正则表达式表现奇怪.net

时间:2012-09-26 08:40:39

标签: c# asp.net regex

我有一些代码会读取CSV文件的每一行,如果该值与正确的值不匹配,它会将其添加到错误列表中,并返回给用户屏幕。我遇到的问题是正则表达式本身。

protected void ReadData(string filePath, bool upload)
{
    StringBuilder sb = new StringBuilder();
    #region upload
    if (upload == true) // CSV file upload chosen
    {
        using (CsvReader csv = new CsvReader(new StreamReader(filePath), true)) // Cache CSV file to memory
        {
            int fieldCount = csv.FieldCount; // Total number of fields per row
            string[] headers = csv.GetFieldHeaders(); // Correct CSV headers stored in array 
            SortedList<int, string> errorList = new SortedList<int, string>(); // This list will contain error values
            bool errorFlag = false;
            int errorCount = 0;

            // Check if headers are correct first before reading data
            if (headers[0] != "first name" || headers[1] != "last name" || headers[2] != "job title" || headers[3] != "email address" || headers[4] != "telephone number" || headers[5] != "company" || headers[6] != "research manager" || headers[7] != "user card number")
            {
                sb.Append("Headers are incorrect");
            }

            else 
            {
                while (csv.ReadNextRecord())
                    try
                    {
                        //Check csv obj data for valid values
                        for (int i = 0; i < fieldCount; i++)
                        {
                            if (i == 0 || i == 1) // FirstName and LastName
                            {
                                if (Regex.IsMatch(csv[i].ToString(), "[a-zA-Z]", RegexOptions.IgnoreCase)) //REGEX letters only min of 5 char max of 20
                                {
                                    errorList.Add(errorCount, csv[i]);
                                    errorCount += 1;
                                    errorFlag = true;
                                    string text = csv[i].ToString();
                                }
                            }

                            else if (i == 5) // Company name
                            {
                                string text = csv[i];
                                text.Replace("&", "and");
                            }
                        } 

                        if (errorFlag == true)
                        {
                            sb.Append("<b>" + "Number of Error: " + errorCount + "</b>");
                            sb.Append("<ul>");
                            foreach (KeyValuePair<int, string> key in errorList)
                            {
                                sb.Append("<li>" + key.Value + "</li>");
                            }
                        }
                        else // All validation checks equaled to false. Create User
                        {

                            ORCLdap.CreateUserAccount(rootLDAPPath, svcUsername, svcPassword, csv[0], csv[1], csv[2], csv[3], csv[4], csv[5], csv[7]);
                            sb.Append("<b>New user data uploaded successfully</b>");
                        }

                    }// end of try

                    catch (Exception ex)
                    {
                        sb.Append(ex.ToString());
                    }

                    finally
                    {
                        lblMessage.Text = sb.ToString();
                        sb.Remove(0, sb.Length);

            }
        }
    }
    #endregion

lblMessage.text包含此html:

  

错误数:4

     
      
  • David1212
  •   
  • 史密斯
  •   
  • Nick444
  •   
  • Gowdy333
  •   

当它应该是3个错误,因为史密斯不包含数字。

有没有人对此提出建议?

2 个答案:

答案 0 :(得分:1)

您还有一个逻辑错误:

if (Regex.IsMatch(csv[i].ToString(), "[a-zA-Z]", RegexOptions.IgnoreCase)) //REGEX letters only min of 5 char max of 20

应该是

if (!Regex.IsMatch(csv[i].ToString(), "^[a-zA-Z]+$", RegexOptions.IgnoreCase)) //REGEX letters only min of 5 char max of 20

因为如果名称中包含其他字符而不是[a-zA-Z],那只是一个错误,对吗?

(如果您使用RegexOptions.IgnoreCase,则不需要[a-zA-Z][a-z]会这样做

答案 1 :(得分:0)

您需要为正则表达式添加单词边界,或者启动“^”并结束“$” 即 ^[a-zA-Z]+$

http://regexr.com?3298g

您当前的正则表达式不正确,并且会匹配任何位置包含a-z或A-Z,任何字母的任何字符串。

http://regexr.com?3298j