the >> name of itemtype in a loop and $themeId the selected themeCode) >> I have created 2 Element range Index, one for element ItemType and one >> for element ThemeCode. How can I get same statistics using 'nested' >> Element range Index and cts:and-query (or other syntax ?) ? >> >> Any suggestions would be appreciated, many thanks >> Lanz >> _______________________________________________ >> General mailing list >> General@developer.marklogic.com >> http://developer.marklogic.com/mailman/listinfo/general >> > _______________________________________________ > General mailing list > General@developer.marklogic.com > http://developer.marklogic.com/mailman/listinfo/general > _______________________________________________ General mailing list General@developer.marklogic.com http://developer.marklogic.com/mailman/listinfo/general
