Javascript - Regex - how to filter characters that are not part of regex

Question

I want to accept words and some special characters, so if my regex does not fully match, let's say I display an error,

var re = /^[[:alnum:]\-_.&\s]+$/;
var string = 'this contains invalid chars like #@';
var valid = string.test(re);

but now I want to "filter" a phrase removing all characters not matching the regex ?

usualy one use replace, but how to list all characters not matching the regex ?

var validString = string.filter(re); // something similar to this

how do I do this ?

regards

Wiktor Stribiżew solution works fine :

regex=/[^a-zA-Z\-_.&\s]+/g;
let s='some bloody-test @rfdsfds';
s = s.replace(/[^\w\s.&-]+/g, '');
console.log(s);

Rajesh solution :

regex=/^[a-zA-Z\-_.&\s]+$/;
let s='some -test @rfdsfds';
s=s.split(' ').filter(x=> regex.test(x));
console.log(s);

JS regex engine does not support POSIX character classes like [:alnum:] (you may use [A-Za-z0-9] instead, but only to match ASCII letters and digits). you may try running s.match(/[^\w\s.&-]/g) to get the chars that do not match letters, digits, _, ., &, whitespace and -. — Wiktor Stribiżew
– Wiktor Stribiżew, Commented Oct 16, 2017 at 13:24
@Rajesh sadfully it does not filter anything, I modified my question — Phil
– Phil, Commented Oct 16, 2017 at 13:33

Wiktor Stribiżew · Accepted Answer · 2017-10-16 13:31:52Z

1

JS regex engine does not support POSIX character classes like [:alnum:]. You may use [A-Za-z0-9] instead, but only to match ASCII letters and digits.

Your current regex matches the whole string that contains allowed chars, and it cannot be used to return the chars that are not matched with [^a-zA-Z0-9_.&\s-].

You may remove the unwanted chars with

var s = 'this contains invalid chars like #@';
var res = s.replace(/[^\w\s.&-]+/g, '');
var notallowedchars = s.match(/[^\w\s.&-]+/g);
console.log(res);
console.log(notallowedchars);

The /[^\w\s.&-]+/g pattern matches multiple occurrences (due to /g) of any one or more (due to +) chars other than word chars (digits, letters, _, matched with \w), whitespace (\s), ., & and -.

edited Oct 16, 2017 at 13:31

answered Oct 16, 2017 at 13:29

Wiktor Stribiżew

631k41 gold badges502 silver badges632 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

Justinas Over a year ago

Why you need []+ if it's global replace?

Wiktor Stribiżew Over a year ago

@Justinas To remove matched streaks of chars in one go. It means fewer replacements made by the regex engine internally.

Justinas Over a year ago

And how much is that faster then simple replace (if faster at all)?

Wiktor Stribiżew Over a year ago

@Justinas Check this fiddle. It is faster with the + quantifier.

Justinas · Accepted Answer · 2017-10-16 13:28:32Z

0

To match all characters that is not alphanumeric, or one of -_.& move ^ inside group []

var str = 'asd.=!_#$%^&*()564';

console.log(
    str.match(/[^a-z0-9\-_.&\s]/gi),
    str.replace(/[^a-z0-9\-_.&\s]/gi, '')
);

answered Oct 16, 2017 at 13:28

Justinas

43.9k5 gold badges72 silver badges108 bronze badges

Collectives™ on Stack Overflow

Javascript - Regex - how to filter characters that are not part of regex

2 Answers 2

4 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

4 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related