Cannot get a stored procedure table output into my .NET Application - c#

I have a stored procedure that looks like this:
DECLARE #A TABLE ( TabYear int, Std decimal(18,2))
DECLARE #B TABLE ( TabYear int, Std decimal(18,2))
BEGIN
DECLARE #FinalTable TABLE (TabYear int, HoldayA decimal(18,2), HolidayB decimal(18,2))
BEGIN
INSERT INTO #A(TabYear, std)
Select ... from ...
INSERT INTO #B(TabYear, Std)
Select ... from ...
END
INSERT INTO #FinalTable(TabYear, HoldayA, HoldayB)
SELECT
A.TabYear, a.Std, U.Std
FROM
#A A
LEFT JOIN
#B U ON a.TabYear = U.TabYear
END
Select * from #FinalTable
Now I want to get that into a DATASET within my .NET application using the TableAdapter Wizard. I choose :"Use existing stored procedure" and choose the procedure (above) but as Data Column I get only Column1. It looks like the DataTableAdapter does not recognize the table. If I let the SP run on the server everything is fine. I get the table as I wish. I also checked if the wizard recognizes other SP's I am working with on the table and that works fine. But I have to admit that all the other SP's I use are straigth Select commands nothing with querying #Tables like in this SP. Could anyone help me get the table into my application. I do not have go go with the DataSet necessarily a code solution getting the data into an array or something would help me too.

You cannot achieve this using the TableAdapter Wizard because your stored procedure has no defined output because its dynamically created. You must manually construct the TableAdapter in your code.

Probably because you SELECT * from a table that's defined within the sproc, so the analyzer can't determine the schema. Either define the columns outside of the wizard or change your sproc to select specific columns (although the wizard won't be able to determine the type, so you'll still have to edit the data table schema).
You could also turn the sproc into one SELECT that uses subqueries instead of filling a table variable, but that may have been done for performance reasons...
I do not have go go with the DataSet necessarily
That's fine, but you'll still hove to map the columns of the result set to a structure in C# code at some point.
Wizards only take you so far; at some point you need to start casting your own spells...

Related

Having a stored procedure affect only the currently focused row

I’m using Visual Studio with DevExpress Winforms in conjunction with Microsoft SQL Server Manager, and I’m currently trying to write a stored procedure that will activate on a button click, copying the focused row to a different table.
I’ve looked into it and haven’t been able to find a solution that does what I need it to. I would prefer not to use checkboxes if possible, but am open to it.
My current code within the stored procedure is:
CREATE PROCEDURE [dbo].[procedurename]
AS
BEGIN
SET NOCOUNT ON;
IF NOT EXISTS (SELECT * FROM tbl_b)
INSERT INTO tbl_b (column names here)
SELECT DISTINCT (column names here)
FROM tbl_a
END
I’ve been able to successfully call this procedure when testing the program with the button, and it functions mostly as intended, moving the data from one table to the other with no duplicates, however it moves the entire table when I only want to move a focused row.
I’m using gridView for this, and I’ve tried solutions that use DataGridView that don’t seem to work.
This is a desktop based program.
You need a parameter in your stored procedure that holds an identification of the row you want to copy
CREATE PROCEDURE [dbo].[procedurename] (#ID int)
AS
BEGIN
SET NOCOUNT ON;
IF NOT EXISTS (SELECT * FROM tbl_b)
INSERT INTO tbl_b (column names here)
SELECT DISTINCT (column names here)
FROM tbl_a
where tbl_A.PrimaryKeyColumn = #ID
END
In the Click event of your button you could then do something like this
var id = myGridView.GetRowCellValue(myGridView.FocusedRowHandle, "YourPrimaryKeyName");
and now you can use this id variable to pass to your stored procedure
However, you do have another problem in your stored procedure
IF NOT EXISTS (SELECT * FROM tbl_b)
This check will never allow you to do your insert once there is 1 or more rows in that table.
I think you need something like
IF NOT EXISTS (SELECT 1 FROM tbl_b where tbl_b.PrimaryKey = #ID)

