#!/usr/bin/perl
#
# by Thomas A. Alspaugh
#
# Licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 2.5 License
# http://creativecommons.org/licenses/by-nc-sa/2.5/
#


use strict;
use warnings;

sub help {
  my $cmdname = $0;
  $cmdname =~ s:.*/::;
  print STDERR "$cmdname -[/hv] file ...\n";
  print STDERR "Check HTML files.\n";
  print STDERR "  -help    Show this help.\n";
  print STDERR "  -/       Require end tags or empty-element tags.\n";
  print STDERR "  -H       List the absolute destinations linked to.\n";
  print STDERR "  -h       List the relative destinations linked to.\n";
  print STDERR "  -#       List the anchors defined.\n";
  print STDERR "  -q       Quiet for correct files\n";
  print STDERR "             (otherwise each file name is echoed.)\n";
  print STDERR "  -v       Verbose output.\n";
  print STDERR "  -x       Check XHTML.\n";
  print STDERR "What is checked:\n";
  print STDERR "  elements and character entities are valid (HTML4.01)\n";
  print STDERR "  start/end elements balance\n";
  print STDERR "  no links to undefined same-file anchors\n";
  print STDERR "  no links to nonexistent local files or anchors in them\n";
  print STDERR "  no links to nonexistent local directory index files\n";
  print STDERR "    or anchors in them\n";
  print STDERR "  no img elements with nonexistent local sources\n";
  print STDERR "  no open comments\n";
  print STDERR "  no closing tags with attributes\n";
  print STDERR "  no -- in comments\n";
}

# Bad character `Ę' (looks like accented E)

if (! scalar @ARGV) {
  &help;
  exit 0;
}

my $HOUT = *STDOUT;
my $anchor = "";
my $arg = "";
my $leaveCrumbs = "y";
my $elt = "";
my $endTagsRequired = "";
my $fname = "";
my $fragment = "";
my $fDirSlash = "";
my $href = "";
my %hrefs = ();
my %ids = ();
my $lineno = 0;
my $linenoComment = 0;
my $pathname = "";
my $showAnchors = "";
my $showLocalRefs = "";
my $showNonLocalRefs = "";
my $tag = "";
my $tagName = "";
my $tagRest = "";
my $v = "";
my %charEnts = ();
my %htmlTags = ();  #  "/" signifies the tag must be closed.

&setup;

foreach $arg (@ARGV) {
  if    ($arg =~ /^-+help/) {
    &help;
    exit 0;
  }
  elsif ($arg =~ /^-+q$/) {
    $leaveCrumbs = "";
  }
  elsif ($arg =~ /^-+v$/) {
    $v = "y";
  }
  elsif ($arg =~ /^-+\/$/ || $arg =~ /^-+x$/) {
    $endTagsRequired = "y";
  }
  elsif ($arg =~ /^-+#$/) {
    $showAnchors = "y";
  }
  elsif ($arg =~ /^-+h$/) {
    $showLocalRefs = "y";
  }
  elsif ($arg =~ /^-+H$/) {
    $showNonLocalRefs = "y";
  }
  elsif ($arg =~ /^-+.*/) {
    print STDERR "Unexpected option \"$arg\".\n";
    exit -1;
  }
  else {
    $fname = $arg;
    if ($fname =~ /\//) {
      $fDirSlash = $fname;
      $fDirSlash =~ s:/[^/]+$:/:;
    }
    else {
      $fDirSlash = "";
    }
    #print STDERR "$fname:\n";
    if (open HIN, "<$fname") {
      if ($leaveCrumbs) {  print STDERR "($fname)\n";  }
      &htmlFile;
      close HIN;
    }
    else {
      print STDERR "Can't open input file \"$arg\".\n";
    }
  }
}


sub htmlFile {
  %ids = ();
  $lineno = 0;
  my $linenoTag = 0;
  #print STDERR "($fname:$lineno):\n";
  my @stack = ();
  my $stackDepth = 0;
  my $top;
  my $topTagName;
  my $topWhere;
  %hrefs = ();
  while (&readLineElideComments) {
    chomp;
    #print STDERR "($fname:$lineno):\n";
    if (/<!DOCTYPE/ && 1 == $lineno) {
      if (/>/) {
        # The entire DOCTYPE was on this line
      }
      else {
        # Find the end of the DOCTYPE
        while (&readLineElideComments) {
          chomp;
          if    (/</) {
            print $HOUT "($fname:$lineno)  '<' while looking for > to close <!DOCTYPE from line $linenoTag.\n";
          }
          if (/>/) {
            $linenoTag = 0;
            last;
          }
        }
        if ($linenoTag) {
          print $HOUT "($fname:$lineno)  EOF while looking for > to close <!DOCTYPE from line $linenoTag.\n";
          exit;
        }
      }
      next;
    }
    # check for bad characters
    &checkChars;
    # check for bad character entities
    my $charCheck = $_;
    while ($charCheck =~ /(&[a-zA-Z0-9]+;)/) {
      $charCheck =~ s/^.*(&[a-zA-Z0-9]+;)/$1/;
      my $entity = $charCheck;
      $entity    =~ s/^(&[a-zA-Z0-9]+;)(.*)/$1/;
      $charCheck =~ s/^(&[a-zA-Z0-9]+;)(.*)/$2/;
