I have always learn that UDFs are not Pre-Compiled while Stored procedures are. But I always had this questions why are UDFs are not Pre-Compiled.
I searched online for the answer but dint get an concrete justification on the same.
But it doesn't explicitly tell wherther Interpreted (disk-based) tables can be accessed by Natively compiled stored procedures.And SQL Server Express edition doesn't allow creating Memory-optimized objects to very this.
View 2 Replies View RelatedHow can I list the stored procedures and user-defined functions that reference a given column? I could search ROUTINE_DEFINITION in INFORMATION_SCHEMA.ROUTINES for '%MyColumnName%' but MyColumnName is not always unique.
Thanks.
I want to know the differences between SQL Server 2000 storedprocedures and oracle stored procedures? Do they have differentsyntax? The concept should be the same that the stored proceduresexecute in the database server with better performance?Please advise good references for Oracle stored procedures also.thanks!!
View 11 Replies View RelatedHi,
This Might be a really simple thing, however we have just installed SQL server 2005 on a new server, and are having difficulties with the set up of the Store Procedures. Every time we try to modify an existing stored procedure it attempts to save it as an SQL file, unlike in 2000 where it saved it as part of the database itself.
Thank you in advance for any help on this matter
I am under the impression that triggers are not pre-compiled like SP's. True or false?
View 3 Replies View Related
Using SQL 2005, SP2. All of a sudden, whenever I create any stored procedures in the master database, they get created as system stored procedures. Doesn't matter what I name them, and what they do.
For example, even this simple little guy:
CREATE PROCEDURE BOB
AS
PRINT 'BOB'
GO
Gets created as a system stored procedure.
Any ideas what would cause that and/or how to fix it?
Thanks,
Jason
Hello Forum
I have an issue that hopefully someone can help me with. We have SQL server running in a 64 bit environment. We are running SQL Server 2005 SP2. The problem I am experiencing is that my SSIS script components produce the following error when the SSIS package is executed from the server.
Code: 0x00000003 Source: Set Perspective Connection Description: Precompiled script failed to load. The computer that is running the Script Task does not have a compiler capable of recompiling the script. To recompile the script, see the Microsoft Knowledge Base article, KB931846 (http://go.microsoft.com/fwlink/?LinkId=81885).
I have reviewed this article and everything points to SP2 as being the fix. Being that I am running SP2 there is something else going on. I was curious to see if anyone else has run into this issue with SP2.
Any input you have is greatly appreciated.
Thanks in advance.
we're about to plan exactly how we install rs reports on our customers' systems and are wondering if there is more of a binary/compiled format that can be deployed as opposed to the xml source we see in .rdl files?
We keep the lowest form of everything (.dlls, sql scripts etc) in deployment directories and wonder if RS2005's lowest (or most efficient for deployment) form is the .rdl itself. I noticed that Reporting Services (when scripting a report in mgt studio) shows what appears to be the .rdl wrapped in VB but dont really know if that form has to be stored or not on the ReportServer DB in order for RS to work. I couldnt find a ReportServer DB table that had this form but have to believe it is there somewhere if Reporting Services can display it. But maybe a format more geared toward execution is stored along side it?
So if .rdl can be considered "source code" I guess the question is "does it need to be deployed in that form for RS to work or is a more binary like format also available"?
1. Create table and 2 triggers with script below (*)
2. Run the following script:
use testjrl;
begin transaction;
update testtable
set workphone = '333' where id = 1;
rollback;
Get following error message:
(1 row(s) affected)
Msg 2801, Level 16, State 1, Procedure trgTestTableInsertUpdate, Line 7
The definition of object 'trgTestTableInsertUpdate' has changed since it was compiled.
The statement has been terminated.
3. Change trgTestTableInsertUpdate trigger to
alter table TestTable disable trigger trgTestTableDelete
select 1
alter table TestTable enable trigger trgTestTableDelete
4. Run the script from step 2 again - works fine.
Questions:
a. Why does first error occur?
b. Why does error go away when you specify trigger name to disable?
