CODEWITHSUNDEEP
regexexpression
SergyL
SergyL

Reputation: 11

How to use regular expression to validate wrong symbol

I'm completely new and have a question. I have an XML parameter form file with field validation like this:

ValidChars="(?=.\*[a-zA-Z0-9@#$%*])" ErrorOnValidationFailure="Entered Invalid Char."

The current expression ((?=.*[a-zA-Z0-9@#$%*])) can generate an error only if the first char is incorrect. Once I enter a valid character, the validation stops working. Is there a regex that will enable me to identify an invalid character in any part of the string?

Thanks

Upvotes: 1

Views: 387

Answers (1)

Wiktor Stribiżew
Wiktor Stribiżew

Reputation: 627190

The point is that (?=.*[a-zA-Z0-9@#$%*]) regex executed once only asserts if the input string contains a character from the specified ranges/sets defined in the character class after 0+ characters (in NFA regex, other than a newline/carriage return/etc. - depends on the regex library).

What you need to use is a regex to check if the whole string consists of your whitelisted characters, so you need a 

^[a-zA-Z0-9@#$%*]*$

See the regex demo

The $ might be replaced with \Z in Python, or \z in PCRE/.NET regex to disallow trailing newline at the end of the string to be accepted.

If the pattern is used by a method that anchors the pattern by default, you need no ^, $, \A, nor \z/\Z.

Upvotes: 1

Related Questions