      if (!exists $charEnts{$entity}) {
        #  It's not an HTML entity.
        print $HOUT "($fname:$lineno)  Not HTML entity: `$entity'.\n";
      }
      else {
      }
    }
    while (/</) {
      #print STDERR "Looping for <'s ...\n";
      # Discard everything up to the first <.
      my $before = $_;
      $before =~ s/^([^<]*)(<.*)/$1/;
      $_      =~ s/^([^<]*)(<.*)/$2/;
      if ($before =~ />/) {
        #  Stray >.
        if ($tagName eq "style") {
          #print $HOUT "($fname:$lineno)  '>' in style.\n";
        }
        else {
          print $HOUT "($fname:$lineno)  Stray '>'.\n";
        }
      }
      $tag = $_;
      if (/>/) {
        # The entire tag was on this line
        $tag =~ s/(<[^>]*>)(.*)/$1/;
        $_   =~ s/(<[^>]*>)(.*)/$2/;
        #\/?[_A-Za-z][-_.A-Za-z0-9]*
      }
      else {
        # End of tag wasn't on this line.
        $linenoTag = $lineno;
        $_ = "";
        # Find the end of the tag
        while (&readLineElideComments) {
          chomp;
          $tag = "$tag $_";   # Perhaps should put newline rather than blank.
          if (/>/) {
            $linenoTag = 0;
            last;
          }
        }
        if ($linenoTag) {
          print $HOUT "($fname:$lineno)  EOF while looking for > to close < from line $linenoTag.\n";
          exit;
        }
        $_ = $tag;
        $tag =~ s/(<[^>]*>)(.*)/$1/;
        $_   =~ s/(<[^>]*>)(.*)/$2/;
      }
      $elt = $tag;
      $elt =~ s/='[^'>]*'/=./g;
      $elt =~ s/="[^">]*"/=../g;
      #print STDERR "for quotes  ($fname:$lineno)  $elt\n";
      if ($elt =~ /'/) {
        print $HOUT "($fname:$lineno)  Runaway single quote.\n";
      }
      if ($elt =~ /"/) {
        print $HOUT "($fname:$lineno)  Runaway double quote.\n";
      }
      $tagName = lc $tag;
      $tagRest = $tag;
      $tagName =~ s/<\/?([_A-Za-z][-_.A-Za-z0-9]*)([ \n\/>](.|\n)*)/$1/;
      $tagRest =~ s/<\/?([_A-Za-z][-_.A-Za-z0-9]*)([ \n\/>](.|\n)*)/$2/;
      $tagRest =~ s/([^>]*)\/?>.*/$1/;
      if ($v) {  print $HOUT "($fname:$lineno)    Found tag `$tag'.\n";  }
      if ($tagRest =~ /</) {
        print $HOUT "($fname:$lineno)  Found '<' in tag while looking for closing '>'.\n";
      }
      #  What kind of tag is it?
      if (!exists $htmlTags{$tagName}) {
        #  It's not an HTML tag.
        print $HOUT "($fname:$lineno)  Not HTML: `$tagName' `$tag'.\n";
      }
      elsif ($tag =~ /^<\//) {
        #  It's a closing tag.
        if ($tagRest =~ /=/) {
          print $HOUT "($fname:$lineno)  Closing tag can't have attributes:  `$tag'.\n";
        }
        #  Figure out which tag it closes.
        while ("true") {
          if (0 < scalar @stack) {
            $top = pop @stack;
            $topTagName = $top;
            $topTagName =~ s/^([^ ]+) (.*)/$1/;
            $top        =~ s/^([^ ]+) (.*)/$2/;
            $topWhere = $top;
            $topWhere =~ s/(\([^:]*:[0-9]*\)).*/$1/;
            if ($tagName eq $topTagName) {
              $stackDepth = 1 + scalar @stack;
              if ($v) {  print $HOUT "($fname:$lineno)    {$stackDepth} Matched <$topTagName> from $topWhere.\n";  }
              last;
            }
            elsif ($endTagsRequired || "/" eq $htmlTags{$topTagName}) {
              if (0 >scalar @stack) {
                my $top1 = pop @stack;
                my $top1TagName = $top1;
                $top1TagName =~ s/^([^ ]+) (.*)/$1/;
                $top1        =~ s/^([^ ]+) (.*)/$2/;
                my $top1Where = $top;
                $top1Where =~ s/(\([^:]*:[0-9]*\)).*/$1/;
                if ($tagName eq $top1TagName) {
                  $stackDepth = 1 + scalar @stack;
                  if ($v) {  print $HOUT "($fname:$lineno)    {$stackDepth} Matched <$top1TagName>  from $top1Where.\n";  }
                  print $HOUT "($fname:$lineno)  Unmatched start tag <$topTagName> from $topWhere.\n";
                }
                else {
                  print $HOUT "($fname:$lineno)  Unmatched end tag $tag (tried 2 from stack).\n";
                  $stackDepth = 1 + scalar @stack;
                  print $HOUT "($fname:$lineno)  {$stackDepth} Unmatched start tag <$topTagName> from $topWhere.\n";
                }
                last;
              } else {
                print $HOUT "($fname:$lineno)  Unmatched end tag $tag.\n";
                $stackDepth = 1 + scalar @stack;
                print $HOUT "($fname:$lineno)  {$stackDepth} Unmatched start tag <$topTagName> from $topWhere.\n";
                last;
              }
            }
            else {
              $stackDepth = 1 + scalar @stack;
              if ($v) {  print $HOUT "$topWhere    {$stackDepth} No closing tag expected for <$topTagName>.\n";  }
            }
          }
          else {
            print $HOUT "($fname:$lineno)  Unmatched end tag $tag (empty stack).\n";
            last;
