I have a a number of sp's that create a temporary table #TempData with various fields. Within these sp's I call some processing sp that operates on #TempData. Temp data processing depends on sp input parameters. SP code is:
CREATE PROCEDURE [dbo].[tempdata_proc]
#ID int,
#NeedAvg tinyint = 0
AS
BEGIN
SET NOCOUNT ON;
if #NeedAvg = 1
Update #TempData set AvgValue = 1
Update #TempData set Value = -1;
END
Then, this sp is called in outer sp with the following code:
USE [BN]
--GO
--DBCC FREEPROCCACHE;
GO
Create table #TempData
(
tele_time datetime
, Value float
--, AvgValue float
)
Create clustered index IXTemp on #TempData(tele_time);
insert into #TempData(tele_time, Value ) values( GETDATE(), 50 ); --sample data
declare
#ID int,
#UpdAvg int;
select
#ID = 1000,
#UpdAvg = 1
;
Exec dbo.tempdata_proc #ID, #UpdAvg ;
select * from #TempData;
drop table #TempData
This code throws an error: Msg 207, Level 16, State 1, Procedure tempdata_proc, Line 8: Invalid column name "AvgValue".
But if only I uncomment declaration AvgValue float - everything works OK.
The question: is there any workaround letting the stored proc code remain the same and providing a tip to the optimizer - skip this because AvgValue column will not be used by the sp due to params passed.
Dynamic SQL is not a welcomed solution BTW. Using alternative to #TempData tablename is undesireable solution according to existing tsql code (huge modifications necessary for that).
Tried SET FMTONLY, tempdb.tempdb.sys.columns, try-catch wrapping without any success.
The way that stored procedures are processed is split into two parts - one part, checking for syntactical correctness, is performed at the time that the stored procedure is created or altered. The remaining part of compilation is deferred until the point in time at which the store procedure is executed. This is referred to as Deferred Name Resolution and allows a stored procedure to include references to tables (not just limited to temp tables) that do not exist at the point in time that the procedure is created.
Unfortunately, when it comes to the point in time that the procedure is executed, it needs to be able to compile all of the individual statements, and it's at this time that it will discover that the table exists but that the column doesn't - and so at this time, it will generate an error and refuse to run the procedure.
The T-SQL language is unfortunately a very simplistic compiler, and doesn't take runtime control flow into account when attempting to perform the compilation. It doesn't analyse the control flow or attempt to defer the compilation in conditional paths - it just fails the compilation because the column doesn't (at this time) exist.
Unfortunately, there aren't any mechanisms built in to SQL Server to control this behaviour - this is the behaviour you get, and anything that addresses it is going to be perceived as a workaround - as evidenced already by the (valid) suggestions in the comments - the two main ways to deal with it are to use dynamic SQL or to ensure that the temp table always contains all columns required.
One way to workaround your concerns about maintenance if you go down the "all uses of the temp table should have all columns" is to move the column definitions into a separate stored procedure, that can then augment the temporary table with all of the required columns - something like:
create procedure S_TT_Init
as
alter table #TT add Column1 int not null
alter table #TT add Column2 varchar(9) null
go
create procedure S_TT_Consumer
as
insert into #TT(Column1,Column2) values (9,'abc')
go
create procedure S_TT_User
as
create table #TT (tmp int null)
exec S_TT_Init
insert into #TT(Column1) values (8)
exec S_TT_Consumer
select Column1 from #TT
go
exec S_TT_User
Which produces the output 8 and 9. You'd put your temp table definition in S_TT_Init, S_TT_Consumer is the inner query that multiple stored procedures call, and S_TT_User is an example of one such stored procedure.
Create the table with the column initially. If you're populating the TEMP table with SPROC output just make it an IDENTITY INT (1,1) so the columns line up with your output.
Then drop the column and re-add it as the appropriate data type later on in the SPROC.
