当使用字符串定义RegEx时,我想知道是否有一种方法可以使我的代码识别目录中包含的文件中的模式。

目标是使用我们的命名约定来重命名这些文件,因此我正在编写一些东西来尝试创建要在RegEx中使用的表达式。

我从这里开始,但是我不认为这是最好的,而且我不确定如何填写RegEx表达式的"{0}"部分。

    private Regex m_regex;

    public string DirPattern(string path, string[] extensions) {
        string result = null;
        int endPos = 0;
        int resLen = 0;
        int startLen = 0;
        var dir = new DirectoryInfo(path);
        foreach (var file in dir.GetFiles()) {
            if (extensions.Contains(file.Extension)) {
                if (!String.IsNullOrEmpty(result)) {
                    int sL = 0;
                    int fileLen = file.Name.Length;
                    string one = null;
                    for (int i = 0; i < resLen && i < fileLen; i++) {
                        if (result[i] == file.Name[i]) {
                            sL = i + 1;
                            if (String.IsNullOrEmpty(one)) {
                                one = file.Name;
                            } else {
                                break;
                            }
                        }
                    }
                    if (!String.IsNullOrEmpty(one)) {
                        int eP = 0;
                        int oneLen = one.Length;
                        for (int i = fileLen - 1; -1 < i; i--) {
                            if (result[i] == file.Name[i]) {
                                eP = i - 1;
                            } else {
                                break;
                            }
                        }
                        if ((0 < endPos) && (eP == endPos)) {
                            if ((0 < startLen) && (sL == startLen)) {
                                result = one.Substring(0, startLen) + "{0}" + one.Substring(endPos);
                            } else if (0 < sL) {
                                startLen = sL;
                            }
                        } else if (0 < sL) {
                            startLen = sL;
                        }
                    }
                } else {
                    result = file.Name;
                    resLen = result.Length;
                }
            }
        }
        return result;
    }

    public bool GenerateRexEx(string path, string[] extensions) {
        var pattern = DirPattern(path, extensions);
        if (!String.IsNullOrEmpty(pattern)) {
            m_regex = new Regex(pattern);
            return true;
        }
        return false;
    }


这是最类似于我们公司文件的文件列表的示例(不允许发布):



更新:

目的是获取具有以下名称的文件:

FOLDER_PATTERN_1 + MixedContent + FOLDER_PATTERN_2


并使用我们的格式将其重命名:

OUR_PATTERN_1 + MixedContent + OUR_PATTERN_2


这样,我们的软件将能够更有效地搜索文件。

最佳答案

我认为,在您的情况下,您只需要查找前缀模式和后缀模式中的字符数即可。然后,您可以简单地用图案替换一些字符。我写了一个简单的代码,我测试并工作了。您可以激发自己,并使用我认为的相同方法。无论如何,都有一些地方可以使它变得更好,但是我希望它足以回答您的问题。

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Text.RegularExpressions;
using System.Threading.Tasks;

namespace ConsoleApplication1
{
    static class Program
    {
        static void Main()
        {
            var inputFilenames = new string[]
            {
                "mtn_flint501-muxed",
                "mtn_flint502-muxed",
                "mtn_flint503-muxed",
                "mtn_flint504-muxed",
                "mtn_flint505-muxed",
                "mtn_flint506-muxed",
                "mtn_flint507-muxed",
                "mtn_flint508-muxed",
                "mtn_flint509-muxed",
                "mtn_flint510-muxed",
                "mtn_flint511-muxed",
                "mtn_flint512-muxed",
            };
            var replacedFilenames = ReplaceFileNames(inputFilenames);

            for (int i = 0; i < inputFilenames.Length; i++)
            {
                Console.WriteLine("{0} >> {1}", inputFilenames[i], replacedFilenames[i]);
            }
            Console.ReadKey();
        }

        private const string OurPrefixPattern = "Prefix_";
        private const string OurPostfixPattern = "_Postfix";