          }
        }
      }
      elsif ($tag =~ /\/>/) {
        #  It's an empty-element tag.
        if ($v) {  print $HOUT "($fname:$lineno)    Found empty-element <$tagName/>.\n";  }
        &checkLinks;
      }
      else {
        #  It's a start tag.
        push @stack, "$tagName ($fname:$lineno)  $tag";
        if ($v) {  print $HOUT "($fname:$lineno)    {" . scalar @stack . "} Pushed <$tagName>  $tag.\n";  }
        &checkLinks;
      }
    }
    if (/>/) {
      #  Stray >.
      if ($tagName eq "style") {
        #print $HOUT "($fname:$lineno)  '>' in style.\n";
      }
      else {
        print $HOUT "($fname:$lineno)  Stray '>'.\n";
      }
    }
  }
  while (scalar @stack) {
    $stackDepth = scalar @stack;
    $top = pop @stack;
    $topTagName = $top;
    $topTagName =~ s/^([^ ]+) (.*)/$1/;
    $top        =~ s/^([^ ]+) (.*)/$2/;
    $topWhere = $top;
    $topWhere =~ s/(\([^:]*:[0-9]*\)).*/$1/;
    print $HOUT "($fname:$lineno)  {$stackDepth} Unmatched start tag <$topTagName> from $topWhere.\n";
  }
  my @hrefList = reverse sort keys %hrefs;
  while (@hrefList) {
    $href = pop @hrefList;
    if    ($href =~ /^#/) {
      # Local id -- can check    :)
      my $id = $href;
      $id =~ s/^#//;
      if (exists $ids{$id}) {
        if ($v) {  print $HOUT "$hrefs{$href}    Link to `$id' defined at $ids{$id}.\n";  }
      }
      elsif ($id eq "") {
        # do nothing -- the empty ID means top-of-file
      }
      else {
        print $HOUT "$hrefs{$href}  Link to undefined id `$id'.\n";
      }
    }
    else {
      die "Unexpected href '$href' -- not local.\n";
    }
  }
}


#  Checks the link attributes of a tag (in $tag)
sub checkLinks {
  #  IDs
  if ($tag =~ /.* id="([^">]+)".*/ ||
      $tag =~ /.* id='([^'>]+)'.*/ ||
      $tag =~ /^<a [^>]*name="([^">]+)".*/ ||
      $tag =~ /^<a [^>]*name='([^'>]+)'.*/) {
    my $id = $tag;
    if    ($tag =~  /.* id="([^">]+)"/) {
      $id       =~ s/.* id="([^">]+)".*/$1/;
    }
    elsif ($tag =~  /.* id='([^'>]+)'/) {
      $id       =~ s/.* id='([^'>]+)'.*/$1/;
    }
    elsif ($tag =~  /^<a [^>]*name="([^">]+)".*/) {
      $id       =~ s/^<a [^>]*name="([^">]+)".*/$1/;
    }
    elsif ($tag =~  /^<a [^>]*name='([^'>]+)'.*/) {
      $id       =~ s/^<a [^>]*name='([^'>]+)'.*/$1/;
    }
    else {
      print STDERR "Something seriously wrong:  $tag\n";
      exit
    }
    if ($showAnchors) {  print STDOUT "$id  defined ($fname:$lineno)\n";  }
    if ($v) {  print $HOUT "($fname:$lineno)    id `$id' defined.\n";  }
    if (exists $ids{$id}) {
      print $HOUT "($fname:$lineno)  id `$id' already defined $ids{$id}.\n";
    }
    $ids{$id} = "($fname:$lineno)";
    if ($id =~ /^[A-Za-z][-_.:A-Za-z0-9]*$/) {}
    else {
      print $HOUT "($fname:$lineno)  Invalid name not matching [A-Za-z][-_.:A-Za-z0-9]*:  \"$id\".\n";
    }
  }
  #  hrefs
  if    ($tag =~  /<(a|link) [^>]*href="([^">]+)"/ ||
         $tag =~  /<(a|link) [^>]*href='([^'>]+)'/) {
    $href = $tag;
    if    ($href =~  /<(a|link) [^>]*href="([^">]+)".*/) {
      $href      =~ s/<(a|link) [^>]*href="([^">]+)".*/$2/;
    }
    elsif ($href =~  /<(a|link) [^>]*href='([^'>]+)'.*/) {
      $href      =~ s/<(a|link) [^>]*href='([^'>]+)'.*/$2/;
    }
    if    ($href =~ /^#/) {
      $hrefs{$href} = "($fname:$lineno)";
