Regex to remove a specific repeated character

Question

I would like to create a regex in C# that removes a specific character if it is repeated and so it is not the last character of the string.

Example:

"a--b-c-" => "a-b-c"
"-a-b--c" => "a-b-c"
"--a--b--c--" => "a-b-c"

I never want the - repeated, and I never want it to be the first or last character of my string. How could I write a regex to do this?

Answer 1

Probably easiest to do this in two steps. First replace each occurrence of one or more "-" with a single "-", then trim any leading/trailing "-".

var reducedString = Regex.Replace(inputString, "-+", "-");

var finalString = reducedString.Trim('-');

Answer 2

For this specific problem, I'd probably not use a regex. Instead, I'd probably use a combination of String.Split and String.Join, which will be simpler and likely faster:

Like this:

string.Join("-", s.Split(new char[] {'-'}, StringSplitOptions.RemoveEmptyEntries));

With tests:

using System;
class Program
{
    static string RemoveDashes(string s)
    {
        return string.Join("-", s.Split(new char[] { '-' }, 
                            StringSplitOptions.RemoveEmptyEntries));
    }
    static void Main(string[] args)
    {
        Tuple<string, string>[] tests = new Tuple<string,string> [] 
        {
            new Tuple<string, string> ("a--b-c-", "a-b-c"),
            new Tuple<string, string> ("-a--b-c-", "a-b-c"),
            new Tuple<string, string> ("--a--b--c--", "a-b-c"),
        };
        foreach (var t in tests)
        {
            string s = RemoveDashes(t.Item1);
            Console.WriteLine("{3}: {0} => Expected: {1}, Actual: {2}", 
                        t.Item1, t.Item2, s, s == t.Item2 ? "PASS" : "FAIL");
        }
    }
}

Answer 3

string tidyText = Regex.Replace(originalText, "^-+|(?<=-)-+|-+$", "");