derobins/wmd

Stack Overflow branch of WMD (No longer maintained - Please use PageDown instead)

wmd's Issues

Multiple textarea on same page

How to use wmd for multiple textarea on same page?

Toolbar style issue with IE8

Hi,

I have found a toolbar style issue with IE8. It may be just me I am not 100% sure.

After a bit of debugging I found that wmd.Util.oldIE was true becuase nav.userAgent contained a bit of text which said (compatibility MSIE 6....

So I added the following:
if (nav.userAgent.indexOf("MSIE 8.")) {
wmd.Util.oldIE = false;
wmd.Util.newIE = false;
}

This didnt really fix anything, so added wmd.Util.IE = false; just to see what happened.

Again nothing happened.

So, after a bit of searching I found that there were several lines which rechecked if nav.userAgent.indexOf("MSIE") != -1 rather than using the wmd.Util.IE value.

After after replacing these the toolbar style fixed.

Sorry just found that seeing the IE oldIE and newIE to false breaks a lot of the mouse dblclick and highlighting function.

So although the toolbar looks good some thing else is broken.

Cheers

Luke

JS does't works whith ajax load content

Hi,

I include JS of wmd (https://raw.githubusercontent.com/derobins/wmd/master/wmd.js), and I load content of my page with ajax call. I append this content to body.
There is a #wmd-editor and #wmd-input div, but when I press keydown, there is no preview, nothing happen.

How can I do in this case?

thx

quick and dirty js implementation of smartypants and javascript function call for href links

//
// showdown.js -- A javascript port of Markdown.
//
// Copyright (c) 2007 John Fraser.
//
// Original Markdown Copyright (c) 2004-2005 John Gruber
// http://daringfireball.net/projects/markdown/
//
// The full source distribution is at:
//
// A A L
// T C A
// T K B
//
// http://www.attacklab.net/
//

//
// Wherever possible, Showdown is a straight, line-by-line port
// of the Perl version of Markdown.
//
// This is not a normal parser design; it's basically just a
// series of string substitutions. It's hard to read and
// maintain this way, but keeping Showdown close to the original
// design makes it easier to port new features.
//
// More importantly, Showdown behaves like markdown.pl in most
// edge cases. So web applications can do client-side preview
// in Javascript, and then build identical HTML on the server.
//
// This port needs the new RegExp functionality of ECMA 262,
// 3rd Edition (i.e. Javascript 1.5). Most modern web browsers
// should do fine. Even with the new regular expression features,
// We do a lot of work to emulate Perl's regex functionality.
// The tricky changes in this file mostly have the "attacklab:"
// label. Major or self-explanatory changes don't.
//
// Smart diff tools like Araxis Merge will be able to match up
// this file with markdown.pl in a useful way. A little tweaking
// helps: in a copy of markdown.pl, replace "#" with "//" and
// replace "$text" with "text". Be sure to ignore whitespace
// and line endings.
//

//
// Showdown usage:
//
// var text = "Markdown rocks.";
//
// var converter = new Attacklab.showdown.converter();
// var html = converter.makeHtml(text);
//
// alert(html);
//
// Note: move the sample code to the bottom of this
// file before uncommenting it.
//

//
// Attacklab namespace
//
var Attacklab = Attacklab || {}

//
// Showdown namespace
//
Attacklab.showdown = Attacklab.showdown || {}

//
// converter
//
// Wraps all "globals" so that the only thing
// exposed is makeHtml().
//
Attacklab.showdown.converter = function() {

//
// Globals:
//

// Global hashes, used by various utility routines
var g_urls;
var g_titles;
var g_html_blocks;

// Used to track when we're inside an ordered or unordered list
// (see _ProcessListItems() for details):
var g_list_level = 0;

this.makeHtml = function(text) {
//
// Main function. The order in which other subs are called here is
// essential. Link and image substitutions need to happen before
// _EscapeSpecialCharsWithinTagAttributes(), so that any *'s or _'s in the
// and tags get encoded.
//
//alert("gogo");
// Clear the global hashes. If we don't clear these, you get conflicts
// from other articles when generating a page which contains more than
// one article (e.g. an index page that shows the N most recent
// articles):
g_urls = new Array();
g_titles = new Array();
g_html_blocks = new Array();
g_repl_blocks = new Array();
// attacklab: Replace ~ with T
// This lets us use tilde as an escape char to avoid md5 hashes
// The choice of character is arbitray; anything that isn't
// magic in Markdown will work.
text = text.replace(//g,"~T");

// attacklab: Replace $ with ~D
// RegExp interprets $ as a special character
// when it's in a replacement string
text = text.replace(/\$/g,"~D");

// Standardize line endings
text = text.replace(/\r\n/g,"\n"); // DOS to Unix
text = text.replace(/\r/g,"\n"); // Mac to Unix

// Make sure text begins and ends with a couple of newlines:
text = "\n\n" + text + "\n\n";

// Convert all tabs to spaces.
text = _Detab(text);

// Strip any lines consisting only of spaces and tabs.
// This makes subsequent regexen easier to write, because we can
// match consecutive blank lines with /\n+/ instead of something
// contorted like /[ \t]*\n+/ .
text = text.replace(/^[ \t]+$/mg,"");


text = _StripLinkDefinitions(text);
// Turn block-level HTML blocks into hash entries

text = _HashHTMLBlocks(text);

// Strip link definitions, store in hashes.


text = _RunBlockGamut(text);

text = _UnescapeSpecialChars(text);

// attacklab: Restore dollar signs
text = text.replace(/~D/g,"$$");

// attacklab: Restore tildes
text = text.replace(/~T/g,"~");
//text = HTMLtoDOMSmartyPants("<div>" + text + "</div>");
text = _RunSmartyPants(text);
return text;

}

var _StripLinkDefinitions = function(text) {
//
// Strips link definitions from text, stores the URLs and titles in
// hash references.
//

// Link defs are in the form: ^[id]: url "optional title"