      if ($v) {  print $HOUT "($fname:$lineno)    Saving '$href' to check later.\n";  }
    }
    elsif ($href =~ /^[^:]+:\/\//) {
      # URL -- can't check it.   :(
      if ($showNonLocalRefs) {
        print $HOUT "$href  linked to from ($fname:$lineno)\n";
      }
    }
    else {
      # Local file or directory  :)
      if ($showLocalRefs) {
        print $HOUT "$href  linked to from ($fname:$lineno)\n";
      }
      $pathname = $href;
      $pathname =~ s/%20/ /g;
      $pathname =~ s/#.*//;
      if ($href =~ /#/) {
        $fragment = $href;
        $fragment =~ s/.*#//;
      }
      else {
        $fragment = "";
      }
      if ($pathname =~ /^file:/) {
        $pathname =~ s/^file://;
        $pathname =~ s/^\/+/\//;
        #print STDERR "# file:pathname $pathname\n";
      }
      if ($pathname =~ /^[^\/]/ && $fDirSlash) {
        #print STDERR "# [^/]pathname $pathname\n";
        $pathname = "$fDirSlash$pathname";
      }
      if ($pathname =~ /\/$/) {
        my $dirname = $pathname;
        $pathname = $pathname . "index.html";
        if (-f "$pathname") {
          if ($v) {  print $HOUT "($fname:$lineno)    Link to existing index file `$pathname'.\n";  }
          if ($fragment) {
            &findAnchor;
          }
        }
        else {
          if (-d "$dirname") {
            if ($v) {  print $HOUT "($fname:$lineno)    Link to existing directory `$dirname'.\n";  }
            if ($fragment) {
              &findAnchor;
            }
          }
          else {
            print $HOUT "($fname:$lineno)  Link to nonexistent index file `$pathname'.\n";
          }
        }
      }
      else {
        if (-f $pathname) {
          if ($v) {  print $HOUT "($fname:$lineno)    Link to existing file `$pathname'.\n";  }
          if ($fragment) {
            &findAnchor;
          }
        }
        else {
          print $HOUT "($fname:$lineno)  Link to nonexistent file `$pathname'.\n";
        }
      }
    }
  }
  #  srcs
  if ($tag =~  /<img [^>]*src="([^"]+)"/ ||
      $tag =~  /<img [^>]*src='([^']+)'/) {
    my $src = $tag;
    if    ($src =~  /<img [^>]*src="([^"]+)".*/) {
      $src      =~ s/<img [^>]*src="([^"]+)".*/$1/;
    }
    elsif ($src =~  /<img [^>]*src='([^']+)'.*/) {
      $src      =~ s/<img [^>]*src='([^']+)'.*/$1/;
    }
    if ($src =~ /^http:\/\//) {
      # URL
      if ($v) {  print $HOUT "($fname:$lineno)    <img> sources external image file `$src'.\n";  }
    }
    else {
      if (-f "$fDirSlash$src") {
        if ($v) {  print $HOUT "($fname:$lineno)    <img> sources existing image file `$src'.\n";  }
      }
      else {
        print $HOUT "($fname:$lineno)  <img> sources nonexistent image file `$src'.\n";
      }
    }
  }
}


#  Reads an HTML file <HANCHOR> searching for an anchor definition
#  Doesn't account for comments.
sub findAnchor {
  my $saved = $_;
  open HANCHOR, "<$pathname";
  while (<HANCHOR>) {
    chomp;
    #print STDERR "### $_\n";
    if (/ (name|id)='$fragment'/ ||
        / (name|id)="$fragment"/) {
    if ($v) {  print $HOUT "($fname:$lineno)    Link to existing anchor '$pathname#$fragment'.\n";  }
      close HANCHOR;
      $_ = $saved;
      return;
    }
  }
  print $HOUT "($fname:$lineno)  Link to nonexistent anchor '$fragment' in existing file '$pathname'.\n";
  close HANCHOR;
  $_ = $saved;
}


