How to prevent adding the empty CDATA elements?

Question

The below is used XSLT and input/output XML. The output XML contains the empty CDATA elements. How to prevent adding it without excluding from cdata-section-elements? XSLT

    <?xml version="1.0" encoding="utf-8"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="2.0">
    <xsl:output cdata-section-elements="first second" indent="yes"/>
    <xsl:strip-space elements="first second"/>
    <xsl:template match="@* | node()">
        <xsl:copy>
            <xsl:apply-templates select="@* | node()"/>
        </xsl:copy>
    </xsl:template>
</xsl:stylesheet>

Input

    <?xml version="1.0" encoding="UTF-8"?>
<top>
    <first>
        <second/>
    </first>
    <first>
        <second><![CDATA[! Please note...]]></second>
    </first>
</top>

Output with strip-space

    <?xml version="1.0" encoding="UTF-8"?>
<top>
    <first>
      <second/>
   </first>
    <first>
      <second><![CDATA[! Please note...]]></second>
   </first>
</top>

Output without strip-space

<?xml version="1.0" encoding="UTF-8"?>
<top>
    <first><![CDATA[
        ]]><second/><![CDATA[
    ]]></first>
    <first><![CDATA[
        ]]><second><![CDATA[! Please note...]]></second><![CDATA[
    ]]></first>
</top>

Which is the output you expect given that you have the first element designated as a CDATA section element and it has a child element? — Martin Honnen
– Martin Honnen, Commented Oct 11, 2016 at 17:19
The above sample reflects real documents structure and content. The first and the second elements may have a CDATA section. The above identity template works as expected if first or second contains CDATA. Also input XML may have nested second in the first without CDATA. I expected the same output as an input while applying identity template. — senleft
– senleft, Commented Oct 12, 2016 at 10:11
I asked about the expected output. If you think that an identity transformation in XSLT is supposed to preserve CDATA sections, then no, it does not, the XML is parsed by an XML parser into a tree model underlying XSLT/XPath which does simply have text nodes, that input tree is transformed into a result tree where text node children of certain elements can be serialized as CDATA sections if asked for with xsl:output cdata-section-elements. But that is a result tree serialization step that then serializes the text nodes as CDATA section, it does not depend on the input markup format. — Martin Honnen
– Martin Honnen, Commented Oct 12, 2016 at 10:19
Sorry for not the clear question and thank you for clarification. I updated the input file and provided output with/without 'strip-space'. The output with strip-space is expected. — senleft
– senleft, Commented Oct 12, 2016 at 12:01

uL1 · Accepted Answer · 2016-10-11 16:51:22Z

1

Keyword to the solution is function strip-space. Go ahead with:

<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">

    <xsl:output cdata-section-elements="first"/>
    <xsl:strip-space elements="first second"/>
    ...

To be precise there are whitespace-text nodes between these two nodes:

<first>
    <second/>

The CDATA can't ignore these whitespaces, otherwise it would change the content. So you have to command the processor, what to do with these text-nodes.

Second possible solution: You address the whitespace-text via template and remove them:

<xsl:template match="first/text()[not(normalize-space())]"/>

edited Oct 11, 2016 at 16:51

answered Oct 11, 2016 at 16:38

uL1

2,1672 gold badges20 silver badges29 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

michael.hor257k · Accepted Answer · 2016-10-12 12:13:15Z

0

I would suggest you try it this way:

<xsl:stylesheet version="2.0" 
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml" version="1.0" encoding="UTF-8" indent="yes" cdata-section-elements="second"/>
<xsl:strip-space elements="*"/>

<!-- identity transform -->
<xsl:template match="@*|node()">
    <xsl:copy>
        <xsl:apply-templates select="@*|node()"/>
    </xsl:copy>
</xsl:template>

</xsl:stylesheet>

Although I don't quite see the point of this exercise, as the output is identical to the input - both semantically and lexically (except possibly the amount of indent).

answered Oct 12, 2016 at 12:13

michael.hor257k

118k6 gold badges36 silver badges55 bronze badges

Collectives™ on Stack Overflow

How to prevent adding the empty CDATA elements?

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related