c# - 多个正则表达式字符串模式(不同字段)

Question

我正在尝试从具有以下格式的 word 文档中提取文本并将数据插入到 SQL 数据库中。

Word 文档

Name of House: Aasleagh Lodge
Townland: Srahatloe
Near: Killary Harbour, Leenane
Status/Public Access: maintained, private fishing lodge
Date Built: 1838-1850, burnt 1923, rebuilt 1928

源代码

        var wordApp = new Microsoft.Office.Interop.Word.Application();
        var wordDoc = wordApp.Documents.Open(@"C:\Users\mhoban\Documents\Book.docx");
        var txt = wordDoc.Content.Text;

        var regex = new Regex(@"(Name of House\: )(.+?)[\r\n]");

        var allMatches = regex.Matches(txt);
        foreach (Match match in allMatches)
        {
            var nameValue = match.Groups[2].Value;
            var townValue = match.Groups[2].Value;

            SqlConnection con = new SqlConnection(ConfigurationManager.ConnectionStrings["ConnectionString"].ToString());
            SqlCommand com = new SqlCommand();

            com.CommandText = "INSERT INTO Houses (Name, Townland) VALUES (@name, @town)";

            com.Parameters.Add("@name", SqlDbType.NVarChar).SqlValue = nameValue;
            com.Parameters.Add("@town", SqlDbType.NVarChar).SqlValue = townValue;

            com.Connection = con;

            con.Open();

            com.ExecuteNonQuery();

            con.Close();
        }

这非常有效，唯一的问题是我将如何编写代码来插入文本的其他字段，例如这一行

var regex = new Regex(@"(Name of House\: )(.+?)[\r\n]");

在本例中插入房屋的名称“Aasleagh Lodge”，但我将如何编写此行以插入城镇？

我尝试用我需要的字段名称替换正则表达式中的“Townland”，但我最终得到的是单个记录，每个记录只包含一个不同的列值。

有没有一种方法可以同时插入数据，也许是通过使用列表或其他方式，这样就不会发生这种情况。

新源代码

var wordApp = new Microsoft.Office.Interop.Word.Application();
            var wordDoc = wordApp.Documents.Open(@"C:\Users\mhoban\Documents\Book.docx");
            var txt = wordDoc.Content.Text;

            using (var sr = new StringReader(txt))
            {
                var s = string.Empty;
                var nameValue = new StringBuilder();
                var townValue = new StringBuilder();
                while ((s = sr.ReadLine()) != null)
                {
                    if (s.StartsWith("Name of House"))
                    {
                        nameValue.Append(s.Split(new[] { ':' })[1].Trim());
                    }
                    else if (s.StartsWith("Townland"))
                    {
                        townValue.Append(s.Split(new[] { ':' })[1].Trim());
                    }

                    if (nameValue.Length > 0 && townValue.Length > 0)
                    {
                        SqlConnection con = new SqlConnection(ConfigurationManager.ConnectionStrings["ConnectionString"].ToString());
                        SqlCommand com = new SqlCommand();

                        com.CommandText = "INSERT INTO Houses (Name, Townland) VALUES (@name, @town)";
                        com.CommandText = "INSERT INTO Houses (Name) VALUES (@name)";

                        com.Parameters.Add("@name", SqlDbType.NVarChar).SqlValue = nameValue;
                        com.Parameters.Add("@town", SqlDbType.NVarChar).SqlValue = townValue;

                        com.Connection = con;

                        con.Open();

                        com.ExecuteNonQuery();

                        con.Close();

                        nameValue.Clear(); townValue.Clear();
                    }
                }
            }

数据库字段

[Id]          NCHAR (10)     NULL,
[Name]        NVARCHAR (MAX) NULL,
[Townland]    NVARCHAR (MAX) NULL,
[Near]        NVARCHAR (MAX) NULL,
[Status]      NVARCHAR (MAX) NULL,
[Built]       NVARCHAR (MAX) NULL,
[Description] NVARCHAR (MAX) NULL,
[Families]    NVARCHAR (MAX) NULL,
[Images]      IMAGE          NULL

Answer 1

这是一个没有正则表达式的解决方案。你真的不需要它。

var txt = "Name of House: Aasleagh Lodge\r\nTownland: Srahatloe\r\nNear: Killary Harbour, Leenane\r\nStatus/Public Access: maintained, private fishing lodge\r\nDate Built: 1838-1850, burnt 1923, rebuilt 1928\r\nName of House: House of Lan\r\nTownland: Another town land\r\nNear: Killary Harbour, Leenane\r\nStatus/Public Access: maintained, private fishing lodge\r\nDate Built: 1838-1850, burnt 1923, rebuilt 1928\r\nName of House: New Lodge\r\nTownland: NewTownLand\r\nNear: Killary Harbour, Leenane\r\nStatus/Public Access: maintained, private fishing lodge\r\nDate Built: 1838-1850, burnt 1923, rebuilt 1928";
using (var sr = new StringReader(txt))
{
   var s = string.Empty;
   var nameOfHouse = new StringBuilder();
   var townland = new StringBuilder();
   while ((s = sr.ReadLine()) != null)
   {
      if (s.StartsWith("Name of House"))
      {
          nameOfHouse.Append(s.Split(new[] {':'})[1].Trim());
      }
      else if (s.StartsWith("Townland"))
      {
           townland.Append(s.Split(new[] { ':' })[1].Trim());
      }

      if (nameOfHouse.Length > 0 && townland.Length > 0)
      { 
          // INSERT THE VALUES AND RESET THEM
          nameOfHouse.Clear(); townland.Clear();
      }
   }
}