#
# webunify converts a WNT developed web to work properly on UNIX
#
# webunify converts
#
#   all file names to lower case
#   all anchors, src="" and href="" references to lowercase
#   all 127+ characters to &#nnn; format
#   all &oumlaut; etc. escapes to &#nnn; format
#       (this is needed for some old browsers to display ISO-Latin 2
#        characters properly)
#
# AUTHOR: Peter Verhas
# DATE: January 1, 1997.
#
# Modified because it killed <pre> to <p> because it thought that
# it was <p param... >
$StartDir = shift;

if( ! -d $StartDir ){
print <<END
webunify converts a whole static NT html web to UNIX file name
compliant web site converting all filenames and references to
lover case letters as well as converting all iso-latin2 escape
characters to code &#nnn; format.

Usage:      webunify directory

where directory is the root directory of the web. (use . for the
current directory)

NOTE: When using NAME="xxx" anchors do not use xxx and XXX as
two different anchors, because this program converts both to lowercase,
and it is dangerous to health.
END
;
exit;
  }

@files = ( $StartDir );

while( $#files > -1 ){

  $File = shift @files;

  if( $File ne lc($File) ){
print "Rename $File to \l$File\n";
    rename $File,lc($File);
    }

  if( -d $File ){
    opendir(D,$File) || next;
    @nfiles = readdir(D);
    closedir D;
    @nfiles = grep(!/^[.]/ && !/^[.][.]/, @nfiles);
    for( @nfiles ){ $_ = "$File/$_"; }
    @files = ( @nfiles , @files );
    next;
    }

  if( $File !~ /\.html?$/i && $File !~ /\.asp$/i){ next; }

  open(F,"<$File") || next;
  @lines = <F>;
  close F;

print "Converting $File\n";

  for( @lines ){

#convert all uml characters to code

         s/\&ouml;/&#246;/g;
         s/\&uuml\;/\&\#252\;/g;
         s/\&oacute\;/\&\#243\;/g;
         s/\&otilde\;/\&\#245\;/g;
         s/\&uacute\;/\&\#250\;/g;
         s/\&eacute\;/\&\#233\;/g;
         s/\&aacute\;/\&\#225\;/g;
         s/\&ucirc\;/\&\#251\;/g;
         s/\&iacute\;/\&\#237\;/g;
         s/\&Ouml\;/\&\#214\;/g;
         s/\&Uuml\;/\&\#220\;/g;
         s/\&Oacute\;/\&\#211\;/g;
         s/\&Otilde\;/\&\#213\;/g;
         s/\&Uacute\;/\&\#218\;/g;
         s/\&Eacute\;/\&\#201\;/g;
         s/\&Aacute\;/\&\#193\;/g;
         s/\&Ucirc\;/\&\#219\;/g;
         s/\&Iacute\;/\&\#205\;/g;

#convert &#nnn; to code
         while( /\&\#(\d+)\;/ ){
           $c = chr($1);
           $k = $1;
           s/\&\#$k\;/$c/g;
           }
#convert tabs to space
         s/\t/\ /g;

#kill all META text
         s/\<\s*meta[^>]*\>//ig;
#kill all </p>
         s/\<\s*\/p\s*\>//ig;
#convert all <P ...> to <p>
         s/\<\s*p\s[^>]*\>/\<p\>/ig;
#delete FONT face directives
         s/\<\s*FONT\s*FACE[^>]*\>/\<FONT\>/gi;
#delete empty lines
         if( /^\n$/ ){ $_ = '' }
#delete duplicated spaces
         while( s/\s\s/\ /g ){}
#delete multiple <P>s
         while( s/<p>\s<p>/<p>/g){}

#    while( /([\x8F-\xFF])/ ){
#      $r = ord($1);
#      s#([\x8F-\xFF])#\&\#$r\;#;#
#      }

    while( /([hH][rR][eE][fF]\s*=\s*\"[^"]*[A-Z][^"]*\")/ ){
      $r = lc $1;
print "$1 -> $r\n";
      s#[hH][rR][eE][fF]\s*=\s*\"[^"]*[A-Z][^"]*\"#$r#;
      }

   while( /([sS][rR][cC]\s*=\s*\"[^"]*[A-Z][^"]*\")/ ){
      $r = lc $1;
print "$1 -> $r\n";
      s#[sS][rR][cC]\s*=\s*\"[^"]*[A-Z][^"]*\"#$r#;
      }
while( /([nN][aA][mM][eE]\s*=\s*\"[^"]*[A-Z][^"]*\")/ ){
      $r = lc $1;
print "$1 -> $r\n";
      s#[nN][aA][mM][eE]\s*=\s*\"[^"]*[A-Z][^"]*\"#$r#;
      }
    }

  open(F,">$File") || next;
  for( @lines ){ print F};
  close F;
  }


__END__