/*
    var text = text.replace(/
            ^[ ]{0,3}\[(.+)\]:  // id = $1  attacklab: g_tab_width - 1
              [ \t]*
              \n?               // maybe *one* newline
              [ \t]*
            <?(\S+?)>?          // url = $2
              [ \t]*
              \n?               // maybe one newline
              [ \t]*
            (?:
              (\n*)             // any lines skipped = $3 attacklab: lookbehind removed
              ["(]
              (.+?)             // title = $4
              [")]
              [ \t]*
            )?                  // title is optional
            (?:\n+|$)
          /gm,
          function(){...});
*/
var text = text.replace(/^[ ]{0,3}\[(.+)\]:[ \t]*\n?[ \t]*<?(\S+?)>?[ \t]*\n?[ \t]*(?:(\n*)["(](.+?)[")][ \t]*)?(?:\n+)/gm,
    function (wholeMatch,m1,m2,m3,m4) {
        m1 = m1.toLowerCase();
        g_urls[m1] = _EncodeAmpsAndAngles(m2);  // Link IDs are case-insensitive
        if (m3) {
            // Oops, found blank lines, so it's not a title.
            // Put back the parenthetical statement we stole.
            return m3+m4;
        } else if (m4) {
            g_titles[m1] = m4.replace(/"/g,"&quot;");
        }

        // Completely remove the definition from the text
        return "";
    }
);

return text;

}

var _HashExcludeSmartyPants = function(text)
{

}

var _HashHTMLBlocks = function(text) {
// attacklab: Double up blank lines to reduce lookaround
text = text.replace(/\n/g,"\n\n");

// Hashify HTML blocks:
// We only want to do this for block-level HTML tags, such as headers,
// lists, and tables. That's because we still want to wrap <p>s around
// "paragraphs" that are wrapped in non-block-level tags, such as anchors,
// phrase emphasis, and spans. The list of tags we're looking for is
// hard-coded:
var block_tags_a = "p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|script|noscript|form|fieldset|iframe|math|ins|del"
var block_tags_b = "p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|script|noscript|form|fieldset|iframe|math"

// First, look for nested blocks, e.g.:
//   <div>
//     <div>
//     tags for inner block must be indented.
//     </div>
//   </div>
//
// The outermost tags must start at the left margin for this to match, and
// the inner nested divs must be indented.
// We need to do this before the next, more liberal match, because the next
// match will start at the first `<div>` and stop at the first `</div>`.

// attacklab: This regex can be expensive when it fails.
/*
    var text = text.replace(/
    (                       // save in $1
        ^                   // start of line  (with /m)
        <($block_tags_a)    // start tag = $2
        \b                  // word break
                            // attacklab: hack around khtml/pcre bug...
        [^\r]*?\n           // any number of lines, minimally matching
        </\2>               // the matching end tag
        [ \t]*              // trailing spaces/tabs
        (?=\n+)             // followed by a newline
    )                       // attacklab: there are sentinel newlines at end of document
    /gm,function(){...}};
*/
text = text.replace(/^(<(code|kbd|pre|script|noscript|iframe|math|ins|del)\b[^\r]*?\n<\/\2>[ \t]*(?=\n+))/gm,hashElement);

/*process text parts correctly with markdown*/  
text = text.replace(/^(<(p|div|h[1-6]|blockquote|table|dl|ol|ul|form|fieldset)\b[^\r]*?\n<\/\2>[ \t]*(?=\n+))/gm,hashAndCashElement);

//
// Now match more liberally, simply from `\n<tag>` to `</tag>\n`
//

/*
    var text = text.replace(/
    (                       // save in $1
        ^                   // start of line  (with /m)
        <($block_tags_b)    // start tag = $2
        \b                  // word break
                            // attacklab: hack around khtml/pcre bug...
        [^\r]*?             // any number of lines, minimally matching
        .*</\2>             // the matching end tag
        [ \t]*              // trailing spaces/tabs
        (?=\n+)             // followed by a newline
    )                       // attacklab: there are sentinel newlines at end of document
    /gm,function(){...}};
*/
text = text.replace(/^(<(p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|script|noscript|form|fieldset|iframe|math)\b[^\r]*?.*<\/\2>[ \t]*(?=\n+)\n)/gm,hashElement);

// Special case just for <hr />. It was easier to make a special case than
// to make the other regex more complicated.  

/*
    text = text.replace(/
    (                       // save in $1
        \n\n                // Starting after a blank line
        [ ]{0,3}
        (<(hr)              // start tag = $2
        \b                  // word break
        ([^<>])*?           // 
        \/?>)               // the matching end tag
        [ \t]*
        (?=\n{2,})          // followed by a blank line
    )
    /g,hashElement);
*/
text = text.replace(/(\n[ ]{0,3}(<(hr)\b([^<>])*?\/?>)[ \t]*(?=\n{2,}))/g,hashElement);

// Special case just for <img />. It was easier to make a special case than
// to make the other regex more complicated.  

/*
    text = text.replace(/
    (                       // save in $1
        \n\n                // Starting after a blank line
        [ ]{0,3}
        (<(img)             // start tag = $2
        \b                  // word break
        ([^<>])*?           // 
        \/?>)               // the matching end tag
        [ \t]*
        (?=\n{2,})          // followed by a blank line
    )
    /g,hashElement);
*/
//text = text.replace(/(([ \t\n])*<(img)\b([^<>])*?\/?>([ \t\n])*)/g,hashElement);


// Special case for standalone HTML comments:

/*
    text = text.replace(/
    (                       // save in $1
        \n\n                // Starting after a blank line
        [ ]{0,3}            // attacklab: g_tab_width - 1
        <!
        (--[^\r]*?--\s*)+
        >
        [ \t]*
        (?=\n{2,})          // followed by a blank line
    )
    /g,hashElement);
*/
text = text.replace(/(\n\n[ ]{0,3}<!(--[^\r]*?--\s*)+>[ \t]*(?=\n{2,}))/g,hashElement);

// PHP and ASP-style processor instructions (<?...?> and <%...%>)

/*
    text = text.replace(/
    (?:
        \n\n                // Starting after a blank line
    )
    (                       // save in $1
        [ ]{0,3}            // attacklab: g_tab_width - 1
        (?:
            <([?%])         // $2
            [^\r]*?
            \2>
        )
        [ \t]*
        (?=\n{2,})          // followed by a blank line
    )
    /g,hashElement);
*/
text = text.replace(/(?:\n\n)([ ]{0,3}(?:<([?%])[^\r]*?\2>)[ \t]*(?=\n{2,}))/g,hashElement);

// attacklab: Undo double lines (see comment at top of this function)
text = text.replace(/\n\n/g,"\n");
return text;

}

var hashElement = function(wholeMatch,m1) {

var blockText = m1;

// Undo double lines
blockText = blockText.replace(/\n\n/g,"\n");
blockText = blockText.replace(/^\n/,"");
// strip trailing blank lines
blockText = blockText.replace(/\n+$/g,"");


