Question 50664 in Javascript Question posted in
A very good W3school tutorial can be found here.

regex to remove accents, special caracteres but preserve dash, underline and extension – Javascript

I’m trying to remove accents and special caracteres except dash(-), underline(_) and preserve the extension the string for exemple:

ÁÉÍÓÚáéíóúâêîôûàèìòùÇãç.,~!@#$%&_-12345.png

to:

AEIOUaeiouaeiouaeiouCac_-12345.png

I came to this result but the problem is that it’s ignoring all dots. I need to ignore only last occurence to preserve the extension from filename.

"ÁÉÍÓÚáéíóúâêîôûàèìòùÇãç.,~!@#$%&-12345.png".normalize(‘NFD’).replace(/[^a-zA-Z0-9-]/g, "")

I already tried negative look behind like this:

/[^a-zA-Z0-9-]+(?<!.)/g

using this reference but I didn’t have success.

"ÁÉÍÓÚáéíóúâêîôûàèìòùÇãç.,~!@#$%&-12.34.5.png".normalize('NFD').replace(/[^a-zA-Z0-9-]+(?<!.)/g, '')

If i have more than a dot in this case it only removes the first .

Tags:

3

Answers


  1. 0

    Instead of checking every charecter for being a file extension, select the whole extension at once

    let nfd = "ÁÉÍÓÚáéíóúâêîôûàèìòùÇãç.,~!@#$%&-12.34.5.png".normalize('NFD')

let exttest = nfd.replace(/(?<extension>.[^.]+$|)(?<badchar>.?)/g, '$1')
console.log({ exttest }) // { "exttest": ".png" }

let result = nfd.replace(/(?<extension>.[^.]+$|)(?<badchar>[^a-zA-Z0-9-]?)/g, '$1')
console.log({ result }) // { "result": "AEIOUaeiouaeiouaeiouCac-12345.png" }

    ((?<name>blah) is just a named (blah) group, it’s named just for explaining)

    Login or Signup to reply.
  2. 0

    A negative lookahead that excludes any . followed by any letter or number which in turn is followed by a non-word character can work.

    /[^a-zA-Z0-9-._]|.(?![a-zA-Z0-9]+b)/g

    An alternative to [^a-zA-Z0-9-._] is [^w.-].

    RegEx101

    Explanation

    Segment Description 
    [^a-zA-Z0-9-._]
    		 Exclude any letter, number, hyphen, underscore, and dot 
    |
    		 OR 
    .(?![a-zA-Z0-9]+b)
    		 Exclude any dot that is NOT followed by one or more letters and/or numbers followed by a non-word character

    Example

    const rgx = /[^a-zA-Z0-9-._]+|.(?![a-zA-Z0-9]+b)/g;

const str = `ÁÉÍÓÚáéíóúâêîôûàèìòùÇãç.,~!@#$%&_-12345.png`;

const file = str.replace(rgx, "");

console.log(file);
    Login or Signup to reply.
  3. 0

    In your pattern you forgot to exclude the _ as you want to keep that in the result.

    You are using a negative lookbehind that asserts that from the current position, there is not a dot directly to the left.

    The negated character class [^a-zA-Z0-9-]+ can match a dot, but the lookbehind (?<!.) fails if it did match a dot, so it will never match any dot.

    What you could do, is to match this character class [^a-zA-Z0-9-]+, but assert that from the current position to the right there is a dot, followed by 1+ chars except a dot or a whitespace char till the end of the string using a positive lookahead (?=.*.[^s.]+$)

    const result = "ÁÉÍÓÚáéí...óúâêîôûàèìòùÇãç.,~!@#$%&_-12345.png"
  .normalize('NFD')
  .replace(/[^a-zA-Z0-9_-]+(?=.*.[^s.]+$)/g, '');
console.log(result);
    Login or Signup to reply.
Please signup or login to give your own answer.

Back To Top
Search