#!/usr/local/bin/perl5.005
# Quick-n-dirty, HIGHLY LIMITED, RTF -> HTML converter
# jason@jclark.org
#
# Only supports bold (as ), italic (as ), underline,
# and paragraphs. Corrects improperly nested tags, handles rtf's
# \plain. Ignores any RTF aside from the list above. Emits valid
# xhtml fragments (i.e., wrap in a ). Quite certainly
# contains bugs. Use as use wish, at your own risk.
foreach my $file (@ARGV) {
open(IN, "< $file");
local $/; #slurp mode
my $rtfdoc = ;
close IN;
print rtf2html($rtfdoc);
print "\n\n";
}
sub rtf2html {
# note: use of '#' as regexp delim just a workaround
# for an xemacs syntax highlight bug.
($_) = @_;
s![\n\r]!!g;
s#^\s*{##; #clobber header
s#{[^}]*?}##g;
s#}\s*$##; #... and footer
s!\\uc\w+!!g;
s!\\pard!!g;
s!\\ulnone!!g;
s!\\b0!!g; #convert closing tags
s!\\i0!!g;
s!\\u0!!g;
s!\\plain!!g;
s!\\b!!g; #convert opening tags
s!\\i!!g;
s!\\ul!!g;
s!\\u!!g;
s!\\par!!g; #convert para's
s!\\[^\s\\<]+(?=[\\ <])!!g; #remove any leftover rtf tags
#fix , wrongly nested tags
my @dom=split /(<.+?>)/;
my @elems;
foreach my $elem (@dom) {
if ($elem =~ m!|||!)
{ unshift(@elems, "$elem")}
elsif ($elem =~ m!
|
||!) {
#find opening tag
my $index;
for ($index=0;$index < @elems;$index++) {
if(closetag($elems[$index]) eq $elem) {
$elems[$index] = "";
last;
}
}
for ($index--;$index >= 0; $index--){
$elem = closetag($elems[$index]) . $elem . $elems[$index]
}
while(@elems > 0 and $elems[0] eq '') { shift @elems;}
}
elsif ($elem =~ m!!) {
$elem = "";
while(scalar(@elems)>0 and ($elems[0] =~ m!<[^/]+?>!) and ($elems[0] ne '')) {
$elem .= closetag(shift(@elems));
}
}
}
#unwind remainder of stack
foreach my $elem (@elems) {
push @dom, closetag($elem);
}
$_ = join('',@dom);
$_ = "
" . $_; #first opening para
1 while s!<(\w+)>\s*!!g ; #eliminate empty tag pairs
return $_;
}
sub closetag {
$tag = shift;
$tag =~ s{<(?!/)}{