The only (or maybe best) way i can thing off beyond dynamic SQL is using checks for database structure.
if exists (Select 1 From tempdb.sys.columns Where object_id=OBJECT_ID('tempdb.dbo.#TTT') and name = 'AvgValue')
begin
--do something AvgValue related
end
maybe create a simple function that takes table name and column or only column if its always #TempTable and retursn 1/0 if the column exists, would be useful in the long run i think
if dbo.TempTableHasField('AvgValue')=1
begin
-- do something AvgValue related
end
EDIT1: Dang, you are right, sorry about that, i was sure i had ... this.... :( let me thing a bit more
Related
I'm using SQL Server 2008.
How can I pass Table Valued parameter to a Stored procedure across different Databases, but same server?
Should I create the same table type in both databases?
Please, give an example or a link according to the problem.
Thanks for any kind of help.
In response to this comment (if I'm correct and that using TVPs between databases isn't possible):
What choice do I have in this situation? Using XML type?
The purist approach would be to say that if both databases are working with the same data, they ought to be merged into a single database. The pragmatist realizes that this isn't always possible - but since you can obviously change both the caller and callee, maybe just use a temp table that both stored procs know about.
I don't believe it's possible - you can't reference a table type from another database, and even with identical type definitions in both DBs, a value of one type isn't assignable to the other.
You don't pass the temp table between databases. A temp table is always stored in tempdb, and is accessible to your connection, so long as the connection is open and the temp table isn't dropped.
So, you create the temp table in the caller:
CREATE TABLE #Values (ID int not null,ColA varchar(10) not null)
INSERT INTO #Values (ID,ColA)
/* Whatever you do to populate the table */
EXEC OtherDB..OtherProc
And then in the callee:
CREATE PROCEDURE OtherProc
/* No parameter passed */
AS
SELECT * from #Values
Table UDTs are only valid for stored procs within the same database.
So yes you would have to create the type on each server and reference it in the stored procs - e.g. just run the first part of this example in both DBs http://msdn.microsoft.com/en-us/library/bb510489.aspx.
If you don't need the efficency you can always use other methods - i.e. pass an xml document parameter or have the s.p. expect a temp table with the input data.
Edit: added example
create database Test1
create database Test2
go
use Test1
create type PersonalMessage as TABLE
(Message varchar(50))
go
create proc InsertPersonalMessage #Message PersonalMessage READONLY AS
select * from #Message
go
use Test2
create type PersonalMessage as TABLE
(Message varchar(50))
go
create proc InsertPersonalMessage #Message PersonalMessage READONLY AS
select * from #Message
go
use Test1
declare #mymsg PersonalMessage
insert #mymsg select 'oh noes'
exec InsertPersonalMessage #mymsg
go
use Test2
declare #mymsg2 PersonalMessage
insert #mymsg2 select 'oh noes'
exec InsertPersonalMessage #mymsg2
Disadvantage is that there are two copies of the data.
But you would be able to run the batch against each database simultaneously.
Whether this is any better than using a table table is really down to what processing/data sizes you have - btw to use a temp table from an s.p. you just access it from the s.p. code (and it fails if it doesn't exist).
Another way to solve this (though not necessarily the correct way) is to only utilize the UDT as a part of a dynamic SQL call.
USE [db1]
CREATE PROCEDURE [dbo].[sp_Db2Data_Sync]
AS
BEGIN
/*
*
* Presumably, you have some other logic here that requires this sproc to live in db1.
* Maybe it's how you get your identifier?
*
*/
DECLARE #SQL VARCHAR(MAX) = '
USE [db2]
DECLARE #db2tvp tableType
INSERT INTO #db2tvp
SELECT dataColumn1
FROM db2.dbo.tblData td
WHERE td.Id = ' + CAST(#YourIdentifierHere AS VARCHAR) '
EXEC db2.dbo.sp_BulkData_Sync #db2tvp
'
EXEC(#SQL)
END
It's definitely not a purist approach, and it doesn't work for every use case, but it is technically an option.
I created a stored proc that creates a temp table, inserts, selects then drops. Executing the stored proc within SQL Server Management Studio works fine and gives the expected result.
CREATE PROCEDURE usp_TempTableTest
-- Add the parameters for the stored procedure here
#color VARCHAR(10)
AS
BEGIN
-- SET NOCOUNT ON added to prevent extra result sets from
-- interfering with SELECT statements.
