How to compare multiple nodes with XPath - xpath

Question

Given the XML below, I am trying to write a XPath which returns only this match:
<TOKEN BEGIN="17" END="19" SENTENCE_BEGIN="0" SENTENCE_END="156" PROP="C" DOUBLE="YES" />
This is the input file:
<?xml version="1.0" encoding="utf-8"?>
<DOCUMENT>
<SECTION>
<PARAGRAPH TRACK="4">
<SENTENCE NAME="PRIMARY" COUNT="4">
<TOKEN BEGIN="9" END="11" SENTENCE_BEGIN="0" SENTENCE_END="156" />
<TOKEN BEGIN="32" END="37" SENTENCE_BEGIN="0" SENTENCE_END="156" />
<TOKEN BEGIN="167" END="169" SENTENCE_BEGIN="158" SENTENCE_END="316" />
<TOKEN BEGIN="210" END="215" SENTENCE_BEGIN="158" SENTENCE_END="316" />
</SENTENCE>
<SENTENCE NAME="SECONDARY" COUNT="2">
<TOKEN BEGIN="139" END="141" SENTENCE_BEGIN="0" SENTENCE_END="156" PROP="A" DOUBLE="YES" />
<TOKEN BEGIN="143" END="145" SENTENCE_BEGIN="0" SENTENCE_END="156" PROP="B" />
</SENTENCE>
<SENTENCE NAME="SECONDARY" COUNT="1">
<TOKEN BEGIN="17" END="19" SENTENCE_BEGIN="0" SENTENCE_END="156" PROP="C" DOUBLE="YES" />
</SENTENCE>
</PARAGRAPH>
</SECTION>
</DOCUMENT>
This is my xpath string:
//TOKEN [#DOUBLE] [#BEGIN <= ../../SENTENCE[#NAME='PRIMARY']/TOKEN/#END] [ (#SENTENCE_BEGIN = ../../SENTENCE[#NAME='PRIMARY']/TOKEN/#SENTENCE_BEGIN) and (#SENTENCE_END = ../../SENTENCE[#NAME='PRIMARY']/TOKEN/#SENTENCE_END) ]
I receive as output two nodes:
<TOKEN BEGIN="17" END="19" SENTENCE_BEGIN="0" SENTENCE_END="156" PROP="C" DOUBLE="YES" />
and
<TOKEN BEGIN="139" END="141" SENTENCE_BEGIN="0" SENTENCE_END="156" PROP="A" DOUBLE="YES" />
The result above is not correct because I want to check that:
the SENTENCE_BEGIN SENTENCE_END values of TOKEN[#DOULBE=YES] and
SENTENCE[#NAME="PRIMARY"]/TOKEN are the same, and
the value of the attribute BEGIN of TOKEN[#DOULBE=YES] is less than the BEGIN of SENTENCE[#NAME="PRIMARY"]/TOKEN
The result
<TOKEN BEGIN="139" END="141" SENTENCE_BEGIN="0" SENTENCE_END="156" PROP="A" DOUBLE="YES" />
is not correct, as the BEGIN value (139) is greater than the corresponding BEGIN values of the tokens with the same SENTENCE_BEGIN SENTENCE_END values:
<TOKEN BEGIN="9" END="11" SENTENCE_BEGIN="0" SENTENCE_END="156" />
<TOKEN BEGIN="32" END="37" SENTENCE_BEGIN="0" SENTENCE_END="156" />
How can I modify the code line to obtain the desired result?

Answer 1

Since you've tagged your question XSLT 1.0, I'd use XSLT instead of pure XPath. That way you can use xsl:key and also current().
Example...
<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output indent="yes"/>
<xsl:strip-space elements="*"/>
<xsl:key name="primary_tokens"
match="SENTENCE[#NAME='PRIMARY']/TOKEN"
use="concat(#SENTENCE_BEGIN,'|',#SENTENCE_END)"/>
<xsl:template match="/*">
<xsl:for-each select=".//TOKEN[#DOUBLE='YES'][key('primary_tokens',concat(#SENTENCE_BEGIN,'|',#SENTENCE_END))]">
<xsl:if test="key('primary_tokens',concat(#SENTENCE_BEGIN,'|',#SENTENCE_END))[#BEGIN > current()/#BEGIN]">
<xsl:copy-of select="."/>
</xsl:if>
</xsl:for-each>
</xsl:template>
</xsl:stylesheet>
Fiddle: http://xsltfiddle.liberty-development.net/gWmuiJ8