#  Reads a line into $_ and returns it, eliding any comments
#  (including those spanning two or more lines),
#  keeps $lineno correct, doesn't chomp the newlines.
sub readLineElideComments {
  $_ = <HIN>;
  ++$lineno;
  if ($_) {
    if ($linenoComment) {
      # Looking for --> before any <!--
      my $comment = $_;
      chomp $comment;
      if    (/-->/) {
        $comment =~ s/^(.*?)-->(.*)/$1/;
        $_       =~ s/^(.*?)-->(.*)/$2/;
        $linenoComment = 0;
      }
      else {
        s/.*//;
      }
      while ($comment =~ /<!--/) {
        print $HOUT "($fname:$lineno)  <!-- in multiline comment.\n";
        $comment =~ s/<!--//;
      }
      if ($comment =~ /--/)   {  print $HOUT "($fname:$lineno)  -- in multiline comment.\n";  }
    }
    if (!$linenoComment) {
      # Looking for <!-- before any -->
      while (/<!--.*?-->/) {
        my $comment = $_;
        chomp $comment;
        my $pre = $comment;
        my $suf = $comment;
        $pre     =~ s/^(.*?)<!--(.*?)-->(.*)/$1/;
        $comment =~ s/^(.*?)<!--(.*?)-->(.*)/$2/;
        $suf     =~ s/^(.*?)<!--(.*?)-->(.*)/$3/;
        if    ($comment =~ /^-/)   {  print $HOUT "($fname:$lineno)  Bad comment:  '<!---'.\n";  }
        if    ($comment =~ /-$/)   {  print $HOUT "($fname:$lineno)  Bad comment:  '--->'.\n";  }
        while ($comment =~ /<!--/) {
          print $HOUT "($fname:$lineno)  '<!--' in comment.\n";
          $comment =~ s/<!--//;
        }
        if ($comment =~ /--/)   {  print $HOUT "($fname:$lineno)  '--' in comment.\n";  }
        if ($pre =~ /<!--/) {
          print $HOUT "($fname:$lineno)  <!-- in comment.\n";
          $pre =~ s/<!--.*//;
        }
        $_ = "$pre$suf\n";
      }
      while (/-->/)   {
        print $HOUT "($fname:$lineno)  Unmatched '-->'.\n";
        s/-->//;
      }
      if (/<!--/) {
        $linenoComment = $lineno;
        my $comment = $_;
        chomp $comment;
        $comment =~ s/^(.*?)<!--(.*)/$2/;
        $_       =~ s/^(.*?)<!--(.*)/$1/;
        if    ($comment =~ /^-/)   {  print $HOUT "($fname:$lineno)  Bad comment:  '<!---'.\n";  }
        while ($comment =~ /<!--/) {
          print $HOUT "($fname:$lineno)  '<!--' in comment.\n";
          $comment =~ s/<!--//;
        }
        if ($comment =~ /--/)   {  print $HOUT "($fname:$lineno)  '--' in comment.\n";  }
      }
    }
  }
  return $_;
}


#  Look for bad characters (quotes, accents, etc.)
sub checkChars {
    if (/Ž/) {  print $HOUT "($fname:$lineno)  Non-ASCII AE ligature.\n";  }
    if (/[įåˁĖ€]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented A.\n";  }
    if (/‚/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented C.\n";  }
   #if (/?/) {  print $HOUT "($fname:$lineno)  Non-ASCII .\n";  }
    if (/[ƒæéč]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented E.\n";  }
    if (/[ęëíė]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented I.\n";  }
    if (/[„]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented N.\n";  }
    if (/[îīņ¯Í…]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented O.\n";  }
   #if (/?/) {  print $HOUT "($fname:$lineno)  Non-ASCII .\n";  }
    if (/[ōķô†]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented U.\n";  }
   #if (/?/) {  print $HOUT "($fname:$lineno)  Non-ASCII .\n";  }
    if (/[‡‰]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented a.\n";  }
    if (/ž/) {  print $HOUT "($fname:$lineno)  Non-ASCII ae ligature.\n";  }
    if (/[ˆŒ‹Š]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented a.\n";  }
    if (//) {  print $HOUT "($fname:$lineno)  Non-ASCII accented c.\n";  }
