(Client side) Perl Syntax Highlighter

After reading this thread: Feature Request: Adding Colors to Source Code i got kindof carried away and made a Javascript rip-off of Cory Johns' excellent Syntax::Highlight::Perl module.
Here it is:
To try, cut&paste into a html file, run it in IE or Mozilla and click anywhere in the perl code you see.
<html>
<head>
<style>
.quotedString
{
    color: #0000FF;
}
.comment
{
    color: #999999;
}
.operator
{
    color: #00CCCC;
}
.builtinVariable
{
    color: #CCCC00;
}
.variableSpecifier
{
    color: #FF0000;
}
.keyword
{
    color: #9900FF;
}
.builtinFunction
{
    color: #FF00FF;
}
.identifier
{
    color: #009900;
}
.number
{
    color: #9999FF;
}
</style>
<script>
var ie = document.all != null;
var moz = !ie && document.getElementById != null && document.layers ==
+ null;
function emulateHTMLModel()
{
    // copied from http://www.webfx.nu/dhtml/ieemu/htmlmodel.html

    // This function is used to generate a html string for the text pr
+operties/methods
    // It replaces '\n' with "<BR"> as well as fixes consecutive white
+ spaces
    // It also repalaces some special characters    
    function convertTextToHTML(s) {
        s = s.replace(/\&/g, "&amp;").replace(/</g, "&lt;").replace(/>
+/g, "&gt;").replace(/\n/g, "<BR>").replace(/\t/g, "    "); //tachyon
        while (/\s\s/.test(s))
            s = s.replace(/\s\s/, "&nbsp; ");
        return s.replace(/\s/g, " ");
    }


    HTMLElement.prototype.__defineSetter__("innerText", function (sTex
+t) {
        this.innerHTML = convertTextToHTML(sText);
        return sText;        
    });

    var tmpGet;
    HTMLElement.prototype.__defineGetter__("innerText", tmpGet = funct
+ion () {
        var r = this.ownerDocument.createRange();
        r.selectNodeContents(this);
        return r.toString();
    });

}

if (moz) 
    emulateHTMLModel();


// Regular Expressions largely copied from Cory Johns (darkness@yossma
+n.net) excellent Syntax::Highlight::Perl module (see http://search.cp
+an.org/~johnsca/)

var re;
var RE  = new Array;

// quoted string
re  = /('|"|`).*?\1/;
RE[0]  = new RegExp(re);

// comment
re  = /\#.*?[\r\n]+/; //tachyon
RE[1]  = new RegExp(re);
 
// operator
re = /xor|\.\.\.|and|not|\|\|\=|cmp|\>\>\=|\<\<\=|\<\=\>|\&\&\=|or|\=\
+>|\!\~|\^\=|\&\=|\|\=|\.\=|x\=|\%\=|\/\=|\*\=|\-\=|\+\=|\=\~|\*\*|\-\
+-|\.\.|\|\||\&\&|\+\+|\-\>|ne|eq|\!\=|\=\=|ge|le|gt|lt|\>\=|\<\=|\>\>
+|\<\<|\,|\=|\:|\?|\^|\||x|\%|\/|\*|\<|\&|\\|\~|\!|\>|\.|\-|\+ /;
RE[2]  = new RegExp(re);

// builtin variables
re = /\$\#?_|\$(?:\^[LAECDFHIMOPRSTWX]|[0-9&`'+*.\/|,\\";#%=\-~^:?!@\$
+<>()\[\]])|\$\#?ARGV(?:\s*\[)?|\$\#?INC\s*\[|\$(?:ENV|SIG|INC)\s*\{|\
+@(?:_|ARGV|INC)|\%(?:INC|ENV|SIG)/;
RE[3] = new RegExp(re);

// variable class specifiers
re = /(?:(?:[\@\%\*]|\$\#?)\$*)/;
RE[4] = new RegExp(re);

// keyword
re  = /(continue|foreach|require|package|scalar|format|unless|local|un
+til|while|elsif|next|last|goto|else|redo|sub|for|use|no|if|my)\b/;
RE[5]  = new RegExp(re);

