C# 假设字符串SQL操作中的行顺序

我有一个存储过程，需要假设行顺序始终相同，以便正确运行，否则我将得到错误的数据。用户提供需要查询的ID字符串，存储过程使用该字符串将这些字符串放入表中_SPLIT@Parameter“，”操作

假设字符串将按照给定给表的顺序放入表中，安全吗？我创建了一个单元测试并运行了10000次，每次都检查表的顺序，结果通过了。因此，从这一点来看，我猜测做出这样的假设是安全的

SQL:

C语言中的单元测试

[TestMethod]
public void SQLOrderTest()
{
    // Arrange
    string testIds = "3, 5, 1, 10";
    string spName = "[dbo].[sp_TestOrder]";
    string ServerName = PARR_DBDataObject.ServerName;
    string DBName = PARR_DBDataObject.DBName;

    // Act 
    bool results = true;

    for(int i = 0; i < 10000; i++)
    {
        DataTable dt = SQL.QueryStoredProcedure(spName, DBName, ServerName, false, 5000,
                                             new Tuple<string, object>("@TestIDs", testIds));
        // First Row            
        if (GetRowNumber(dt, "3") != 1) results = false;

        // Second Row            
        if (GetRowNumber(dt, "5") != 2) results = false;

        // Third Row
        if (GetRowNumber(dt, "1") != 3) results = false;

        // Forth Row 
        if (GetRowNumber(dt, "10") != 4) results = false;
    }

    Assert.AreEqual(results, true);
}

private int GetRowNumber(DataTable dt, string TestIDValue)
{
    var row = from r in dt.AsEnumerable()
              where r.Field<string>("TestIDs") == TestIDValue
              select new
              {
                  TestID = r.Field<string>("TestIDs"),
                  rowNum = r.Field<Int32>("RowNumber")

              };
    return row.First().rowNum;
}

SQL表表示无序集。假设数据是按特定顺序排列从来都不是一个好主意，除非您使用ORDERBY。通常，单个页面上的小数据集甚至会像您期望的那样一致地排序。但使用记录为不正确的功能仍然是危险的

如果要保证排序，则可以插入带有order by和ordering值的记录。以下是使用递归CTE的方法：

with cte as (
      select convert(varchar(max), left(@testids, charindex(',', @testids + ',') - 1)) as id,
             convert(varchar(max), stuff(@testids, 1, charindex(',', @testids + ','), '')) as rest,
             1 as lev
      union all
      select convert(varchar(max), left(rest, charindex(',', rest + ',') - 1)) as id,
             convert(varchar(max), stuff(rest, 1, charindex(',', rest + ','), '')),
             lev + 1 as lev
      from cte
      where rest <> ''
     )
select trim(id) as id, lev
into #testnames
from cte
where trim(id) <> ''
order by lev;  -- not really needed but handy if you have an identity column

---

SQL Server是多线程的，数据可以随机保存和读取。要始终以相同的顺序获取数据，唯一的方法是使用order BY。如果您需要检索特定顺序的数据，请不要添加带有序列号的列。感谢您的解释，即使使用我的小表，我也不会使用这种假设。

with cte as (
      select convert(varchar(max), left(@testids, charindex(',', @testids + ',') - 1)) as id,
             convert(varchar(max), stuff(@testids, 1, charindex(',', @testids + ','), '')) as rest,
             1 as lev
      union all
      select convert(varchar(max), left(rest, charindex(',', rest + ',') - 1)) as id,
             convert(varchar(max), stuff(rest, 1, charindex(',', rest + ','), '')),
             lev + 1 as lev
      from cte
      where rest <> ''
     )
select trim(id) as id, lev
into #testnames
from cte
where trim(id) <> ''
order by lev;  -- not really needed but handy if you have an identity column

select tn.*
from #testnames tn
order by lev;