// Replace the element text with a marker ("~KxK" where x is its key)
blockText = "\n\n~K" + (g_html_blocks.push(blockText)-1) + "K\n\n";

return blockText;

};

var hashAndCashElement = function(wholeMatch,m1) {

var blockText = m1;

// Undo double lines
blockText = blockText.replace(/\n\n/g,"\n");
blockText = blockText.replace(/^\n/,"");
// strip trailing blank lines
blockText = blockText.replace(/\n+$/g,"");

/*test block level transformations*/
/*blockText = _StripLinkDefinitions(blockText);*/

//text = _FormParagraphs(text);
blockText = _RunBlockGamut(blockText);



blockText = _UnescapeSpecialChars(blockText);


/*test end*/



// Replace the element text with a marker ("~KxK" where x is its key)
blockText = "\n\n~K" + (g_html_blocks.push(blockText)-1) + "K\n\n";

return blockText;

};

var _RunBlockGamut = function(text) {
//
// These are all the transformations that form block-level
// tags like paragraphs, headers, and list items.
//
text = _DoHeaders(text);

// Do Horizontal Rules:
var key = hashBlock("<hr />");
text = text.replace(/^[ ]{0,2}([ ]?\*[ ]?){3,}[ \t]*$/gm,key);
text = text.replace(/^[ ]{0,2}([ ]?-[ ]?){3,}[ \t]*$/gm,key);
text = text.replace(/^[ ]{0,2}([ ]?_[ ]?){3,}[ \t]*$/gm,key);

text = _DoLists(text);
text = _DoCodeBlocks(text);
text = _DoBlockQuotes(text);

// We already ran _HashHTMLBlocks() before, in Markdown(), but that
// was to escape raw HTML in the original Markdown source. This time,
// we're escaping the markup we've just created, so that we don't wrap
// <p> tags around block-level tags.
text = _HashHTMLBlocks(text);
text = _FormParagraphs(text);

return text;

}

var _RunSpanGamut = function(text) {
//
// These are all the transformations that occur within block-level
// tags like paragraphs, headers, and list items.
//

text = _DoCodeSpans(text);
text = _EscapeSpecialCharsWithinTagAttributes(text);
text = _EncodeBackslashEscapes(text);


// Process anchor and image tags. Images must come first,
// because ![foo][f] looks like an anchor.
text = _DoImages(text);
text = _DoAnchors(text);

// Make links out of things like `<http://example.com/>`
// Must come after _DoAnchors(), because you can use < and >
// delimiters in inline links like [this](<url>).
text = _DoAutoLinks(text);
text = _EncodeAmpsAndAngles(text);
text = _DoItalicsAndBold(text);

// Do hard breaks:
text = text.replace(/  +\n/g," <br />\n");

return text;

}

var _EscapeSpecialCharsWithinTagAttributes = function(text) {
//
// Within tags -- meaning between < and > -- encode [\ ` * _] so they
// don't conflict with their use in Markdown for code, italics and strong.
//

// Build a regex to find HTML tags and comments.  See Friedl's 
// "Mastering Regular Expressions", 2nd Ed., pp. 200-201.
var regex = /(<[a-z\/!$]("[^"]*"|'[^']*'|[^'">])*>|<!(--.*?--\s*)+>)/gi;

text = text.replace(regex, function(wholeMatch) {
    var tag = wholeMatch.replace(/(.)<\/?code>(?=.)/g,"$1`");
    tag = escapeCharacters(tag,"\\`*_");
    return tag;
});

return text;

}

var _DoAnchors = function(text) {
//
// Turn Markdown link shortcuts into XHTML tags.
//
//
// First, handle reference-style links: [link text] [id]
//

/*
    text = text.replace(/
    (                           // wrap whole match in $1
        \[
        (
            (?:
                \[[^\]]*\]      // allow brackets nested one level
                |
                [^\[]           // or anything else
            )*
        )
        \]

        [ ]?                    // one optional space
        (?:\n[ ]*)?             // one optional newline followed by spaces

        \[
        (.*?)                   // id = $3
        \]
    )()()()()                   // pad remaining backreferences
    /g,_DoAnchors_callback);
*/
text = text.replace(/(\[((?:\[[^\]]*\]|[^\[\]])*)\][ ]?(?:\n[ ]*)?\[(.*?)\])()()()()/g,writeAnchorTag);

//
// Next, inline-style links: [link text](url "optional title")
//

/*
    text = text.replace(/
        (                       // wrap whole match in $1
            \[
            (
                (?:
                    \[[^\]]*\]  // allow brackets nested one level
                |
                [^\[\]]         // or anything else
            )
        )
        \]
        \(                      // literal paren
        [ \t]*
        ()                      // no id, so leave $3 empty
        <?(.*?)>?               // href = $4
        [ \t]*
        (                       // $5
            (['"])              // quote char = $6
            (.*?)               // Title = $7
            \6                  // matching quote
            [ \t]*              // ignore any spaces/tabs between closing quote and )
        )?                      // title is optional
        \)
    )
    /g,writeAnchorTag);
*/
text = text.replace(/(\[((?:\[[^\]]*\]|[^\[\]])*)\]\([ \t]*()<?(.*?)>?[ \t]*((['"])(.*?)\6[ \t]*)?\))/g,writeAnchorTag);

//
// Last, handle reference-style shortcuts: [link text]
// These must come last in case you've also got [link test][1]
// or [link test](/foo)
//

/*
    text = text.replace(/
    (                           // wrap whole match in $1
        \[
        ([^\[\]]+)              // link text = $2; can't contain '[' or ']'
        \]
    )()()()()()                 // pad rest of backreferences
    /g, writeAnchorTag);
*/
text = text.replace(/(\[([^\[\]]+)\])()()()()()/g, writeAnchorTag);

return text;

}

var writeAnchorTag = function(wholeMatch,m1,m2,m3,m4,m5,m6,m7) {
if (m7 == undefined) m7 = "";
var whole_match = m1;
var link_text = m2;
var link_id = m3.toLowerCase();
var url = m4;
var title = m7;

if (url == "") {
    if (link_id == "") {
        // lower-case and turn embedded newlines into spaces
        link_id = link_text.toLowerCase().replace(/ ?\n/g," ");
    }
    url = "#"+link_id;

    if (g_urls[link_id] != undefined) {
        url = g_urls[link_id];
        if (g_titles[link_id] != undefined) {
            title = g_titles[link_id];
        }
    }
    else {
        if (whole_match.search(/\(\s*\)$/m)>-1) {
            // Special case for explicit empty url
            url = "";
        } else {
            return whole_match;
        }
    }
}   
var result = "";
url = escapeCharacters(url,"*_");
if (url.search(/javascript/i) > -1)
{
    result = "<a href=\"" + url;// + "(" +  + "\"";
    if (title != "") {
        title = title.replace(/"/g,"&quot;");
        title = escapeCharacters(title,"*_");
        result += "(" + title + ")\"";
    }

