XSL select all node where not node of another node

Question

Thse is my xml

<element1> <subel1/> </element1> <element2> <subel2/> </element2> <element3> <subel3/> </element3> <criteria> <subel3/> </criteria>

how i can select all node with xsl that not are in criteria subnodes? like these

 <subel1/> <subel2/>

How is this done?

If the xml is formated as:

<element1> 
<el> subel1 </el>
</element1>
 <element2> 
<el> subel2 
</el> 
</element2> 
<element3> 
<el> subel3 </el> 
</element3> 
<criteria> 
<subel3/> 
</criteria>

Answer 1

arent you missing the xmlcode? would be easier to help if you put in the code and in the editor mark it as sourcecode (the button with "101 010")

Answer 2

The easiest way to do this is to construct an xpath expression which cross checks against the criteria nodes. For instance:

Step 1: The following will select all nodes that are not in the criteria area (So all nodes that do not have a parent named "criteria")

//*[name(..) != 'criteria'] 

Step 2: Filter out any nodes that also appear in critera section

//*[name(..) != 'criteria' and not(name(//criteria/*) = name(.))]

Now that last statement will select all the nodes that don't match a criteria node. But you just want sub nodes, so we can modify the selector to only grab elements that are "sub elements" or leaf elements that have no child nodes.

//*[name(..) != 'criteria' and not(name(//criteria/*) = name(.)) and not(*)]

So here is the breakdown of each of our conditionals one more time:

name(..) != 'criteria' -- Limits to nodes that are not in the criteria section

not(name(//criteria/*) = name(.)) -- Limits to nodes that do not have the same name as a node in the criteria section

and not(*) - Limits to nodes that do not have any child nodes (so the leaf nodes that you want.)

So if you were do something like:

<xsl:for-each select="//*[name(..) != 'criteria' and not(name(//criteria/*) = name(.)) and not(*)]">
 <xsl:value-of select="name(current())"/> :
</xsl:for-each>

For your above example this would print:

subel1 : subel2

Hope this helps.

Cheers,

Casey

Answer 3

This transformation:

<xsl:stylesheet version="1.0"
 xmlns:xsl="http://www.w3.org/1999/XSL/Transform"&gt;
 <xsl:output omit-xml-declaration="yes" indent="yes"/>

 <xsl:template match="/">
  <xsl:for-each select=
  "/*/*[not(self::criteria)]/*">
   <xsl:copy-of select=
   "self::node()[not(/*/criteria/*[name()=name(current())])]"/>
  </xsl:for-each>
 </xsl:template>
</xsl:stylesheet>

when applied on this XML document (based on the provided XML fragment, but wrapped the XML fragmentwithin a top element and added one more child to criteria to make the problem less trivial):

<t>
    <element1>
        <subel1/>
    </element1>
    <element2>
        <subel2/>
    </element2>
    <element3>
        <subel3/>
    </element3>
    <criteria>
        <subel3/>
        <subel1/>
    </criteria>
</t>

produces the wanted result (no other published answer at the time of writing this produces a correct result):

<subel2/>

Step-by-step explanation:

1. The XPath expression:

   /*/*[not(self::criteria)]/*

selects every element whose parent is an element not named "criteria" and which is a child of the top element of the document.

1. Within the <xsl:for-each> instruction we check if the current node has a name that is not one of the names of any children of criteria and only copy such a node:

self::node()[not(/*/criteria/*[name()=name(current())])]

This XPath expression selects the current node ( self::node() ) only if there doesn't exist a child of /*/criteria, whose name is the same as the name of the current node ( not(/*/criteria/*[name()=name(current())]) ).

Here we use the fact that not(someNode-Set) is false() only if the nodeset someNode-Set is empty.