The goal is to group elements starting with different heading levels into sections nested according to those levels.
Problem is similar to XSLT: moving a grouping html elements into section levels. The difference here is that heading levels are not in strict order.
To give a simplified example, I want to transform an input like
<body>
<p>0.1</p>
<p>0.2</p>
<h2>h2.1</h2>
<h3>h3.1</h3>
<p>3.1</p>
<p>3.2</p>
<h1>h1.1</h1>
<p>1.1</p>
<h3>h3.2</h3>
<p>3a.1</p>
<p>3a.2</p>
</body>
into this desired output:
<document>
<body>
<p>0.1</p>
<p>0.2</p>
<section level="2">
<h2>h2.1</h2>
<section level="3">
<h3>h3.1</h3>
<p>3.1</p>
<p>3.2</p>
</section>
</section>
<section level="1">
<h1>h1.1</h1>
<p>1.1</p>
<section level="3">
<h3>h3.2</h3>
<p>3a.1</p>
<p>3a.2</p>
</section>
</section>
</body>
</document>
This is what I have tried so far, using some modifications to the solution given in XSLT: moving a grouping html elements into section levels:
<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:xs="http://www.w3.org/2001/XMLSchema"
xmlns:mf="http://example.com/mf"
exclude-result-prefixes="xs mf"
version="2.0">
<xsl:output indent="yes"/>
<xsl:template match="body">
<document>
<xsl:copy>
<xsl:apply-templates select="@*"/>
<xsl:sequence select="mf:group(*, 1)"/>
</xsl:copy>
</document>
</xsl:template>
<xsl:template match="@* | node()">
<xsl:copy>
<xsl:apply-templates select="@*, node()"/>
</xsl:copy>
</xsl:template>
<xsl:function name="mf:group" as="node()*">
<xsl:param name="elements" as="element()*"/>
<xsl:param name="level" as="xs:integer"/>
<xsl:for-each-group select="$elements"
group-starting-with="*[
mf:isHead(local-name()) and
(mf:getHLevel(local-name()) = $level or
count(preceding::*[mf:isHead(local-name())]) = 0
)
]">
<xsl:choose>
<xsl:when test="self::*[mf:getHLevel(local-name()) < 999]">
<xsl:variable name="myLevel"
select="mf:getHLevel(local-name())"/>
<section level="{$myLevel}">
<xsl:copy>
<xsl:apply-templates select="@*, node()"/>
</xsl:copy>
<xsl:sequence
select="mf:group(current-group() except ., $myLevel + 1)"/>
</section>
</xsl:when>
<xsl:otherwise>
<xsl:apply-templates select="current-group()"/>
</xsl:otherwise>
</xsl:choose>
</xsl:for-each-group>
</xsl:function>
<!-- Functions:
mf:isHead(string): tests whether string is a headline-name (h1, h2,...)
mf:getHLevel(string): gets level of heading (h1 -> 1, h2 -> 2, ..., no heading -> 999)
-->
<xsl:function name="mf:getHLevel" as="xs:integer">
<xsl:param name="s"/>
<xsl:value-of>
<xsl:choose>
<xsl:when test="mf:isHead($s)">
<xsl:value-of select="xs:integer(replace($s,'.*?(d+).*','$1'))"/>
</xsl:when>
<xsl:otherwise>
<xsl:value-of select="999"/>
</xsl:otherwise>
</xsl:choose>
</xsl:value-of>
</xsl:function>
<xsl:function name="mf:isHead" as="xs:boolean">
<xsl:param name="s"/>
<xsl:value-of select="matches($s,'hd+')"/>
</xsl:function>
</xsl:stylesheet>
I'm pretty sure that the conditions in @group-starting-with are wrong. Namely, count(preceding::*[mf:isHead(local-name())]) = 0 seems to not check, whether a heading-element is the first within the current sequence of elements. But I can't figure out what modifications are needed to achieve the desired output, so any help is appreciated.
