Karinto/CsvReader.cs

   1 /*\r
   2  *      Karinto Library Project\r
   3  *\r
   4  *      This software is distributed under a zlib-style license.\r
   5  *      See license.txt for more information.\r
   6  */\r
   7 \r
   8 using System;\r
   9 using System.IO;\r
  10 using System.Collections.Generic;\r
  11 using System.Text;\r
  12 using System.Text.RegularExpressions;\r
  13 \r
  14 \r
  15 namespace Karinto\r
  16 {\r
  17     public class CsvReader<TDataSet> where TDataSet : IDataSet, new()\r
  18     {\r
  19         private string separator = ",";\r
  20 \r
  21         public string Separator\r
  22         {\r
  23             get { return separator; }\r
  24         }\r
  25 \r
  26         /// <summary>\r
  27         ///     区切り文字を指定する\r
  28         ///     カンマ以外だったらCSVじゃないような...?\r
  29         /// </summary>\r
  30         /// <param name="separator">区切り文字(正規表現可)</param>\r
  31         public void SetSeparator(string separator)\r
  32         {\r
  33             this.separator = separator;\r
  34         }\r
  35 \r
  36         /// <summary>\r
  37         ///     CSVファイルを読み込む\r
  38         /// </summary>\r
  39         /// <param name="path">CSVファイルのパス</param>\r
  40         /// <param name="proto">行の雛型</param>\r
  41         /// <returns>指定した型のリスト</returns>\r
  42         public List<TDataSet> Read(string path)\r
  43         {\r
  44             return Read(path, new TDataSet());\r
  45         }\r
  46 \r
  47         /// <summary>\r
  48         ///     雛型を指定してCSVファイルを読み込む\r
  49         /// </summary>\r
  50         /// <param name="path">CSVファイルのパス</param>\r
  51         /// <param name="proto">行の雛型</param>\r
  52         /// <returns>指定した型のリスト</returns>\r
  53         public List<TDataSet> Read(string path, TDataSet proto)\r
  54         {\r
  55             List<TDataSet> records = new List<TDataSet>();\r
  56 \r
  57             try\r
  58             {\r
  59                 using (FileStream fs = new FileStream(path, FileMode.Open, FileAccess.Read, FileShare.ReadWrite))\r
  60                 {\r
  61                     using (StreamReader reader = new StreamReader(fs))\r
  62                     {\r
  63                         string line = "";\r
  64                         // 数値にマッチする正規表現\r
  65                         string rfloat = @"\s*([-+]?(?:[0-9]+(\.[0-9]*)?|(\.[0-9]+))([eE][-+]?[0-9]+)?)\s*";\r
  66                         Regex rCsv = new Regex(rfloat + separator, RegexOptions.Compiled);\r
  67 \r
  68                         while (reader.Peek() >= 0)\r
  69                         {\r
  70                             line = reader.ReadLine() + separator;\r
  71                             //Console.WriteLine("> " + line);\r
  72                             MatchCollection mc = rCsv.Matches(line);\r
  73 \r
  74                             // 現状ではNaNが含まれる行は受理しない\r
  75                             if (!proto.Width.Includes(mc.Count)) continue;\r
  76 \r
  77                             if (proto.KeyRange != null)\r
  78                             {\r
  79                                 double key = Double.Parse(mc[0].Groups[1].Value);\r
  80                                 if (!proto.KeyRange.Includes(key)) continue;\r
  81                             }\r
  82 \r
  83                             TDataSet c = new TDataSet();\r
  84                             foreach (Match m in mc)\r
  85                             {\r
  86                                 c.Push(Double.Parse(m.Groups[1].Value));\r
  87                             }\r
  88                             records.Add(c);\r
  89                         }\r
  90                     }\r
  91                 }\r
  92             }\r
  93             catch (Exception e)\r
  94             {\r
  95                 Console.WriteLine("The file could not be read:");\r
  96                 Console.WriteLine("> " + path);\r
  97                 Console.WriteLine(e.Message);\r
  98                 throw e;\r
  99             }\r
 100 \r
 101             return records;\r
 102         }\r
 103     }\r
 104 }\r