0

在以下函数（完美运行）中，我现在面临的挑战是让它不仅返回找到匹配项的位置，而且返回匹配项是什么......代码：

txtFilePattern 是一个管道分隔的文件扩展名列表。txtKeywords 是我正在寻找的关键字的多行文本框 txtPatterns 与 txtKeywords 相同，但适用于正则表达式模式。

这是我自己对 C# Grep 的小实验。

private List<Tuple<String, Int32, String>> ScanDocuments2()
    {
        Regex searchPattern = new Regex(@"$(?<=\.(" + txtFilePattern.Text + "))", RegexOptions.IgnoreCase);
        string[] keywordtext = txtKeywords.Lines;
        List<string> keywords = new List<string>();
        List<Regex> patterns = new List<Regex>();
        for (int i = 0; i < keywordtext.Length; i++)
        {
            if (keywordtext[i].Length > 0)
            {
                keywords.Add(keywordtext[i]);
            }
        }
        string[] patterntext = txtPatterns.Lines;
        for (int j = 0; j < patterntext.Length; j++)
        {
            if (patterntext[j].Length > 0)
            {
                patterns.Add(new Regex(patterntext[j]));
            }
        }
        try
        {
            var files = Directory.EnumerateFiles(txtSelectedDirectory.Text, "*.*", SearchOption.AllDirectories).Where(f => searchPattern.IsMatch(f));
            //fileCount = files.Count();
            var lines = files.Aggregate(
                   new List<Tuple<String, Int32, String>>(),
                   (accumulator, file) =>
                   {
                       fileCount++;
                       using (var reader = new StreamReader(file))
                       {
                           var counter = 0;
                           String line;
                           while ((line = reader.ReadLine()) != null)
                           {
                               if (keywords.Any(keyword => line.ToLower().Contains(keyword.ToLower())) || patterns.Any(pattern => pattern.IsMatch(line)))
                               {
                                   //cleans up the file path for grid
                                   string tmpfile = file.Replace(txtSelectedDirectory.Text, "..");
                                   accumulator.Add(Tuple.Create(tmpfile, counter, line));
                               }
                               counter++;
                           }
                       }
                       return accumulator;
                   },
                   accumulator => accumulator
            );
            return lines;
        }
        catch (UnauthorizedAccessException UAEx)
        {
            Console.WriteLine(UAEx.Message);
            throw UAEx;
        }
        catch (PathTooLongException PathEx)
        {
            Console.WriteLine(PathEx.Message);
            throw PathEx;
        }
    }

问题是 - 我如何确定通过哪个关键字或模式与我返回的元组匹配？

4

2 回答 2

1

您如何引入一个新变量来保存匹配模式，并使用FirstOrDefault而不是Any. 然后，只要新变量不是null你有匹配的模式，你就可以在你的元组中返回它。

例如

...
        new List<Tuple<String, Int32, String, Regex>>()
...
                       while ((line = reader.ReadLine()) != null)
                       {
                           Regex matchingReg = patterns.FirstOrDefault(pattern => pattern.IsMatch(line));

                           if (keywords.Any(keyword => line.ToLower().Contains(keyword.ToLower())) || matchingReg != null)
                           {
                               //cleans up the file path for grid
                               string tmpfile = file.Replace(txtSelectedDirectory.Text, "..");
                               accumulator.Add(Tuple.Create(tmpfile, counter, line, matchingReg));
                           }
                           counter++;
                       }
...
于 2013-02-27T18:22:34.090 回答
1

这是一些重构的代码。肯尼斯的想法是对的。

private IEnumerable<LineMatch> ScanDocuments2()
{
    string[] keywordtext = txtKeywords.Lines;
    string[] patterntext = txtPatterns.Lines;

    Regex searchPattern = GetSearchPattern();
    var keywords = GetKeywords(keywordtext).ToList();
    var patterns = GetPatterns(patterntext).ToList();

    try
    {
        var files = GetFiles(searchPattern);

        var lines = files.Aggregate(
               new List<LineMatch>(),
               (accumulator, file) =>
               {

                foreach(var item in EnumerateFile(file, keywords, patterns))
                {
                  accumulator.Add(item);
                }

                   return accumulator;
               },
               accumulator => accumulator
        );
        return lines;
    }
    catch (UnauthorizedAccessException UAEx)
    {
        Console.WriteLine(UAEx.Message);
        throw;
    }
    catch (PathTooLongException PathEx)
    {
        Console.WriteLine(PathEx.Message);
        throw;
    }
}

private LineMatch EnumerateFile(string file, IEnumerable<string> keywords, IEnumerable<Regex> patterns)
{
  var counter = 0;

  foreach(var line in File.ReadLines(file))
  {
    var matchingRegex = patterns.FirstOrDefault(p => p.IsMatch(line));
    var keyword = keywords.FirstOrDefault(k => line.ToLower().Contains(k.ToLower()));

    if(keyword == null && matchingRegex == null) continue;

    string tmpfile = file.Replace(txtSelectedDirectory.Text, "..");

    yield return new LineMatch
    {
        Counter = counter,
        File = tmpfile,
        Line = line,
        Pattern = matchingRegex == null ? null : matchingRegex.Pattern,
        Keyword = keyword
    };

    counter++;
  }
}

private IEnumerable<string> GetFiles(Regex searchPattern)
{
  return Directory.EnumerateFiles(txtSelectedDirectory.Text, "*.*", SearchOption.AllDirectories).Where(f => searchPattern.IsMatch(f));
}

private IEnumerable<string> GetKeywords(IEnumerable<string> keywordtext)
{
  foreach(var keyword in keywordtext)
  {
    if(keyword.Length <= 0) continue;

    yield return keyword;
  }
}

private IEnumerable<string> GetPatterns(IEnumerable<string> patterntext)
{
  foreach(var pattern in patterntext)
  {
    if(pattern.Length <= 0) continue;

    yield return new Regex(pattern);
  }
}

private Regex GetSearchPattern()
{
  return new Regex(string.Format(@"$(?<=\.({0}))",  txtFilePattern.Text), RegexOptions.IgnoreCase);
}

public class LineMatch
{
  public int Counter { get; set; }
  public string File { get; set; }
  public string Line { get; set; }
  public string Pattern { get; set; }
  public string Keyword { get; set; }
}
于 2013-02-27T20:33:14.290 回答