Passing an array of objects to SQL stored procedure

I have 2 tables
payment (payment_id, otherCosts, GarageCosts)
spareparts (payment_id, sparepartId, sparePartQty)
In payment table payment_id is autogenerated. Apart from otherCosts and garagecosts values, in my C# asp.net application there is an array of objects with
{ sparepartId : 'Somevalue', sparePartQty : 'somevalue' }
What I need to do is in a stored procedure first enter the record into payment table with garage costs and others costs value. Then return the last generated payment ID and enter it to spareParts table as paymentId for each of the value pairs in the array.
What is the way to achieve this? Please help.
Based on your tags I am going to assume we are talking about SQL Server / T-SQL.
You could do all this in one stored procedure:
CREATE PROCEDURE dbo.Foo ... /* input parameters */
AS
BEGIN
DECLARE #PaymentId int
INSERT INTO payment(otherCosts, GarageCosts) VALUES (...)
SET #PaymentId = SCOPE_IDENTITY()
INSERT INTO spareparts(payment_id, sparepartId, sparePartQty) VALUES(#PaymentId, ...)
END
GO
You may want to also look into ##IDENTITY but make sure you read about ##IDENTITY and SCOPE_IDENTITY and understand the risks associated with the first one.
If you need to have two separate sprocs you can do that too and here is how the first sproc would look like. Note that the #PaymentId is an output parameter which means that the caller can retrieve it and pass it to the second procedure.
CREATE PROCEDURE dbo.Foo
/* input parameters */
#PaymentId int OUT
AS
BEGIN
INSERT INTO payment(otherCosts, GarageCosts) VALUES (...)
SET #PaymentId = SCOPE_IDENTITY()
END
GO
Edit - after the scope of the question was clarified:
If you need to call the second stored procedure and pass it an array of parameters, with SQL Server 2008 or newer you can use TVP (Table Value Parameters). To see how you can use them in stored procedures and how you can pass them from C# code see Table Value Parameters in SQL Server 2008 and .NET (C#) or Table-Valued Parameters.
You can also use TVPs with the solution where you only have one sproc.
To solve your problem try this
First insert your data in payment table with otherCosts and GarageCosts.
Then create a procedure to get the latest stored payment_id from payment table
create procedure select_last_payment_id
as
begin
select top 1 payment_id
from payment
order by payment_id desc
end
Lastly get that payment_id by running stored procedure and assigning it to payment_id of spareparts table and storing spareparts data.
Hope it works for you.

Stored Procedure Return Type Can Not Be Detected

I am trying to drag this procedure over to the dbml in VS 2012 and I am getting the return type cannot be detected message.
I have tried these:
LINQ to SQL - Stored Procedure Return Type Error
The return types for the following stored procedures could not be detected
I tried re-writing the procedure as a CTE and also as union to remove the OR, but it is giving me the same message.
The only return type in the designer properties for the method is int32.
Here is my procedure:
ALTER PROCEDURE [dbo].[GetStringFromFiles]
#SearchWord NVARCHAR(100) = null
AS
BEGIN
SET NOCOUNT ON
SET #SearchWord = UPPER(#SearchWord);
Select
a.FileId <---Guid
, a.FileData <---Binary
, a.BaselineId <---Guid
, a.FileName <---NVARCHAR
, a.FileExtension <---NVARCHAR
, b.FileByItemId <----Guid
, b.ItemId <---Guid
From FileTable a
Inner Join
FileByItem b on a.FileId = b.FileId
WHERE CONTAINS(a.FileData,'FORMSOF(INFLECTIONAL, #SearchWord)') or FREETEXT(a.FileData, #SearchWord)
RETURN 1
END
UPDATE:
A. I can add it if I comment out the entire Where Clause -- Auto-Generated-ReturnType
B. If I take away the or and just use:
WHERE CONTAINS(a.FileData,'FORMSOF(INFLECTIONAL, #SearchWord)')
it lets me add it -- Auto-Generated-ReturnType
C. If I just use
WHERE FREETEXT(a.FileData, #SearchWord)
It throws the error so it doesnt like the FREETEXT
If I comment the where clause out and add it and let it generate the return type then alter the procedure in the db are there any issues with that?
This is somewhat related. I stumbled across the same error today but for a different reason. If your query uses a temporary table in its results then it is unable to automatically assign the return type.
I changed it to a table variable (performance was not a major concern/and was not impacted)
Hope this helps someone browsing.
The line
RETURN 1
is the culprit in both the cases. Lose it and Linq should be able to detect the types your result set.
Linq to Sql won't get the results directly from a Stored procedure.
Instead an ISingleResult will be returned. You have to enumerate
through the result set to get the values returned from the stored
procedure
If the stored procedure have any error also then this error can occur, for me its happened once since one of the table name was changed.
Another factor is to set Primary Keys and table relations if possible
A simple way to solve this issue is (December 2019)
Just making double # precede #tmp => ##tmp
Comment out DROP TABLE #tmp => --DROP TABLE #tmp
Execute stored procedure
Drag stored procedure again and That's it, It will generate return type
Last, Turn your store back to the first situation and then save.
Hope I can help.

