Re: [xsl] perfomance question about selecting from child-node

Subject: Re: [xsl] perfomance question about selecting from child-node
From: "Braumüller, Hans" <H.Braumueller@xxxxxxxxxxxx>
Date: Fri, 23 May 2003 10:44:19 +0200
Hello,

regarding Michael comment about my silly question: is it not obvious in my case(i shortcut my question), because  i need 
from the xml-information-source for each node row some data, but i output a html-rows from the template rule for DARKONTRAKT (a child from each row).

To summarize a walking on a parent node seems no big difference with other methods, like passing parameters.
It seems that like post David that major selections like
<snip comment="David Carliste">
  If you were searching up arbitrary amounts,
  ancestor::*[this and that and the other] then not doing that search and
  passing down a parameter will probably start to win
</snip>
one should go better with passing parameters.

Thanks for your time,

Hans



It might be a silly question, but I would have thought the obvious way
of coding this was:

>   <xsl:template match="row">
>     <xsl:value-of select="tannr"/>
>     <xsl:value-of select="tscreate"/>
>     <xsl:value-of select="tsupdate"/>
>     <xsl:apply-templates select="DARKONTRAKT"/>
>   </xsl:template>

In any case, the only way to get answers to this kind of question is to
try it on your chosen XSLT processor and see. I would expect the
difference to be very small on most processors.

Michael Kay

> -----Original Message-----
> From: owner-xsl-list@xxxxxxxxxxxxxxxxxxxxxx 
> [mailto:owner-xsl-list@xxxxxxxxxxxxxxxxxxxxxx] On Behalf Of 
> Braumüller, Hans
> Sent: 22 May 2003 15:36
> To: xsl-list@xxxxxxxxxxxxxxxxxxxxxx
> Subject: [xsl] perfomance question about selecting from child-node
> 
> 
> Hello friends,
> 
> a short perfomance question about selecting from child-node
> 
> XML (... = etc)
> /********************/
> <DARLEHEN> 
>  ...
>     <row num="16">
>       <tannr>20010013</tannr>
>       <tscreate>2001-06-01-12.17.02.162000</tscreate>
>       <tsupdate>2001-06-01-14.46.15.125000</tsupdate>
>       <geschart ztp="true"/>
>       <DARKONTRAKT num="16">
>         <credit_line>1000000,00</credit_line>
>         <isonominal>EUR</isonominal>
>         <kontraktnr>1</kontraktnr>
>       	
>       
>       </DARKONTRAKT>
> 	<DARKONTRAKT num="17">
>       ...
>       </DARKONTRAKT>
>     </row>
>     <row num="17">
> 	...
>     </row>
>   ...
> </DARLEHEN>
> XSLT (... = etc)
> /********************/
> 
> What is better for getting a faster transformation in 
> general? 1. When i am transcurring node-set "DARLEHEN" doing 
> an applying template for "row" with rule
>   <xsl:template match="row">
>     <xsl:apply-templates select="DARKONTRAKT">
>       <xsl:with-param name="tannr" select="tannr"/>
>       <xsl:with-param name="tscreate" select="tscreate"/>
>       <xsl:with-param name="tsupdate" select="tsupdate"/>
>       <!-- §_010 -->
>       <xsl:with-param name="geschartZTP" select="geschart/@ztp"/>
>     </xsl:apply-templates>
>   </xsl:template>
> 
> then putting  the values of the parameters on the node-set 
> DARKONTRAKT out
> 
> 2.
> Or make a direct applying template DARKONTRAKT and from there 
> getting the tree one step up, like
>  	<xsl:value-of select="../tannr"/></xsl:attribute>
>       <xsl:value-of select="../tscreate"/>
>    	<xsl:value-of select="../tsupdate"/>
>      	<xsl:value-of select="../geschart/@ztp"/>
> 
> Imagine hundreds of rows in the xml-source!
> 
> Thanks,
> 
> Hans Braumüller 
> Systementwickler Web-Design 
> Hanse Orga AG
> Oldesloer Straße 63
> D-22457 Hamburg 
> Telefon: (+49) 040 51 48 08-62 
> Telefax: (+49) 040 51 48 08-88 
> E-Mail: h.braumueller@xxxxxxxxxxxx 
> 
> 
> 
>  XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list
> 


 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