可以将文章内容翻译成中文,广告屏蔽插件可能会导致该功能失效(如失效,请关闭广告屏蔽插件后再试):
问题:
In my code I have a page that includes information from 3 different tables. To show this information I make 3 SQL select calls and unite them in one list to pass as Model to my view. Can I do it with one SQL call? Data has no connection with one another.
My code:
public ActionResult Index()
{
StorePageData PageData = new StorePageData();
return View(PageData);
}
public class StorePageData
{
public List<Table1Data> Table1 { get; set; }
public List<Table2Data> Table2 { get; set; }
public List<Table3Data> Table3 { get; set; }
public StorePageData()
{
Table1 = //loading from Database1
Table2 = //loading from Database2
Table3 = //loading from Database3
}
}
public class Table1Data
{
public int Id { get; set; }
public double Info1 { get; set; }
public string Info2 { get; set; }
}
public class Table2Data
{
public int Id { get; set; }
public List<int> Info1 { get; set; }
public List<int> Info2 { get; set; }
}
public class Table3Data
{
public int Id { get; set; }
public List<string> Info1 { get; set; }
public List<string> Info2 { get; set; }
}
If there is a way to load all 3 tables in one SQL request it will improve significantly the load time of this page.
Thank you.
回答1:
You can get multiple result sets in a single request using a DataReader
. You can use it with or without entity framework.
If you are using Entity Framework, you can pass a DbDataReader
to ObjectContext.Translate
method to translate multiple result set to requested object types. The command which is used to create the data reader can be a stored procedure, or you can simply use a command containing your queries to shape multiple result set.
Example
List<Table1> list1;
List<Table2> list2;
using (var cn = new SqlConnection(@"Connection String"))
{
cn.Open();
using (var cmd = cn.CreateCommand())
{
cmd.CommandText = "SELECT * FROM Table1; SELECT * FROM Table2";
var reader = cmd.ExecuteReader();
using (var db = new YourDbContext())
{
var context = ((IObjectContextAdapter)db).ObjectContext;
list1 = context.Translate<Table1>(reader).ToList();
reader.NextResult();
list2 = context.Translate<Table2>(reader).ToList();
}
}
}
If you are using SqlDataAdapter
, you can simply pass a command containing your queries and then using Fill
, fill a data set. The data adapter itself will use DataReader
behind the scene.
Example
var connectionString = @"Connection String";
var commandText = "SELECT * FROM Table1; SELECT * FROM Table2;";
var ds = new DataSet();
using (var da = new SqlDataAdapter(commandText, connectionString))
{
da.Fill(ds);
}
Then you can shape the results to List<Table1>
and List<Table2>
.
回答2:
DO NOT USE UNION.
DataAdapter is weapon of choise.
var commandText = "SELECT * FROM Table1; SELECT * FROM Table2;";
var ds = new DataSet();
using (var da = new SqlDataAdapter(commandText, "your cn"))
{
da.Fill(ds);
}
Using:
ds.Tables["Table1"]...
ds.Tables["Table2"]...
回答3:
You can use UNION ALL
to merge multiple queries.
Do something like this:
SELECT * FROM Table1
UNION ALL
SELECT * FROM Table2
Edit:
You can do this if you want to know where a single record is from:
SELECT *, 1 AS TableName FROM Table1
UNION ALL
SELECT *, 2 AS TableName FROM Table2
This will add another column that can be used to split the array into 3 lists.
回答4:
Assuming you are using EntityFramwork you can use EF Stored procedure that returns multiple result sets. Then map the result tables to your class object. Take a look here or google it. It will take one round to the DB.
回答5:
You can do something like this
SELECT Info1, Info2
FROM Table1Data
UNION ALL
SELECT Id, Info2
FROM Table2Data;
Then you can parse the result.
Depends on the case but you can also consider using a transaction.
回答6:
May be this can helpful to you.
Select the three tables in a single procedure as a separate result sets, and then in the C# side get the result sets and convert it into JSON format, and then from the JSON you can get the each table data into your list.
DataSet dataSet = new DataSet("dataSet");
dataSet.Namespace = "NetFrameWork";
DataTable table = new DataTable();
DataColumn idColumn = new DataColumn("id", typeof(int));
idColumn.AutoIncrement = true;
DataColumn itemColumn = new DataColumn("item");
table.Columns.Add(idColumn);
table.Columns.Add(itemColumn);
dataSet.Tables.Add(table);
for (int i = 0; i < 2; i++)
{
DataRow newRow = table.NewRow();
newRow["item"] = "item " + i;
table.Rows.Add(newRow);
}
dataSet.AcceptChanges();
string json = JsonConvert.SerializeObject(dataSet, Formatting.Indented);
Console.WriteLine(json);
//{
// "Table1": [
// {
// "id": 0,
// "item": "item 0"
// },
// {
// "id": 1,
// "item": "item 1"
// }
// ]
//"Table2": [
// {
// "id": 0,
// "item": "item 0",
// "rate": 200.00
// },
// {
// "id": 1,
// "item": "item 1",
// "rate": 225.00
//}
// ]
// "Table3": [
// {
// "id": 0,
// "item": "item 0",
// "rate": 200.00,
// "UOM" : "KG"
// },
// {
// "id": 1,
// "item": "item 1",
// "rate": 225.00,
// "UOM" : "LTR"
// }
// ]
//}
回答7:
you can use below generic code snippet
/// <method>
/// Select Query
/// </method>
public DataTable executeSelectQuery(String _query, SqlParameter[] sqlParameter)
{
SqlCommand myCommand = new SqlCommand();
DataTable dataTable = new DataTable();
dataTable = null;
DataSet ds = new DataSet();
try
{
myCommand.Connection = openConnection();
myCommand.CommandText = _query;//or proc
myCommand.Parameters.AddRange(sqlParameter);
myAdapter.SelectCommand = myCommand;
myAdapter.Fill(ds);
dataTable = ds.Tables[0];//as per your requirement
}
catch (SqlException e)
{//any logger
Console.Write("Error - Connection.executeSelectQuery - Query: " + _query + " \nException: " + e.StackTrace.ToString());
return null;
}
return dataTable;
}