        /// <summary>
        /// Method which will find the filename's pattern and replace it with our pattern
        /// </summary>
        /// <param name="fileNames"></param>
        /// <returns></returns>
        public static string[] ReplaceFileNames(params string[] fileNames)
        {
            //At first, we will find count of characters, which are same for
            //all filenames as prefix and store it to prefixCount variable and
            //we will find count of characters which are same for all filenames
            //as postfix and store it to postfixCount variable
            var prefixCount = int.MaxValue;
            var postfixCount = int.MaxValue;
            //We will use first filename as the reference one (we will be comparing)
            //all filenames with this one
            var referenceFilename = fileNames[0];
            var reversedReferenceFilename = referenceFilename.ReverseString();
            //Lets find the prefixCount and postfixCount
            foreach (var filename in fileNames)
            {
                if (filename == referenceFilename)
                {
                    continue;
                }
                //Check for prefix count
                var firstDifferenceIndex = referenceFilename.GetFirstDifferentIndexWith(filename);
                if (firstDifferenceIndex < prefixCount)
                {
                    prefixCount = firstDifferenceIndex;
                }
                //For postfix count we will do the same, but with reversed strings
                firstDifferenceIndex = reversedReferenceFilename.GetFirstDifferentIndexWith(filename.ReverseString());
                if (firstDifferenceIndex < postfixCount)
                {
                    postfixCount = firstDifferenceIndex;
                }
            }
            //So now replace given filnames with our prefix and post fix.
            //Our regex determines only how many characters should be replaced
            var prefixRegexToReplace = string.Format("^.{{{0}}}", prefixCount);
            var postfixRegexToReplace = string.Format(".{{{0}}}$", postfixCount);
            var result = new string[fileNames.Length];
            for (int i = 0; i < fileNames.Length; i++)
            {
                //Replace the prefix
                result[i] = Regex.Replace(fileNames[i], prefixRegexToReplace, OurPrefixPattern);
                //Replace the postfix
                result[i] = Regex.Replace(result[i], postfixRegexToReplace, OurPostfixPattern);
            }
            return result;
        }

        /// <summary>
        /// Gets the first index in which the strings has different character
        /// </summary>
        /// <param name="value"></param>
        /// <param name="stringToCompare"></param>
        /// <returns></returns>
        private static int GetFirstDifferentIndexWith(this string value, string stringToCompare)
        {
            return value.Zip(stringToCompare, (c1, c2) => c1 == c2).TakeWhile(b => b).Count();
        }

        /// <summary>
        /// Revers given string
        /// </summary>
        /// <param name="value">String which should be reversed</param>
        /// <returns>Reversed string</returns>
        private static string ReverseString(this string value)
        {
            char[] charArray = value.ToCharArray();
            Array.Reverse(charArray);
            return new string(charArray);
        }
    }
}


控制台输出看起来像这样


  mtn_flint501-muxed >> Prefix_01_Postfix
  
  mtn_flint502-muxed >> Prefix_02_Postfix
  
  mtn_flint503-muxed >> Prefix_03_Postfix
  
  mtn_flint504-muxed >> Prefix_04_Postfix
  
  mtn_flint505-muxed >> Prefix_05_Postfix
  
  mtn_flint506-muxed >> Prefix_06_Postfix
  
  mtn_flint507-muxed >> Prefix_07_Postfix
  
  mtn_flint508-muxed >> Prefix_08_Postfix
  
  mtn_flint509-muxed >> Prefix_09_Postfix
  
  mtn_flint510-muxed >> Prefix_10_Postfix
  
  mtn_flint511-muxed >> Prefix_11_Postfix
  
  mtn_flint512-muxed >> Prefix_12_Postfix

关于c# - 检测文件名模式以创建RegEx,我们在Stack Overflow上找到一个类似的问题:https://stackoverflow.com/questions/25453438/

10-09 01:00