#!/usr/bin/perl -w
#
# fhtchars  --crb3 24apr01
# file-flow filter to fix Micro$soft crap
#
if(defined($ARGV[0]) and defined($ARGV[1])){ 
  $infile = $ARGV[0];     
  $outfile = $ARGV[1]; 
}else{      
  die "Usage: htchars infile outfile"; 
} 
open(IFIL,"<$infile") or die "can't find infile $infile\n"; 
open(OFIL,">$outfile") or die "can't make outfile $outfile\n";

#
while(defined($inline=<IFIL>)){
  chomp($inline);
  $inline =~ s/\r//;				# dos -> unix eol
#
# multiline span tag? pack it together.
#
  while($inline =~ /\<span\s+/ and $inline !~ /\<span\s+.+\>{1}?/){
    $inline .= $newline = <IFIL>;
    chomp $inline;
    $inline =~ s/\r//;
    last unless defined($newline);  
  }

  $inline =~ s/\<span\s.*\>{1}?//gi;		# kill all span tags
  $inline =~ s/\<\/span\>//gi;			# kill all /span tags
  $inline =~ s/\<p\s?class\=.*\>/<p>/gi;	# chop out classes in <P> 
  $inline =~ s/\<div\s+.*\>{1}?//gi;		# chop out divs 
  $inline =~ s/\<\/div\>//gi;			# chop out divs 
  $inline =~ s/\<\/?font\s?[^>]*\>{1}?//gi;	# kill all font tags

#  $inline =~ s($oldendp)($endp);		# <o:p></o:p> -> </P>
  $inline =~ s/\<o\:p\>//gi;
  $inline =~ s/\<\/o\:p\>/<\/p>/gi;
#  $inline =~ s/\<o\:p\>\<\/o\:p\>/$endp/gi;	# <o:p></o:p> -> </P>
  $inline =~ tr/\x91\x92\x93\x94/''""/;		# fix the damn quotes
  $inline =~ s/\x85/.../g;
  $inline =~ s/\x96/--/g;
  $inline =~ s/\xBD/1\/2/g;
 
  print OFIL "$inline\n";			# still got its \n
}
close(IFIL);
close(OFIL);

__END__

^R+80h = 0x92 -> apostrophe
^S+80h = 0x93 -> start quote
^T+80h = 0x94 -> end quote
^V+80h = 0x96 -> em-dash (doubledash)
         0xBD -> 1/2

kill all span tags.
kill all class= declarations within <P> tags.
kill all font tags
kill <o:p></o:p> crap... convert to </P>.

locase all tags. (they are, but...)

kill all ^M.