Find all but the first occurrence of a character with REGEX

Question

I'm building a .Net application and I need to strip any non-decimal character from a string (excluding the first '.'). Essentially I'm cleaning user input to force a real number result.

So far I've been using online RegEx tools to try and achieve this in a single pass, but I'm not getting very far.

I wish to accomplish this:

asd123.asd123.123.123 = 123.123123123

Unfortunately I've only managed to get to the stage where

asd123.asd123.123.123 = 123.123.123.123

by using this code.

System.Text.RegularExpressions.Regex.Replace(str, "[^\.|\d]*", "")

But I am stuck trying to remove all but the first decimal-point.

Can this be done in a single pass?
Is there a better-way™?

Answer 1

This can be done in a single regex, at least in .NET which supports infinite repetition inside lookbehind assertions:

resultString = Regex.Replace(subjectString, @"(?<!^[^.]*)\.|[^\d.]", "");

Explanation:

(?<!^[^.]*) # Either match (as long as there is at least one dot before it)
\.          # a dot
|           # or
[^\d.]      # any characters except digits or dots.

(?<!^[^.]*) means: Assert that it's impossible to match a string that starts at the beginning of the input string and consists solely of characters other than dots. This condition is true for all dots following the first one.