Cannot alter table with LINQ TO SQL and stored procedure

I have to alter the table Statistic when I add a new metric in the table Metric I add a column in table Statistic.
I used a stored procedure that allows me to alter the table Statistic so the code :
CREATE PROCEDURE dbo.addnewmetricInstat
(
#MetricName varchar(254),
#TypeMetric varchar(254)
)
AS
IF (#TypeMetric='int')
Begin
alter table Statistic
add #MetricName int null
end
ELSE if (#TypeMetric='string')
begin
alter table Statistic
add #MetricName varchar(254) null
end
Then I successfully called the stored procedure but the columns is not added. The code I used in C# for calling this stored procedure is:
using (DataClassesDataContext db = new DataClassesDataContext("Data Source=EMEA-TUN-WS0367\\SQLEXPRESS;Initial Catalog=Perfgas;Integrated Security=True"))
{
db.addnewmetricInstat(metric.MetricName, metric.Type);
db.SubmitChanges();
}
First you shoud not be trying to call the sp from your application until you have tested it in SSMS. That way you know if the problem is your call or the sp. It will save you much debugging time if you do this.
Your proc is the problem. You will need to use dynamic SQl for this. Right now you are trying to add a column called MetericName because you didn't use the variable. This would work once, but of course the second time you run it, you will get an error becasue teh column already exists. However you can;t just throw a variable into an alter table statement, you must use dynamic SQL.