}
else
{
    result = "<a href=\"" + url + "\"";

    if (title != "") {
        title = title.replace(/"/g,"&quot;");
        title = escapeCharacters(title,"*_");
        result +=  " title=\"" + title + "\"";
    }
}

result += ">" + link_text + "</a>";

return result;

}

var _DoImages = function(text) {
//
// Turn Markdown image shortcuts into tags.
//

//
// First, handle reference-style labeled images: ![alt text][id]
//

/*
    text = text.replace(/
    (                       // wrap whole match in $1
        !\[
        (.*?)               // alt text = $2
        \]

        [ ]?                // one optional space
        (?:\n[ ]*)?         // one optional newline followed by spaces

        \[
        (.*?)               // id = $3
        \]
    )()()()()               // pad rest of backreferences
    /g,writeImageTag);
*/
text = text.replace(/(!\[(.*?)\][ ]?(?:\n[ ]*)?\[(.*?)\])()()()()/g,writeImageTag);

//
// Next, handle inline images:  ![alt text](url "optional title")
// Don't forget: encode * and _

/*
    text = text.replace(/
    (                       // wrap whole match in $1
        !\[
        (.*?)               // alt text = $2
        \]
        \s?                 // One optional whitespace character
        \(                  // literal paren
        [ \t]*
        ()                  // no id, so leave $3 empty
        <?(\S+?)>?          // src url = $4
        [ \t]*
        (                   // $5
            (['"])          // quote char = $6
            (.*?)           // title = $7
            \6              // matching quote
            [ \t]*
        )?                  // title is optional
    \)
    )
    /g,writeImageTag);
*/
text = text.replace(/(!\[(.*?)\]\s?\([ \t]*()<?(\S+?)>?[ \t]*((['"])(.*?)\6[ \t]*)?\))/g,writeImageTag);

return text;

}

var writeImageTag = function(wholeMatch,m1,m2,m3,m4,m5,m6,m7) {
var whole_match = m1;
var alt_text = m2;
var link_id = m3.toLowerCase();
var url = m4;
var title = m7;

if (!title) title = "";

if (url == "") {
    if (link_id == "") {
        // lower-case and turn embedded newlines into spaces
        link_id = alt_text.toLowerCase().replace(/ ?\n/g," ");
    }
    url = "#"+link_id;

    if (g_urls[link_id] != undefined) {
        url = g_urls[link_id];
        if (g_titles[link_id] != undefined) {
            title = g_titles[link_id];
        }
    }
    else {
        return whole_match;
    }
}   

alt_text = alt_text.replace(/"/g,""");
url = escapeCharacters(url,"*_");
var result = "<img src=\"" + url + "\" alt=\"" + alt_text + "\"";

// attacklab: Markdown.pl adds empty title attributes to images.
// Replicate this bug.

//if (title != "") {
    title = title.replace(/"/g,"&quot;");
    title = escapeCharacters(title,"*_");
    result +=  " title=\"" + title + "\"";
//}

result += " />";

return result;

}

var _DoHeaders = function(text) {

// Setext-style headers:
//  Header 1
//  ========
//  
//  Header 2
//  --------
//
text = text.replace(/^(.+)[ \t]*\n=+[ \t]*\n+/gm,
    function(wholeMatch,m1){return hashBlock("<h1>" + _RunSpanGamut(m1) + "</h1>");});

text = text.replace(/^(.+)[ \t]*\n-+[ \t]*\n+/gm,
    function(matchFound,m1){return hashBlock("<h2>" + _RunSpanGamut(m1) + "</h2>");});

// atx-style headers:
//  # Header 1
//  ## Header 2
//  ## Header 2 with closing hashes ##
//  ...
//  ###### Header 6
//

/*
    text = text.replace(/
        ^(\#{1,6})              // $1 = string of #'s
        [ \t]*
        (.+?)                   // $2 = Header text
        [ \t]*
        \#*                     // optional closing #'s (not counted)
        \n+
    /gm, function() {...});
*/

text = text.replace(/^(\#{1,6})[ \t]*(.+?)[ \t]*\#*\n+/gm,
    function(wholeMatch,m1,m2) {
        var h_level = m1.length;
        return hashBlock("<h" + h_level + ">" + _RunSpanGamut(m2) + "</h" + h_level + ">");
    });

return text;

}

// This declaration keeps Dojo compressor from outputting garbage:
var _ProcessListItems;

var _DoLists = function(text) {
//
// Form HTML ordered (numbered) and unordered (bulleted) lists.
//

// attacklab: add sentinel to hack around khtml/safari bug:
// http://bugs.webkit.org/show_bug.cgi?id=11231
text += "~0";

// Re-usable pattern to match any entirel ul or ol list:

/*
    var whole_list = /
    (                                   // $1 = whole list
        (                               // $2
            [ ]{0,3}                    // attacklab: g_tab_width - 1
            ([*+-]|\d+[.])              // $3 = first list item marker
            [ \t]+
        )
        [^\r]+?
        (                               // $4
            ~0                          // sentinel for workaround; should be $
        |
            \n{2,}
            (?=\S)
            (?!                         // Negative lookahead for another list item marker
                [ \t]*
                (?:[*+-]|\d+[.])[ \t]+
            )
        )
    )/g
*/
var whole_list = /^(([ ]{0,3}([*+\-]|\d+[.])[ \t]+)[^\r]+?(~0|\n{2,}(?=\S)(?![ \t]*(?:[*+\-]|\d+[.])[ \t]+)))/gm;

if (g_list_level) {
    text = text.replace(whole_list,function(wholeMatch,m1,m2) {
        var list = m1;
        var list_type = (m2.search(/[*+\-]/g)>-1) ? "ul" : "ol";

        // Turn double returns into triple returns, so that we can make a
        // paragraph for the last item in a list, if necessary:
        list = list.replace(/\n{2,}/g,"\n\n\n");;
        var result = _ProcessListItems(list);

