Question

I have the following xml:

<?xml version="1.0" encoding="UTF-8"?>
<abc1 formName="Form">
    <Level1>
        <Element1>ZZZ</Element1>
        <Element2>
            <SubElement1>Apples</SubElement1>
            <SubElement2>Oranges</SubElement2>
            <SubElement3>Pears</SubElement3>
            <SubElement4>Blueberries</SubElement4>
            <SubElement5>Milkshakes</SubElement5>
        </Element2>
    </Level1>
    <Level1>
        <Element1>XXX</Element1>
        <Element2>
            <SubElement1>Apples</SubElement1>
            <SubElement2>Oranges</SubElement2>
            <SubElement3>Kiwifruit</SubElement3>
            <SubElement4>Blueberries</SubElement4>
            <SubElement5>Soda</SubElement5>
        </Element2>
    </Level1>
</abc1>

and what I need to be able to do is take a look at the values in some nodes, determine if they are different, and if so write some different html code with a page break between each section.

So, I need to compare the <Element1> values and see if they are different. The <Element1> values could be any text, and the number of <Element1>tags could be unlimited in the xml.

So in this case, we have two different <Element1> values: ‘ZZZ’ and ‘XXX’.

So something like the following if there are two ”:

<xsl:choose>
<xsl:when test="differentvalues = 'true'">
<!--  SomeHTMLCode!-->
<p style="page-break-after: always"/>
<!--  SomeHTMLCode!-->
</xsl:when>
<xsl:otherwise><!--  SomeHTMLCode!--></xsl:otherwise>
</xsl:choose>

but using the for-loop to search through the xml, and something like

<xsl:choose>
<xsl:when test="differentvalues = 'true'">
<!--  SomeHTMLCode!-->
<p style="page-break-after: always"/>
<!--  SomeHTMLCode!-->
<p style="page-break-after: always"/>
<!--  SomeHTMLCode!-->
</xsl:when>
<xsl:otherwise><!--  SomeHTMLCode!--></xsl:otherwise>
</xsl:choose>

if there are three different </Element1> values.

I’m not even sure if this can be done.
Thanks in advance for any help you can provide me with.

Answer 1

This XPath expression gives you the wanted count of distinct values for Element1 (supposing an Element1 cannot have a descendent also Element1 (there is more efficient way to handle such problems using keys, but I wouldn’t talk about keys to a newbie):

count(//Element1[not(. = preceding::Element1)])

You can define a variable specifying the above value in its select attribute and then use this variable in the test attribute` of your conditional instructions.

Even better, don’t use conditional instructions at all and specify the above XPath expression as part of a template match pattern:

<xsl:stylesheet version="1.0"
 xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
 <xsl:output omit-xml-declaration="yes" indent="yes"/>
 <xsl:strip-space elements="*"/>

 <xsl:template match="/*[count(//Element1[not(. = preceding::Element1)]) = 2]">
     Two processing
 </xsl:template>

 <xsl:template match="/*[count(//Element1[not(. = preceding::Element1)]) = 3]">
     Three processing
 </xsl:template>
</xsl:stylesheet>

When this transformation is applied on the provided XML document:

<abc1 formName="Form">
    <Level1>
        <Element1>ZZZ</Element1>
        <Element2>
            <SubElement1>Apples</SubElement1>
            <SubElement2>Oranges</SubElement2>
            <SubElement3>Pears</SubElement3>
            <SubElement4>Blueberries</SubElement4>
            <SubElement5>Milkshakes</SubElement5>
        </Element2>
    </Level1>
    <Level1>
        <Element1>XXX</Element1>
        <Element2>
            <SubElement1>Apples</SubElement1>
            <SubElement2>Oranges</SubElement2>
            <SubElement3>Kiwifruit</SubElement3>
            <SubElement4>Blueberries</SubElement4>
            <SubElement5>Soda</SubElement5>
        </Element2>
    </Level1>
</abc1>

the wanted, correct result is produced:

 Two processing

When the same transformation is applied on the following XML document:

<abc1 formName="Form">
    <Level1>
        <Element1>ZZZ</Element1>
        <Element2>
            <SubElement1>Apples</SubElement1>
            <SubElement2>Oranges</SubElement2>
            <SubElement3>Pears</SubElement3>
            <SubElement4>Blueberries</SubElement4>
            <SubElement5>Milkshakes</SubElement5>
        </Element2>
    </Level1>
    <Level1>
        <Element1>YYY</Element1>
        <Element2>
            <SubElement1>Apples</SubElement1>
            <SubElement2>Oranges</SubElement2>
            <SubElement3>Pears</SubElement3>
            <SubElement4>Blueberries</SubElement4>
            <SubElement5>Milkshakes</SubElement5>
        </Element2>
    </Level1>
    <Level1>
        <Element1>XXX</Element1>
        <Element2>
            <SubElement1>Apples</SubElement1>
            <SubElement2>Oranges</SubElement2>
            <SubElement3>Kiwifruit</SubElement3>
            <SubElement4>Blueberries</SubElement4>
            <SubElement5>Soda</SubElement5>
        </Element2>
    </Level1>
</abc1>

again the correct result is produced:

 Three processing

Note:

You need, of course, to substitute the bodies of the templates with your desired processing.