我正在读取*.csv文件。

*.csv-file由分号(";")分隔的两列组成。

我能够阅读*.csv文件使用StreamReader,并能够通过使用Split()函数分离每一行。我想将每一列存储到一个单独的数组中,然后显示它。

有可能做到吗?


当前回答

我花了几个小时寻找合适的库,但最后我写了自己的代码:) 你可以用任何你想要的工具读取文件(或数据库),然后应用下面的例程到每一行:

private static string[] SmartSplit(string line, char separator = ',')
{
    var inQuotes = false;
    var token = "";
    var lines = new List<string>();
    for (var i = 0; i < line.Length; i++) {
        var ch = line[i];
        if (inQuotes) // process string in quotes, 
        {
            if (ch == '"') {
                if (i<line.Length-1 && line[i + 1] == '"') {
                    i++;
                    token += '"';
                }
                else inQuotes = false;
            } else token += ch;
        } else {
            if (ch == '"') inQuotes = true;
            else if (ch == separator) {
                lines.Add(token);
                token = "";
                } else token += ch;
            }
    }
    lines.Add(token);
    return lines.ToArray();
}

其他回答

我一直在使用csvreader.com(付费组件)多年,我从来没有遇到过问题。它结实、小巧、快速,但你必须为此付费。您可以将分隔符设置为您喜欢的任何值。

using (CsvReader reader = new CsvReader(s) {
    reader.Settings.Delimiter = ';';
    reader.ReadHeaders();  // if headers on a line by themselves.  Makes reader.Headers[] available
    while (reader.ReadRecord())
        ... use reader.Values[col_i] ...
}

如果你需要跳过(头部)行和/或列,你可以使用它来创建一个2维数组:

    var lines = File.ReadAllLines(path).Select(a => a.Split(';'));
    var csv = (from line in lines               
               select (from col in line
               select col).Skip(1).ToArray() // skip the first column
              ).Skip(2).ToArray(); // skip 2 headlines

如果您需要在进一步处理数据之前对其进行塑形,这是非常有用的(假设前两行由标题组成,而第一列是行标题—您不需要在数组中拥有行标题,因为您只想考虑数据)。

注意:您可以通过使用以下代码轻松获得标题和第一列:

    var coltitle = (from line in lines 
                    select line.Skip(1).ToArray() // skip 1st column
                   ).Skip(1).Take(1).FirstOrDefault().ToArray(); // take the 2nd row
    var rowtitle = (from line in lines select line[0] // take 1st column
                   ).Skip(2).ToArray(); // skip 2 headlines

这个代码示例假设你的*.csv文件的结构如下:

注意:如果你需要跳过空行——有时这很方便,你可以通过插入来实现

    where line.Any(a=>!string.IsNullOrWhiteSpace(a))

在上面的LINQ代码示例中的from和select语句之间。

LINQ:

var lines = File.ReadAllLines("test.txt").Select(a => a.Split(';'));
var csv = from line in lines
          select (from piece in line
                  select piece);

^^错误-尼克编辑

原来的应答者似乎试图用一个2维数组填充csv -一个包含数组的数组。第一个数组中的每一项都包含一个表示行号的数组,嵌套数组中的每一项都包含该特定列的数据。

var csv = from line in lines
          select (line.Split(',')).ToArray();

这是我的2个简单的静态方法,将文本从csv文件转换为列表<列表<字符串>>,反之亦然。每种方法都使用行转换器。

这段代码应该考虑csv文件的所有可能性。您可以定义自己的csv分隔符,该方法尝试纠正转义双引号字符,并处理当所有文本在引号中是一个单元格,csv分隔符在引号字符串中,包括一个单元格中的多行,可以忽略空行。

最后一种方法仅用于测试。所以你可以忽略它,或者用这个测试方法测试你自己的或别人的解决方案:)。为了测试,我使用了这个硬csv, 4行2行:

