outermost()

Description

outermost() filters a sequence of nodes to keep only those that are not descendants of any other node in the same sequence. It removes any node that has an ancestor also in the sequence, keeping only the highest-level nodes.

The result is returned in document order. outermost() is the complement of innermost(): where innermost() retains the deepest nodes, outermost() retains the shallowest ancestors. The function is useful for deduplicating overlapping selections and for determining the roots of a set of subtrees.

A common use case is normalizing the result of multiple union expressions where some selected nodes are contained within others — outermost() strips the redundant descendants, leaving only the minimal covering set.

Parameters

Return value

node()* — the subset of input nodes that have no ancestors in the input sequence, in document order.

Examples

Finding the top-level selected nodes

Input XML:

<?xml version="1.0" encoding="UTF-8"?>
<report>
  <chapter id="ch1">
    <section id="s1"><para>Text</para></section>
  </chapter>
  <chapter id="ch2"/>
</report>

Stylesheet:

<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="3.0"
  xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
  <xsl:output method="xml" indent="yes"/>

  <xsl:template match="/report">
    <xsl:variable name="mixed" select="chapter | chapter/section | chapter/section/para"/>
    <outermost-nodes>
      <xsl:for-each select="outermost($mixed)">
        <node id="{@id}" name="{name()}"/>
      </xsl:for-each>
    </outermost-nodes>
  </xsl:template>
</xsl:stylesheet>

Output:

<outermost-nodes>
  <node id="ch1" name="chapter"/>
  <node id="ch2" name="chapter"/>
</outermost-nodes>

Deduplicating subtree roots

Stylesheet:

<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="3.0"
  xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
  <xsl:output method="text"/>

  <xsl:template match="/report">
    <xsl:variable name="all" select="chapter | chapter/section"/>
    <xsl:value-of select="count(outermost($all))"/>
    <xsl:text> outermost nodes</xsl:text>
  </xsl:template>
</xsl:stylesheet>

Output:

2 outermost nodes

Notes

• If no node in the sequence is a descendant of any other, outermost() returns all nodes in document order.
• outermost() is particularly useful when preparing a node-set for xsl:copy-of where copying an ancestor would duplicate its descendants.
• The result is always in document order regardless of the input order.
• The function was introduced in XPath 3.0 / XSLT 3.0 alongside the general expansion of sequence manipulation functions.

See also

• innermost()
• has-children()
• path()