Xref: utzoo comp.sources.wanted:16826 alt.sources.wanted:1295
Path: utzoo!utgpu!watserv1!watmath!att!linac!pacific.mps.ohio-state.edu!zaphod.mps.ohio-state.edu!think.com!spool.mu.edu!agate!sunkist.berkeley.edu!raymond
From: raymond@math.berkeley.edu (Raymond Chen)
Newsgroups: comp.sources.wanted,alt.sources.wanted
Subject: Re: seeking PostScript->ASCII filter
Message-ID: <1991May24.195036.27185@agate.berkeley.edu>
Date: 24 May 91 19:50:36 GMT
References: <May.1.07.47.30.1991.24030@pilot.njin.net> <1991May24.173214.14681@iti.org>
Sender: root@agate.berkeley.edu (Charlie Root)
Reply-To: raymond@math.berkeley.edu (Raymond Chen)
Followup-To: comp.sources.wanted
Organization: U.C. Berkeley
Lines: 64
In-Reply-To: dhw@iti.org (David H. West)
Originator: raymond@sunkist.berkeley.edu

The only way to do it (short of writing your own Postscript interpreter)
is to customize the parser for the Postscript file itself.  Here's one
I wrote that handles groff Postscript output.

#!/usr/unsupported/perl
# Skip the leading glop

while (<> !~ /^%%Page: 1 2/) { ; }

@stack = ();
$y = 0;

main: while(<>){ chop;
while (s/\\$//) { $_ .= <>; chop; }
next if /^%/;
s/\\\(/\\050/g;
s/\\\)/\\051/g;
  while ($_) {
  s/^\s*//;#nuke leading whitespace
  if (s/^([\d.-]+)//) { # a number
  push(@stack, $1); }
  elsif (s/^\/[@_a-zA-Z-]+//) { # a literal
    push(@stack, ""); }
  elsif (s/^\(([^)]*)\)//) { # a string
    push(@stack, $1); }
  elsif (s/^(\w+)//) { # a command
    $c = $1;
    if ($c eq "C") { print &spaceout($stack[2]); }
    elsif ($c eq "E") { print "~" if $stack[1] > 0; print &fixup($stack[0]); }
    elsif ($c eq "F") { print &fixup($stack[1]); }
    elsif ($c eq "F2") { ; }
    elsif ($c eq "G") { print &spaceout($stack[1]); }
    elsif ($c eq "H") { print &fixup($stack[2]); }
    elsif ($c eq "Q") { &moveshow; }
    elsif ($c eq "R") { shift(@stack); &moveshow; }
    elsif ($c eq "S") { shift(@stack); &spaceout($stack[0]); &moveshow; }
    elsif ($c eq "T") { shift(@stack); shift(@stack); &moveshow; }
    elsif ($c eq "BP") { }
    elsif ($c eq "EP") { print "\n", "-" x 40, "\n"; }
    elsif ($c eq "SF") { }
    elsif ($c eq "end") { last main; }
  else { print STDERR "\7", join(":", @stack), " <$c>?\n"; }
  @stack = ();
  }
  elsif (s/^<(.*)>//) { # a hex string
     $c = ""; $d = $1;
     while ($d =~ s/(..)//) { $c .= sprintf("%c", hex($1)); }
    push(@stack, $c); }
  else { print STDERR "\7How to parse $_?\n"; }
  }
  }

sub moveshow {
	if ($y != $stack[2]) { $y = $stack[2]; print "\n"; }
	else { print "~"; }
	print &fixup($stack[0]); 
}

sub spaceout { @t = split(//, $_[0]); $_[0] = &fixup(join(" ", @t)); }

sub fixup { $_[0] =~ s/\.\.\.\.[.]*/\.\.\.\./; 
$_[0] =~ s/\\(\d\d\d)/sprintf("%c",oct($1))/eg;
$_[0] =~ s/\214/fi/g; $_[0] =~ s/\215/fl/g; 
$_[0]; }