0,a,""bc,d
"e, f",g,"this,is, o
ne ""lo
ng, cell""",h

这是最终代码。为了简单起见,我删除了所有的try catch块。

using System;
using System.Collections.Generic;
using System.Linq;

public static class Csv {
  public static string FromListToString(List<List<string>> csv, string separator = ",", char quotation = '"', bool returnFirstRow = true)
  {
    string content = "";
    for (int row = 0; row < csv.Count; row++) {
      content += (row > 0 ? Environment.NewLine : "") + RowFromListToString(csv[row], separator, quotation);
    }
    return content;
  }

  public static List<List<string>> FromStringToList(string content, string separator = ",", char quotation = '"', bool returnFirstRow = true, bool ignoreEmptyRows = true)
  {
    List<List<string>> csv = new List<List<string>>();
    string[] rows = content.Split(new string[] { Environment.NewLine }, StringSplitOptions.None);
    if (rows.Length <= (returnFirstRow ? 0 : 1)) { return csv; }
    List<string> csvRow = null;
    for (int rowIndex = 0; rowIndex < rows.Length; rowIndex++) {
      (List<string> row, bool rowClosed) = RowFromStringToList(rows[rowIndex], csvRow, separator, quotation);
      if (rowClosed) { if (!ignoreEmptyRows || row.Any(rowItem => rowItem.Length > 0)) { csv.Add(row); csvRow = null; } } // row ok, add to list
      else { csvRow = row; } // not fully created, continue
    }
    if (!returnFirstRow) { csv.RemoveAt(0); } // remove header
    return csv;
  }

  public static string RowFromListToString(List<string> csvData, string separator = ",", char quotation = '"')
  {
    csvData = csvData.Select(element =>
    {
      if (element.Contains(quotation)) {
        element = element.Replace(quotation.ToString(), quotation.ToString() + quotation.ToString());
      }
      if (element.Contains(separator) || element.Contains(Environment.NewLine)) {
        element = "\"" + element + "\"";
      }
      return element;
    }).ToList();
    return string.Join(separator, csvData);
  }

  public static (List<string>, bool) RowFromStringToList(string csvRow, List<string> continueWithRow = null, string separator = ",", char quotation = '"')
  {
    bool rowClosed = true;
    if (continueWithRow != null && continueWithRow.Count > 0) {
      // in previous result quotation are fixed so i need convert back to double quotation
      string previousCell = quotation.ToString() + continueWithRow.Last().Replace(quotation.ToString(), quotation.ToString() + quotation.ToString()) + Environment.NewLine;
      continueWithRow.RemoveAt(continueWithRow.Count - 1);
      csvRow = previousCell + csvRow;
    }

    char tempQuote = (char)162;
    while (csvRow.Contains(tempQuote)) { tempQuote = (char)(tempQuote + 1); }
    char tempSeparator = (char)(tempQuote + 1);
    while (csvRow.Contains(tempSeparator)) { tempSeparator = (char)(tempSeparator + 1); }

    csvRow = csvRow.Replace(quotation.ToString() + quotation.ToString(), tempQuote.ToString());
    if(csvRow.Split(new char[] { quotation }, StringSplitOptions.None).Length % 2 == 0) { rowClosed = !rowClosed; }
    string[] csvSplit = csvRow.Split(new string[] { separator }, StringSplitOptions.None);

    List<string> csvList = csvSplit
      .ToList()
      .Aggregate("",
          (string row, string item) => {
              if (row.Count((ch) => ch == quotation) % 2 == 0) { return row + (row.Length > 0 ? tempSeparator.ToString() : "") + item; }
              else { return row + separator + item; }
          },
          (string row) => row.Split(tempSeparator).Select((string item) => item.Trim(quotation).Replace(tempQuote, quotation))
      ).ToList();
    if (continueWithRow != null && continueWithRow.Count > 0) {
      return (continueWithRow.Concat(csvList).ToList(), rowClosed);
    }
    return (csvList, rowClosed);
  }

  public static bool Test()
  {
    string csvText = "0,a,\"\"bc,d" + Environment.NewLine + "\"e, f\",g,\"this,is, o" + Environment.NewLine + "ne \"\"lo" + Environment.NewLine + "ng, cell\"\"\",h";
    List<List<string>> csvList = new List<List<string>>() { new List<string>() { "0", "a", "\"bc", "d" }, new List<string>() { "e, f", "g", "this,is, o" + Environment.NewLine + "ne \"lo" + Environment.NewLine + "ng, cell\"", "h" } };

