"Word Mode" - The definition of a 'word' about google-diff-match-patch HOT 6 CLOSED

This is why DMP doesn't have a built-in word mode; no two people can agree on 
what a word is, what to do with punctuation, etc.  By writing your own function 
as the linked page describes, you get to define your own delimiters.

To implement your desired behaviour, you might want something like this in your 
diff_linesToWords function:

if (text.charAt(0).match(/\s/) {
  // The next token will be whitespace.
  lineEnd = text.indexOf(/\S/, lineStart);
} else {
  // The next token will be a word.
  lineEnd = text.indexOf(/\s/, lineStart);
}

Oops, I forgot that text isn't being chopped at each iteration.  That first 
line should be:

if (text.charAt(lineStart).match(/\s/) {

Currently working on something else but I will get back to my issue and try 
this out sometime soon. Thanks a lot for the help, very much appreciated!

I'm trying to use this but am not getting desired results...

1. First off, I don't think indexOf() takes regex, not according to 
specification anyway apparently. I am using a custom "regexIndexOf()" function 
i found online, here it is:
===CODE===
String.prototype.regexIndexOf = function(regex, startpos) {
 var indexOf = this.substring(startpos || 0).search(regex);
 return (indexOf >= 0) ? (indexOf + (startpos || 0)) : indexOf;
}
===/CODE===


2. It doesn't seem to work, I still see whitespace (in my case ' ' space 
character) considered part of its adjacent previous word. To illustrate, 
consider this string "Hello world people"
I need DMP to diff it by words as follows:
[Hello] [ ] [world] [ ] [people]
Instead, I get:
[Hello ] [world ] [people]

Thanks in advance for your time if you get around to this... :)

*
(when I said "it doesn't seem to work" I am referring to your snippet you 
posted above, including your follow-up correction)

Hm... Why not just Add string/array with word delimiters and enum such like:

Enum TreatWord
{
DelimeterOnStart
DelimeterOnEnd
DelimeterOnBothSides
}

And depending on this make diff for words. This will cover all situations "on 
what a word is".

By the way, the linked example is obsolete and for adding word diff 
functionality you have to change two methods as I can remember.