// builtin function
re  = /(getprotobynumber|getprotobyname|getservbyname|gethostbyaddr|ge
+thostbyname|getservbyport|getnetbyaddr|getnetbyname|getsockname|getpe
+ername|setpriority|getprotoent|setprotoent|getpriority|endprotoent|ge
+tservent|setservent|endservent|sethostent|socketpair|getsockopt|getho
+stent|endhostent|setsockopt|setnetent|quotemeta|localtime|prototype|g
+etnetent|endnetent|rewinddir|wantarray|getpwuid|closedir|getlogin|rea
+dlink|endgrent|getgrgid|getgrnam|shmwrite|shutdown|readline|endpwent|
+setgrent|readpipe|formline|truncate|dbmclose|syswrite|setpwent|getpwn
+am|getgrent|getpwent|ucfirst|sysread|setpgrp|shmread|sysseek|sysopen|
+telldir|defined|opendir|connect|lcfirst|getppid|binmode|syscall|sprin
+tf|getpgrp|readdir|seekdir|waitpid|reverse|unshift|symlink|dbmopen|se
+mget|msgrcv|rename|listen|chroot|msgsnd|shmctl|accept|unpack|exists|f
+ileno|shmget|system|unlink|printf|gmtime|msgctl|semctl|values|rindex|
+substr|splice|length|msgget|select|socket|return|caller|delete|alarm|
+ioctl|index|undef|lstat|times|srand|chown|fcntl|close|write|umask|rmd
+ir|study|sleep|chomp|untie|print|utime|mkdir|atan2|split|crypt|flock|
+chmod|BEGIN|bless|chdir|semop|shift|reset|link|stat|chop|grep|fork|du
+mp|join|open|tell|pipe|exit|glob|warn|each|bind|sort|pack|eval|push|k
+eys|getc|kill|seek|sqrt|send|wait|rand|tied|read|time|exec|recv|eof|c
+hr|int|ord|exp|pos|pop|sin|log|abs|oct|hex|tie|cos|vec|END|ref|map|di
+e|\-C|\-b|\-S|\-u|\-t|\-p|\-l|\-d|\-f|\-g|\-s|\-z|uc|\-k|\-e|\-O|\-T|
+\-B|\-M|do|\-A|\-X|\-W|\-c|\-R|\-o|\-x|lc|\-w|\-r)\b/;
RE[6]  = new RegExp(re);

// identifier (variable, subroutine, packages)
re = /(?:(?:[A-Za-z_]|::)(?:\w|::)*)/;
RE[7] = new RegExp(re);

// number
re = /0x[\da-fA-F]+|[_.\d]+([eE][-+]?\d+)?/;
RE[8] = new RegExp(re);


var classes  = new Array("quotedString", "comment", "operator", "built
+inVariable", "variableSpecifier", "keyword", "builtinFunction", "iden
+tifier", "number");


/* This is the actual highlighting function.
 * Takes an html object as argument
 * returns nothing
 * replaces the text inside the html object with colored text using <s
+pan>'s
 * css is defined separately. See the array classes to find out the cs
+s class names.
 */
function HighlightCode(object)
{
    codeText = object.innerText; //HTML.replace(/<.*?>/g, "");
    object.innerHTML = '';
    var left;
    var match;
     var right;
    while (codeText.length > 0)
    {
        var mode = -1 ;
        var index = 999999999;
        for (var i = 0; i < RE.length; i++)
        {
            if ((codeText.match(RE[i])) && (RegExp.leftContext.length 
+< index))
            {
                left  = RegExp.leftContext;
                match = RegExp.lastMatch;
                right = RegExp.rightContext;
                index = RegExp.leftContext.length;
                 mode  = i;
            }
        }
        if (mode == -1)
        {
            object.appendChild(document.createTextNode(codeText)); //.
+replace(/\r\n/g, "\r")));
            codeText = '';
        }
        else
        {
            // append the plain text to the <code> block
            object.appendChild(document.createTextNode(left)); //.repl
+ace(/\r\n/g, "\r")));

            // create a new <span> with the current code
            var span = document.createElement("span");
            span.setAttribute("className", classes[mode]); // ie
            span.setAttribute("class", classes[mode]); //mozilla
            span.appendChild(document.createTextNode(match));
            object.appendChild(span);    

            codeText  = right;
        }
    }
}
</script>
</head>
<body>
<pre onClick ="HighlightCode(this);">
%EXPORT_TAGS = (
    crlf    => [qw(CR LF CRLF $CR $LF $CRLF)],
    all     => [@EXPORT, @EXPORT_OK],
);

BEGIN {
    sub CR   () {"\015"}
    sub LF   () {"\012"}
    sub CRLF () {"\015\012"}
}

*CR   = \CR();
*LF   = \LF();
*CRLF = \CRLF();

sub sockaddr_in {
    if (@_ == 6 && !wantarray) { # perl5.001m compat; use this && die
    my($af, $port, @quad) = @_;
    warnings::warn "6-ARG sockaddr_in call is deprecated" 
        if warnings::enabled();
    pack_sockaddr_in($port, inet_aton(join('.', @quad)));
    } elsif (wantarray) {
    croak "usage:   (port,iaddr) = sockaddr_in(sin_sv)" unless @_ == 1
+;
        unpack_sockaddr_in(@_);
    } else {
    croak "usage:   sin_sv = sockaddr_in(port,iaddr))" unless @_ == 2;
        pack_sockaddr_in(@_);
    }
}

# comment blaat
my $i = 2345;

sub sockaddr_un {
    if (wantarray) {
    croak "usage:   (filename) = sockaddr_un(sun_sv)" unless @_ == 1;
        unpack_sockaddr_un(@_);
    } else {
    croak "usage:   sun_sv = sockaddr_un(filename)" unless @_ == 1;
        pack_sockaddr_un(@_);
    }
}
</pre>
</body>
</html>
[download]
At the moment it's rather colorfull but that's just to show what it can.

Perhaps it could be discussed here if we could use it on the monestary.
If you find bugs or have suggestions for improvements, you might want to pm/mail/call/write me.
Back to Perl Monks Discussion