JavaScript - replace square brackets in a string

Try this code:

string.replace(/(\[~)(\w+)(\])/g, function(match, p1, p2, p3) {
  // p1 = [~
  // p2 = theme / media / whateverWord
  // p3 = ]
  return '@' + p2
  // Returns @whateverWord
})

In the 1st group:

• \[ will select a [
• ~ will select a ~

In the 2nd group:

• \w will select any alphanumeric character or an _
• The + states that the alphanumeric character must appear at least once, i.e. there must be at least 1 letter between the [~ and ]

In the 3rd group:

• \] will select any ]

In the function:

• match is not used in the output, but it contains the whole matched substring
• p1 contains the [~
• p2 contains the word between the [~ and ], i.e. theme or media
• p3 contains the ]

The return statement returns an @, followed by the word between the [~ and ]

This will replace all [~ with @

Here is a working example:

var string = 'Completely engineer client-based strategic [~theme] areas before cross [~media] technology. Also, [this tag] [will be kept]'

document.body.innerHTML = string.replace(/(\[~)(\w+)(\])/g, function(match, p1, p2, p3) {
  return '@' + p2
})

Edit: Actually, you can make it simpler:

string.replace(/(\[~)(\w+)(\])/g, '@$2')

Check out the demo below:

var string = 'Completely engineer client-based strategic [~theme] areas before cross [~media] technology. Also, [this tag] [will be kept]'

document.body.innerHTML = string.replace(/(\[~)(\w+)(\])/g, '@$2')

The $2 contains the contents of the second capture group, and the second capture group contains the text between the [~ and ]. So the output is an @, followed by the text.

This is simpler and faster than the version above, and takes up less space

You could use RegExp /(\[~)|(\])/g

var string = 'Completely engineer client-based strategic [~theme] areas before cross [~media] technology';

var res = string.replace(/(\[~)|(\])/g, function(match, p1, p2) {
  if (p1) return "@";
  if (p2) return ""
});

document.body.textContent = res;

The code below uses Regexp to find all group of sequences that start with [~ and end with ]. It also captures the word in between. $1 in the second parameter of replace function references the found word.

var string = 'Completely engineer client-based strategic [~theme] areas before cross [~media] technology'


document.body.innerHTML = string.replace(/\[~(\w+)\]/g, '@$1');

"Completely engineer client-based strategic [~theme] areas before cross [~media] technology [with some] other bits.".replace(/\[~([^\]]+)\]/g,"@$1");

You need to qualify the tilde in the search. I'm surprised at all the crazy down voting. People are trying to be helpful here. If someone has a problem with the post and the answers, it's more helpful to explain WHY you downvote and not just go willy-nilly on the down-votes without explaining yourself.

As others have said, google is a friend. Try searching through www.regular-expressions.info for more help.

I think this is what you are looking for https://jsfiddle.net/DIRTY_SMITH/7oe5kh9L/33/

string.replace(/\[~/g, '@').replace(/\]/g, '');

var string = 'Completely engineer client-based strategic [~theme] areas before cross [~media] technology';
alert("Starting string:" +string);

var res = string.replace(/\[~/g, '@').replace(/\]/g, '');
alert("Final string: "+res)