Adding stored procedures complex types in Entity Framework

I am trying to use a stored procedure in Entity Framework that returns nothing.
I did the following:
Added a function (right click on stored procedure -> add -> function import-> Complex Type -> Get column information -> create New Complex-Type)
My function name: summarySP_Result. After building the project the entity class is not generated in Generated_code (BusinessAccount.web.g.cs)
But entity classes for tables and views are all created but nor for stored procedure.
Can anybody give the idea why it is not generated entity class in BusinessAccount.web.g.cs?
Update :
Let me confirm ReturnDataFromTemTable_result entity class created in your XXXXXX.web.g.cs class.
Like :
[DataContract(Namespace="http://schemas.datacontract.org/2004/07/BizFramework.Web.Model")]
public sealed partial class ReturnDataFromTemTable_Result : Entity
{
-------------------
}
OK - here's the step-by-step way of doing this:
(1) add your stored procedure to the EDMX file (when you first create it, or later on by using Update model from database and picking that stored procedure)
(2) once you have the stored procedure in your model - use the Model Browser to add a Function Import :
(3) the next dialog that pops up is vitally important - you need to (1) define that the stored procedure returns a collection of complex types, then you need to (2) get the column info from that stored procedure to know what columns it will return, then (3) you tell Visual Studio to generate a new complex type based on that column info:
(4) once you've done that - you should now see the stored procedure in your conceptual model section in the Model Browser, and the newly generated complex type should show up there, too:
This is for Ross Brigoli
Try adding this line to the beginning of your stored procedure:
SET FMTONLY OFF
You can remove this after you have finished importing.
Source:-
Why can't Entity Framework see my Stored Procedure's column information?
If this is still unresolved, after you Add the Function Import, go to the Solution Explorer, right click your {name}.Context.tt file and do "Run Custom Tool". The method will now show up in your derived Context class.
This seems like a bug in Visual Studio 2012, which is what I am using, I haven't applied Update 1, I will try to see if that fixes it.
As Sandeep said,
EF doesn't support importing stored procedures which build result set from Dynamic queries or Temporary tables.
But you don't have to rewrite your whole SP.
Just write another one, with the same name, that returns the correct row format without using dynamic SQL or a temp table. Then use the EF SP adding function, which will now auto generate the complex type.
Edit: It's actually easier to make a comment at the top of the SP that immediately selects the desired row with all the data types specified with CASTS. When you need to import the SP into EF, just uncomment the code.
e.g.
CREATE PROCEDURE myProc ()
AS
BEGIN
-- uncomment the following row to import:
-- SELECT CAST( 0 AS int) AS column1Name, CAST( 'a' AS varchar(50)) AS clumn2name
-- comment out the SP content when you want to import it.
< proper SP content >
END
Then drop the stored proc and create the original.
Save this temporary importing SP you have made in case you need it again, though.
EF doesn't support importing stored procedures which build result set from:
Dynamic queries
Temporary tables
Re-write your stored procedure to use a table variable instead.
remember to drop the stored procudure and function import from your model before updating as it wont generate the complex type unless it also adds the stored procedure. or go to function import properties and use the get column information feature after updating your stored procedure.
create procedure [dbo].[usp_InsertOrUpdate]
/*if your table(tbl_InsertOrUpdate) as 3 columns like uniqueid,col1,col2*/
#uniqueId bigint NULL,/*if insert send value as null or 0*/
#col1 bigint null,
#col2 [varchar](500) NULL
as
begin
set nocount ON
SET FMTONLY OFF
/* for giving result which column updated(uniqueId) and is it insert or update(IsInsert)*/
declare #varResult table (uniqueId bigint ,IsInsert bit )
/*create a var table before inserting original table*/
declare #varInsertOrUpdate table (
uniqueId bigint ,
col1 [bigint] ,
col2 [varchar]
)
/*default we are feel as update only*/
insert into #varResult (uniqueId,IsInsert) values (#uniqueId,0)
/*insert into var table*/
INSERT INTO #varInsertOrUpdate (uniqueId,col1,col2)
VALUES
(#uniqueId,#col1,#col2)
/*Insert into original table with where condition without if else*/
INSERT INTO tbl_InsertOrUpdate (col1,col2)
select col1,col2 from #varInsertOrUpdate
where uniqueId!=0;
/*if its insert updating result returning table*/
update #varResult set
uniqueId=IDENT_CURRENT('tbl_InsertOrUpdate'),
IsInsert=1 where #uniqueId=0;
/*updating table where #uniqueid is null or empty*/
UPDATE tbl_InsertOrUpdate
SET col1=#col1,
col2=#col2,
WHERE uniqueId=#uniqueId and #uniqueId!=0
select * from #varResult
end
To add complex type correctly, go Model browser, right click on function, then display edit, click edit an fill the dialog box. The name of the function should be same as name of the stored procedure. Click OK button. Now function is created. Then right click on the created function and go edit again.There is a update button aside of complex type button.Update it using that update button. Now complex type is created completely.
This is my SP to Implement the multiple search
***************************************************
CREATE PROCEDURE [dbo].[uspSEARCH_POSITIONS]