        // Trim any trailing whitespace, to put the closing `</$list_type>`
        // up on the preceding line, to get it past the current stupid
        // HTML block parser. This is a hack to work around the terrible
        // hack that is the HTML block parser.
        result = result.replace(/\s+$/,"");
        result = "<"+list_type+">" + result + "</"+list_type+">\n";
        return result;
    });
} else {
    whole_list = /(\n\n|^\n?)(([ ]{0,3}([*+\-]|\d+[.])[ \t]+)[^\r]+?(~0|\n{2,}(?=\S)(?![ \t]*(?:[*+\-]|\d+[.])[ \t]+)))/g;
    text = text.replace(whole_list,function(wholeMatch,m1,m2,m3) {
        var runup = m1;
        var list = m2;

        var list_type = (m3.search(/[*+\-]/g)>-1) ? "ul" : "ol";
        // Turn double returns into triple returns, so that we can make a
        // paragraph for the last item in a list, if necessary:
        var list = list.replace(/\n{2,}/g,"\n\n\n");;
        var result = _ProcessListItems(list);
        result = runup + "<"+list_type+">\n" + result + "</"+list_type+">\n";   
        return result;
    });
}

// attacklab: strip sentinel
text = text.replace(/~0/,"");

return text;

}

_ProcessListItems = function(list_str) {
//
// Process the contents of a single ordered or unordered list, splitting it
// into individual list items.
//
// The $g_list_level global keeps track of when we're inside a list.
// Each time we enter a list, we increment it; when we leave a list,
// we decrement. If it's zero, we're not in a list anymore.
//
// We do this because when we're not inside a list, we want to treat
// something like this:
//
// I recommend upgrading to version
// 8. Oops, now this line is treated
// as a sub-list.
//
// As a single paragraph, despite the fact that the second line starts
// with a digit-period-space sequence.
//
// Whereas when we're inside a list (or sub-list), that line will be
// treated as the start of a sub-list. What a kludge, huh? This is
// an aspect of Markdown's syntax that's hard to parse perfectly
// without resorting to mind-reading. Perhaps the solution is to
// change the syntax rules such that sub-lists must start with a
// starting cardinal number; e.g. "1." or "a.".

g_list_level++;

// trim trailing blank lines:
list_str = list_str.replace(/\n{2,}$/,"\n");

// attacklab: add sentinel to emulate \z
list_str += "~0";

/*
    list_str = list_str.replace(/
        (\n)?                           // leading line = $1
        (^[ \t]*)                       // leading whitespace = $2
        ([*+-]|\d+[.]) [ \t]+           // list marker = $3
        ([^\r]+?                        // list item text   = $4
        (\n{1,2}))
        (?= \n* (~0 | \2 ([*+-]|\d+[.]) [ \t]+))
    /gm, function(){...});
*/
list_str = list_str.replace(/(\n)?(^[ \t]*)([*+\-]|\d+[.])[ \t]+([^\r]+?(\n{1,2}))(?=\n*(~0|\2([*+\-]|\d+[.])[ \t]+))/gm,
    function(wholeMatch,m1,m2,m3,m4){
        var item = m4;
        var leading_line = m1;
        var leading_space = m2;

        if (leading_line || (item.search(/\n{2,}/)>-1)) {
            item = _RunBlockGamut(_Outdent(item));
        }
        else {
            // Recursion for sub-lists:
            item = _DoLists(_Outdent(item));
            item = item.replace(/\n$/,""); // chomp(item)
            item = _RunSpanGamut(item);
        }

        return  "<li>" + item + "</li>\n";
    }
);

// attacklab: strip sentinel
list_str = list_str.replace(/~0/g,"");

g_list_level--;
return list_str;

}

var _DoCodeBlocks = function(text) {
//
// Process Markdown <pre><code> blocks.
//

/*
    text = text.replace(text,
        /(?:\n\n|^)
        (                               // $1 = the code block -- one or more lines, starting with a space/tab
            (?:
                (?:[ ]{4}|\t)           // Lines must start with a tab or a tab-width of spaces - attacklab: g_tab_width
                .*\n+
            )+
        )
        (\n*[ ]{0,3}[^ \t\n]|(?=~0))    // attacklab: g_tab_width
    /g,function(){...});
*/

// attacklab: sentinel workarounds for lack of \A and \Z, safari\khtml bug
text += "~0";

text = text.replace(/(?:\n\n|^)((?:(?:[ ]{4}|\t).*\n+)+)(\n*[ ]{0,3}[^ \t\n]|(?=~0))/g,
    function(wholeMatch,m1,m2) {
        var codeblock = m1;
        var nextChar = m2;

        codeblock = _EncodeCode( _Outdent(codeblock));
        codeblock = _Detab(codeblock);
        codeblock = codeblock.replace(/^\n+/g,""); // trim leading newlines
        codeblock = codeblock.replace(/\n+$/g,""); // trim trailing whitespace

        codeblock = "<pre><code>" + codeblock + "\n</code></pre>";

        return hashBlock(codeblock) + nextChar;
    }
);

// attacklab: strip sentinel
text = text.replace(/~0/,"");

return text;

}

var hashBlock = function(text) {
text = text.replace(/(^\n+|\n+$)/g,"");
return "\n\n~K" + (g_html_blocks.push(text)-1) + "K\n\n";
}

var _DoCodeSpans = function(text) {
//
// * Backtick quotes are used for spans.
//
// * You can use multiple backticks as the delimiters if you want to
// include literal backticks in the code span. So, this input:
//
// Just type foo bar baz at the prompt.
//
// Will translate to:
//
//

Just type foo bar baz at the prompt.

