Question 277088 in Javascript Question posted in
A very good W3school tutorial can be found here.

Javascript – Regex accept only Latin (only English ) characters and remove extended ASCII characters

I have the regex below

^(?=.*[A-Z])(?=.*[a-z])(?=.*d)(?=.*W)[A-Za-z0-9W][x00-x7F]{8,}$

am expecting from the regex the below

  • not less than 8 in length
  • at least one capital letter
  • at least one small letter
  • at least one digit
  • at least one non-alphanumeric symbol
  • all characters should be Latin only (no extended char like Ş in Turkish or special letters in German like Ü and no Arabic or such languages )

what am facing is the regex does match many strings, but, as a string like 123aB!ВЕ it’s not matching. Another example is the string June!12 It does not match either, but it could be matched if extended to be 9 letters in length June!12sd ???

for example, Fth4&yusÖ should not match because of the letter Ö but when removed it should match, but, it is not matching unless it’s extended to be more than 9 characters.

any help is appreciated

Tags:

2

Answers


  1. 0

    It sounds like you’re just doing password validation, and regex is a tool you choose to use rather than a requirement. This sort of thing can be better (faster + easier to read) without regex.

    Here’s a C# solution without using regex.

    private static bool IsPasswordValid(string input)
{
    if (input.Length < 8)
        return false;

    bool hasUppercaseLetter = false;
    bool hasLowercaseLetter = false;
    bool hasDigit = false;
    bool hasNonAlphaNumeric = false;

    foreach (var c in input)
    {
        // non-ASCII means immediate failure
        if (!char.IsAscii(c))
            return false;

        else if (char.IsAsciiDigit(c))
            hasDigit = true;

        else if (char.IsAsciiLetterUpper(c))
            hasUppercaseLetter = true;

        else if (char.IsAsciiLetterLower(c))
            hasLowercaseLetter = true;

        else if (c is '!' or '@' or '#' or '$' or '%' or '^' or '&' or '*') // add more as needed
            hasNonAlphaNumeric = true;
    }

    return hasUppercaseLetter && hasLowercaseLetter && hasDigit && hasNonAlphaNumeric;
}

    This works by processing each character one by one and updating some booleans based on if it passes a rule or not. Because you don’t have any "max number of …" rules, simply setting the boolean to true every time you have a matching character is fine.

    It passes all your examples:

    • 123aB!ВЕ fails because the last two characters are Cyrillic.
    • 123aB!BE passes – this is an altered version of your example to make the last two characters use Latin letters.
    • June!12 fails as it less than 8 characters
    • June!12sd passes
    • Fth4&yusÖ fails because of the last character.
    Login or Signup to reply.
  2. 0

    This is a good way to do your specs.
    Note that white space is excluded as are other control codes.

    ^(?=[x21-x7e]{8,}$)(?=.*[W_])(?=.*[A-Z])(?=.*[a-z])(?=.*d).+$

    Explained

    ^ 
(?= [x21-x7e]{8,} $ )       # ASCII only x21-x7e range 8 or more
(?= .* [W_] )                # 1 Special Character 
(?= .* [A-Z] )                # 1  A-Z Character
(?= .* [a-z] )                # 1  a-z  Character
(?= .* d )                   # 1  Digit Character
.+                            # Passed get them all
$
    Login or Signup to reply.
Please signup or login to give your own answer.

Back To Top
Search