xslt-lab/site/content/xslt/functions/xpath-tokenize.md

title, description, date, version, versionLabel, category, syntax, tags

title	description	date	version	versionLabel	category	syntax	tags
tokenize()	Splits a string into a sequence of substrings using a regular expression as the delimiter pattern.	2026-04-18T00:00:00Z	2.0	XSLT 2.0	string function	tokenize(string, pattern, flags?)	xslt reference xslt2 xpath

Description

tokenize() splits a string into a sequence of substrings wherever the pattern (a regular expression) matches. The matched delimiters are not included in the result. The function returns a sequence of xs:string items.

If the input string begins or ends with the delimiter pattern, the result includes an empty string at the start or end of the sequence respectively. If the input is an empty string and the pattern matches the empty string, the result is an empty sequence.

tokenize() is the inverse of string-join(): where string-join() assembles strings from a sequence, tokenize() disassembles a string into a sequence. This makes them natural complements for round-tripping delimited data.

Parameters

Parameter	Type	Required	Description
string	xs:string	Yes	The input string to split.
pattern	xs:string	Yes	A regular expression matching the delimiter.
flags	xs:string	No	Regex flags: i (case-insensitive), m (multiline), s (dot-all), x (extended).

Return value

xs:string* — a sequence of substrings split at each match of pattern.

Examples

Split a comma-separated list

Input XML:

<?xml version="1.0" encoding="UTF-8"?>
<data>
  <csv>apple,banana,cherry,date</csv>
</data>

Stylesheet:

<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
  <xsl:output method="xml" indent="yes"/>

  <xsl:template match="/data">
    <items>
      <xsl:for-each select="tokenize(csv, ',')">
        <item><xsl:value-of select="."/></item>
      </xsl:for-each>
    </items>
  </xsl:template>
</xsl:stylesheet>

Output:

<items>
  <item>apple</item>
  <item>banana</item>
  <item>cherry</item>
  <item>date</item>
</items>

Split on whitespace

Stylesheet:

<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
  <xsl:output method="xml" indent="yes"/>

  <xsl:template match="/data">
    <words>
      <!-- \s+ matches one or more whitespace characters -->
      <xsl:for-each select="tokenize(normalize-space(csv), '\s+')">
        <word><xsl:value-of select="."/></word>
      </xsl:for-each>
    </words>
  </xsl:template>
</xsl:stylesheet>

Count tokens

<!-- Count the number of comma-separated values -->
<xsl:value-of select="count(tokenize(csv, ','))"/>

Access a specific token by position

<!-- Get the second CSV field -->
<xsl:value-of select="tokenize(csv, ',')[2]"/>

Notes

• The pattern argument is a regular expression, not a plain string. Characters like ., *, +, ?, (, ) must be escaped with \ if used literally (e.g., to split on a literal ., use '\.').
• tokenize() returns zero or more strings; iterate the result with xsl:for-each or index it with [n].
• If pattern can match an empty string (e.g., '.*'), Saxon raises an error — the delimiter must have non-zero length.
• For simple fixed-character splitting (comma, pipe), tokenize() is the idiomatic choice in XSLT 2.0+. In XSLT 1.0, use recursive named templates.

See also

• matches()
• string-join()