Как структурировать плохо структурированный контент, основанный на следующих предикатах предиката/предшествующего словаря sibling?

Question

Основываясь на исходном XML ниже, я хочу захватить текстовые узлы и <emphasis> элементов, которые не содержат <emphasis bold="yes"> ПРИМЕЧАНИЕ: </emphasis> с их соответствующими <emphasis bold="yes"> ПРИМЕЧАНИЕ #: </emphasis> элемент.

Источник XML:

<section> 
    <para> 
    <emphasis bold="yes">NOTE1:</emphasis> This is the text of the first note 1 <emphasis bold="yes">that should only be in the <emphasis italic="yes">first</emphasis> subsection occurance of note one.</emphasis>. This is the second sentence of the first note one. <emphasis italic="yes">Here is some other text</emphasis> that can appear. <emphasis bold="yes">Marvin Gaye is an excellent musician1.</emphasis> Play it for your girlfriend1 <emphasis italic="yes">now1.</emphasis>. 
    <emphasis bold="yes">NOTE2:</emphasis> This is the text of the first note two2.1 <emphasis italic="yes">The Isley Brothers are also good.2.1</emphasis> 
    <emphasis bold="yes">NOTE1:</emphasis> This is the text of the second note one.1.2 <emphasis italic="yes">My girlfriend loves them1.2</emphasis> 
    <emphasis bold="yes">NOTE3:</emphasis> This is the text of the first note three3.1. 
    <emphasis bold="yes">NOTE1:</emphasis> This is the text of the third note one.1.3<emphasis italic="yes">She is going to make me dinner tonight1.3</emphasis> 
    <emphasis bold="yes">NOTE3:</emphasis> This is the text of the second note three.3.2<emphasis italic="yes">Steak and potatos3.2</emphasis> 
    <emphasis bold="yes">NOTE2:</emphasis> This is the text of the second note two.2.2<emphasis italic="yes">And then some wine2.2</emphasis> 
    </para> 
</section> 

Текущий XSLT:

<xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"> 

    <xsl:output method="xml" version="1.0" encoding="UTF-8" indent="yes"/> 

    <xsl:template match="*|@*|text()"> 
    <xsl:copy> 
     <xsl:apply-templates select="*|@*|text()"/> 
    </xsl:copy> 
    </xsl:template> 

    <xsl:template match="section"> 
    <root> 
     <xsl:apply-templates select="*|@*|text()"/> 
    </root> 
    </xsl:template> 

    <xsl:template match="para/emphasis[preceding-sibling::emphasis[@bold='yes' and text()='NOTE1:']]"/> 
    <xsl:template match="para/emphasis[preceding-sibling::emphasis[@bold='yes' and text()='NOTE2:']]"/> 
    <xsl:template match="para/emphasis[preceding-sibling::emphasis[@bold='yes' and text()='NOTE3:']]"/> 
    <xsl:template match="para/text()[preceding-sibling::emphasis[@bold='yes' and text()='NOTE1:']]"/> 
    <xsl:template match="para/text()[preceding-sibling::emphasis[@bold='yes' and text()='NOTE2:']]"/> 
    <xsl:template match="para/text()[preceding-sibling::emphasis[@bold='yes' and text()='NOTE3:']]"/> 

    <xsl:template match="para/emphasis[@bold='yes' and text()='NOTE1:' and not(position()=last())]"> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE1:</emphasis> 
     <xsl:copy-of 
      select="following-sibling::text()[not(preceding-sibling::emphasis[@bold='yes' and text()='NOTE2:']) and not(preceding-sibling::emphasis[@bold='yes' and text()='NOTE3:'])] |following-sibling::emphasis[not(contains(string(), 'NOTE1:')) and not(contains(string(), 'NOTE2:')) and not(contains(string(), 'NOTE3:'))]" 
     /> 
     </para> 
    </subsection> 
    </xsl:template> 

    <xsl:template match="para/emphasis[position()=last() and position() > 1 and @bold='yes' and text()='NOTE1:']"> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE1:</emphasis> 
     <xsl:copy-of 
      select="following-sibling::text()[position() = 1]|following-sibling::emphasis[not(contains(string(), 'NOTE1:')) and not(contains(string(), 'NOTE2:')) and not(contains(string(), 'NOTE3:'))]" 
     /> 
     </para> 
    </subsection> 
    </xsl:template> 

    <xsl:template match="para/emphasis[@bold='yes' and text()='NOTE2:' and not(position()=last())]"> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE2:</emphasis> 
     <xsl:copy-of 
      select="following-sibling::text()[not(preceding-sibling::emphasis[@bold='yes' and text()='NOTE1:']) and not(preceding-sibling::emphasis[@bold='yes' and text()='NOTE3:'])] |following-sibling::emphasis[not(contains(string(), 'NOTE1:')) and not(contains(string(), 'NOTE2:')) and not(contains(string(), 'NOTE3:'))]" 
     /> 
     </para> 
    </subsection> 
    </xsl:template> 

