How to catch the remainder of a regex .match()?

Question

Is it possible, in JavaScript, to catch the remainder of a regex .match()?

For example:

var string = "505 Swanton St";

var m = string.match(/\d+/);

I can access "505" with m[0], but there are no other elements in m's array.

Is there a way to catch the remainder of string?

Answer 1

You can catch everything before and after the pattern using capture groups:

var string = "505 Swanton St";

var m = string.match(/^(.*?)(\d+)(.*)$/);

Here:

m[0] == "505 Swanton St";  // the entire matched pattern
m[1] == "";                // first group
m[2] == "505";             // second group
m[3] == " Swanton St";     // third group

var string = "505 Swanton St";

var m = string.match(/^(.*?)(\d+)(.*)$/);

console.log(JSON.stringify(m));

Answer 2

string.match and regexp.exec return special arrays with the added index property which is the index of the first character of the match. So you could also get the remainder with:

var string = "505 Swanton St";
var m = string.match(/\d+/);
var remainder = string.substring(0, m.index) + string.substring(m[0].length);

If used with a regex with the /g flag and the regex.exec function instead, you'll need to keep concatenating parts the between the end of a match and the beginning of the next.

function parseInput(input) {
  parseInput.regex.lastIndex = 0;
  var matches = [];
  var remainder = "";
  var prevIndex = 0;

  while((match = parseInput.regex.exec(input)) !== null) {
    matches.push(match);
    remainder += input.substring(prevIndex, match.index);
    prevIndex = parseInput.regex.lastIndex;
  }

  if(matches.length) {
    remainder += input.substring(prevIndex); // after last result
    return {"matches": matches, "remainder": remainder};
  } else {
    return {"matches": null, "remainder": input};
  }
}
parseInput.regex = /\d+/g;

I used this with a much more complex regex to parse all kinds of units. If there's an extraneous decimal point, my regex won't grab it and it will be in the remainder, so I can tell there's an error in the input by simply looking for points in the remainder.

* EDIT: Fixed the parseInput function so it also gets the remainder after the last match *