c# Regex.Replace [^\w ] that also removes underscores?

Question

So I have spent far too long on this and have tried tons of things with no luck. I think I am just bad at regex. I am trying to clean a string of ALL non alpha numeric characters but leaving spaces. I DO NOT WANT TO USE [^A-Za-z0-9 ]+ due language concerns.

Here are a few things I have tried:

cleaned_string = Regex.Replace(input_string, @"[^\w ]+[_]+);

cleaned_string = Regex.Replace(input_string, ([^\w ]+)([_]+));

cleaned_string = Regex.Replace(input_string, [^ \w?<!_]+);

Edit: Solved thanks to a very helpful person below.

My final product ended up being this: [_]+|[^\w\s]+

Thanks for all the help!

Answer 1

This should work for you

// Expression: _|[^\w\d ]
cleaned_string = Regex.Replace(input_string, @"/_|[^\w\d ]", "");

Answer 2

You may use

var res = Regex.Replace(s, @"[\W_-[\s]]+", string.Empty);

See the regex demo.

Look at \W pattern: it matches any non-word chars. Now, you want to exclude a whitespace matching pattern from \W - use character class subtraction: [\W-[\s]]. This matches any char \W matches except what \s matches. And to also match a _, just add it to the character class. Add + quantifier to remove whole consecutive chunks of matching chars at one go.

Details

• [ - start of a character class
  • \W_ - any non-word or _ chars
  • -[\s] - except for chars matched with \s (whitespace) pattern
• ] - end of the character class
• + - one or more times.