136 lines
3.4 KiB
Bash
Executable File
136 lines
3.4 KiB
Bash
Executable File
#!/bin/sh
|
||
# @(#$Id: rclps,v 1.10 2007-06-08 13:51:09 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 postscript file by executing pstotext or ps2ascii.
|
||
#
|
||
# The default is to use pstotext which can deal with accents, but in a
|
||
# partially broken way (it always outputs iso8859-1, when it should use utf.
|
||
#
|
||
# OTOH, ps2ascii is much faster, comes with ghostscript, and sometimes work
|
||
# better (ie: on some openoffice output files).
|
||
#
|
||
#================================================================
|
||
|
||
# set variables
|
||
LANG=C ; export LANG
|
||
LC_ALL=C ; export LC_ALL
|
||
progname="rclps"
|
||
decoder=pstotext
|
||
#decoder=ps2ascii
|
||
filetype=postscript
|
||
|
||
|
||
#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 $decoder iconv awk
|
||
|
||
# output the result
|
||
# The strange 'BEGIN' setup is to prevent 'file' from thinking this file
|
||
# is an awk program
|
||
$decoder "$infile" |
|
||
awk 'BEGIN'\
|
||
' {
|
||
printf("<html><head><title></title>\n")
|
||
printf("<meta http-equiv=\"Content-Type\" content=\"text/html;charset=UTF-8\">\n")
|
||
printf("</head>\n<body><p>");
|
||
doescape = 1
|
||
cont = ""
|
||
}
|
||
{
|
||
$0 = cont $0
|
||
cont = ""
|
||
|
||
if ($0 == "\f") {
|
||
print "</p>\n<hr>\n\f<p>"
|
||
next
|
||
} else if ($0 ~ /<2F>$/) {
|
||
# Note : soft-hyphen is iso8859 0xad
|
||
# Break at last whitespace
|
||
match($0, "[ \t][^ \t]+$")
|
||
line = substr($0, 0, RSTART)
|
||
cont = substr($0, RSTART, RLENGTH)
|
||
$0 = line
|
||
gsub("<22>", "", cont)
|
||
}
|
||
|
||
if(doescape > 0) {
|
||
gsub(/&/, "\\&", $0)
|
||
gsub(/</, "\\<", $0)
|
||
gsub(/>/, "\\>", $0)
|
||
}
|
||
print $0 "<br>"
|
||
}
|
||
END {
|
||
print "</p></body></html>"
|
||
}' | iconv -f iso-8859-1 -t UTF-8 -c -s
|
||
|