Reduce (UTF-8) strings to alphanumeric

urlify.js

// port of https://gist.github.com/909692
// TODO: check char-map of https://github.com/jbroadway/urlify/blob/master/URLify.php for characters we've missed
// TODO: check performance against http://stackoverflow.com/questions/1946040/using-javascript-to-normalize-url-rewriting-strings-entered-by-users

var Urlifyer = function(options) {
    var _key, _code, i, _source, _sources;
    
    // Allow instantiation without the 'new' keyword
    if (!(this instanceof Urlifyer)) {
        return new Urlifyer(options);
    }
    
    if (!options) {
        options = {};
    }
    
    this._trim = options.trim || Urlifyer.trim;
    this._lower = options.lower || Urlifyer.lower;
    this._allow = {};
    this._replace = {};
    
    // allowed characters
    _sources = [Urlifyer.allow, options.allow];
    for (i=0; (_source = _sources[i]) !== undefined; i++) {
        if (!_source) {
            continue;
        }
        
        var index = _source.length;
        while (index--) {
            _key = _source.charCodeAt(index);
            this._allow[_key] = true;
        }
    }
    
    // replacements
    _sources = [Urlifyer.replace, options.replace];
    for (i=0; (_source = _sources[i]) !== undefined; i++) {
        if (!_source) {
            continue;
        }
        
        for (_key in _source) {
            if (Object.prototype.hasOwnProperty.call(_source, _key)) {
                console.log(_key);
                // allow passing either unicode-points or characters
                var __key = _key.match(/^[0-9]+$/) ? _key : _key.charCodeAt(0);
                this._replace[__key] = _source[_key];
            }
        }
    }

    return this;
};
Urlifyer.prototype.urlify = function(string) {
    var index = string.length;
    
    while (index--) {
        var _code = string.charCodeAt(index),
            _replacement = null;
        
        if ((_code >= 97 && _code <= 122) || (_code >= 65 && _code <= 90) || (_code >= 48 && _code <= 57) || _code === 95 || _code === 45) {
            // skip normalization for alphanumeric characters [a-zA-Z0-9_-]
            continue;
        } else if (this._allow && this._allow[_code]) {
            // skip normalization for allowed characters
            continue;
        } else if (this._replace[_code]) {
            // replace as defined
            _replacement = this._replace[_code];
        } else if ((_code >= 0xC0 && _code <= 0xC6) || (_code >= 0xE0 && _code <= 0xE6) || (_code >= 0x100 && _code <= 0x105)) {
            _replacement = "a";
        } else if (_code == 0xC7 || _code == 0xE7 || (_code >= 0x106 && _code <= 0x10D)) {
            _replacement = "c";
        } else if (_code == 0xD0 || (_code >= 0x10E && _code <= 0x111)) {
            _replacement = "d";
        } else if ((_code >= 0xC8 && _code <= 0xCB) || (_code >= 0xE8 && _code <= 0xEB) || (_code >= 0x112 && _code <= 0x11B)) {
            _replacement = "e";
        } else if ((_code >= 0xCC && _code <= 0xCF) || (_code >= 0xEC && _code <= 0xEF)) {
            _replacement = "i";
        } else if (_code == 0xD1 || _code == 0xF1) {
            _replacement = "n";
        } else if ((_code >= 0xD2 && _code <= 0xD6) || (_code >= 0xF2 && _code <= 0xF6) || _code == 0xD8 || _code == 0xF8 ) {
            _replacement = "o";
        } else if ((_code >= 0xD9 && _code <= 0xDC) || (_code >= 0xF9 && _code <= 0xFC)) {
            _replacement = "u";
        } else if (_code == 0xDD || _code == 0xFD || _code == 0xFF) {
            _replacement = "y";
        } else {
            _replacement = "-";
        }
        
        string = string.slice(0, index)
            + _replacement
            + string.slice(index + 1);
    }
    
    if (this._trim) {
        // reduce to single dash
        string = string.replace(/-{2,}/g, '-');
        // trim dash from beginning and end
        string = string.replace(/^-|-$/g, '');
    }
    
    if (this._lower) {
        string = string.toLowerCase();
    }
    
    return string;
};

Urlifyer.lower = false;
Urlifyer.trim = false;
Urlifyer.allow = "";
Urlifyer.replace = {
    0xE4: "ae", // ä
    0xC4: "Ae", // Ä
    0xF6: "oe", // ö
    0xD6: "Oe", // Ö
    0xFC: "ue", // ü
    0xDC: "Ue", // Ü
    0xDF: "ss", // ß
    0xE6: "ae", // æ
    0xC6: "AE" // Æ
};



var u = new Urlifyer({trim: true, lower: true, allow: "õ", replace: {"ñ":"#"}});
console.log(u);
console.log(u.urlify("häLLö w(orl)d ñõ so?"));