#OBJ_TYPE_REQUEST varchar(2000),--'FIRST_NAME;SEARCHVALUE|LAST_NAME;SEARCHVALUE|JOB_DESCRIPTION;SEARCHVALUE'
#DELIMITER varchar(10) --'| Which seperates the col;searchvalue|col;searchvalue
AS
BEGIN
SET FMTONLY OFF
DECLARE
#lLastName varchar(100),
#lFirstName varchar(100),
#lPositionNumber varchar(20),
#lJobDescription varchar(50),
#lJobCode varchar(20),
#lOccupancyIndicator varchar(50),
#ldeleimitercolsearchval varchar(10)
SET #ldeleimitercolsearchval =';'
CREATE TABLE #TempTable (ColSearchValues VARCHAR(2000))
INSERT INTO #TempTable
SELECT * FROM [dbo].[fnSplit](#OBJ_TYPE_REQUEST,#DELIMITER)--'fname;searchvalfname|lname;searchvallname|jobcode;searchvaljobcode','|')
SELECT #lLastName=SUBSTRING(ColSearchValues,CHARINDEX(#ldeleimitercolsearchval ,ColSearchValues)+1,LEN(ColSearchValues)) from #TempTable where lower(ColSearchValues) like '%last%'
SELECT #lFirstName =SUBSTRING(ColSearchValues,CHARINDEX(#ldeleimitercolsearchval ,ColSearchValues)+1,LEN(ColSearchValues)) from #TempTable where lower(ColSearchValues) like '%first%'
SELECT #lPositionNumber =SUBSTRING(ColSearchValues,CHARINDEX(#ldeleimitercolsearchval ,ColSearchValues)+1,LEN(ColSearchValues)) from #TempTable where lower(ColSearchValues) like '%position%'
SELECT #lJobDescription=SUBSTRING(ColSearchValues,CHARINDEX(#ldeleimitercolsearchval ,ColSearchValues)+1,LEN(ColSearchValues)) from #TempTable where lower(ColSearchValues) like '%jobd%'
SELECT #lJobCode=SUBSTRING(ColSearchValues,CHARINDEX(#ldeleimitercolsearchval ,ColSearchValues)+1,LEN(ColSearchValues)) from #TempTable where lower(ColSearchValues) like '%jobc%'
SELECT #lOccupancyIndicator=SUBSTRING(ColSearchValues,CHARINDEX(#ldeleimitercolsearchval ,ColSearchValues)+1,LEN(ColSearchValues)) from #TempTable where lower(ColSearchValues) like '%ccupancy%'
SELECT [PS].[POSITION_NUMBER]
,[PS].[COST_CENTER]
,[PS].[JOB_CODE]
,[PS].[JOB_CODE_DESCRIPTION]
,[PS].[SITE_CODE]
,[EMP].[EMPLOYEE_ID]
,[EMP].[EIN]
,[EMP].[GRADE]
,[EMP].[LOGIN_ID]
,[EMP].[FIRST_NAME]
,[EMP].[LAST_NAME]
,LTRIM(RTRIM(ISNULL([EMP].[LAST_NAME],''))) + ',' +LTRIM(RTRIM(ISNULL([EMP].[FIRST_NAME],''))) AS [FULL_NAME]
,[EMP].[DISTRICT]
,[EMP].[SUPERVISOR_EIN]
,COUNT(*) OVER() AS TOTAL_RECORD_COUNT
FROM [DBSERVER].[dbo].[uvwPOSITION_SEARCH] PS
LEFT JOIN [DBSERVER].[dbo].[uvwEMPLOYEES] EMP
ON PS.POSITION_NUMBER=EMP.POSITION_NUMBER
WHERE
(#lLastName IS NULL OR [LAST_NAME] LIKE '%' + #lLastName + '%')
AND (#lFirstName IS NULL OR [FIRST_NAME] LIKE '%' + #lFirstName + '%')
AND (#lPositionNumber IS NULL OR [PS].[POSITION_NUMBER] LIKE '%' + #lPositionNumber + '%')
AND (#lJobDescription IS NULL OR [PS].[JOB_CODE_DESCRIPTION] LIKE '%' + #lJobDescription + '%')
AND (#lJobCode IS NULL OR [PS].[JOB_CODE] LIKE '%' + #lJobCode + '%')
AND (#lOccupancyIndicator IS NULL OR [EMP].[FILLED_VACANT] LIKE '%' + #lOccupancyIndicator + '%')
END
Now you can consume above SP in edmx using below
Adding stored procedures complex types in Entity Framework
Why can't Entity Framework see my Stored Procedure's column information?
And in case you have to update your SP below worked for me.
Updating Complex Type if Stored Procedure Updates
How Do I Get Entity Framework To Update Complex Types?
For me, Im having problems where importing my Stored Procedure into EF is not generating the Complex Entity return object (automatically). I found however, after commenting out sections of my sproc (aka stored procedure), that when I then re-imported the stored procedure back in (ie refreshed using the Get Column Information button in the Function Import Edit screen), that the Complex type could then be generated!
In short, there could be a where clause (or maybe something else) causing EF to not generate the Complex Type. Try commenting out sections of your sproc and re-importing the sproc to
UPDATE:
Further to my investigation above, I found that the reason the Complex Entity was not being generated was because my sproc was using a view (instead of a typical table). For curiosity sake, I changed the view to another table just to see what would happen, and the complex entity generated.
So, in short, it looks like Complex Entities might not generate automatically if you have a view. To try, I ripped out the view temporarily, re-import the sproc, generated the Complex Entity, then put the view back in. But now my code gives exceptions.
Will update on this later when I learn more =)
UPDATE:
Fixed the issue. Really silly mistake! The viewname that I was using was not spelled right =D. Im sort of angry that an error wasnt thrown by Sql Server when I created the sproc..... I guess that is life :) Alas, problem now fixed!
The issue of complex type not appearing may happen due to a different reason as well which is what I faced in our case. The issue was due to a syntax error in the SPROC where temp table was defined as below -
create table #temp(
col1 int,
col2 nvarchar(100),
col3 nvarchar(100), -- Notice the comma in the end
);
Surprisingly, SQL Server doesn't throw any error when you compile the sproc. Removing the comma fixed the problem for us.
In short, while some of the above solutions might work depending on the specific issue, my suggestion is to check your sproc for such syntactical errors that SQL might ignore but could be the underlying reason for this problem. Thanks.
Go to the Model Browser
If you need to modify existing function
Under the Function Imports >> Select the function to be modified >> Click Edit
You will need to update the function to refresh and you can see the columns need to be added

Categories