Extract all allowed characters from a regular expression - regex

Question

I need to extract a list of all allowed characters from a given regular expression.
So for example, if the regex looks like this (some random example):
[A-Z]*\s+(4|5)+
the output should be
ABCDEFGHIJKLMNOPQRSTUVWXYZ45
(omitting the whitespace)
One obvious solution would be to define a complete set of allowed characters, and use a find method, to return the corresponding subsequence for each character. This seems to be a bit of a dull solution though.
Can anyone think of a (possibly simple) algorithm on how to implement this?

Answer 1

One thing you can do is:
split the regex by subgroup
test the char panel against the subgroup
See the following example (not perfect yet) c#:
static void Main(String[] args)
{
Console.WriteLine($"-->{TestRegex(#"[A-Z]*\s+(4|5)+")}<--");
}
public static string TestRegex(string pattern)
{
string result = "";
foreach (var subPattern in Regex.Split(pattern, #"[*+]"))
{
if(string.IsNullOrWhiteSpace(subPattern))
continue;
result += GetAllCharCoveredByRegex(subPattern);
}
return result;
}
public static string GetAllCharCoveredByRegex(string pattern)
{
Console.WriteLine($"Testing {pattern}");
var regex = new Regex(pattern);
var matches = new List<char>();
for (var c = char.MinValue; c < char.MaxValue; c++)
{
if (regex.IsMatch(c.ToString()))
{
matches.Add(c);
}
}
return string.Join("", matches);
}
Which outputs:
Testing [A-Z]
Testing \s
Testing (4|5)
-->ABCDEFGHIJKLMNOPQRSTUVWXYZ
? ? ???????? 45<--