I've seen a number of posts about using tokenize() to split strings, but they all involve a single field. I have a slightly different situation and I'm not sure how to approach it.
My XML can resemble something like this:
<somefield>
<code>A,B,C</code>
<description>Description of A; Description of B; Description of C</description>
</somefield>
I want to transform that into something like this:
{
"somefield": [
{
"code": "A",
"description": "Description of A"
},
{
"code": "B",
"description": "Description of B"
},
{
"code": "C",
"description": "Description of C"
}
]
}
I don't quite see how to split both fields simultaneously like that. Any ideas?
In XSLT 2.0 you could do something like
<xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="xml" indent="yes"/>
<xsl:template match="/somefield">
<array key="somefield">
<xsl:variable name="descr" select="tokenize(description, ';')"/>
<xsl:for-each select="tokenize(code, ',')">
<xsl:variable name="i" select="position()"/>
<map>
<string key="code">
<xsl:value-of select="."/>
</string>
<string key="description">
<xsl:value-of select="$descr[$i]"/>
</string>
</map>
</xsl:for-each>
</array>
</xsl:template>
</xsl:stylesheet>
to get:
<?xml version="1.0" encoding="UTF-8"?>
<array key="somefield">
<map>
<string key="code">A</string>
<string key="description">Description of A</string>
</map>
<map>
<string key="code">B</string>
<string key="description"> Description of B</string>
</map>
<map>
<string key="code">C</string>
<string key="description"> Description of C</string>
</map>
</array>
(Sorry, I don't have the time to adjust this to output JSON.)