c# - 解析具有多个表的制表符分隔文件

Question

谁能推荐一个使用 c#（可能是 filehelper）的好方法，它可以让我在 c# 中解析像这样格式化的文件？

%T  person
%F  id  name    address city
%R  1   Bob 999 Main St Burbank
%R  2   Sara    829 South st    Pasadena
%T  houses
%F  id  personid    housetype  Color
%R  25  1       House   Red
%R  26  2       condo   Green

我想把这两个表变成一个数据表或者我可以用 linq 查询的东西。

该文件是制表符分隔的

Answer 1

此类数据的示例解析器

public IEnumerable<Dictionary<string, string>> Parse(TextReader reader)
{
    var state = new State { Handle = ExpectTableTitle };
    return GenerateFrom(reader)
        .Select(line => state.Handle(line.Split('\t'), state))
        .Where(returnIt => returnIt)
        .Select(returnIt => state.Row);
}

private bool ExpectTableTitle(string[] lineParts, State state)
{
    if (lineParts[0] == "%T")
    {
        state.TableTitle = lineParts[1];
        state.Handle = ExpectFieldNames;
    }
    else
    {
        Console.WriteLine("Expected %T but found '"+lineParts[0]+"'");
    }
    return false;
}

private bool ExpectFieldNames(string[] lineParts, State state)
{
    if (lineParts[0] == "%F")
    {
        state.FieldNames = lineParts.Skip(1).ToArray();
        state.Handle = ExpectRowOrTableTitle;
    }
    else
    {
        Console.WriteLine("Expected %F but found '" + lineParts[0] + "'");
    }
    return false;
}

private bool ExpectRowOrTableTitle(string[] lineParts, State state)
{
    if (lineParts[0] == "%R")
    {

        state.Row = lineParts.Skip(1)
            .Select((x, i) => new { Value = x, Index = i })
            .ToDictionary(x => state.FieldNames[x.Index], x => x.Value);
        state.Row.Add("_tableTitle",state.TableTitle);
        return true;
    }
    return ExpectTableTitle(lineParts, state);
}

public class State
{
    public string TableTitle;
    public string[] FieldNames;
    public Dictionary<string, string> Row;
    public Func<string[], State, bool> Handle;
}

private static IEnumerable<string> GenerateFrom(TextReader reader)
{
    string line;
    while ((line = reader.ReadLine()) != null)
    {
        yield return line;
    }
}

然后只需根据 _tableTitle 条目将每个结果字典转换/映射到您的域对象之一。

这是使用您的示例数据的测试工具。要从文件中读取，请传入 StreamReader 而不是 StringReader。

const string data = @"%T\tperson
%F\tid\tname\taddress\tcity
%R\t1\tBob\t999 Main St\tBurbank
%R\t2\tSara\t829 South st\tPasadena
%T\thouses
%F\tid\tpersonid\thousetype\tColor
%R\t25\t1\tHouse\tRed
%R\t26\t2\tcondo\tGreen";

var reader = new StringReader(data.Replace("\\t","\t"));

var rows = Parse(reader);
foreach (var row in rows)
{
    foreach (var entry in row)
    {
        Console.Write(entry.Key);
        Console.Write('\t');
        Console.Write('=');
        Console.Write('\t');
        Console.Write(entry.Value);
        Console.WriteLine();
    }
    Console.WriteLine();
}

输出：

id    =    1
name    =    Bob
address    =    999 Main St
city    =    Burbank
_tableTitle    =    person

id    =    2
name    =    Sara
address    =    829 South st
city    =    Pasadena
_tableTitle    =    person

id    =    25
personid    =    1
housetype    =    House
Color    =    Red
_tableTitle    =    houses

id    =    26
personid    =    2
housetype    =    condo
Color    =    Green
_tableTitle    =    houses