Question

I have a textarea with unfiltered user input, which includes line returns, spaces, punctuation marks, etc. I would like to get all the distinct lowercased words, their occurrence, sorted by occurrence. I haven’t found a straight forward way to extract words when the strip() string is variable. Any ideas how to achieve this?

For example:

WORD1 Word2 word1
Word1, …
word2 HELLO …
. . hello .hi

would become

val array = {
    word1 : 3,
    word2 : 2,
    hello : 2,
       hi : 1
};

Thanks for your help!

Answer 1

Here is one possible solution:

var result = {},
    value = $("textarea").val(),
    res = value.match(/\b([a-z0-9]+)\b/g) || [];

for (var i = 0; i < res.length; i++) {
    result[res[i]] = (value.match(new RegExp(res[i], "ig")) || []).length;
}

console.log(result);​

DEMO: http://jsfiddle.net/mmFgE/