SET NOCOUNT ON;
CREATE TABLE #tmptable (
color VARCHAR(10)
)
INSERT INTO #tmptable (color) VALUES (#color)
SELECT color FROM #tmptable
DROP TABLE #tmptable
END
GO
However, when creating in the Import/Export tool and using that stored proc as the data source, it gives me the error:
Invalid object name '#tmptable'.
Any idea why this would happen? If I change it to a table variable it seems work fine with Import/Export, but I don't understand why it is not working with a temp table.
When I run a mimicked stored procedure, like yours above, in SSMS, I can get the data returned like you mentioned in the procedure. However, if I try the #tmptable, like you did, I also get the same error because the DROP TABLE removes it. From what I can tell, the import/export is basically a final INSERT process. The reason it works with the table variable is because the data still exist on the final insert; in the case of the DROP TABLE, it does not. For instance, when I remove the DROP TABLE, it works.
I might be wrong here, but it seems the logic when it's an import or export in the case of the above procedure is
INSERT data
SELECT data
DROP data
INSERT (import/export): this generates the "Invalid object name tmptable'"
With the variable (or no DROP), it's
INSERT data
SELECT data
INSERT (import/export)
In the second case, the data still exist. In the first case, they're gone. One way around it if you want to use the #tmptable, start your code with:
IF OBJECT_ID('tempdb..#tmptable') IS NOT NULL DROP TABLE #tmptable
Put "SET FMTONLY OFF;" right above "SET NOCOUNT ON"
http://msdn.microsoft.com/en-us/library/ms173839.aspx
If I try to execute the following code, I get the errors
Msg 207, Level 16, State 1, Line 3 Invalid column name 'Another'. Msg
207, Level 16, State 1, Line 4 Invalid column name 'Another'.
even though the predicate for both IF statements always evaluates to false.
CREATE TABLE #Foo (Bar INT)
GO
IF (1=0)
BEGIN
SELECT Another FROM #Foo
END
GO
IF (1=0)
BEGIN
ALTER TABLE #Foo ADD Another INT
SELECT Another FROM #Foo
END
GO
DROP TABLE #Foo
This is probably over-simplified for the sake of the example; in reality what I need to do is select the values from a column, but only if the column exists. If it doesn't exist, I don't care about it. In the problem that drove me to ask this question, my predicate was along the lines of EXISTS (SELECT * FROM sys.columns WHERE object_id = #ID AND name = #Name). Is there a way to achieve this without resorting to my arch-enemy Dynamic SQL? I understand that my SQL must always be well-formed (i.e. conform to grammar) - even within a block that's never executed - but I'm flabbergasted that I'm also being forced to make it semantically correct too!
EDIT:
Though I'm not sure the code below adds much to the code above, it's a further example of the problem. In this scenario, I only want to set the value of Definitely (which definitely exists as a column) with the value from Maybe (which maybe exists as a column) if Maybe exists.
IF EXISTS (SELECT * FROM sys.columns WHERE object_id = OBJECT_ID('dbo.TableName', 'U') AND name = 'Maybe')
BEGIN
UPDATE dbo.TableName SET Definitely = Maybe
END
SQL Server doesn't execute line by line. It isn't procedural like .net or Java code. So there is no "non-executed block"
The batch is compiled in one go. At this point, the column doesn't exist but it knows the table will be. Table does not have a column called "Another". Fail.
Exactly as expected.
Now, what is the real problem you are trying to solve?
Some options:
2 tables or one table with both columns
use Stored procedures to decouple scope
not use temp tables (maybe not needed; it could be your procedural thinking...)
dynamic SQL (from Mitch's deleted answer)
Edit, after comment;
Why not hide schema changes behind a view, rather than changing all code to work with columns that may/may not be there?
You can use EXEC to handle it. It's not really dynamic SQL if the code never actually changes.