    <xsl:template match="para/emphasis[position()=last() and position() > 1 and @bold='yes' and text()='NOTE2:']"> 
    <note> 
     <para> 
     <emphasis bold="yes">NOTE2:</emphasis> 
     <xsl:copy-of 
      select="following-sibling::text()[position() = 1]|following-sibling::emphasis[not(contains(string(), 'NOTE1:')) and not(contains(string(), 'NOTE2:')) and not(contains(string(), 'NOTE3:')) ]" 
     /> 
     </para> 
    </note> 
    </xsl:template> 

    <xsl:template match="para/emphasis[@bold='yes' and text()='NOTE3:' and not(position()=last())]"> 
    <note> 
     <para> 
     <emphasis bold="yes">NOTE3:</emphasis> 
     <xsl:copy-of 
      select="following-sibling::text()[not(preceding-sibling::emphasis[@bold='yes' and text()='NOTE2:']) and not(preceding-sibling::emphasis[@bold='yes' and text()='NOTE1:'])] | following-sibling::emphasis[not(contains(string(), 'NOTE1:')) and not(contains(string(), 'NOTE2:')) and not(contains(string(), 'NOTE3:'))]" 
     /> 
     </para> 
    </note> 
    </xsl:template> 

    <xsl:template match="para/emphasis[position()=last() and position() > 1 and @bold='yes' and text()='NOTE3:']"> 
    <note> 
     <para> 
     <emphasis bold="yes">NOTE3:</emphasis> 
     <xsl:copy-of select="following-sibling::text()[position() = 1]|following-sibling::emphasis[not(contains(string(), 'NOTE1:')) and not(contains(string(), 'NOTE2:')) and not(contains(string(), 'NOTE3:')) ]"/> 
     </para> 
    </note> 
    </xsl:template> 

</xsl:stylesheet> 

Токовый выход:

<root> 
    <para> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE1:</emphasis> This is the text of the first note 1 <emphasis bold="yes">that should only be in the <emphasis italic="yes">first</emphasis> subsection occurance of note one.</emphasis>. This is the second sentence of the first note one. <emphasis italic="yes">Here is some other text</emphasis> that can appear. <emphasis bold="yes">Marvin Gaye is an excellent musician1.</emphasis> Play it for your girlfriend1 <emphasis italic="yes">now1.</emphasis>. 
     <emphasis italic="yes">The Isley Brothers are also good.2.1</emphasis> 
     <emphasis italic="yes">My girlfriend loves them1.2</emphasis> 
     <emphasis italic="yes">She is going to make me dinner tonight1.3</emphasis> 
     <emphasis italic="yes">Steak and potatos3.2</emphasis> 
     <emphasis italic="yes">And then some wine2.2</emphasis> 
     </para> 
    </subsection> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE2:</emphasis> 
     <emphasis italic="yes">The Isley Brothers are also good.2.1</emphasis> 
     <emphasis italic="yes">My girlfriend loves them1.2</emphasis> 
     <emphasis italic="yes">She is going to make me dinner tonight1.3</emphasis> 
     <emphasis italic="yes">Steak and potatos3.2</emphasis> 
     <emphasis italic="yes">And then some wine2.2</emphasis> 
     </para> 
    </subsection> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE1:</emphasis> 
     <emphasis italic="yes">My girlfriend loves them1.2</emphasis> 
     <emphasis italic="yes">She is going to make me dinner tonight1.3</emphasis> 
     <emphasis italic="yes">Steak and potatos3.2</emphasis> 
     <emphasis italic="yes">And then some wine2.2</emphasis> 
     </para> 
    </subsection> 
    <note> 
     <para> 
     <emphasis bold="yes">NOTE3:</emphasis> 
     <emphasis italic="yes">She is going to make me dinner tonight1.3</emphasis> 
     <emphasis italic="yes">Steak and potatos3.2</emphasis> 
     <emphasis italic="yes">And then some wine2.2</emphasis> 
     </para> 
    </note> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE1:</emphasis> 
     <emphasis italic="yes">She is going to make me dinner tonight1.3</emphasis> 
     <emphasis italic="yes">Steak and potatos3.2</emphasis> 
     <emphasis italic="yes">And then some wine2.2</emphasis> 
     </para> 
    </subsection> 
    <note> 
     <para> 
     <emphasis bold="yes">NOTE3:</emphasis> 
     <emphasis italic="yes">Steak and potatos3.2</emphasis> 
     <emphasis italic="yes">And then some wine2.2</emphasis> 
     </para> 
    </note> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE2:</emphasis> 
     <emphasis italic="yes">And then some wine2.2</emphasis> 
     </para> 
    </subsection> 
    </para> 
</root> 


Desired output: 