//
// There's no arbitrary limit to the number of backticks you
// can use as delimters. If you need three consecutive backticks
// in your code, use four for delimiters, etc.
//
// * You can use spaces to get literal backticks at the edges:
//
// ... type bar ...
//
// Turns to:
//
// ... type bar ...
//

/*
    text = text.replace(/
        (^|[^\\])                   // Character before opening ` can't be a backslash
        (`+)                        // $2 = Opening run of `
        (                           // $3 = The code block
            [^\r]*?
            [^`]                    // attacklab: work around lack of lookbehind
        )
        \2                          // Matching closer
        (?!`)
    /gm, function(){...});
*/

text = text.replace(/(^|[^\\])(`+)([^\r]*?[^`])\2(?!`)/gm,
    function(wholeMatch,m1,m2,m3,m4) {
        var c = m3;
        c = c.replace(/^([ \t]*)/g,""); // leading whitespace
        c = c.replace(/[ \t]*$/g,"");   // trailing whitespace
        c = _EncodeCode(c);
        return m1+"<code>"+c+"</code>";
    });

return text;

}

var _EncodeCode = function(text) {
//
// Encode/escape certain characters inside Markdown code runs.
// The point is that in code, these characters are literals,
// and lose their special Markdown meanings.
//
// Encode all ampersands; HTML entities are not
// entities within a Markdown code span.
text = text.replace(/&/g,"&");

// Do the angle bracket song and dance:
text = text.replace(/</g,"&lt;");
text = text.replace(/>/g,"&gt;");

// Now, escape characters that are magic in Markdown:
text = escapeCharacters(text,"\*_{}[]\\",false);

// jj the line above breaks this:
//---

//* Item

// 1. Subitem

// special char: *
//---

return text;

}

var _DoItalicsAndBold = function(text) {

// <strong> must go first:
text = text.replace(/(\*\*|__)(?=\S)([^\r]*?\S[\*_]*)\1/g,
    "<strong>$2</strong>");

text = text.replace(/(\*|_)(?=\S)([^\r]*?\S)\1/g,
    "<em>$2</em>");

return text;

}

var markPants = function(wholeMatch,m1) {

var blockText = m1;

// Undo double lines
blockText = blockText.replace(/\n\n/g,"\n");
blockText = blockText.replace(/^\n/,"");
// strip trailing blank lines
blockText = blockText.replace(/\n+$/g,"");
return blockText;

};

var educatePants = function(wholeMatch,m1,m2,m3,m4,m5,m6) {

var blockText = m5;
if (m5.search(/(<([a-zA-Z1-6]*)([^\r>]?)(>)([^\r]*?)(<\/\2>))/mg)>-1)
{
    //alert(blockText);
    blockText = blockText.replace(/(<)([a-zA-Z1-6]*)([^\r>]?)(>)([^\r]*?)(<\/\2>)/mg, educatePants);

}
if (m5.search(/(<([a-zA-Z1-6]*)([^\r>]?)(>)([^\r]*?)(<\/\2>))/g)>-1)
    blockText = blockText.replace(/(<)([a-zA-Z1-6]*)([^\r>]?)(>)([^\r]*?)(<\/\2>)/g, educatePants);
else blockText = _applyPants(blockText);
return m1 + m2 + m3 + m4 + blockText + m6;

};

var _applyImgPant = function(wholeMatch, m1)
{

var blockText = m1;
//blockText = m1.replace(/\/>/g, "</img>");// + "</img>";   
blockText = blockText.replace(/&\#8220;/g, "\"");
blockText = blockText.replace(/&\#8221;/g, "\"");
blockText = blockText.replace(/&\#8216;/g, "'");
blockText = blockText.replace(/&\#8217;/g, "'");
blockText = blockText.replace(/&\#8230;/g, "...");

return blockText;

}

var _applyPants = function(text)
{

    text = text.replace (/^\x27(?=[!\x22#\$\%\x27()*+,\-.\/:;<=>?\@\[\\]\^_`{|}~]\B)/g, "&#8216;");
    text = text.replace (/^\x22(?=[!\x22#\$\%\x27()*+,\-.\/:;<=>?\@\[\\]\^_`{|}~]\B)/g, "&#8220;");
    text = text.replace(/^\x22(?=\w)/g, "&#8220;");
    text = text.replace(/^\x27(?=\w)/g, "&#8216;");

    text = text.replace(/\x22\x27(?=\w)/g, "&#8220;&#8216;");
    text = text.replace(/\x27\x22(?=\w)/g, "&#8216;&#8220;");

    //  Special case for decade abbreviations (the '80s):
    text = text.replace(/\x27(?=\d{2}s)/g, "&#8217;");


    text = text.replace(/(>|\t|\n|\s|&nbsp;|--|&[mn]dash;|&\#8211;|&\#8212;|&\#x201[34];)\x27(?=\w)/g, "$1&#8216;");

    text = text.replace(/([^<>\\ \t\r\n\[\{\(\-])\x27(?=\s | s\b)/g, "$1&#8217;");


    // Any remaining single quotes should be opening ones:
    text = text.replace (/\x27/g, "&#8217;");

    text = text.replace(/(>|\t|\n|\s|&nbsp;|--|&[mn]dash;|&\#8211;|&\#8212;|&\#x201[34];)\x22(?=\w)/g, "$1&#8220;");

    text = text.replace (/([^<>\\ \t\r\n\[\{\(\-])\x22(?=\s | s\b)/g, "$1&#8221;");


    text = text.replace (/\x22/ig, "&#8221;");
    text.replace(/\.\.\./g, "&#8230;").replace (/\.\s\.\s\./g, "&#8230;");
    return text;

}

var _RunSmartyPants = function(text) {

text = text.replace(/^(<([a-zA-Z1-6]*)[^\r]?\n<\/\2>[ \t]*(?=\n+))/gm,markPants);
text = text.replace(/(<)([a-zA-Z1-6]*)([^\r>]?)(>)([^\r]*?)(<\/\2>)/gm, educatePants);
//clean everything inside html tags
text = text.replace(/(<([a-zA-Z1-6]*)\b([^\r>]*?)(\/)?>)/g, _applyImgPant);
//clean out replacements inside special tags
text = text.replace(/((<)(code|kbd|pre|script|noscript|iframe|math|ins|del|pre)([^\r]?)(>)([^\r]*?)(<\/)(code|kbd|pre|script|noscript|iframe|math|ins|del|pre)(>))/gm, _applyImgPant);


return text;

}

var _DoBlockQuotes = function(text) {

/*
    text = text.replace(/
    (                               // Wrap whole match in $1
        (
            ^[ \t]*>[ \t]?          // '>' at the start of a line
            .+\n                    // rest of the first line
            (.+\n)*                 // subsequent consecutive lines
            \n*                     // blanks
        )+
    )
    /gm, function(){...});
*/

text = text.replace(/((^[ \t]*>[ \t]?.+\n(.+\n)*\n*)+)/gm,
    function(wholeMatch,m1) {
        var bq = m1;