   #if (/Ö/) {  print $HOUT "($fname:$lineno)  Non-ASCII .\n";  }
    if (/[Ž]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented e.\n";  }
   #if (/?/) {  print $HOUT "($fname:$lineno)  Non-ASCII .\n";  }
    if (/‘/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented e.\n";  }
    if (/[’”“•]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented i.\n";  }
    if (/–/) {  print $HOUT "($fname:$lineno)  Non-ASCII fi ligature.\n";  }
    if (/[—™˜ŋ›š]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented o.\n";  }
    if (/§/) {  print $HOUT "($fname:$lineno)  Non-ASCII ess-tzet.\n";  }
   #if (/?/) {  print $HOUT "($fname:$lineno)  Non-ASCII .\n";  }
   #if (/?/) {  print $HOUT "($fname:$lineno)  Non-ASCII .\n";  }
    if (/[œžŸ]/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented u.\n";  }
   #if (/?Ø/) {  print $HOUT "($fname:$lineno)  Non-ASCII accented y.\n";  }
    if (/[ŌĶÔÕ]/) {  print $HOUT "($fname:$lineno)  Non-ASCII quote.\n";  }
    if (/Ĩ/) {  print $HOUT "($fname:$lineno)  Non-ASCII bullet.\n";  }
    if (/Ņ/) {  print $HOUT "($fname:$lineno)  Non-ASCII em-dash.\n";  }
}


sub setup {
  $charEnts{"&AElig;"} = "y";
  $charEnts{"&Aacute;"} = "y";
  $charEnts{"&Acirc;"} = "y";
  $charEnts{"&Agrave;"} = "y";
  $charEnts{"&Alpha;"} = "y";
  $charEnts{"&Aring;"} = "y";
  $charEnts{"&Atilde;"} = "y";
  $charEnts{"&Auml;"} = "y";
  $charEnts{"&Beta;"} = "y";
  $charEnts{"&Ccedil;"} = "y";
  $charEnts{"&Chi;"} = "y";
  $charEnts{"&Dagger;"} = "y";
  $charEnts{"&Delta;"} = "y";
  $charEnts{"&ETH;"} = "y";
  $charEnts{"&Eacute;"} = "y";
  $charEnts{"&Ecirc;"} = "y";
  $charEnts{"&Egrave;"} = "y";
  $charEnts{"&Epsilon;"} = "y";
  $charEnts{"&Eta;"} = "y";
  $charEnts{"&Euml;"} = "y";
  $charEnts{"&Gamma;"} = "y";
  $charEnts{"&Iacute;"} = "y";
  $charEnts{"&Icirc;"} = "y";
  $charEnts{"&Igrave;"} = "y";
  $charEnts{"&Iota;"} = "y";
  $charEnts{"&Iuml;"} = "y";
  $charEnts{"&Kappa;"} = "y";
  $charEnts{"&Lambda;"} = "y";
  $charEnts{"&Mu;"} = "y";
  $charEnts{"&Ntilde;"} = "y";
  $charEnts{"&Nu;"} = "y";
  $charEnts{"&OElig;"} = "y";
  $charEnts{"&Oacute;"} = "y";
  $charEnts{"&Ocirc;"} = "y";
  $charEnts{"&Ograve;"} = "y";
  $charEnts{"&Omega;"} = "y";
  $charEnts{"&Omicron;"} = "y";
  $charEnts{"&Oslash;"} = "y";
  $charEnts{"&Otilde;"} = "y";
  $charEnts{"&Ouml;"} = "y";
  $charEnts{"&Phi;"} = "y";
  $charEnts{"&Pi;"} = "y";
  $charEnts{"&Prime;"} = "y";
  $charEnts{"&Psi;"} = "y";
  $charEnts{"&Rho;"} = "y";
  $charEnts{"&Scaron;"} = "y";
  $charEnts{"&Sigma;"} = "y";
  $charEnts{"&THORN;"} = "y";
  $charEnts{"&Tau;"} = "y";
  $charEnts{"&Theta;"} = "y";
  $charEnts{"&Uacute;"} = "y";
  $charEnts{"&Ucirc;"} = "y";
  $charEnts{"&Ugrave;"} = "y";
  $charEnts{"&Upsilon;"} = "y";
  $charEnts{"&Uuml;"} = "y";
  $charEnts{"&Xi;"} = "y";
  $charEnts{"&Yacute;"} = "y";
  $charEnts{"&Yuml;"} = "y";
  $charEnts{"&Zeta;"} = "y";
  $charEnts{"&aacute;"} = "y";
  $charEnts{"&acirc;"} = "y";
  $charEnts{"&acute;"} = "y";
  $charEnts{"&aelig;"} = "y";
  $charEnts{"&agrave;"} = "y";
  $charEnts{"&alefsym;"} = "y";
  $charEnts{"&alpha;"} = "y";
  $charEnts{"&amp;"} = "y";
  $charEnts{"&and;"} = "y";
  $charEnts{"&ang;"} = "y";
  $charEnts{"&aring;"} = "y";
  $charEnts{"&asymp;"} = "y";
  $charEnts{"&atilde;"} = "y";
  $charEnts{"&auml;"} = "y";
  $charEnts{"&bdquo;"} = "y";
  $charEnts{"&beta;"} = "y";
  $charEnts{"&brvbar;"} = "y";
  $charEnts{"&bull;"} = "y";
  $charEnts{"&cap;"} = "y";
  $charEnts{"&ccedil;"} = "y";
  $charEnts{"&cedil;"} = "y";
  $charEnts{"&cent;"} = "y";
  $charEnts{"&chi;"} = "y";
