% $Id: tex4ht-xhtmml-xtpipes.tex 740 2020-06-13 22:35:32Z karl $

% $Id: tex4ht-xhtmml-xtpipes.tex 740 2020-06-13 22:35:32Z karl $
% htlatex tex4ht-xhtmml-xtpipes "xhtml,next,3" "" "-d./"

% Copyright 2009-2020 TeX Users Group
% Copyright 1996-2009 Eitan M. Gurari
% Released under LPPL 1.3c+.
% See tex4ht-cpright.tex for license text.

\documentclass{article}
\Configure{ProTex}{log,<<<>>>,title,list,`,[[]]}
\usepackage{url}
\input{common.tex}
\begin{document}
\input tex4ht-cpright.tex
\input tex4ht-dir

%%%%%%%%%%%%%%%%%%
\part{Post Processing for HTML Output Mode}
%%%%%%%%%%%%%%%%%%

%%%%%%%%%%%%%%%%%%
\section{Outline}
%%%%%%%%%%%%%%%%%%

\AtEndDocument{\OutputCodE\<xhtmml.4xt\>}

\Needs{"xmllint --valid --noout xhtmml.4xt"}

\<xhtmml.4xt\><<<
<?xml version="1.0" encoding="UTF-8" ?>
<!DOCTYPE xtpipes SYSTEM "xtpipes.dtd" >

<xtpipes preamble="yes" signature="xhtmml.4xt (`version)">
<sax content-handler="xtpipes.util.ScriptsManager"
lexical-handler="xtpipes.util.ScriptsManagerLH" >
`<p script`>
`<li script`>
`<table script`>
`<mrow script`>
`<math script`>
`<mstyle script`>
</sax>
</xtpipes>
>>>

%
\AtEndDocument{\OutputCodE\<XhtmmlUtilities.java\>}

\ifdojava
\Needs{"
if [ ! -d \XTPIPES\space]; then exit 1; fi
;
javac XhtmmlUtilities.java -d \XTPIPES
"}
\fi

\<XhtmmlUtilities.java\><<<
package tex4ht;
/* XhtmmlUtilities.java (`version), generated from `jobname.tex
Copyright (C) 2009-2010 TeX Users Group
Copyright (C) `CopyYear.2008. Eitan M. Gurari
`<TeX4ht copyright`> */

import org.w3c.dom.*;
public class XhtmmlUtilities {
`<static void p(dom)`>
`<static void li(dom)`>
`<static void table(dom)`>
`<static void mrow(dom)`>
`<static void math(dom)`>
`<void barsIntoFenced(dom)`>
`<void digitsIntoNumbers(dom)`>
}
>>>

\<mrow script\><<<
<script element="mrow" >
<dom name="." xml="." method="mrow" class="tex4ht.XhtmmlUtilities" />
</script>
>>>

\<static void mrow(dom)\><<<
public static void mrow(Node dom) {
barsIntoFenced( dom );
digitsIntoNumbers( dom );
}
>>>

\<math script\><<<
<script element="math" >
<dom name="." xml="." method="math" class="tex4ht.XhtmmlUtilities" />
`<math-mstyle script`>
</script>
>>>

\<static void math(dom)\><<<
public static void math(Node dom) {
barsIntoFenced( dom );
digitsIntoNumbers( dom );
}
>>>

%%%%%%%%%%%%%
\section{Empty Paragraphs}
%%%%%%%%%%%%%

\<p script\><<<
<script element="p" >
<dom name="." xml="." method="p" class="tex4ht.XhtmmlUtilities" />
</script>
>>>

\<p script\><<<
<script element="div" >
<dom name="." xml="." method="p" class="tex4ht.XhtmmlUtilities" />
</script>
>>>

\<static void p(dom)\><<<
public static void p(Node dom) {
Node pNode = dom.getFirstChild();
if( pNode.hasChildNodes() ){
boolean drop = true;
Node child = pNode.getFirstChild();
while( child != null ){
short type = child.getNodeType();
if(
(type == Node.ELEMENT_NODE)
||
(type == Node.CDATA_SECTION_NODE)
||
(type == Node.TEXT_NODE)
&&
!((Text) child).getWholeText().trim().equals("")
){
drop = false; break;
}
child = child.getNextSibling();
}
if( drop ){
dom.removeChild( pNode );
}
} else {
dom.removeChild( pNode );
}
}
>>>

