sheercms/static/js/htmlbeautify.js

Sheer Cliff CMS is a simple and powerful content management system (CMS) for Node JS.

/*
Copyright Arantius.com
http://tools.arantius.com/tabifier
*/

//gogo global variable
var level=0;
var LOOP_SIZE=100;

function finishTabifier(code) {
    code=code.replace(/\n\s*\n/g, '\n');  //blank lines
    code=code.replace(/^[\s\n]*/, ''); //leading space
    code=code.replace(/[\s\n]*$/, ''); //trailing space
    level=0;
    return code;
}

function cleanHTML(code, callback) {
    var i=0;
    function cleanAsync() {
        var iStart=i;
        for (; i<code.length && i<iStart+LOOP_SIZE; i++) {
            point=i;

            //if no more tags, copy and exit
            if (-1==code.substr(i).indexOf('<')) {
                out+=code.substr(i);
                callback(finishTabifier(out));
                return;
            }

            //copy verbatim until a tag
            while (point<code.length && '<'!=code.charAt(point)) point++;
            if (i!=point) {
                cont=code.substr(i, point-i);
                if (!cont.match(/^\s+$/)) {
                    if ('\n'==out.charAt(out.length-1)) {
                        out+=tabs();
                    } else if ('\n'==cont.charAt(0)) {
                        out+='\n'+tabs();
                        cont=cont.replace(/^\s+/, '');
                    }
                    cont=cont.replace(/\s+/g, ' ');
                    out+=cont;
                } if (cont.match(/\n/)) {
                    out+='\n'+tabs();
                }
            }
            start=point;

            //find the end of the tag
            while (point<code.length && '>'!=code.charAt(point)) point++;
            tag=code.substr(start, point-start);
            i=point;

            //if this is a special tag, deal with it!
            if ('!--'==tag.substr(1,3)) {
                if (!tag.match(/--$/)) {
                    while ('-->'!=code.substr(point, 3)) point++;
                    point+=2;
                    tag=code.substr(start, point-start);
                    i=point;
                }
                if ('\n'!=out.charAt(out.length-1)) out+='\n';
                out+=tabs();
                out+=tag+'>\n';
            } else if ('!'==tag[1]) {
                out=placeTag(tag+'>', out);
            } else if ('?'==tag[1]) {
                out+=tag+'>\n';
            } else if (t=tag.match(/^<(script|style)/i)) {
                t[1]=t[1].toLowerCase();
                tag=cleanTag(tag);
                out=placeTag(tag, out);
                end=String(code.substr(i+1)).toLowerCase().indexOf('</'+t[1]);
                if (end) {
                    cont=code.substr(i+1, end);
                    i+=end;
                    out+=cont;
                }
            } else {
                tag=cleanTag(tag);
                out=placeTag(tag, out);
            }
        }

        if (i<code.length) {
            setTimeout(cleanAsync, 0);
        } else {
            callback(finishTabifier(out));
        }
    }

    var point=0, start=null, end=null, tag='', out='', cont='';
    cleanAsync();
}

function tabs() {
    var s='';
    for (var j=0; j<level; j++) s+='\t';
    return s;
}

function cleanTag(tag) {
    var tagout='';
    tag=tag.replace(/\n/g, ' ');       //remove newlines
    tag=tag.replace(/[\s]{2,}/g, ' '); //collapse whitespace
    tag=tag.replace(/^\s+|\s+$/g, ' '); //collapse whitespace
    var suffix='';
    if (tag.match(/\/$/)) {
        suffix='/';
        tag=tag.replace(/\/+$/, '');
    }
    var m, partRe = /\s*([^= ]+)(?:=((['"']).*?\3|[^ ]+))?/;
    while (m = partRe.exec(tag)) {
        if (m[2]) {
            tagout += m[1].toLowerCase() + '=' + m[2];
        } else if (m[1]) {
            tagout += m[1].toLowerCase();
        }
        tagout += ' ';

        // Why is this necessary?  I thought .exec() went from where it left off.
        tag = tag.substr(m[0].length);
    }
    return tagout.replace(/\s*$/, '')+suffix+'>';
}

/////////////// The below variables are only used in the placeTag() function
/////////////// but are declared global so that they are read only once
//opening and closing tag on it's own line but no new indentation level
var ownLine=['area', 'body', 'head', 'hr', 'i?frame', 'link', 'meta',
    'noscript', 'style', 'table', 'tbody', 'thead', 'tfoot'];

//opening tag, contents, and closing tag get their own line
//(i.e. line before opening, after closing)
var contOwnLine=['li', 'dt', 'dt', 'h[1-6]', 'option', 'script'];

//line will go before these tags
var lineBefore=new RegExp(
        '^<(/?'+ownLine.join('|/?')+'|'+contOwnLine.join('|')+')[ >]'
);

//line will go after these tags
lineAfter=new RegExp(
        '^<(br|/?'+ownLine.join('|/?')+'|/'+contOwnLine.join('|/')+')[ >]'
);

//inside these tags (close tag expected) a new indentation level is created
var newLevel=['blockquote', 'div', 'dl', 'fieldset', 'form', 'frameset',
    'map', 'ol', 'p', 'pre', 'select', 'td', 'th', 'tr', 'ul'];
newLevel=new RegExp('^</?('+newLevel.join('|')+')[ >]');
function placeTag(tag, out) {
    var nl=tag.match(newLevel);
    if (tag.match(lineBefore) || nl) {
        out=out.replace(/\s*$/, '');
        out+="\n";
    }

    if (nl && '/'==tag.charAt(1)) level--;
    if ('\n'==out.charAt(out.length-1)) out+=tabs();
    if (nl && '/'!=tag.charAt(1)) level++;

    out+=tag;
    if (tag.match(lineAfter) || tag.match(newLevel)) {
        out=out.replace(/ *$/, '');
        out+="\n";
    }
    return out;
}