Subject: Re: [xsl] performance issues saxon From: "Vasu Chakkera" <vasucv@xxxxxxxxxxx> Date: Tue, 18 Feb 2003 15:16:15 -0000 |
well yeah I have actually used the keys for the second step..( was a mistake when I reproduced the actual problem) Any Idea about the performance of Saxon Under Tru64 Unix Box? Thanks Vasu ----- Original Message ----- From: "Michael Kay" <michael.h.kay@xxxxxxxxxxxx> To: <xsl-list@xxxxxxxxxxxxxxxxxxxxxx> Sent: Tuesday, February 18, 2003 1:49 PM Subject: RE: [xsl] performance issues saxon > > Ok if the requirement is to group all the products of the > > same id together, I would follow the following steps. 1. Get > > all the unique nodes 2. Group all the products with the id as > > defined by the id of the unique node. > > > > > > Is there any other way we could avoid using two nested loops > > for a requirement like this? Thanks Vasu > > Absolutely. Muenchian grouping solves this in O(n log n) time. See > www.jenitennison.com/xslt/grouping for the algorithm. > > Actually it's essentially the algorithm you described above. Step 1 > requires a serial scan of the document, Step 2 requires an index lookup > using a key. Your mistake was to do step 2 with a (second, nested) > serial scan. > > Michael Kay > Software AG > home: Michael.H.Kay@xxxxxxxxxxxx > work: Michael.Kay@xxxxxxxxxxxxxx > > > > XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list > XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
RE: [xsl] performance issues saxon, Michael Kay | Thread | [xsl] document() passed as argument, William S. |
RE: xslt core and intuition was RE:, Tony Graham | Date | Re: [xsl] Weird problem. Xalan Stop, Ted Stresen-Reuter |
Month |