<root> 
    <para> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE1:</emphasis> This is the text of the first note 1 <emphasis bold="yes">that should only be in the <emphasis italic="yes">first</emphasis> subsection occurance of note one.</emphasis>. This is the second sentence of the first note one. <emphasis italic="yes">Here is some other text</emphasis> that can appear. <emphasis bold="yes">Marvin Gaye is an excellent musician1.</emphasis> Play it for your girlfriend1 <emphasis italic="yes">now1.</emphasis>. 
     </para> 
    </subsection> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE2:</emphasis>This is the text of the first note two2.1<emphasis italic="yes">The Isley Brothers are also good.2.1</emphasis> 
     </para> 
    </subsection> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE1:</emphasis> This is the text of the second note one.1.2 <emphasis italic="yes">My girlfriend loves them1.2</emphasis> 
     </para> 
    </subsection> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE3:</emphasis> This is the text of the first note three3.1. 
     </para> 
    </subsection> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE1:</emphasis> This is the text of the third note one.1.3<emphasis italic="yes">She is going to make me dinner tonight1.3</emphasis> 
     </para> 
    </subsection> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE3:</emphasis> This is the text of the second note three.3.2<emphasis italic="yes">Steak and potatos3.2</emphasis> 
     </para> 
    </subsection> 
    <subsection> 
     <para> 
     <emphasis bold="yes">NOTE2:</emphasis> This is the text of the second note two.2.2<emphasis italic="yes">And then some wine2.2</emphasis> 
     </para> 
    </subsection> 
    </para> 
</root> 

Answer 1

Если я правильно понимаю ваши требования правильно, я думаю, что вы можете иметь несколько усложненной ваш XSLT , Похоже, вы просто хотите, чтобы сгруппировать узлы, с началом каждой группы представляется неким акцент узла с текстом состояния «Примечание»

Предполагая, что вы используете XSLT 2.0 (как текущий XSLT помечается версии = «2.0»), тогда поздороваться с полезной командой xsl: for-each-group, которая должна быть именно то, что вам нужно. В вашем случае, группируя узлы, начиная с соответствующим акцентом элементы:

<xsl:for-each-group 
    select="node()" 
    group-starting-with="emphasis[starts-with(text(), 'NOTE')]"> 

внутри в этом, вы бы выходной ваши подраздел и пункт элементы, а затем итерацию по элементам в группе, чтобы скопировать их

<xsl:apply-templates select="current-group()" /> 

Вы бы использовать XSLT identity transform для вывода диакритических самих элементов.

Здесь счастливо много-короче XSLT

<xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"> 
    <xsl:output method="xml" indent="yes"/> 
    <xsl:strip-space elements="*"/> 

    <xsl:template match="para"> 
     <para> 
     <xsl:for-each-group select="node()" group-starting-with="emphasis[starts-with(text(), 'NOTE')]"> 
      <subsection> 
       <para> 
        <xsl:apply-templates select="current-group()" /> 
       </para> 
      </subsection> 
     </xsl:for-each-group> 
     </para> 
    </xsl:template> 

    <xsl:template match="@*|node()"> 
     <xsl:copy> 
     <xsl:apply-templates select="@*|node()"/> 
     </xsl:copy> 
    </xsl:template> 
</xsl:stylesheet> 

Применительно к вашему XML, следующий выход

<section> 
    <para> 
     <subsection> 
     <para> 
      <emphasis bold="yes">NOTE1:</emphasis> This is the text of the first note 1 
      <emphasis bold="yes">that should only be in the 
       <emphasis italic="yes">first</emphasis>subsection occurance of note one. 
      </emphasis>. This is the second sentence of the first note one. 
      <emphasis italic="yes">Here is some other text</emphasis>that can appear. 
      <emphasis bold="yes">Marvin Gaye is an excellent musician1.</emphasis>Play it for your girlfriend1 
      <emphasis italic="yes">now1.</emphasis>. 
     </para> 
     </subsection> 
     <subsection> 
     <para> 
      <emphasis bold="yes">NOTE2:</emphasis>This is the text of the first note two2.1 
      <emphasis italic="yes">The Isley Brothers are also good.2.1</emphasis></para> 
     </subsection> 
     <subsection> 
     <para> 
      <emphasis bold="yes">NOTE1:</emphasis>This is the text of the second note one.1.2 
      <emphasis italic="yes">My girlfriend loves them1.2</emphasis></para> 
     </subsection> 
     <subsection> 
     <para> 
      <emphasis bold="yes">NOTE3:</emphasis>This is the text of the first note three3.1. </para> 
     </subsection> 
     <subsection> 
     <para> 
      <emphasis bold="yes">NOTE1:</emphasis>This is the text of the third note one.1.3 
      <emphasis italic="yes">She is going to make me dinner tonight1.3</emphasis></para> 
     </subsection> 
     <subsection> 
     <para> 
      <emphasis bold="yes">NOTE3:</emphasis> This is the text of the second note three.3.2 
      <emphasis italic="yes">Steak and potatos3.2</emphasis></para> 
     </subsection> 
     <subsection> 
     <para> 
      <emphasis bold="yes">NOTE2:</emphasis> This is the text of the second note two.2.2 
      <emphasis italic="yes">And then some wine2.2</emphasis></para> 
     </subsection> 
    </para> 
</section> 

См http://www.xml.com/lpt/a/1314 больше примеров о том, как группа в XSLT 2.0