Subject: Re: [xsl] problem with handling conditional for loops|
From: Jon Gorman <jonathan.gorman@xxxxxxxxx>
Date: Wed, 27 Apr 2005 10:35:47 -0500
> Theres only one problem though in this solution. For instance the > $clsName returns values 'Hello', 'How', 'Are', 'You' each time the > for-loop executes, extracting a node value from First.xml one at a time. > My XPaths differ in how I look for the existence of 'Hello' etc. in > SubConcept and the rest of the SubConcept's. > > (1) First I use the XPath - > document(Second.xml)/SubConcepts/SubConcept/Value and match if text() > = $clsName > > (2) If 'Hello' etc. isnt available in (1) then I look for it using the > XPath - document(Second.xml)/SubConcepts/SubConcept/@name and match if > @name=$clsName > Ah, typing too quickly and I missed that we were using clsName for comparison, not as a target variable to fill. I don't have much time, so I"ll only give it one more shot before concluding that I'll have to wait till I have time to look at it closer. Like I said, the outer parent loop seems to be a red herring. It has little to do with the problem at hand. Might be misreading the matter. But since you need to apply different xpaths we can do something similar...might be some scoping problems here (might have to move it so the <xsl:variable wraps around) but something like this should work (although I'm suspicious of the text(), it would seem to me you need . or something similar). <xsl:choose> <xsl:when test="document(Second.xml)/SubConcepts/SubConcept/Value[$clsName=text()]"> <xsl:variable name="nodeToProcess" select="document(Second.xml)/SubConcepts/SubConcept/Value[$clsName=text()] "/> </xsl:when> <xsl:otherwise> <xsl:variable name="Values" select="document(Second.xml)/SubConcepts/SubConcept/Value/@name[.=$clsName]"/ > <xsl:variable name="nodeToProcess" select="$Values"> </xsl:otherwise> </xsl:when> I wouldn't be surprised if there are some gotchas in this. But it should be a direction to look at. >Moreover Second.xml is a > very large file and I would prefer to save execution time by avoiding > duplicate match finds. Sorry, not sure how you could save your time here. It would depend heavily on the parser/processor. Flipping through each node in a for-each loop might even take more time than selecting them as a group. There are others that might be able to respond.