Hi all,

Is there any better/quickest way of converting string of tag to its its value and later while re-transforming back it to orginal converting those attribute values back to string. Let me be more clear with a dummy example. Below is what my xml looks like:

<details>
<code>XXX</code>
<issn type="print">0000-0000</issn>
<issn type="electronic">0000-0000</issn>
<det>Xvvvv TTTTT v 145- 149</det>
</details.

All I am trying to do is to convert all tag string to dummy attribute value like below.

<details>
<code dummy="XXX"/>
<issn type="print" dummy="0000-0000"/>
<issn type="electronic" dummy="0000-0000"/>
<det dummy="Xvvvv TTTTT v 145- 149">
</details.


Well converting string to attribute values and vice versa in XSLT is not an issue for me but I am trying to have smart way of doing this as there are hundreds of elements (at different levels) in my actual whose string needs to suppressed in view and needs to be reverted back again re-transforming with another XSLT.

My worry is if start suppressing all these elements at different level I will end up with a giant stylesheet. That's the reason I am looking for smart way of achieving this.

At this stage I think I need to do this trick with <xsl:call-template>, but unfortunately I have never used it earlier. Can somebody guide whether I am on right direction or not or help me with snippet of code.

Sorry for asking without trying it and thanks in advance for any help.

Pankaj















Pankaj

__________________
Pankaj

Assuming you don't have any mixed content in your document, you can do the conversion using:

<xsl:template match="*[text()]">
  <xsl:copy>
  <xsl:copy-of select="@*"/>
  <xsl:attribute name="dummy">
    <xsl:value-of select="."/>
  </xsl:attribute>
  </xsl:copy>
</xsl:template>

<xsl:template match="*[not(text())]">
  <xsl:copy>
  <xsl:copy-of select="@*"/>
  <xsl:apply-templates/>
  </xsl:copy>
</xsl:template>

Michael Kay
http://www.saxonica.com/
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference

Well I do have the mixed content in my xml and also I want to perform this on certain elements and not on all.

Your code seems to be applying on each and every element of xml document.

Correct me if I am wrong.

Thanks.

Pankaj

Pankaj

If you only want to apply it to some elements then change "*[text()]" to match the element you want to perform it on.

And by 'mixed content' we mean something like this: <detail>Some text <i>goes here</i>.</detail>

<detail> is considered to have mixed content, i.e. some of its child elements are text() nodes, and some are elements (in this case it has two text nodes and the <i> element as its children).

/- Sam Judson : Wrox Technical Editor -/

My code was applying one rule for elements with text nodes, another for elements without (I should have added <xsl:strip-space elements="*"/> to avoid complications with whitespace text nodes). That should handle the example you've shown us.

If your rules are more complex than that then you will need to use more specific match patterns.

Michael Kay
http://www.saxonica.com/
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference

Thanks Sam and Michael. I think I need to match the element along with exact path.

Sam in case of mixed content case I think I need to use string() instead of text(). Will take work or there is any other content-markup sort of function. Content can be with entities too.

Thanks


Pankaj