        // attacklab: hack around Konqueror 3.5.4 bug:
        // "----------bug".replace(/^-/g,"") == "bug"

        bq = bq.replace(/^[ \t]*>[ \t]?/gm,"~0");   // trim one level of quoting

        // attacklab: clean up hack
        bq = bq.replace(/~0/g,"");

        bq = bq.replace(/^[ \t]+$/gm,"");       // trim whitespace-only lines
        bq = _RunBlockGamut(bq);                // recurse

        bq = bq.replace(/(^|\n)/g,"$1  ");
        // These leading spaces screw with <pre> content, so we need to fix that:
        bq = bq.replace(
                /(\s*<pre>[^\r]+?<\/pre>)/gm,
            function(wholeMatch,m1) {
                var pre = m1;
                // attacklab: hack around Konqueror 3.5.4 bug:
                pre = pre.replace(/^  /mg,"~0");
                pre = pre.replace(/~0/g,"");
                return pre;
            });

        return hashBlock("<blockquote>\n" + bq + "\n</blockquote>");
    });
return text;

}

var _FormParagraphs = function(text) {
//
// Params:
// $text - string to process with html

tags
//

// Strip leading and trailing lines:
text = text.replace(/^\n+/g,"");
text = text.replace(/\n+$/g,"");

var grafs = text.split(/\n{2,}/g);
var grafsOut = new Array();

//
// Wrap <p> tags.
//
var end = grafs.length;
for (var i=0; i<end; i++) {
    var str = grafs[i];

    // if this is an HTML marker, copy it
    if (str.search(/~K(\d+)K/g) >= 0) {
        grafsOut.push(str);
    }
    else if (str.search(/\S/) >= 0) {

        str = _RunSpanGamut(str);

        str = str.replace(/^([ \t]*)/g,"<p>");
        str += "</p>"
        grafsOut.push(str);
    }

}

//
// Unhashify HTML blocks
//
end = grafsOut.length;
for (var i=0; i<end; i++) {
    // if this is a marker for an html block...
    while (grafsOut[i].search(/~K(\d+)K/) >= 0) {
        var blockText = g_html_blocks[RegExp.$1];
        blockText = blockText.replace(/\$/g,"$$$$"); // Escape any dollar signs
        grafsOut[i] = grafsOut[i].replace(/~K\d+K/,blockText);
    }
}

return grafsOut.join("\n\n");

}

var _EncodeAmpsAndAngles = function(text) {
// Smart processing for ampersands and angle brackets that need to be encoded.

// Ampersand-encoding based entirely on Nat Irons's Amputator MT plugin:
//   http://bumppo.net/projects/amputator/
text = text.replace(/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)/g,"&");

// Encode naked <'s
text = text.replace(/<(?![a-z\/?\$!])/gi,"&lt;");

return text;

}

var _EncodeBackslashEscapes = function(text) {
//
// Parameter: String.
// Returns: The string, with after processing the following backslash
// escape sequences.
//

// attacklab: The polite way to do this is with the new
// escapeCharacters() function:
//
//  text = escapeCharacters(text,"\\",true);
//  text = escapeCharacters(text,"`*_{}[]()>#+-.!",true);
//
// ...but we're sidestepping its use of the (slow) RegExp constructor
// as an optimization for Firefox.  This function gets called a LOT.

text = text.replace(/\\(\\)/g,escapeCharacters_callback);
text = text.replace(/\\([`*_{}\[\]()>#+\-.!])/g,escapeCharacters_callback);
return text;

}

var _DoAutoLinks = function(text) {

text = text.replace(/<((https?|ftp|dict):[^'">\s]+)>/gi,"<a href=\"$1\">$1</a>");

// Email addresses: <[email protected]>

/*
    text = text.replace(/
        <
        (?:mailto:)?
        (
            [-.\w]+
            \@
            [-a-z0-9]+(\.[-a-z0-9]+)*\.[a-z]+
        )
        >
    /gi, _DoAutoLinks_callback());
*/
text = text.replace(/<(?:mailto:)?([-.\w]+\@[-a-z0-9]+(\.[-a-z0-9]+)*\.[a-z]+)>/gi,
    function(wholeMatch,m1) {
        return _EncodeEmailAddress( _UnescapeSpecialChars(m1) );
    }
);

return text;

}

var _EncodeEmailAddress = function(addr) {
//
// Input: an email address, e.g. "[email protected]"
//
// Output: the email address as a mailto link, with each character
// of the address encoded as either a decimal or hex entity, in
// the hopes of foiling most address harvesting spam bots. E.g.:
//
// foo
// @example.com
//
// Based on a filter by Matthew Wickline, posted to the BBEdit-Talk
// mailing list: http://tinyurl.com/yu7ue
//

// attacklab: why can't javascript speak hex?
function char2hex(ch) {
    var hexDigits = '0123456789ABCDEF';
    var dec = ch.charCodeAt(0);
    return(hexDigits.charAt(dec>>4) + hexDigits.charAt(dec&15));
}

var encode = [
    function(ch){return "&#"+ch.charCodeAt(0)+";";},
    function(ch){return "&#x"+char2hex(ch)+";";},
    function(ch){return ch;}
];

addr = "mailto:" + addr;

addr = addr.replace(/./g, function(ch) {
    if (ch == "@") {
        // this *must* be encoded. I insist.
        ch = encode[Math.floor(Math.random()*2)](ch);
    } else if (ch !=":") {
        // leave ':' alone (to spot mailto: later)
        var r = Math.random();
        // roughly 10% raw, 45% hex, 45% dec
        ch =  (
                r > .9  ?   encode[2](ch)   :
                r > .45 ?   encode[1](ch)   :
                            encode[0](ch)
            );
    }
    return ch;
});

addr = "<a href=\"" + addr + "\">" + addr + "</a>";
addr = addr.replace(/">.+:/g,"\">"); // strip the mailto: from the visible part

return addr;

}

var _UnescapeSpecialChars = function(text) {
//
// Swap back in all the special characters we've hidden.
//
text = text.replace(/~E(\d+)E/g,
function(wholeMatch,m1) {
var charCodeToReplace = parseInt(m1);
return String.fromCharCode(charCodeToReplace);
}
);
return text;
}

var _Outdent = function(text) {
//
// Remove one level of line-leading tabs or spaces
//

// attacklab: hack around Konqueror 3.5.4 bug:
// "----------bug".replace(/^-/g,"") == "bug"

text = text.replace(/^(\t|[ ]{1,4})/gm,"~0"); // attacklab: g_tab_width

// attacklab: clean up hack
text = text.replace(/~0/g,"")

return text;

}

var _Detab = function(text) {
// attacklab: Detab's completely rewritten for speed.
// In perl we could fix it by anchoring the regexp with \G.
// In javascript we're less fortunate.