    List<List<string>> csvTextAsList = Csv.FromStringToList(csvText);
    bool ok = Enumerable.SequenceEqual(csvList[0], csvTextAsList[0]) && Enumerable.SequenceEqual(csvList[1], csvTextAsList[1]);
    string csvListAsText = Csv.FromListToString(csvList);
    return ok && csvListAsText == csvText;
  }
}

使用例子:

// get List<List<string>> representation of csv
var csvFromText = Csv.FromStringToList(csvAsText);

// read csv file with custom separator and quote
// return no header and ignore empty rows
var csvFile = File.ReadAllText(csvFileFullPath);
var csvFromFile = Csv.FromStringToList(csvFile, ";", '"', false, false);

// get text representation of csvData from List<List<string>>
var csvAsText = Csv.FromListToString(csvData);

注: 这:char tempQuote = (char)162;是ASCI表中的第一个稀有字符。脚本会搜索这个字符,或者不是文本中的第一个接下来的几个ascii字符,并将其用作临时转义和引用字符。

看看这个

使用CsvFramework;

使用System.Collections.Generic;

命名空间 CvsParser {

public class Customer
{
    public int Id { get; set; }
    public string Name { get; set; }
    public List<Order> Orders { get; set; }        
}

public class Order
{
    public int Id { get; set; }

    public int CustomerId { get; set; }
    public int Quantity { get; set; }

    public int Amount { get; set; }

    public List<OrderItem> OrderItems { get; set; }

}

public class Address
{
    public int Id { get; set; }
    public int CustomerId { get; set; }

    public string Name { get; set; }
}

public class OrderItem
{
    public int Id { get; set; }
    public int OrderId { get; set; }

    public string ProductName { get; set; }
}

class Program
{
    static void Main(string[] args)
    {

        var customerLines = System.IO.File.ReadAllLines(@"Customers.csv");
        var orderLines = System.IO.File.ReadAllLines(@"Orders.csv");
        var orderItemLines = System.IO.File.ReadAllLines(@"OrderItemLines.csv");

        CsvFactory.Register<Customer>(builder =>
        {
            builder.Add(a => a.Id).Type(typeof(int)).Index(0).IsKey(true);
            builder.Add(a => a.Name).Type(typeof(string)).Index(1);
            builder.AddNavigation(n => n.Orders).RelationKey<Order, int>(k => k.CustomerId);

        }, false, ',', customerLines);

        CsvFactory.Register<Order>(builder =>
        {
            builder.Add(a => a.Id).Type(typeof(int)).Index(0).IsKey(true);
            builder.Add(a => a.CustomerId).Type(typeof(int)).Index(1);
            builder.Add(a => a.Quantity).Type(typeof(int)).Index(2);
            builder.Add(a => a.Amount).Type(typeof(int)).Index(3);
            builder.AddNavigation(n => n.OrderItems).RelationKey<OrderItem, int>(k => k.OrderId);

        }, true, ',', orderLines);


        CsvFactory.Register<OrderItem>(builder =>
        {
            builder.Add(a => a.Id).Type(typeof(int)).Index(0).IsKey(true);
            builder.Add(a => a.OrderId).Type(typeof(int)).Index(1);
            builder.Add(a => a.ProductName).Type(typeof(string)).Index(2);


        }, false, ',', orderItemLines);



        var customers = CsvFactory.Parse<Customer>();


    }
}

}