For example:
CREATE TABLE dbo.Test (definitely INT NOT NULL)
INSERT INTO dbo.Test (definitely) VALUES (1), (2), (3)
IF EXISTS (SELECT *
FROM sys.columns
WHERE object_id = OBJECT_ID('dbo.Test', 'U') AND
name = 'Maybe')
BEGIN
EXEC('UPDATE dbo.Test SET definitely = maybe')
END
SELECT * FROM dbo.Test
ALTER TABLE dbo.Test ADD maybe INT NOT NULL DEFAULT 999
IF EXISTS (SELECT *
FROM sys.columns
WHERE object_id = OBJECT_ID('dbo.Test', 'U') AND
name = 'Maybe')
BEGIN
EXEC('UPDATE dbo.Test SET definitely = maybe')
END
SELECT * FROM dbo.Test
DROP TABLE dbo.Test
You can also try Martin Smith's "Workaround" using a non-existing table to get "deferred name resolution" for columns.
I had the same issue.
We are creating a script for all changes for years and this is the first time that we have this issue.
I've tried all your answers and didn't find the issue.
In my case it was because of temporary table within the script that I'm using also within a stored procedure, although every sentence has go.
I've found that if I'm adding if exists with drop to the temporary table after the script is using the temporary table, it is working correctly.
Best regards,
Chen
Derived from the answer by #gbn.
What i did to solve the issue was to use 'GO' between the ALTER query and the query that uses the column added by ALTER. This will make the 2 queries to be run as separate batches thereby ensuring your 'Another' column is there before the SELECT query.
I have a Stored Procedure in MSSQL 2008, inside of this i've created a Temporary Table, and then i executed several inserts into the temporary Table.
How can i select all the columns of the Temporary Table outside the stored procedure? I Mean, i have this:
CREATE PROCEDURE [dbo].[LIST_CLIENTS]
CREATE TABLE #CLIENT(
--Varchar And Numeric Values goes here
)
/*Several Select's and Insert's against the Temporary Table*/
SELECT * FROM #CLIENT
END
In another Query i'm doing this:
sp_configure 'Show Advanced Options', 1
GO
RECONFIGURE
GO
sp_configure 'Ad Hoc Distributed Queries', 1
GO
RECONFIGURE
GO
SELECT *
INTO #CLIENT
FROM OPENROWSET
('SQLOLEDB','Server=(local);Uid=Cnx;pwd=Cnx;database=r8;Trusted_Connection=yes;
Integrated Security=SSPI',
'EXEC dbo.LIST_CLIENTS ''20110602'', NULL, NULL, NULL, NULL, NULL')
But i get this error:
Msg 208, Level 16, State 1, Procedure LIST_CLIENTS, Line 43
Invalid object name '#CLIENT'.
I've tried with Global Temporary Tables and It doesn't work.
I know that is the scope of the temporary table, but, how can i get the table outside the scope of the SP?
Thanks in advance
I think there is something deeper going on here.
One idea is to use a table variable inside the stored procedure instead of a #temp table (I have to assume you're using SQL Server 2005+ but it's always nice to state this up front). And use OPENQUERY instead of OPENROWSET. This works fine for me:
USE tempdb;
GO
CREATE PROCEDURE dbo.proc_x
AS
BEGIN
SET NOCOUNT ON;
DECLARE #x TABLE(id INT);
INSERT #x VALUES(1),(2);
SELECT * FROM #x;
END
GO
SELECT *
INTO #client
FROM OPENQUERY
(
[loopback linked server name],
'EXEC tempdb.dbo.proc_x'
) AS y;
SELECT * FROM #client;
DROP TABLE #client;
DROP PROCEDURE dbo.proc_x;
Another idea is that perhaps the error is occurring even without using SELECT INTO. Does the stored procedure reference the #CLIENT table in any dynamic SQL, for example? Does it work when you call it on its own or when you just say SELECT * FROM OPENROWSET instead of SELECT INTO? Obviously, if you are working with the #temp table in dynamic SQL you're going to have the same kind of scope issue working with a #table variable in dynamic SQL.
At the very least, name your outer #temp table something other than #CLIENT to avoid confusion - then at least nobody has to guess which #temp table is not being referenced correctly.
