仅匹配给定字符串中单词序列中的字符

Question

I am trying to find a closest match for a word by giving a specific string, for example:

so I would have:

"jonston" x "john"  => "jo" //only "jo" is the part that matches
"joshua" x "john" => "jo" 
"mark" x "marta"    => "mar"

as you can see I only would like to retrieve the characters in sequence matching, that's why joshua and john only would have jo in common sequence and not joh since both have the letter h

I've tried that with regular expression by using the following:

"john".match(/["joshua"]+/) //=> outputs ["joh"] and not ["jo"]

is there any way I could match only the first chars that match?

I will be using javascript for the implementation

I hope that makes sense

Thanks in advance

Answer 1

var a = "john";
var b = "joshua";
var x = "";

for (var i = 0; i < a.length; i++) {
    if (x == "" && i > 0) break;
    else if (a[i] == b[i]) x += a[i];
    else if (x != "") break;
}

console.log(x);

DEMO: http://jsfiddle.net/jMuDm/

Answer 2

Yet another solution:

if(typeof String.prototype.commonFirstChars !== 'function') {
    String.prototype.commonFirstChars = function(s) {
        var common = "";
        for(var i=0; i<this.length; i++) {
            if(this[i] !== s[i]) {
                return common;
            }
            common += this[i];           
        }
    };
}

You can use it like this:

var commonFirstChars = "john".commonFirstChars("joshua");
// "john".commonFirstChars("joshua") === "joshua".commonFirstChars("john")

This will return:

jo

Answer 3

initLCS = function(a, b) {
    for (var i = 0; i < a.length && a[i] == b[i]; i++);
    return a.substr(0, i);
}


initLCS("jonston", "john") // jo
initLCS("jonston", "j111") // j
initLCS("xx", "yy") // ""

If you insist on using regular expressions, it goes like this:

initLCS = function(a, b) {

    function makeRe(x) {
        return x.length ? "(" + x.shift() + makeRe(x) + ")?" : "";
    }

    var re = new RegExp('^' + makeRe(b.split("")), "g");
    return a.match(re)[0];
}

This creates an expression like /^(j(o(h(n)?)?)?)?/g from the second string and applies it to the first one. Not that it makes much sense, just for the heck of it.

Answer 4

You can not really do this with regex. Why dont you just loop through both string and compare the indexes? You can select the chars until you hit a char at the same index with a different value.

Answer 5

I'd do this in a recursive function like this:

EDIT: Updated example to make it more readable.

var testWords = [
    ['ted', 'terminator'],
    ['joe', 'john'],
    ['foo', 'bar']
];

var matches = testWords.map(function(wordPair) {
    return (function matchChars(word1, word2, matches) {
        if (word1[0] !== word2[0]) { 
            return [wordPair[0], wordPair[1], matches];
        }

        matches = matches || '';
        matches += word1[0];
        return matchChars(word1.slice(1), word2.slice(1), matches);
    }(wordPair[0], wordPair[1]));
});


console.log(matches.map(function(match) { return match.join(', '); }).join('\n'));
​

Fiddle (updated): http://jsfiddle.net/VU5QT/2/