Search code examples
regex.net

How do I check if a filename matches some wildcard pattern?


I've got a wildcard pattern, perhaps "*.txt" or "POS??.dat".

I also have list of filenames in memory that I need to compare to that pattern.

How would I do that, keeping in mind I need exactly the same semantics that IO.DirectoryInfo.GetFiles(pattern) uses?

Blindly translating this into a regex will not work.


Solution

  • I have a complete answer in code for you that's 95% like FindFiles(string).

    The 5% that isn't there is the short names/long names behavior in the second note on the MSDN documentation for this function.

    If you would still like to get that behavior, you'll have to complete a computation of the short name of each string you have in the input array, and then add the long name to the collection of matches if either the long or short name matches the pattern.

    Here is the code:

    using System;
    using System.Collections.Generic;
    using System.Linq;
    using System.Text;
    using System.Text.RegularExpressions;
    
    namespace FindFilesRegEx
    {
        class Program
        {
            static void Main(string[] args)
            {
                string[] names = { "hello.t", "HelLo.tx", "HeLLo.txt", "HeLLo.txtsjfhs", "HeLLo.tx.sdj", "hAlLo20984.txt" };
                string[] matches;
                matches = FindFilesEmulator("hello.tx", names);
                matches = FindFilesEmulator("H*o*.???", names);
                matches = FindFilesEmulator("hello.txt", names);
                matches = FindFilesEmulator("lskfjd30", names);
            }
    
            public string[] FindFilesEmulator(string pattern, string[] names)
            {
                List<string> matches = new List<string>();
                Regex regex = FindFilesPatternToRegex.Convert(pattern);
                foreach (string s in names)
                {
                    if (regex.IsMatch(s))
                    {
                        matches.Add(s);
                    }
                }
                return matches.ToArray();
            }
    
            internal static class FindFilesPatternToRegex
            {
                private static Regex HasQuestionMarkRegEx   = new Regex(@"\?", RegexOptions.Compiled);
                private static Regex IllegalCharactersRegex  = new Regex("[" + @"\/:<>|" + "\"]", RegexOptions.Compiled);
                private static Regex CatchExtentionRegex    = new Regex(@"^\s*.+\.([^\.]+)\s*$", RegexOptions.Compiled);
                private static string NonDotCharacters      = @"[^.]*";
                public static Regex Convert(string pattern)
                {
                    if (pattern == null)
                    {
                        throw new ArgumentNullException();
                    }
                    pattern = pattern.Trim();
                    if (pattern.Length == 0)
                    {
                        throw new ArgumentException("Pattern is empty.");
                    }
                    if(IllegalCharactersRegex.IsMatch(pattern))
                    {
                        throw new ArgumentException("Pattern contains illegal characters.");
                    }
                    bool hasExtension = CatchExtentionRegex.IsMatch(pattern);
                    bool matchExact = false;
                    if (HasQuestionMarkRegEx.IsMatch(pattern))
                    {
                        matchExact = true;
                    }
                    else if(hasExtension)
                    {
                        matchExact = CatchExtentionRegex.Match(pattern).Groups[1].Length != 3;
                    }
                    string regexString = Regex.Escape(pattern);
                    regexString = "^" + Regex.Replace(regexString, @"\\\*", ".*");
                    regexString = Regex.Replace(regexString, @"\\\?", ".");
                    if(!matchExact && hasExtension)
                    {
                        regexString += NonDotCharacters;
                    }
                    regexString += "$";
                    Regex regex = new Regex(regexString, RegexOptions.Compiled | RegexOptions.IgnoreCase);
                    return regex;
                }
            }
        }
    }