Re: [xsl] Splitting a paragraph into sentences and keep markup

["Imsieke, Gerrit, le-tex gerrit.imsieke@xxxxxxxxx" <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

Therebs a package for splitting at arbitrarily deeply nested nodes. It 
is part of a paper that I presented at XML Prague this year: 
https://archive.xmlprague.cz/2019/files/xmlprague-2019-proceedings.pdf#page=347

The package itself is at 
https://subversion.le-tex.de/common/presentations/2019-02-09_xmlprague_xslt-upward-projection/lib/split.xsl

Using this package, Martin's p-matching template becomes:

<xsl:template match="p[node()]">
  <xsl:variable name="p-with-markers" as="element(p)">
    <xsl:apply-templates select="." mode="insert-marker"/>
  </xsl:variable><!-- this hasn't changed -->
  <xsl:variable name="chunks" as="document-node(element(split:chunks))">
    <xsl:apply-templates select="$p-with-markers"
      mode="split:split-entrypoint"><!-- mode provided by
        lib/split.xsl -->
      <xsl:with-param name="group-start-exp" as="xs:string"
        select="'self::eos'"/><!-- Will be evaluated as an XPath
          expression for each node in a 
for-each-group[@group-starting-with]
          population. If a population node satisfies the expression, it 
will
          start a group.-->
      <xsl:with-param name="keep-splitting-node" as="xs:boolean"
        select="false()"/><!-- remove <eos/> after splitting -->
    </xsl:apply-templates>
  </xsl:variable>
  <xsl:copy-of select="$chunks/split:chunks/split:chunk/p[node()]"
    copy-namespaces="no"/>
</xsl:template>

The complete stylesheet is at 
https://gist.github.com/gimsieke/529dab000386a45d6136e850a80ac726

Applying it to your input, David, will yield:

<?xml version="1.0" encoding="UTF-8"?><root>
    <p>This has one <span class="zzz">sentence? </span></p><p><span 
class="zzz">Actually, it has
<emphasis>two</emphasis>.  </span></p><p><span class="zzz">No,</span> it 
has three.</p>
</root>

Gerrit


On 24.11.2019 15:32, David Carlisle d.p.carlisle@xxxxxxxxx wrote:
> can we assume the easy case (as in your example) where all the
> sentences end at the top level?
>
> a more challenging example is
>
> <root>
>      <p>This has one <span class="zzz">sentence? Actually, it has
> <emphasis>two</emphasis>.  No,</span> it has three.</p>
> </root>
>
> as then you need to force-close any open elements at the sentence end
> and re-open them in the new sentence so something like
>
>    <p>This has one <span class="zzz">sentence?</span></p>
>    <p><span class="zzz">Actually, it has <emphasis>two</emphasis>.</span></p>
>   <p><span class="zzz">No,</span> it has three.</p>
>
> David
>
> On Sun, 24 Nov 2019 at 13:34, Rick Quatro rick@xxxxxxxxxxxxxx
> <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> wrote:
> > Hi All,
> >
> >
> >
> > I have a situation where I want to split a short paragraph into sentences and use them in different parts of my output. I am using <xsl:analyze-string> because I want to account for a sentence ending with a . or ?. This will work except if there are any children of the paragaph, like the <emphasis> in the second sentence. Can I split a paragraph into sentences and still keep the markup?
> >
> >
> >
> > Here is my input document:
> >
> >
> >
> > <?xml version="1.0" encoding="UTF-8"?>
> >
> > <root>
> >
> >      <p>This has one sentence? Actually, it has <emphasis>two</emphasis>. No, it has three.</p>
> >
> > </root>
> >
> >
> >
> > My stylesheet:
> >
> >
> >
> > <?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:rq="http://www.frameexpert.com";
> >
> >      exclude-result-prefixes="xs rq"
> >
> >      version="2.0">
> >
> >
> >
> >      <xsl:output indent="yes"/>
> >
> >      <xsl:strip-space elements="root"/>
> >
> >
> >
> >      <xsl:template match="/root">
> >
> >          <xsl:copy>
> >
> >              <xsl:apply-templates/>
> >
> >          </xsl:copy>
> >
> >      </xsl:template>
> >
> >
> >
> >      <xsl:template match="p">
> >
> >          <xsl:variable name="sentences" select="rq:splitParagraphIntoSentences(.)"/>
> >
> >          <p><xsl:value-of select="$sentences[1]"/></p>
> >
> >          <note>Something in between.</note>
> >
> >          <p><xsl:value-of select="$sentences[position()&gt;1]"/></p>
> >
> >      </xsl:template>
> >
> >
> >
> >      <xsl:function name="rq:splitParagraphIntoSentences">
> >
> >          <xsl:param name="paragraph"/>
> >
> >          <xsl:analyze-string select="$paragraph" regex=".+?[\.\?](\s+|$)">
> >
> >              <xsl:matching-substring>
> >
> >                  <sentence><xsl:value-of select="replace(.,'\s+$','')"/></sentence>
> >
> >              </xsl:matching-substring>
> >
> >          </xsl:analyze-string>
> >
> >      </xsl:function>
> >
> > </xsl:stylesheet>
> >
> >
> >
> > My output:
> >
> >
> >
> > <?xml version="1.0" encoding="UTF-8"?>
> >
> > <root>
> >
> >     <p>This has one sentence?</p>
> >
> >     <note>Something in between.</note>
> >
> >     <p>Actually, it has two. No, it has three.</p>
> >
> > </root>
> >
> >
> >
> > What I want is this:
> >
> >
> >
> > <?xml version="1.0" encoding="UTF-8"?>
> >
> > <root>
> >
> >     <p>This has one sentence? </p>
> >
> >     <note>Something in between.</note>
> >
> >     <p>Actually, it has <emphasis>two</emphasis>. No, it has three. </p>
> >
> > </root>
> >
> >
> >
> > Any suggestions will be appreciated.
> >
> >
> >
> > Rick