// expand first n-1 tabs
text = text.replace(/\t(?=\t)/g,"    "); // attacklab: g_tab_width

// replace the nth with two sentinels
text = text.replace(/\t/g,"~A~B");

// use the sentinel to anchor our regex so it doesn't explode
text = text.replace(/~B(.+?)~A/g,
    function(wholeMatch,m1,m2) {
        var leadingText = m1;
        var numSpaces = 4 - leadingText.length % 4;  // attacklab: g_tab_width

        // there *must* be a better way to do this:
        for (var i=0; i<numSpaces; i++) leadingText+=" ";

        return leadingText;
    }
);

// clean up sentinels
text = text.replace(/~A/g,"    ");  // attacklab: g_tab_width
text = text.replace(/~B/g,"");

return text;

}

//
// attacklab: Utility functions
//

var escapeCharacters = function(text, charsToEscape, afterBackslash) {
// First we have to escape the escape characters so that
// we can build a character class out of them
var regexString = "([" + charsToEscape.replace(/([[]])/g,"$1") + "])";

if (afterBackslash) {
    regexString = "\\\\" + regexString;
}

var regex = new RegExp(regexString,"g");
text = text.replace(regex,escapeCharacters_callback);

return text;

}

var escapeCharacters_callback = function(wholeMatch,m1) {
var charCodeToEscape = m1.charCodeAt(0);
return "~E"+charCodeToEscape+"E";
}

} // end of Attacklab.showdown.converter

// Version 0.9 used the Showdown namespace instead of Attacklab.showdown
// The old namespace is deprecated, but we'll support it for now:
var Showdown = Attacklab.showdown;

// If anyone's interested, tell the world that this file's been loaded
if (Attacklab.fileLoaded) {
Attacklab.fileLoaded("showdown.js");
}

Bug when wmd is hidden

When using inside tabs, wmd does not play nice when the tab is hidden (firefox 3.5 with firebug enabled will throw countless "ns_error_failure 0x80004005".

The error seems to be related to line 485:

if (!util.isVisible(inputArea)) {
    return;
}

// Update the selection start and end, text. (HERE IS WHERE THE EXCEPTION OCCURS)
if (inputArea.selectionStart || inputArea.selectionStart === 0) {

The implementation of isVisible() only takes into consideration the textarea state, and not of any containers. Changing the implementation to this (ported from scripty2) solves this issue:

util.isVisible = function (elem) {
    while (elem && elem.parentNode) {
        if (elem.style.display == 'none' || elem.style.visibility == 'hidden')
                return false;
        elem = elem.parentNode;
    }
    return true;
};

AngularJS directive

there are any angularjs implementation for this editor?

Quote not working in Google Chrome

Blockquote is not working in Google Chrome. It creates four duplicates of the string and does not indent correctly.

For example:
"test" becomes:

"test
"test >test test
test"

toggle Headings

command.doHeading = function(chunk, postProcessing, useDefaultText){

    //chunk.trimWhitespace();
    // Remove leading/trailing whitespace and reduce internal spaces to single spaces.
    var chunkOrigSel = chunk.selection;
    chunk.selection = chunk.selection.replace(/\s+/g, " ");
    chunk.selection = chunk.selection.replace(/(^\s+|\s+$)/g, "");
    var firstChr = chunk.selection.substring(0, 1);
    // If we clicked the button with no selected text, we just
    // make a level 2 hash header around some default text.
    if(!chunk.selection){
        chunk.findTags(/[^\n]/, /[^\n]/);       
        if(/[^\n]/.test(chunk.startTag))
            chunk.startTag += "\n\n## ";
        else chunk.startTag = "## ";
        chunk.selection = "Heading";
        if(/[^\n]/.test(chunk.startTag))
            chunk.endTag = " ##\n\n" + chunk.endTag;
        else chunk.endTag = " ##";
        return;
    }

    var headerLevel = 0;        // The existing header level of the selected text.

    // Remove any existing hash heading markdown and save the header level.
    chunk.findTags(/#+[ ]*/, /[ ]*#+/);
    if(/#+/.test(chunk.startTag)){
        headerLevel = re.lastMatch.length;
    }
    chunk.startTag = chunk.endTag = "";

    var addNewLine = "";
    if (headerLevel == 0)
    {
        chunk.findTags(/[^\n]/, null);      
        if(/[^\n]/.test(chunk.startTag)){
            addNewLine = chunk.startTag.substring(0,1);
        }
        chunk.startTag = chunk.endTag = "";
    }

    var removeNewLines = false;
    if (headerLevel < 6) headerLevel++;
    else if (headerLevel == 6) 
    {
        headerLevel = 0;
        removeNewLines = true;
    }

    var headerChar = "#";
    var len = headerLevel;
    if (headerLevel > 0)
    {
        if (headerLevel == 1 && addNewLine != "") 
        {
            chunk.startTag = addNewLine.replace(/\s+/g, "") + "\n\n";
        }
        chunk.endTag = " ";
        while(len--){
            chunk.endTag += headerChar;
            chunk.startTag += headerChar;
        }
        chunk.startTag += " ";

        if (headerLevel == 1) 
        {
            chunk.endTag += "\n\n";
            chunk.selection = firstChr + chunk.selection;
        }

    }
    else
    {
        chunk.findTags(/[\n]+/, /[\n]+/);       
        if(/[\n]+/.test(chunk.startTag)){
            chunk.startTag = "";
        }
        if(/[\n]+/.test(chunk.endTag)){
            chunk.endTag = "";
        }
        chunk.selection = chunkOrigSel;     
        chunk.findTags(/[^\n]/, /[^\n]/);       
        if(/[^\n]/.test(chunk.startTag)){
            chunk.startTag = chunk.startTag.substring(0,1) + " ";
            chunk.startTag = chunk.startTag.replace(/\s+/g, " ");
        }
        if(/[^\n]/.test(chunk.endTag)){
            chunk.endTag = " " + chunk.endTag.substring(0,1);
            chunk.endTag = chunk.endTag.replace(/\s+/g, " ");
        }
        chunk.selection = chunkOrigSel;         
    }

};  

ul and ol not working in google chrome

this issue was also reported here: http://code.google.com/p/wmd-new/issues/detail?id=9#c2

when using google chrome, neither the ordered lists nor the unordered lists are working. when pressing enter, wmd inserts some weird garbage.