converted to xslt

src/filters/rclsvg

@@ -1,10 +1,5 @@
 #!/bin/sh
-# @(#$Id: rclsvg,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 a Scalable Vector Graphics file
 #================================================================
@@ -82,56 +77,51 @@ umask 77
 # !! Leave the following line unmodified !
 #ENDRECFILTCOMMONCODE
 
-checkcmds iconv sed
+checkcmds xsltproc
 
-encoding=`sed -ne '/<?xml/s/.*encoding="\([^"]*\).*/\1/p' < $infile`
+xsltproc - $infile <<EOF
+<?xml version="1.0"?>
+<xsl:stylesheet version="1.0"
+  xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+  xmlns:svg="http://www.w3.org/2000/svg"
+  exclude-result-prefixes="svg"
+  >
 
-if test X$encoding = X ; then encoding=UTF-8;fi
+<xsl:output method="html" encoding="UTF-8"/>
 
-# We use several sed instances to make our life easier. Not good for
-# performance, and a sed guru might be able to do better.
-#
-# The first sed makes sure each tag starts on a new line
-# The second one selects the tags we're interested in.
-# The last strips the tags, leaving only text.
-#
-# The whole thing wholly ignore issues like '<' inside quoted strings.
-#
-# We could/should add code to explicitly separate title and other
-# metadata elements.
+<xsl:template match="/">
+  <html>
+  <head>
+  <xsl:apply-templates select="svg:svg/svg:title"/>
+  <xsl:apply-templates select="svg:svg/svg:desc"/>
+  </head>
+  <body>
+  <xsl:apply-templates select="//svg:text"/>
+  </body>
+  </html>
+</xsl:template>
 
-# Insert new line before each tag
-sptagonline='s/</\
-</g'
+<xsl:template match="svg:desc"> 
+  <meta>
+  <xsl:attribute name="name">keywords</xsl:attribute>
+  <xsl:attribute name="content">
+     <xsl:value-of select="."/>
+  </xsl:attribute>
+  </meta><xsl:text>
+</xsl:text>
+</xsl:template>
 
-# Select tags
-spselecttags='/<title/,/<\/title>/p
-/<desc/,/<\/desc>/p
-/<metadata/,/<\/metadata>/p
-/<text/,/<\/text>/p'
+<xsl:template match="svg:title"> 
+  <title><xsl:value-of select="."/></title><xsl:text>
+  </xsl:text>
+</xsl:template>
+	    
+<xsl:template match="svg:text"> 
+  <p><xsl:value-of select="."/></p><xsl:text>
+  </xsl:text>
+</xsl:template>
 
-# Strip tags
-spstriptags='#n
-/</{
-    :c
-     />/!{
-	N
-	b c
-     }
-     />/s/<.*>//g
-}
-/^[ 	]*$/!p'
+</xsl:stylesheet>
+EOF
 
-content=`sed -e "$sptagonline" < $infile | sed -ne "$spselecttags" | \
-    sed -ne "$spstriptags"`
-
-(echo '<html><head>'
-echo '<meta http-equiv="Content-Type" content="text/html;charset=UTF-8">'
-echo '</head><body><pre>'
-echo "$content" 
-echo '</pre></body></html>') \
-| iconv -f $encoding -t UTF-8 -c -s 
-
-
-# exit normally
 exit 0