Thanks
Jim
*Script to create table and triggers:
CREATE TABLE [dbo].[TestTable](
[id] [int] IDENTITY(1,1) NOT NULL,
[WorkPhone] [nchar](10) COLLATE SQL_Latin1_General_CP1_CI_AS NULL,
CONSTRAINT [PK_TestTable] PRIMARY KEY CLUSTERED
(
[id] ASC
)WITH (PAD_INDEX = OFF, IGNORE_DUP_KEY = OFF) ON [PRIMARY]
) ON [PRIMARY]
GO
CREATE TRIGGER [dbo].[trgTestTableInsertUpdate] ON [dbo].[TestTable]
FOR INSERT, UPDATE
AS
alter table TestTable disable trigger all
select 1
alter table TestTable enable trigger all
GO
CREATE TRIGGER [dbo].[trgTestTableDelete] ON [dbo].[TestTable]
FOR DELETE
AS
select 1
GO
I have some stored procedure and there is a cursor inside it. I added some new columns to the table and those columns I included in the cursor declaration and fetch statement. In the cursor declaration I forgot to add comma (,) in between the new columns. So SQL Server it considers as a alias name for that column so syntactically it is correct. But logically in the cursor declaration having less number of columns than the columns in the fetch statement. So it should throw an error. But the procedure is getting compiled without raising any error. But if I execute the procedure that time it is throwing the error.
For example, below I have given the sample procedure. In this procedure, in the cursor declaration I removed the comma (,) between DOB and DOJ. If I compile this procedure it is getting compiled. But when execute that time only it is throwing the error. So I am interested in if any option is available to know the error in the compilation time itself.
ALTER PROCEDURE Test
AS
BEGIN
BEGIN TRY
DECLARE @empId INT,
@fname VARCHAR(50),
@dob DATE,
@doj DATE
[code]....
How would I go about tracing UDF performance in profiler? I'd like to
specifically know the impact of the UDF without having to dig into the
execution plan of the statement containing it. Is this possible?
Can changes made to replicated stored procedures (and User-defined Functions) on the Publisher update the Subscriber database?
If I ALTER a SP (or UDF) on Database1 with peer-to-peer Transaction Replication to Database2, will the alter replicate to the Subscriber(s)?
Thomas
ThomBeaux
I know there is a lot of information already out there on this topic,but given the following scenario...--------------------------------------------------------------------------------------------------------------------------------------Create a view like so ( pardon the pseudo-code )...CREATE View vwContactAddressesSelect * FROM Contact INNER JOIN Address ON Contact.ContactID =Address.ContactIDAnd then do a sargable select from the view using a stored procedureCREATE STORED PROCEDURE spSelect_ContactAddresses@ContactID intASSelect * FROM vwContactAddresses WHERE ContactID = @ContactID--------------------------------------------------------------------------------------------------------------------------------------In my understanding, "vwContactAddresses" would be substituted with theactual SQL join statement when the view is accessed.So for the stored procedure in question an execution plan for"Select * FROM Contact INNER JOIN Address ON Contact.ContactID =Address.ContactID WHERE ContactID = @ContactID" would be cached.Correct?With regards to execution plan caching, is this not the same ascreating an inline UDF that takes parameters or just creating a storedprocedure that would do the join w/out the view reference?
View 3 Replies View RelatedHi,
What is the difference between a computed column and a UDF?
Is a computed column the same as the "Formula" field under Design Table in Enterprise Manager?
Also, what is the proper syntax for the Formula field? Can I use regular SQL on it or is there more to it?
thanks,
Frank
-- Author: Igor Nikiforov, Montreal, EMail: udfs@sympatico.ca
-- GETWORDCOUNT() User-Defined Function Counts the words in a string.
-- GETWORDCOUNT(@cString[, @cDelimiters])
-- Parameters
-- @cString nvarchar(4000) - Specifies the string whose words will be counted.
-- @cDelimiters nvarchar(256) - Optional. Specifies one or more optional characters used to separate words in @cString.
