Programatic Accent Reduction in JavaScript (aka text normalization or unaccenting)

I need to compare 2 strings as equal such as these:

Lubeck == Lübeck

In JavaScript.

Why? Well, I have an auto-completion field that's going out to a Java service using Lucene, where place names are stored naturally (as Lübeck), but also indexed as normalized text,

import sun.text.Normalizer;
oDoc.setNameLC = Normalizer.normalize(oLocName, Normalizer.DECOMP, 0)
    .toLowerCase().replaceAll("[^\\p{ASCII}]","");

This way some-one who doesn't know to type "Mèxico" can type "mexico" and get a match which returns "Mèxico" (among a lot of other possible hits, like "Café Mèxico, Dubai, UAE").

Now the thing is I don't have the ability to change the service to do any highlighting on the server side, therefore I am highlighting on the client JavaScript side with something like:

return result.replace( input.replace(/[aeiou]/g,"."), "<b>$1</b>");

It's a little more fancy because I am escaping special regex characters in the input. This is fine for simple one word matches at the beginning of a hit, but it really breaks down if you suddenly wish to support multi-word matches like "london cafe":

input = input.strip().toLowerCase(); //fyi prototype's strip is like trim
re = new RegEx(input.replace(/[aeiou]/g,".").replace(/\s+/g,"|"),"gi");
return result.replace(re, "<b>$1</b>");

This doesn't work for say "london ca" (was typing london cafe), because it would mark "Jack London Cabin, Dawson City, Canada" as: "Jack London cabin, Dawson City, Canada" [note the "ck" and "Ci" particularly]

Therefore I'm sort of looking for something that's not as crazy as:

input = input.strip().toLowerCase();
input = input.replace(/a/g,"[ÀàÁáÂâÃãÄäÅåÆæĀāĂăĄą]");
input = input.replace(/e


                    
                    
                        
                        javascript
                        
                        unicode
                        
                        normalization
                        
                        accent-insensitive
                        
                        text-normalization
                        
                    
                    
                        
                        
                        
                        
                        
                        Edit
                        
                        
                        
                            
                        
                        
                        Report

1 Answer

/**
 * Creates a RegExp that matches the words in the search string.
 * Case and accent insensitive.
 */
function make_pattern(search_string) {
    // escape meta characters
    search_string = search_string.replace(/([|()[{.+*?^$\\])/g,"\\$1");

    // split into words
    var words = search_string.split(/\s+/);

    // sort by length
    var length_comp = function (a,b) {
        return b.length - a.length;
    };
    words.sort(length_comp);

    // replace characters by their compositors
    var accent_replacer = function(chr) {
        return accented[chr.toUpperCase()] || chr;
    }
    for (var i = 0; i < words.length; i++) {
        words[i] = words[i].replace(/\S/g,accent_replacer);
    }

    // join as alternatives
    var regexp = words.join("|");
    return new RegExp(regexp,'g');
}

var accented = {
    'A': '[Aa\xaa\xc0-\xc5\xe0-\xe5\u0100-\u0105\u01cd\u01ce\u0200-\u0203\u0226\u0227\u1d2c\u1d43\u1e00\u1e01\u1e9a\u1ea0-\u1ea3\u2090\u2100\u2101\u213b\u249c\u24b6\u24d0\u3371-\u3374\u3380-\u3384\u3388\u3389\u33a9-\u33af\u33c2\u33ca\u33df\u33ff\uff21\uff41]',
    'B': '[Bb\u1d2e\u1d47\u1e02-\u1e07\u212c\u249d\u24b7\u24d1\u3374\u3385-\u3387\u33c3\u33c8\u33d4\u33dd\uff22\uff42]',
    'C': '[Cc\xc7\xe7\u0106-\u010d\u1d9c\u2100\u2102\u2103\u2105\u2106\u212d\u216d\u217d\u249e\u24b8\u24d2\u3376\u3388\u3389\u339d\u33a0\u33a4\u33c4-\u33c7\uff23\uff43]',
    'D': '[Dd\u010e\u010f\u01c4-\u01c6\u01f1-\u01f3\u1d30\u1d48\u1e0a-\u1e13\u2145\u2146\u216e\u217e\u249f\u24b9\u24d3\u32cf\u3372\u3377-\u3379\u3397\u33ad-\u33af\u33c5\u33c8\uff24\uff44]',
    'E': '[Ee\xc8-\xcb\xe8-\xeb\u0112-\u011b\u0204-\u0207\u0228\u0229\u1d31\u1d49\u1e18-\u1e1b\u1eb8-\u1ebd\u2091\u2121\u212f\u2130\u2147\u24a0\u24ba\u24d4\u3250\u32cd\u32ce\uff25\uff45]',
    'F': '[Ff\u1da0\u1e1e\u1e1f\u2109\u2131\u213b\u24a1\u24bb\u24d5\u338a-\u338c\u3399\ufb00-\ufb04\uff26\uff46]',
    'G': '[Gg\u011c-\u0123\u01e6\u01e7\u01f4\u01f5\u1d33\u1d4d\u1e20\u1e21\u210a\u24a2\u24bc\u24d6\u32cc\u32cd\u33


                        
                        
                            answered 2008-10-23T00:17:58.520

Programatic Accent Reduction in JavaScript (aka text normalization or unaccenting)

1 Answer

Your Answer