  $charEnts{"&circ;"} = "y";
  $charEnts{"&clubs;"} = "y";
  $charEnts{"&cong;"} = "y";
  $charEnts{"&copy;"} = "y";
  $charEnts{"&crarr;"} = "y";
  $charEnts{"&cup;"} = "y";
  $charEnts{"&curren;"} = "y";
  $charEnts{"&dArr;"} = "y";
  $charEnts{"&dagger;"} = "y";
  $charEnts{"&darr;"} = "y";
  $charEnts{"&deg;"} = "y";
  $charEnts{"&delta;"} = "y";
  $charEnts{"&diams;"} = "y";
  $charEnts{"&divide;"} = "y";
  $charEnts{"&eacute;"} = "y";
  $charEnts{"&ecirc;"} = "y";
  $charEnts{"&egrave;"} = "y";
  $charEnts{"&empty;"} = "y";
  $charEnts{"&emsp;"} = "y";
  $charEnts{"&ensp;"} = "y";
  $charEnts{"&epsilon;"} = "y";
  $charEnts{"&equiv;"} = "y";
  $charEnts{"&eta;"} = "y";
  $charEnts{"&eth;"} = "y";
  $charEnts{"&euml;"} = "y";
  $charEnts{"&euro;"} = "y";
  $charEnts{"&exist;"} = "y";
  $charEnts{"&fnof;"} = "y";
  $charEnts{"&forall;"} = "y";
  $charEnts{"&frac12;"} = "y";
  $charEnts{"&frac14;"} = "y";
  $charEnts{"&frac34;"} = "y";
  $charEnts{"&frasl;"} = "y";
  $charEnts{"&gamma;"} = "y";
  $charEnts{"&ge;"} = "y";
  $charEnts{"&gt;"} = "y";
  $charEnts{"&hArr;"} = "y";
  $charEnts{"&harr;"} = "y";
  $charEnts{"&hearts;"} = "y";
  $charEnts{"&hellip;"} = "y";
  $charEnts{"&iacute;"} = "y";
  $charEnts{"&icirc;"} = "y";
  $charEnts{"&iexcl;"} = "y";
  $charEnts{"&igrave;"} = "y";
  $charEnts{"&image;"} = "y";
  $charEnts{"&infin;"} = "y";
  $charEnts{"&int;"} = "y";
  $charEnts{"&iota;"} = "y";
  $charEnts{"&iquest;"} = "y";
  $charEnts{"&isin;"} = "y";
  $charEnts{"&iuml;"} = "y";
  $charEnts{"&kappa;"} = "y";
  $charEnts{"&lArr;"} = "y";
  $charEnts{"&lambda;"} = "y";
  $charEnts{"&lang;"} = "y";
  $charEnts{"&laquo;"} = "y";
  $charEnts{"&larr;"} = "y";
  $charEnts{"&lceil;"} = "y";
  $charEnts{"&ldquo;"} = "y";
  $charEnts{"&le;"} = "y";
  $charEnts{"&lfloor;"} = "y";
  $charEnts{"&lowast;"} = "y";
  $charEnts{"&loz;"} = "y";
  $charEnts{"&lrm;"} = "y";
  $charEnts{"&lsaquo;"} = "y";
  $charEnts{"&lsquo;"} = "y";
  $charEnts{"&lt;"} = "y";
  $charEnts{"&macr;"} = "y";
  $charEnts{"&mdash;"} = "y";
  $charEnts{"&micro;"} = "y";
  $charEnts{"&middot;"} = "y";
  $charEnts{"&minus;"} = "y";
  $charEnts{"&mu;"} = "y";
  $charEnts{"&nabla;"} = "y";
  $charEnts{"&nbsp;"} = "y";
  $charEnts{"&ndash;"} = "y";
  $charEnts{"&ne;"} = "y";
  $charEnts{"&ni;"} = "y";
  $charEnts{"&not;"} = "y";
  $charEnts{"&notin;"} = "y";
  $charEnts{"&nsub;"} = "y";
  $charEnts{"&ntilde;"} = "y";
  $charEnts{"&nu;"} = "y";
  $charEnts{"&oacute;"} = "y";
  $charEnts{"&ocirc;"} = "y";
  $charEnts{"&oelig;"} = "y";
  $charEnts{"&ograve;"} = "y";
  $charEnts{"&oline;"} = "y";
  $charEnts{"&omega;"} = "y";
  $charEnts{"&omicron;"} = "y";
  $charEnts{"&oplus;"} = "y";
  $charEnts{"&or;"} = "y";
  $charEnts{"&ordf;"} = "y";
  $charEnts{"&ordm;"} = "y";
  $charEnts{"&oslash;"} = "y";
  $charEnts{"&otilde;"} = "y";
  $charEnts{"&otimes;"} = "y";
  $charEnts{"&ouml;"} = "y";
  $charEnts{"&para;"} = "y";
  $charEnts{"&part;"} = "y";
  $charEnts{"&permil;"} = "y";
  $charEnts{"&perp;"} = "y";
  $charEnts{"&phi;"} = "y";
  $charEnts{"&pi;"} = "y";
  $charEnts{"&piv;"} = "y";
  $charEnts{"&plusmn;"} = "y";
  $charEnts{"&pound;"} = "y";
  $charEnts{"&prime;"} = "y";
  $charEnts{"&prod;"} = "y";
  $charEnts{"&prop;"} = "y";
  $charEnts{"&psi;"} = "y";
  $charEnts{"&quot;"} = "y";
  $charEnts{"&rArr;"} = "y";
  $charEnts{"&radic;"} = "y";
  $charEnts{"&rang;"} = "y";
  $charEnts{"&raquo;"} = "y";
  $charEnts{"&rarr;"} = "y";
  $charEnts{"&rceil;"} = "y";
  $charEnts{"&rdquo;"} = "y";
  $charEnts{"&real;"} = "y";
  $charEnts{"&reg;"} = "y";
  $charEnts{"&rfloor;"} = "y";
