textdiff-with-options.xpl

<?xml version="1.0" encoding="utf-8"?>
<p:declare-step xmlns:p="http://www.w3.org/ns/xproc" xmlns:xs="http://www.w3.org/2001/XMLSchema"
  xmlns:c="http://www.w3.org/ns/xproc-step" xmlns:cx="http://xmlcalabash.com/ns/extensions"
  xmlns:dc="http://www.delightfulcomputing.com/ns/" version="1.0" name="textdiff-with-options" type="dc:textdiff-with-options">
  <p:documentation/>

  <p:option name="whitespace" select="'observe'">
    <p:documentation>Alternatives: ignore, normalize. Whether inline whitespace is ignored or contracted to a single space
      before comparison.</p:documentation>
  </p:option>
  <p:option name="original-uri" select="''"/>
  <p:option name="original-text" select="''"/>
  <p:option name="revised-uri" select="''"/>
  <p:option name="revised-text" select="''"/>
  <p:option name="encoding" select="''"/>

  <p:output port="result" primary="true"/>
  <p:serialization port="result" omit-xml-declaration="false" indent="true"/>

  <p:output port="space" sequence="true">
    <p:pipe port="space" step="space-modified"/>
  </p:output>
  <p:serialization port="space" omit-xml-declaration="false" indent="true"/>

  <p:import href="http://xmlcalabash.com/extension/steps/library-1.0.xpl"/>
  <p:import href="textdiff.xpl"/>

  <p:variable name="original-string" 
    select="if ($original-uri)
            then if ($encoding)
                 then unparsed-text($original-uri, $encoding)
                 else unparsed-text($original-uri)
            else $original-text"/>
  <p:variable name="revised-string" 
    select="if ($revised-uri)
            then if ($encoding)
                 then unparsed-text($revised-uri, $encoding)
                 else unparsed-text($revised-uri)
            else $revised-text"/>

  <dc:textdiff name="diff">
    <p:input port="original"><p:empty/></p:input>
    <p:input port="revised"><p:empty/></p:input>
    <p:with-option name="original-text" select="$original-string"/>
    <p:with-option name="revised-text" select="$revised-string"/>
  </dc:textdiff>

  <p:choose name="space-modified">
    <p:when test="$whitespace = ('ignore', 'normalize')">
      <p:output port="result" primary="true"/>
      <p:output port="space" sequence="true">
        <p:pipe port="result" step="space-modified-diff"/>
      </p:output>
      <p:variable name="space-modified-original" 
        select="if ($whitespace = 'ignore') 
                then replace($original-string, '[ \t]+', '', 'm')
                else replace($original-string, '[ \t]+', ' ', 'm')"/>
      <p:variable name="space-modified-revised" 
        select="if ($whitespace = 'ignore') 
                then replace($revised-string, '[ \t]+', '', 'm')
                else replace($revised-string, '[ \t]+', ' ', 'm')"/>
      <dc:textdiff name="space-modified-diff">
        <p:input port="original"><p:empty/></p:input>
        <p:input port="revised"><p:empty/></p:input>
        <p:with-option name="original-text" select="$space-modified-original"/>
        <p:with-option name="revised-text" select="$space-modified-revised"/>
      </dc:textdiff>
      <p:sink/>
      <p:xslt name="consolidate">
        <p:input port="source">
          <p:pipe port="result" step="diff"/>
          <p:pipe port="result" step="space-modified-diff"/>
        </p:input>
        <p:input port="stylesheet">
          <p:inline>
            <xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="2.0">
              <xsl:variable name="diff" as="document-node(element(dc:textdiff))" select="collection()[1]"/>
              <xsl:variable name="space-modified-diff" as="document-node(element(dc:textdiff))" select="collection()[2]"/>
              <xsl:template match="* | @*">
                <xsl:copy>
                  <xsl:apply-templates select="@*, node()"/>
                </xsl:copy>
              </xsl:template>
              <xsl:key name="change-by-orig" match="dc:delta[@type = 'change']" 
                use="string-join(dc:original/(@position, @size), '_')"/>
              <xsl:template match="dc:delta[@type = 'change']
                                           [empty(key('change-by-orig', 
                                                      string-join(dc:original/(@position, @size), '_'),
                                                      $space-modified-diff))]">
                <xsl:variable name="next-pos" as="xs:integer?" 
                  select="following-sibling::dc:delta[dc:original][1]/dc:original/@position"/>
                <xsl:apply-templates 
                  select="$space-modified-diff/dc:textdiff/dc:delta[dc:original/@position >= current()/dc:original/@position]
                                                                   [not(dc:original/@position >= $next-pos)]"/>
              </xsl:template>
            </xsl:stylesheet>
          </p:inline>
        </p:input>
        <p:input port="parameters"><p:empty/></p:input>
      </p:xslt>
    </p:when>
    <p:otherwise>
      <p:output port="result" primary="true"/>
      <p:output port="space" sequence="true">
        <p:empty/>
      </p:output>
      <p:identity/>
    </p:otherwise>
  </p:choose>
  
</p:declare-step>