-- The default delimiters are space, tab, carriage return, and line feed. Note that GETWORDCOUNT( ) uses each of the characters in @cDelimiters as individual delimiters, not the entire string as a single delimiter.
-- Return Value smallint
-- Remarks GETWORDCOUNT() by default assumes that words are delimited by spaces or tabs. If you specify another character as delimiter, this function ignores spaces and tabs and uses only the specified character.
-- If you use 'AAA aaa, BBB bbb, CCC ccc.' as the target string for dbo.GETWORDCOUNT(), you can get all the following results.
-- declare @cString nvarchar(4000)
-- set @cString = 'AAA aaa, BBB bbb, CCC ccc.'
-- select dbo.GETWORDCOUNT(@cString, default) -- 6 - character groups, delimited by ' '
-- select dbo.GETWORDCOUNT(@cString, ',') -- 3 - character groups, delimited by ','
-- select dbo.GETWORDCOUNT(@cString, '.') -- 1 - character group, delimited by '.'
-- See Also GETWORDNUM() User-Defined Function
-- UDF the name and functionality of which correspond to the same built-in function of Visual FoxPro
CREATE function GETWORDCOUNT (@cSrting nvarchar(4000), @cDelimiters nvarchar(256) )
returns smallint
as
begin
-- if no break string is specified, the function uses spaces, tabs and line feed to delimit words.
set @cDelimiters = isnull(@cDelimiters, space(1)+char(9)+char(10))
declare @p smallint, @end_of_string smallint, @wordcount smallint
select @p = 1, @wordcount = 0
select @end_of_string = 1 + datalength(@cSrting)/(case SQL_VARIANT_PROPERTY(@cSrting,'BaseType') when 'nvarchar' then 2 else 1 end) -- for unicode
while dbo.CHARINDEX_BIN(substring(@cSrting, @p, 1), @cDelimiters, 1) > 0 and @end_of_string > @p -- skip opening break characters, if any
set @p = @p + 1
if @p < @end_of_string
begin
set @wordcount = 1 -- count the one we are in now count transitions from 'not in word' to 'in word'
-- if the current character is a break char, but the next one is not, we have entered a new word
while @p < @end_of_string
begin
if @p +1 < @end_of_string and dbo.CHARINDEX_BIN(substring(@cSrting, @p, 1), @cDelimiters, 1) > 0 and dbo.CHARINDEX_BIN(substring(@cSrting, @p+1, 1), @cDelimiters, 1) = 0
select @wordcount = @wordcount + 1, @p = @p + 1 -- Skip over the first character in the word. We know it cannot be a break character.
set @p = @p + 1
end
end
return @wordcount
end
GO
-- Author: Igor Nikiforov, Montreal, EMail: udfs@sympatico.ca
-- GETWORDNUM() User-Defined Function
-- Returns a specified word from a string.
-- GETWORDNUM(@cString, @nIndex[, @cDelimiters])
-- Parameters @cString nvarchar(4000) - Specifies the string to be evaluated
-- @nIndex smallint - Specifies the index position of the word to be returned. For example, if @nIndex is 3, GETWORDNUM( ) returns the third word (if @cString contains three or more words).
-- @cDelimiters nvarchar(256) - Optional. Specifies one or more optional characters used to separate words in @cString.
-- The default delimiters are space, tab, carriage return, and line feed. Note that GetWordNum( ) uses each of the characters in @cDelimiters as individual delimiters, not the entire string as a single delimiter.
-- Return Value nvarchar(4000)
-- Remarks Returns the word at the position specified by @nIndex in the target string, @cString. If @cString contains fewer than @nIndex words, GETWORDNUM( ) returns an empty string.