  $charEnts{"&rho;"} = "y";
  $charEnts{"&rlm;"} = "y";
  $charEnts{"&rsaquo;"} = "y";
  $charEnts{"&rsquo;"} = "y";
  $charEnts{"&sbquo;"} = "y";
  $charEnts{"&scaron;"} = "y";
  $charEnts{"&sdot;"} = "y";
  $charEnts{"&sect;"} = "y";
  $charEnts{"&shy;"} = "y";
  $charEnts{"&sigma;"} = "y";
  $charEnts{"&sigmaf;"} = "y";
  $charEnts{"&sim;"} = "y";
  $charEnts{"&spades;"} = "y";
  $charEnts{"&sub;"} = "y";
  $charEnts{"&sube;"} = "y";
  $charEnts{"&sum;"} = "y";
  $charEnts{"&sup1;"} = "y";
  $charEnts{"&sup2;"} = "y";
  $charEnts{"&sup3;"} = "y";
  $charEnts{"&sup;"} = "y";
  $charEnts{"&supe;"} = "y";
  $charEnts{"&szlig;"} = "y";
  $charEnts{"&tau;"} = "y";
  $charEnts{"&there4;"} = "y";
  $charEnts{"&theta;"} = "y";
  $charEnts{"&thetasym;"} = "y";
  $charEnts{"&thinsp;"} = "y";
  $charEnts{"&thorn;"} = "y";
  $charEnts{"&tilde;"} = "y";
  $charEnts{"&times;"} = "y";
  $charEnts{"&trade;"} = "y";
  $charEnts{"&uArr;"} = "y";
  $charEnts{"&uacute;"} = "y";
  $charEnts{"&uarr;"} = "y";
  $charEnts{"&ucirc;"} = "y";
  $charEnts{"&ugrave;"} = "y";
  $charEnts{"&uml;"} = "y";
  $charEnts{"&upsih;"} = "y";
  $charEnts{"&upsilon;"} = "y";
  $charEnts{"&uuml;"} = "y";
  $charEnts{"&weierp;"} = "y";
  $charEnts{"&xi;"} = "y";
  $charEnts{"&yacute;"} = "y";
  $charEnts{"&yen;"} = "y";
  $charEnts{"&yuml;"} = "y";
  $charEnts{"&zeta;"} = "y";
  $charEnts{"&zwj;"} = "y";
  $charEnts{"&zwnj;"} = "y";

  # / end tag required
  # ? end tag optional
  # - end tag forbidden
  $htmlTags{"a"}          = "/";
  $htmlTags{"area"}       = "-";
  $htmlTags{"b"}          = "/";
  $htmlTags{"blockquote"} = "/";
  $htmlTags{"body"}       = "/";
  $htmlTags{"br"}         = "-";
  $htmlTags{"caption"}    = "/";
  $htmlTags{"center"}     = "/";
  $htmlTags{"cite"}       = "/";
  $htmlTags{"code"}       = "/";
  $htmlTags{"col"}        = "-";
  $htmlTags{"colgroup"}   = "?";
  $htmlTags{"dd"}         = "?";
  $htmlTags{"del"}        = "/";
  $htmlTags{"dfn"}        = "/";
  $htmlTags{"div"}        = "/";
  $htmlTags{"dl"}         = "/";
  $htmlTags{"dt"}         = "?";
  $htmlTags{"em"}         = "?";
  $htmlTags{"font"}       = "/";
  $htmlTags{"frame"}      = "-";
  $htmlTags{"frameset"}   = "/";
  $htmlTags{"h1"}         = "/";
  $htmlTags{"h2"}         = "/";
  $htmlTags{"h3"}         = "/";
  $htmlTags{"h4"}         = "/";
  $htmlTags{"h5"}         = "/";
  $htmlTags{"h6"}         = "/";
  $htmlTags{"head"}       = "/";
  $htmlTags{"hr"}         = "-";
  $htmlTags{"html"}       = "/";
  $htmlTags{"i"}          = "/";
  $htmlTags{"img"}        = "-";
  $htmlTags{"ins"}        = "/";
  $htmlTags{"li"}         = "?";
  $htmlTags{"link"}       = "-";
  $htmlTags{"map"}        = "/";
  $htmlTags{"meta"}       = "-";
  $htmlTags{"noframes"}   = "/";
  $htmlTags{"noscript"}   = "/";
  $htmlTags{"object"}     = "/";
  $htmlTags{"ol"}         = "/";
  $htmlTags{"p"}          = "?";
  $htmlTags{"pre"}        = "/";
  $htmlTags{"q"}          = "?";
  $htmlTags{"samp"}       = "/";
  $htmlTags{"script"}     = "/";
  $htmlTags{"small"}      = "/";
  $htmlTags{"span"}       = "/";
  $htmlTags{"strike"}     = "/";
  $htmlTags{"strong"}     = "/";
  $htmlTags{"style"}      = "/";
  $htmlTags{"sub"}        = "/";
  $htmlTags{"sup"}        = "/";
  $htmlTags{"table"}      = "/";
  $htmlTags{"tbody"}      = "?";
  $htmlTags{"thead"}      = "?";
  $htmlTags{"td"}         = "?";
  $htmlTags{"th"}         = "?";
  $htmlTags{"title"}      = "/";
  $htmlTags{"tr"}         = "?";
  $htmlTags{"tt"}         = "/";
  $htmlTags{"ul"}         = "/";
  $htmlTags{"u"}          = "/";
}