%%%%%%%%%%%%%
\section{Tables}
%%%%%%%%%%%%%

Remove Empty Trailling Row

\<table script\><<<
<script element="table" >
<dom name="." xml="." method="table" class="tex4ht.XhtmmlUtilities" />
</script>
>>>

\<static void table(dom)\><<<
public static void table(Node dom) {
Node tableNode = dom.getFirstChild();
if( tableNode.hasChildNodes() ){
Node trChild = tableNode.getLastChild();
while( (trChild != null)
&&
!trChild.getNodeName().equals("tr") ){
trChild = trChild.getPreviousSibling();
}
if( (trChild != null) && trChild.hasChildNodes() ){
Node tdChild = trChild.getLastChild();
while( (tdChild != null)
&&
!tdChild.getNodeName().equals("td") ){
tdChild = tdChild.getPreviousSibling();
}
if( !tdChild.hasChildNodes() ){
tableNode.removeChild( trChild );
} else {
Node child = tdChild.getFirstChild();
if(
(child.getNodeType() == Node.TEXT_NODE)
&&
((Text) child).getWholeText().trim().equals("")
&&
(child.getNextSibling() == null)
){
tableNode.removeChild( trChild );
} } } } }
>>>

%%%%%%%%%%%%%
\section{List Items}
%%%%%%%%%%%%%

%%%%%%%%%%%%%
\subsection{Outline}
%%%%%%%%%%%%%

\<li script\><<<
<script element="li" >
<dom name="." xml="." method="li" class="tex4ht.XhtmmlUtilities" />
</script>
>>>

\<static void li(dom)\><<<
public static void li(Node dom) {
Node liNode = dom.getFirstChild();
Node liChild;
boolean hasBlock = false, hasInline = false;
`<hasBlock, hasInline := exist in li?`>
if( hasBlock && hasInline ){
`<p := inline code`>
`<insert p node as first child`>
}
}
>>>

%%%%%%%%%%%%%
\subsection{Place Inline Code into Paragraphs}
%%%%%%%%%%%%%

\<p := inline code\><<<
liChild = liNode.getLastChild();
Element g = ((Document) dom).createElement("p");
g.setAttribute("class", "noindent");
while( liChild != null ){
short type = liChild.getNodeType();
if( `<liChild == block?`> ){
`<insert p node after block node`>
liChild = liChild.getPreviousSibling();
} else {
Node nextChild = liChild;
liChild = liChild.getPreviousSibling();
`<move node into p`>
} }
>>>

\<move node into p\><<<
nextChild = liNode.removeChild(nextChild);
type = nextChild.getNodeType();
if(
(type != Node.COMMENT_NODE)
&&
(
(type != Node.TEXT_NODE)
||
!((Text) nextChild).getWholeText().trim().equals("")
)
){
if( g.hasChildNodes() ){
g.insertBefore( nextChild, g.getFirstChild() );
} else {
g.appendChild( nextChild );
} }
>>>

%%%%%%%%%%%%%
\subsection{Place Paragraphs into List Items}
%%%%%%%%%%%%%

\<insert p node as first child\><<<
if( liNode.hasChildNodes() ){
liNode.insertBefore( g, liNode.getFirstChild() );
} else {
liNode.appendChild( g );
}
>>>

\<insert p node after block node\><<<
if( g.hasChildNodes() ){
Node nextChild = liChild.getNextSibling();
if( nextChild == null ){
liNode.appendChild( g );
} else {
liNode.insertBefore( g, nextChild );
}
g = ((Document) dom).createElement("p");
g.setAttribute("class", "noindent");
}
>>>

%%%%%%%%%%%%%
\subsection{Check Nature of Item Content}
%%%%%%%%%%%%%

\<hasBlock, hasInline := exist in li?\><<<
liChild = liNode.getFirstChild();
while( liChild != null ){
short type = liChild.getNodeType();
if( `<liChild == block?`> ){ hasBlock = true; }
else if( type == Node.TEXT_NODE ){
if( !((Text) liChild).getWholeText().trim().equals("") ){
hasInline = true;
} }
else if(
(type != Node.COMMENT_NODE)
&&
(type != Node.PROCESSING_INSTRUCTION_NODE )
){
hasInline = true;
}
liChild = liChild.getNextSibling();
}
>>>

\<liChild == block?\><<<
(type == Node.ELEMENT_NODE)
&&
(
liChild.getNodeName().equals("p")
||
liChild.getNodeName().equals("ol")
||
liChild.getNodeName().equals("ul")
||
liChild.getNodeName().equals("div")
||
liChild.getNodeName().equals("table")
)
>>>

