Parent:
[a8e6ce]
(diff)
Child:
[8ad41f]
(diff)
Download this file
rclopxml
239 lines (201 with data), 6.2 kB
#!/bin/sh
# @(#$Id: rclopxml,v 1.3 2008-10-08 08:27:34 dockes Exp $ (C) 2004 J.F.Dockes
#================================================================
# Extract text from an openxml msword file (will be extended for spreadsheets)
# TODO: Also process docProps/core.xml for attributes, and word/endnotes.xml
#================================================================
# set variables
LANG=C ; export LANG
LC_ALL=C ; export LC_ALL
progname=rclopxml
filetype=openxml
#RECFILTCOMMONCODE
##############################################################################
# !! Leave the previous line unmodified!! Code imported from the
# recfiltcommon file
# Utility code common to all shell filters. This could be sourced at run
# time, but it's slightly more efficient to include the code in the
# filters at build time (with a sed script).
# Describe error in a way that can be interpreted by our caller
senderror()
{
echo RECFILTERROR $*
# Also alert on stderr just in case
echo ":2:$progname::: $*" 1>&2
exit 1
}
iscmd()
{
cmd=$1
case $cmd in
*/*)
if test -x $cmd -a ! -d $cmd ; then return 0; else return 1; fi ;;
*)
oldifs=$IFS; IFS=":"; set -- $PATH; IFS=$oldifs
for d in $*;do test -x $d/$cmd -a ! -d $d/$cmd && return 0;done
return 1 ;;
esac
}
checkcmds()
{
for cmd in $*;do
if iscmd $cmd
then
a=1
else
senderror HELPERNOTFOUND $cmd
fi
done
}
# show help message
if test $# -ne 1 -o "$1" = "--help"
then
echo "Convert a $filetype file to HTML text for Recoll indexing."
echo "Usage: $progname [infile]"
exit 1
fi
infile="$1"
# check the input file existence (may be '-' for stdin)
if test "X$infile" != X- -a ! -f "$infile"
then
senderror INPUTNOSUCHFILE "$infile"
fi
# protect access to our temp files and directories
umask 77
##############################################################################
# !! Leave the following line unmodified !
#ENDRECFILTCOMMONCODE
checkcmds xsltproc unzip
# We need a temporary directory
if test z"$RECOLL_TMPDIR" != z; then
ttdir=$RECOLL_TMPDIR
elif test z"$TMPDIR" != z ; then
ttdir=$TMPDIR
else
ttdir=/tmp
fi
tmpdir=$ttdir/rclopxml_tmp$$
mkdir $tmpdir || exit 1
mkdir $tmpdir/rclopxmltmp || exit 1
cleanup()
{
# Note that we're using a constant part (rclopxmltmp), that hopefully
# guarantees that we can't do big mistakes here.
rm -rf $tmpdir/rclopxmltmp
rmdir $tmpdir
}
trap cleanup EXIT HUP QUIT INT TERM
# Unzip the input file and change to the unzipped directory
unzip -q -d $tmpdir/rclopxmltmp "$infile"
cd $tmpdir/rclopxmltmp
echo '<html>
<head>'
xsltproc - docProps/core.xml <<EOF
<?xml version="1.0"?>
<xsl:stylesheet
xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0"
xmlns:cp="http://schemas.openxmlformats.org/package/2006/metadata/core-properties"
xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:dcterms="http://purl.org/dc/terms/"
xmlns:dcmitype="http://purl.org/dc/dcmitype/"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<!-- <xsl:output method="text"/> -->
<xsl:output omit-xml-declaration="yes"/>
<xsl:template match="cp:coreProperties">
<xsl:text> </xsl:text>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
<xsl:text> </xsl:text>
<xsl:apply-templates/>
</xsl:template>
<xsl:template match="dc:creator">
<meta>
<xsl:attribute name="name">
<!-- <xsl:value-of select="name()"/> pour sortir tous les meta avec
le meme nom que dans le xml (si on devenait dc-natif) -->
<xsl:text>author</xsl:text>
</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta>
<xsl:text> </xsl:text>
</xsl:template>
<xsl:template match="dcterms:modified">
<meta>
<xsl:attribute name="name">
<xsl:text>date</xsl:text>
</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta>
<xsl:text> </xsl:text>
</xsl:template>
<xsl:template match="*">
</xsl:template>
</xsl:stylesheet>
EOF
echo '</head>
<body>'
filename=''
if test -f word/document.xml ; then
filenames=word/document.xml
tagmatch="w:p"
xmlns_decls='
xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0"
xmlns:ve="http://schemas.openxmlformats.org/markup-compatibility/2006"
xmlns:o="urn:schemas-microsoft-com:office:office"
xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"
xmlns:m="http://schemas.openxmlformats.org/officeDocument/2006/math"
xmlns:v="urn:schemas-microsoft-com:vml"
xmlns:wp="http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing"
xmlns:w10="urn:schemas-microsoft-com:office:word"
xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"
xmlns:wne="http://schemas.microsoft.com/office/word/2006/wordml"
'
elif test -f xl/sharedStrings.xml ; then
filenames=xl/sharedStrings.xml
tagmatch='x:t'
xmlns_decls='
xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0"
xmlns:x="http://schemas.openxmlformats.org/spreadsheetml/2006/main"
'
elif test -f ppt/slides/slide1.xml ; then
filenames=`echo ppt/slides/slide*.xml`
tagmatch='a:t'
xmlns_decls='
xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0"
xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main"
xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"
xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main"
'
# I want to suppress text output for all except a:t, don't know how to do it
# help ! At least get rid of these:
moretemplates='
<xsl:template match="p:attrName">
</xsl:template>
'
else
# ??
exit 1
fi
for filename in $filenames;do
xsltproc - $filename <<EOF
<?xml version="1.0"?>
<xsl:stylesheet $xmlns_decls >
<xsl:output omit-xml-declaration="yes"/>
<xsl:template match="/">
<div>
<xsl:apply-templates/>
</div>
</xsl:template>
<xsl:template match="$tagmatch">
<p>
<xsl:value-of select="."/>
</p>
</xsl:template>
$moretemplates
</xsl:stylesheet>
EOF
done
echo '</html>'