#!/usr/bin/perl
#
# Convert Damon's HTML resume to text!  Woohoo!
# Copyright (C) 1999 Damon Harper

$MARGIN=76;
@splitheads=('References');

readentities();

if($ENV{GATEWAY_INTERFACE}=~m{^CGI/}) {
  print "Content-type: text/plain\n\n";
  if($ENV{QUERY_STRING} eq 'source') {
    if(!open(IN, 'resume_damon_text.cgi')) {
      print "Couldn't open resume_damon_text.cgi!\n";
      die "Couldn't open resume_damon_text.cgi!\n";
    }
    while(<IN>) {
      print;
    }
    close(IN);
    exit;
  }
  if(!open(IN, 'index.html')) {
    print "Couldn't open index.html!\n";
    die "Couldn't open index.html!";
  }
} else {
  my($file)=$ARGV[0];
  $file ne '' or $file='index.html';
  open(IN, $file) or die "Couldn't open $file!";
}

$text=join('', <IN>);
$text=~s/<!--.*?-->//sg;

foreach(split(/\n/, $text)) {
  if(s/(.*)\<\/tr\>//i && $intr) {
    $tr.=$1;
    my($left, $right)=split(/\<\/td\>(?:\<td\>)?/i, $tr);
    if(!($left=~s/^.*?\<[ui]\>([^<]+).*/$1/si)) {
      $left='';
    }
    if($left!~/^\s*$/ || $right!~/^\s*$/) {
      my(@t)=split(/ /, $left);
      $leftwidth=max($leftwidth, map(length, @t));
      $right=totext($right);
      for($c=0; $c<@splitheads; $c++) {
	if($left eq $splitheads[$c]) {
	  push(@rows, [$left, '']);
	  push(@rows, ['', $right]);
	  last;
	}
      }
      $copyright=($right=~s/(Copyright \(C\) .*)\Z(?!\n)//s) ? $1 : '';
      push(@rows, [$left, $right]) if $c>=@splitheads;
      if($copyright) {
	$copyright=~s{You can generate a plaintext version .*}{This resume is also available online at http://www.jumeaux.bc.ca/damon/resume/\n\nThis plaintext version was generated from the canonical html file via a custom Perl script.};
        push(@rows, ['', $copyright]);
      }
    }
    $tr='';
    $intr=0;
  }
  if(s/.*\<tr[^>]*\>//i && !$intr) {
    $intr=1;
  }
  $tr.=$_ if $intr;
}
close(IN);

for($c=0; $c<@rows; $c++) {
  $_=$rows[$c];
  if(length($_->[0])>$leftwidth) {
    $_->[0]=join("\n", split(/\s+/, $_->[0]));
  }
}

for($c=0; $c<@rows; $c++) {
  my($left, $right)=@{$rows[$c]};
  if($c==$#rows) {
#    printsep();
    printblank();
  }
  printsep() if !$right;
  printsection($left, wordwrap($right, $MARGIN-$leftwidth-3));
  printblank(2) if !$right && $rows[$c+1]->[0];
  printblank() if $right;
}


sub wordwrap {
  my($text, $maxwidth, $minwidth)=@_;
  my($res);

  $minwidth=1 if !$minwidth;
  my($maxwidthminus)=$maxwidth-1;

  while(length($text)>$maxwidth) {
    while($text=~s/^(.{0,$maxwidth}(?:\n|\Z(?!\n)))//) {
      $res.=$1;
      last if $text eq '';
    }
    last if $text eq '';
    $text=~s/^(.{$minwidth,$maxwidthminus}\S)\s+//;
    if($1) {
      $res.=$1;
    } else {
      $text=~s/^(.{$minwidth,$maxwidthminus})//;
      $res.=$1;
      $res.='-';
    }
    $res.="\n" if $text ne '';
  }
  $res.=$text;

  return $res;
}

sub printsection {
  my($left, $right)=@_;
  my(@left)=split(/\n/, $left);
  my(@right)=split(/\n/, $right);
  my($c);

  for($c=0; $c<@left || $c<@right; $c++) {
    print ' ' x ($leftwidth-length($left[$c])) if $right;
    print $left[$c];
    print ' ' if $right;
    print ' ' x ($leftwidth-length($left[$c])+1) if !$right;
    print "| ";
    print $right[$c];
    print "\n";
  }
}    

sub printblank {
  my($n)=@_;
  $n=1 if !$n;
  while($n) {
    printsection(' ');
    $n--;
  }
}

sub printsep {
  print '-' x ($leftwidth+1);
  print '+';
  print '-' x ($MARGIN-$leftwidth-1);
  print "\n";
}

sub max {
  my($max);
  foreach(@_) {
    $max=$_ if $_>$max;
  }
  return $max;
}

sub totext {
  ($_)=@_;

  s/[\r\n]//g;
  s/\s+/ /g;
  s/\<P\>\s*/\n\n/ig;
  s/\<BR\>\s*/\n/ig;
  s/\s*\<LI\>/  * /i;
  s/\s*\<LI\>/\n  * /ig;
  s/\<[^>]+\>//g;
  s/^\s+//;
  s/\s+$//;
  s/&([A-Za-z]{2,6});/exists($entity{$1}) ? $entity{$1} : "&$1;"/sge;
  return $_;
}

sub readentities {
%entity=( 'copy' => '(C)',
          'trade' => '(TM)',
          'reg' => '(R)',
          'gt' => '>',
          'lt' => '<',
          'quot' => '"',
          'amp' => '&',
          'nbsp' => ' ',
          'Agrave' => 'A',
          'agrave' => 'a',
          'Aacute' => 'A',
          'aacute' => 'a',
          'Acirc' => 'A',
          'acirc' => 'a',
          'Atilde' => 'A',
          'atilde' => 'a',
          'Auml' => 'A',
          'auml' => 'a',
          'Aring' => 'A',
          'aring' => 'a',
          'AElig' => 'AE',
          'aelig' => 'ae',
          'Ccedil' => 'C',
          'ccedil' => 'c',
          'Egrave' => 'E',
          'egrave' => 'e',
          'Eacute' => 'E',
          'eacute' => 'e',
          'Ecirc' => 'E',
          'ecirc' => 'e',
          'Euml' => 'E',
          'euml' => 'e',
          'Igrave' => 'I',
          'igrave' => 'i',
          'Iacute' => 'I',
          'iacute' => 'i',
          'Icirc' => 'I',
          'icirc' => 'i',
          'Iuml' => 'I',
          'iuml' => 'i',
          'ETH' => 'Th',
          'eth' => 'th',
          'Ntilde' => 'N',
          'ntilde' => 'n',
          'Ograve' => 'O',
          'ograve' => 'o',
          'Oacute' => 'O',
          'oacute' => 'o',
          'Ocirc' => 'O',
          'ocirc' => 'o',
          'Otilde' => 'O',
          'otilde' => 'o',
          'Ouml' => 'O',
          'ouml' => 'o',
          'Oslash' => 'O',
          'oslash' => 'o',
          'Ugrave' => 'U',
          'ugrave' => 'u',
          'Uacute' => 'U',
          'uacute' => 'u',
          'Ucirc' => 'U',
          'ucirc' => 'u',
          'Uuml' => 'U',
          'uuml' => 'u',
          'Yacute' => 'Y',
          'yacute' => 'y',
          'yuml' => 'y',
          'THORN' => 'Th',
          'thorn' => 'th',
          'szlig' => 'ss'
        );
}