Question

I’m trying to split words up so I can count the syllable in each one, but i can’t seem to isolate the words at the moment, i can only count the whole number of syllables in the sentence,

        var text = $(this).val();
        var words = new Array(text.replace(/\s/g, ' ').split(' '));
        function countSyllables(words) {
            word = words.toString();
            var lWords = word.toLowerCase();
            var wordsArray=lWords.split(",");
            for (var z = 0; z < lWords.length; z++) {
                console.log(wordsArray, "- array of words");
                console.log(wordsArray[z], "- one item[z]");

My problem seems to be that i have to convert the array to a string to lower it, but then i can’t refer to each individual word, in both cases, to count the syllables:

[16:04:02.726] ["this", "is", "test", "text", "", "the", "words", "need", "to", "be", "separately", "analysed.", "hello"] - array of words    
[16:04:02.726] o - one item[z] 

This is the last character, when i want to refer to the word

I’m sure it’s a simple solution but i can’t wrap my head around it

Answer 1

This line is wrong on many levels:

var words = new Array(text.replace(/\s/g, ' ').split(' '));

What you’re after is, AFAIK, something like this:

var words = text.split(/[^a-z0-9\-]+/gi);

This splits the string, using all white-space, dots, brackets and other punctuation-things as delimiters. I’ve excluded the dash (-), to avoid counting words like ice-cream as two words. If you don’t care about that, or if you feel that the result is more reliable when you do treat dashes as delimiters, just leave them out:

var words = text.split(/[^a-z0-9]+/gi);
console.log(text + ' is approx. ' + words.length + ' words long');

On the converting to lower-case thing: why not convert the entire text to lower-case from the off?

var words = text.toLowerCase().split(/[^a-z0-9\-]+/gi);

Oh, and the output of this code does not leave you with empty strings in the array:

'Foo... bar, Foobar and lorum ipsorum'.toLowerCase().split(/[^a-z0-9\-]+/gi);
//output: ["foo", "bar", "foobar", "and", "lorum", "ipsorum"]
//length returns 6, which I think is correct