Parent:
[8ad41f]
(diff)
Download this file
rclabw
180 lines (156 with data), 4.6 kB
#!/bin/sh
# @(#$Id: rclabw,v 1.3 2008-10-08 08:27:34 dockes Exp $ (C) 2004 J.F.Dockes
# Parts taken from Estraier:
#================================================================
# Estraier: a personal full-text search system
# Copyright (C) 2003-2004 Mikio Hirabayashi
#================================================================
#================================================================
# Extract text from an abiword file
#================================================================
# set variables
LANG=C ; export LANG
LC_ALL=C ; export LC_ALL
progname="rclabw"
filetype=abiword
#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
xsltproc --nonet --novalid - "$infile" <<EOF
<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:ab="http://www.abisource.com/awml.dtd"
exclude-result-prefixes="ab"
>
<xsl:output method="html" encoding="UTF-8"/>
<xsl:template match="/">
<html>
<head>
<xsl:apply-templates select="ab:abiword/ab:metadata"/>
</head>
<body>
<!-- This is for the older abiword format with no namespaces -->
<xsl:for-each select="abiword/section">
<xsl:apply-templates select="p"/>
</xsl:for-each>
<!-- Newer namespaced format -->
<xsl:for-each select="ab:abiword/ab:section">
<xsl:for-each select="ab:p">
<p><xsl:value-of select="."/></p><xsl:text>
</xsl:text>
</xsl:for-each>
</xsl:for-each>
</body>
</html>
</xsl:template>
<xsl:template match="p">
<p><xsl:value-of select="."/></p><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="ab:metadata">
<xsl:for-each select="ab:m">
<xsl:choose>
<xsl:when test="@key = 'dc.creator'">
<meta>
<xsl:attribute name="name">author</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:when>
<xsl:when test="@key = 'abiword.keywords'">
<meta>
<xsl:attribute name="name">keywords</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:when>
<xsl:when test="@key = 'dc.subject'">
<meta>
<xsl:attribute name="name">keywords</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:when>
<xsl:when test="@key = 'dc.description'">
<meta>
<xsl:attribute name="name">abstract</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:when>
<xsl:when test="@key = 'dc.title'">
<title><xsl:value-of select="."/></title><xsl:text>
</xsl:text>
</xsl:when>
<xsl:otherwise>
</xsl:otherwise>
</xsl:choose>
</xsl:for-each>
</xsl:template>
</xsl:stylesheet>
EOF
# exit normally
exit 0