%%%%%%%%%%%%%%%%%%
\section{Scripted Math |...|}
%%%%%%%%%%%%%%%%%%

\<void barsIntoFenced(dom)\><<<
static void barsIntoFenced(Node dom) {
Node rightBarNode = null;
Node mrowNode = dom.getFirstChild();
Node mrowChild = mrowNode.getLastChild();
while( mrowChild != null ){
if(
mrowChild.getNodeName().equals("msub")
||
mrowChild.getNodeName().equals("msup")
||
mrowChild.getNodeName().equals("msubsup")
){
Node firstChild = mrowChild.getFirstChild();
if(
(firstChild.getChildNodes().getLength() == 1)
&&
firstChild.getTextContent().equals( "|" )
){
rightBarNode = mrowChild;
} }
else
if(
(rightBarNode != null)
&&
mrowChild.getNodeName().equals("mo")
&&
(mrowChild.getChildNodes().getLength() == 1)
&&
mrowChild.getTextContent().equals( "|" )
){
if(mrowChild.getNextSibling() != rightBarNode){
`<set mfenced sub,sup`>
}
rightBarNode = null;
}
mrowChild = mrowChild.getPreviousSibling();
} }
>>>

\begin{verbatim}
A + |B|^2 + C
<mrow>
<mi>A</mi>
<mo class="MathClass-bin">+</mo>
<mo class="MathClass-rel">|</mo> #
<mi>B</mi> #
<msup> #
<mrow> #
<mo class="MathClass-rel">|</mo> #
</mrow> #
<mrow> #
<mn>2</mn> #
</mrow> #
</msup> #
<mo class="MathClass-bin">+</mo>
<mi>C</mi>
</mrow>
\end{verbatim}

\<set mfenced sub,sup\><<<
rightBarNode.removeChild( rightBarNode.getFirstChild() );
Node sub = rightBarNode.getFirstChild();
Node mfenced = ((Document) dom).createElement( "mfenced" );
rightBarNode.insertBefore( mfenced, sub );
((Element) mfenced).setAttribute("open","|");
((Element) mfenced).setAttribute("close","|");
((Element) mfenced).setAttribute("separator","");
Node node = mrowChild.getNextSibling();
while( node != rightBarNode ){
Node next = node.getNextSibling();
mrowNode.removeChild( node );
mfenced.appendChild( node );
node = next;
}
mrowNode.removeChild( mrowChild );
mrowChild = mrowNode.getLastChild();
>>>

%%%%%%%%%%%%%%%%%%
\section{Digits into Numbers}
%%%%%%%%%%%%%%%%%%

\<void digitsIntoNumbers(dom)\><<<
static void digitsIntoNumbers(Node dom){
Node mrowNode = dom.getFirstChild();
Node mrowChild = mrowNode.getFirstChild();
short state = `<init state`>;
Node fromNode = null,
toNode = null;
while( mrowChild != null ){
switch( state ){
case `<init state`>:
if( mrowChild.getNodeName().equals("mn") ){
state = `<mn state`>;
fromNode = mrowChild;
toNode = mrowChild;
}
break;
case `<mn state`>:
boolean bool = true;
if( mrowChild.getNodeName().equals("mn") ){
toNode = mrowChild;
bool = ( mrowChild.getNextSibling() == null );
}
else
{ `<handle scripted numbers`> }
if( bool )
{ `<bool := just single digits?`>
if( bool ){
`<num := merge mn's`>
node.replaceChild( ((Document) dom).createTextNode(num),
node.getFirstChild()
);
}
state = `<init state`>;
fromNode = null;
toNode = null;
} }
mrowChild = mrowChild.getNextSibling();
} }
>>>

\<init state\><<<
0
>>>

\<mn state\><<<
1
>>>

\<num := merge mn's\><<<
Node next;
String num = "";
Node node = fromNode;
while( node != toNode ){
num += node.getTextContent();
next = node.getNextSibling();
mrowNode.removeChild( node );
node = next;
}
num += node.getTextContent();
>>>

\<bool := just single digits?\><<<
for( Node node = fromNode; ; node = node.getNextSibling()){
String str = node.getTextContent();
if( str.length() > 1 ){ bool = false; break; }
if( !str.replaceAll("[0-9]","").equals("") ){ bool = false; break; }
if( node == toNode ){ break; }
}
>>>