Since the global temp table failed, use a real table, run this when you start your create script and drop the temp table once you are done to make sure.
IF OBJECT_ID('dbo.temptable', 'U') IS NOT NULL
BEGIN
DROP TABLE dbo.temptable
END
CREATE TABLE dbo.temptable
( ... )
You need to run the two queries within the same connection and use a global temp table.
In SQL Server 2008 you can declare User-Defined Table Types which represent the definition of a table structure. Once created you can create table parameters within your procs and pass them a long and be able to access the table in other procs.
I guess the reason for such behavior is that when you call OPENROWSET from another server it firstly and separately requests the information about procedure output structure (METADATA). And the most interesting thing is that this output structure is taken from the first SELECT statement found in the procedure. Moreover, if the SELECT statement follows the IF-condition the METADATA request ignores this IF-condition, because there is no need to run the whole procedure - the first met SELECT statement is enough. (By the way, to switch off that behavior, you can include SET FMTONLY OFF in the beginning of your procedure, but this might increase the procedure execution time).
The conclusions:
— when the METADATA is being requested from a temp table (created in a procedure) it does not actually exists, because the METADATA request does not actually run the procedure and create the temp table.
— if a temp table can be replaced with a table variable it solves the problem
— if it is vital for the business to use temp table, the METADATA request can be fed with fake first SELECT statement, like:
declare #t table(ID int, Name varchar(15));
if (0 = 1) select ID, Name from #t; -- fake SELECT statement
create table #T (ID int, Name varchar(15));
select ID, Name from #T; -- real SELECT statement
— and one more thing is to use a common trick with FMTONLY (that is not my idea) :
declare #fmtonlyOn bit = 0;
if 1 = 0 set #fmtonlyOn = 1;
set fmtonly off;
create table #T (ID int, Name varchar(15));
if #fmtonlyOn = 1 set fmtonly on;
select ID, Name from #T;
The reason you're getting the error is because the temp table #Client was not declared before you ran the procedure to insert into it. If you declare the table, then execute the list proc and use direct insert -
INSERT INTO #Client
EXEC LIST_CLIENTS
Microsoft SQL Server seems to check column name validity, but not table name validity when defining stored procedures. If it detects that a referenced table name exists currently, it validates the column names in a statement against the columns in that table. So, for example, this will run OK:
CREATE PROCEDURE [dbo].[MyProcedure]
AS
BEGIN
SELECT
Col1, Col2, Col3
FROM
NonExistentTable
END
GO
... as will this:
CREATE PROCEDURE [dbo].[MyProcedure]
AS
BEGIN
SELECT
ExistentCol1, ExistentCol2, ExistentCol3
FROM
ExistentTable
END
GO
... but this fails, with 'Invalid column name':
CREATE PROCEDURE [dbo].[MyProcedure]
AS
BEGIN
SELECT
NonExistentCol1, NonExistentCol2, NonExistentCol3
FROM
ExistentTable
END
GO
Why does SQL Server check columns, but not tables, for existence? Surely it's inconsistent; it should do both, or neither. It's useful for us to be able to define SPs which may refer to tables AND/OR columns which don't exist in the schema yet, so is there a way to turn off SQL Server's checking of column existence in tables which currently exist?
This is called deferred name resolution.
There is no way of turning it off. You can use dynamic SQL or (a nasty hack!) add a reference to a non existent table so that compilation of that statement is deferred.
CREATE PROCEDURE [dbo].[MyProcedure]
AS
BEGIN
CREATE TABLE #Dummy (c int)
SELECT
NonExistantCol1, NonExistantCol2, NonExistantCol3
FROM
ExistantTable
WHERE NOT EXISTS(SELECT * FROM #Dummy)
DROP TABLE #Dummy
END
GO
This article in MSDN should answer your question.
From the article:
When a stored procedure is executed for the first time, the query
processor reads the text of the stored procedure from the
sys.sql_modules catalog view and checks that the names of the objects
used by the procedure are present. This process is called deferred
name resolution because table objects referenced by the stored
procedure need not exist when the stored procedure is created, but
only when it is executed.