-- See Also
-- GETWORDCOUNT() User-Defined Function
-- UDF the name and functionality of which correspond to the same built-in function of Visual FoxPro
CREATE function GETWORDNUM (@cSrting nvarchar(4000), @nIndex smallint, @cDelimiters nvarchar(256) )
returns nvarchar(4000)
as
begin
-- if no break string is specified, the function uses spaces, tabs and line feed to delimit words.
set @cDelimiters = isnull(@cDelimiters, space(1)+char(9)+char(10))
declare @i smallint, @j smallint, @p smallint, @q smallint, @qmin smallint, @end_of_string smallint, @LenDelimiters smallint, @outstr nvarchar(4000)
select @i = 1, @p = 1, @q = 0, @outstr = ''
select @end_of_string = 1 + datalength(@cSrting)/(case SQL_VARIANT_PROPERTY(@cSrting,'BaseType') when 'nvarchar' then 2 else 1 end) -- for unicode
select @LenDelimiters = datalength(@cDelimiters)/(case SQL_VARIANT_PROPERTY(@cDelimiters,'BaseType') when 'nvarchar' then 2 else 1 end) -- for unicode
while @i <= @nIndex
begin
while dbo.CHARINDEX_BIN(substring(@cSrting, @p, 1), @cDelimiters, 1) > 0 and @end_of_string > @p -- skip opening break characters, if any
set @p = @p + 1
if @p >= @end_of_string
break
select @j = 1, @qmin = @end_of_string -- find next break character it marks the end of this word
while @j <= @LenDelimiters
begin
set @q = dbo.CHARINDEX_BIN(substring(@cDelimiters, @j, 1), @cSrting, @p)
set @j = @j + 1
if @q > 0 and @qmin > @q
set @qmin = @q
end
if @i = @nIndex -- this is the actual word we are looking for
begin
set @outstr = substring(@cSrting, @p, @qmin-@p)
break
end
set @p = @qmin + 1
if (@p >= @end_of_string)
break
set @i = @i + 1
end
return @outstr
end
GO
-- Is similar to the built-in function Transact-SQL charindex, but regardless of collation settings,
-- executes case-sensitive search
-- Author: Igor Nikiforov, Montreal, EMail: udfs@sympatico.ca
CREATE function CHARINDEX_BIN(@expression1 nvarchar(4000), @expression2 nvarchar(4000), @start_location smallint = 1)
returns nvarchar(4000)
as
begin
return charindex( cast(@expression1 as nvarchar(4000)) COLLATE Latin1_General_BIN, cast(@expression2 as nvarchar(4000)) COLLATE Latin1_General_BIN, @start_location )
end
GO
Hi!I need to refresh an entire database.I can recompile SPs with sp_recompile (or DBCC FLUSHPROCINDB), andrefresh views with sp_refreshView, but I cannot find any way torefresh my user-defined functions (some of them are like views, withparameters).Any help appreciated :) !Ben
View 5 Replies View RelatedHow do I search for and print all stored procedure names in a particular database? I can use the following query to search and print out all table names in a database. I just need to figure out how to modify the code below to search for stored procedure names. Can anyone help me out?
SELECT TABLE_SCHEMA + '.' + TABLE_NAME
FROM INFORMATION_SCHEMA.TABLES
WHERE TABLE_TYPE = 'BASE TABLE'
Seems like I'm stealing all the threads here, : But I need to learn :) I have a StoredProcedure that needs to return values that other StoredProcedures return.Rather than have my DataAccess layer access the DB multiple times, I would like to call One stored Procedure, and have that stored procedure call the others to get the information I need. I think this way would be more efficient than accessing the DB multiple times. One of my SP is:SELECT I.ItemDetailID, I.ItemDetailStatusID, I.ItemDetailTypeID, I.Archived, I.Expired, I.ExpireDate, I.Deleted, S.Name AS 'StatusName', S.ItemDetailStatusID, S.InProgress as 'StatusInProgress', S.Color AS 'StatusColor',T.[Name] AS 'TypeName', T.Prefix, T.Name AS 'ItemDetailTypeName', T.ItemDetailTypeID FROM [Item].ItemDetails I INNER JOIN Item.ItemDetailStatus S ON I.ItemDetailStatusID = S.ItemDetailStatusID INNER JOIN [Item].ItemDetailTypes T ON I.ItemDetailTypeID = T.ItemDetailTypeID However, I already have StoredProcedures that return the exact same data from the ItemDetailStatus table and ItemDetailTypes table.Would it be better to do it above, and have more code to change when a new column/field is added, or more checks, or do something like:(This is not propper SQL) SELECT I.ItemDetailID, I.ItemDetailStatusID, I.ItemDetailTypeID, I.Archived, I.Expired, I.ExpireDate, I.Deleted, EXEC [Item].ItemDetailStatusInfo I.ItemDetailStatusID, EXEC [Item].ItemDetailTypeInfo I.ItemDetailTypeID FROM [Item].ItemDetails IOr something like that... Any thoughts?