\<handle scripted numbers\><<<
String s = null;
Node base;
if(
( mrowChild.getNodeName().equals("msub")
||
mrowChild.getNodeName().equals("msup")
||
mrowChild.getNodeName().equals("msubsup")
)
&&
((s = (base = mrowChild.getFirstChild())
. getTextContent()).length() == 1 )
&&
s.replaceAll("[0-9]","").equals("")
){
`<bool := just single digits?`>
`<num := merge mn's`>
mrowNode.removeChild( node );
num += s;
Node mn = ((Document) dom).createElement("mn");
mn.appendChild( ((Document) dom).createTextNode(num) );
mrowChild.replaceChild( mn, base);

bool = false;
state = `<init state`>;
fromNode = null;
toNode = null;
}
>>>

%%%%%%%%%%%%%%%%%%
\section{mstyle}
%%%%%%%%%%%%%%%%%%

%%%%%%%%%%%%%
\subsection{Mstyle as Parent of an Element}
%%%%%%%%%%%%%

Merge mstyle into its child. For instance

\begin{verbatim}
<mstyle mathvariant="bold">
<mi>F</mi>
</mstyle>
\end{verbatim}

into

\begin{verbatim}
<mi mathvariant="bold">F</mi>
\end{verbatim}

\<mstyle script\><<<
<script element="mstyle" >
<set name="mstyle" >
`<open xslt script`>
`<mstyle templates`>
`<close xslt script`>
</set>
<xslt name="." xml="." xsl="mstyle" />
</script>
>>>

\<mstyle templates\><<<
<xsl:template match=" m:mstyle [
@mathvariant
and
child::m:*[ not(@mathvariant) ]
and
(count(child::m:*) = 1)
] " >
<xsl:element name="{name(child::m:*[1])}">
<xsl:attribute name="mathvariant" >
<xsl:value-of select="@mathvariant" />
</xsl:attribute>
<xsl:apply-templates select="child::m:*/@*" />
<xsl:apply-templates select="
child::m:*/*
| child::m:*/text()
| child::m:*/comment()
"
/>
</xsl:element>
</xsl:template>
>>>

%%%%%%%%%%%%%
\subsection{Mstyle as Parent of Text}
%%%%%%%%%%%%%

\begin{verbatim}
<mi>
<mstyle mathvariant="bold">div</mstyle>
</mi>
\end{verbatim}

into

\begin{verbatim}
<mi mathvariant="bold">div</mi>
\end{verbatim}

\<math-mstyle script\><<<
<set name="math-mstyle" >
`<open xslt script`>
`<math-mstyle templates`>
`<close xslt script`>
</set>
<xslt name="." xml="." xsl="math-mstyle" />
>>>

\<math-mstyle templates\><<<
<xsl:template match=" m:* [
not(self::m:math)
and
child::m:mstyle[
@mathvariant
and
child::text()
and
(count(child::m:*) = 0)
]
and
(count(child::m:*) = 1)
and
not(@mathvariant)
] " >
<xsl:copy>
<xsl:apply-templates select="@*" />
<xsl:apply-templates select="m:mstyle/@*" />
<xsl:apply-templates select="m:mstyle/text()" />
</xsl:copy>
</xsl:template>
>>>

%%%%%%%%%%%%%%%%%%
\section{Shared}
%%%%%%%%%%%%%%%%%%

\<open xslt script\><<<
<![CDATA[
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:h="http://www.w3.org/1999/xhtml"
xmlns:m="http://www.w3.org/1998/Math/MathML"
>
<xsl:output omit-xml-declaration = "yes" method="xml" />
>>>

\<close xslt script\><<<
<xsl:template match="*|@*|text()|comment()" >
<xsl:copy>
<xsl:apply-templates select="*|@*|text()|comment()" />
</xsl:copy>
</xsl:template>
</xsl:stylesheet>
]]>
>>>

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% END END END END END END END END END END END END END END END END
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\ifdojava
\AtEndDocument{\Needs{%
"pushd \XTPIPES || exit 1
;
jar cf tex4ht.jar *
;
popd
;
if [ ! -d \TEXMFTEXivBIN\space]; then exit 1; fi
;
mv \XTPIPES tex4ht.jar \TEXMFTEXivBIN
;
if [ ! -d \TEXMFTEXivXTPIPES\space]; then exit 1; fi
;
cp \XTPIPES xtpipes/lib/*
\TEXMFTEXivXTPIPES
;
"}}
\fi

%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\end{document}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%