Find and detect a pattern in a string

Question

I'm looking for a function that detects repeated patterns on a string, for example, if the input is:

var string = "HelloHelloHelloHelloHelloHello";
var string2 = "Hello this is a repeated pattern Hello this is a repeated pattern Hello this is a repeated pattern Hello this is a repeated pattern";

it should return true as there are many times repeated "Hello" or "Hello this is a repeated pattern"

I have tried something like this:

function detectPattern(string) {
 string = string.replace(/\s+/g,"_");
 return /(\S)(1円{15,})/g.test(string);
}

but that will only detect if the text contains the same character many times.

Any idea or suggestion?

• Wyck: Why do you need a regular expression for this? Are you open to the idea of other JavaScript-based solutions?
• Mathéo RB: Yup, If you have one I would be glad to read it!
• Wyck: Please note that the space character does not occur at the end of string 2 making it not a repeating pattern. Are you intending to treat the space character after the word pattern specially? e.g.: a a is not a repeating pattern just like axa is not a repeating pattern. But a a is, just like axax is.
• Mathéo RB: Well, that's true. By "repeated pattern" I meant a string2-like string, maybe I expressed wrongly myself. Also note that finding a repeated pattern as you said, will work for me.

Answer 1

A regex that would work is the following:

^(.+?)( ?1円)+$

This will match:

• ^: start of string
• (.+?): the least amount of characters (at least one), followed by
• ( ?1円)+: an optional space and the very same characters of the first group, multiple times
• $: end of string

In order to extract your repeated pattern, it's sufficient to extract the content of Group 1.

Check the demo here.

• Mathéo RB: For some reason when I execute this regex as: function detectPatterns(string){ let regex = /(.*?)( ?1円)+/g; if(string.match(regex)) return true; else return false; } it returns true allways for example with "Hello" or "Hello I'm Mathéo". But when I test it out in regex101 or regexr it's correct. Did I mispell something?
• lemon: I have updated the answer so that it will catch exactly only repeated characters from start to end, yet at least one. Does it solve your problem?
• Mathéo RB: Yep! That works perfectly. I have another related question with this, as there exist some words that by default are detected by this expression. Do you have any kind of idea that could help me to exempt these words without actually writing them one by one?
• lemon: Can you craft some examples? If you are talking about "Palindromic" words, then you are forced to assume there's always a space between your words, and in that case you would remove the ? after the space inside the regex. If there's no way for you to tell where a word ends and another starts, then there may nothing possible for that to be solved with a regex straight (as regex see words just as strings of characters).