Re: ACCESSING SPECIFIC CDATA FIELD

Thanks Jon and friends,
for your explanation about CDATA nodes.

Thank again.
Best regards

Hello Fellows,
I've one XML and trying to access the specific CDATA on XML but when do
this, I capture all CDATAs nodes. See the XML and XSL code bellow:

First of all, there is no such thing as a CDATA node.  CDATA is just a
nice way to write something so that you don't have to worry about
escaping all the < and other special characters.  There's only text
nodes.  So what you really appear to be asking for is a way to select
specific text nodes.

And the answer is yes, of course.  Just select the text nodes and
apply-templates or get their value.

 // -----------------------------xml -----------------------------
<?xml version="1.0" encoding="UTF-8"?>
<scripts>
   <etps total="1">
       <etp id="200">
           <![CDATA[Trying accessing only this]]>
           <nvgs>
               <nvg id="201"><![CDATA[blablablablabla]]></nvg>
               <nvg id="202"><![CDATA[blablablablabla]]></nvg>
           </nvgs>
       </etp>
   </etps>
</scripts>

 // -----------------------------xsl -----------------------------
//code's resume
<xsl:choose>
<xsl:when test="scripts/etapas/@total>0">
<table width="100%" id="SCRIPT">
   <tr valign="top" style="font-weight:bold;">
       <th width="40%">etp</th>
       <th width="20%">nvg</th>
   </tr>

<xsl:for-each select="scripts/etapas/etapa">
  <tr valign="top">
       <td comment="ETP">
           <xsl:value-of select="."/><br/> <!--When I put the prefix "."
            <!--the XSL return all CDATA nodes, including "nvgs"-->

No, it gets the value of the element, which is defined in the specs.
A quick summary would be all the text nodes of this element and all
the text nodes of its descendents.  If you don't want to do that, get
<xsl:value-of select="text()" /> although my guess in this case you
want to get <xsl:value-of select="text()[1]" />.   (or just
apply-templates on those if you do anything special with the text).

Is it necessary to create a node to CDATA ?

No, since it'll be turned into just escaped text via the parser.  As
far as the processor is concerned, it's just a typical text node.