View 3 Replies View RelatedGreetings:
I have MSSQL 2005. On earlier versions of MSSQL saving a stored procedure wasn't a confusing action. However, every time I try to save my completed stored procedure (parsed successfully ) I'm prompted to save it as a query on the hard drive.
How do I cause the 'Save' action to add the new stored procedure to my database's list of stored procedures?
Thanks!
Hi,
I have written an udf that checks for validations of an email address and returns 0 / 1 as per the validations. I came across an artical on MSDN that describes the steps to make my own system udf. http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnsqlpro01/html/sql01l1.asp
with help of that artical, I made my udf as a system udf. and now I can use it as follows.
select * , fn_isvalidemail(email) from #tp
now, I wanted to use this as a declarative check constraints in my tables. so I tried
create table testtable
( email_addr varchar (255) null CONSTRAINT CK_email_chk CHECK ( fn_isvalidemail(email) = 1) )
although this is syntax is logically correct, it does not get compiled
it gives me an error saying,
Server: Msg 195, Level 15, State 10, Line 2
'fn_isvalidemail' is not a recognized function name.
the strange part is that i made the function as another normal udf, It works in declarative constraints.
create table testtable
( email varchar (255) null CONSTRAINT CK_1p12 CHECK ( dbo.isvalidemail(email) = 1) )
Any thoughts on this ?
- Amit
Hello,
How can I call UDFs on linked servers in SQL Server 2005?
More details:
I have some UDFs on Instance "VAXPServerA" in Database "DB1", for example "getSomething(number)".
It might look like this (for example):
----
CREATE FUNCTION [dbo].[getSomething] (
@theNumber int
)
RETURNS int AS
BEGIN
DECLARE @result int
SET @result = theNumber * 5
RETURN @result
END
----
I can call this function inside of the DB by typing:
----
SELECT dbo.getSomething(5)
----
I also can call the function from another DB on instance "VAXPServerA" by typing:
----
SELECT DB1.dbo.getSomething(5)
----
I have a second instance called "VAXPServerB", and "VAXPServerA" is a linked server in "VAXPServerB". I can do selects and stuff, it works fine.
But when I want to call the function...
----
SELECT [VAXPServerA].DB1.dbo.getSomething(5)
----
I obtain the next error:
Msg. 207, Level 16, State 1, Line 1
The column name 'VAXPServerA' is not valid.
Any hint?
Thanks in advance
David
We recently upgraded to SQL Server 2005. We had several stored procedures in the master database and, rather than completely rewriting a lot of code, we just recreated these stored procedures in the new master database.
For some reason, some of these stored procedures are getting stored as "System Stored Procedures" rather than just as "Stored Procedures". Queries to sys.Objects and sys.Procedures shows that these procs are being saved with the is_ms_shipped field set to 1, even though they obviously were not shipped with the product.
I can't update the sys.Objects or sys.Procedures views in 2005.
What effect will this flag (is_ms_shipped = 1) have on my stored procedures?
Can I move these out of "System Stored Procedures" and into "Stored Procedures"?
Thanks!
Hello friends......How are you ? I want to ask you all that how can I do the following ?
I want to now that how many ways are there to do this ?
How can I call one or more stored procedures into perticular one Stored Proc ? in MS SQL Server 2000/05.
Hello
I'm start to work with SSIS.
We have a lot (many hundreds) of old (SQL Server2000) procedures on SQL 2005.
Most of the Stored Procedures ends with the following commands:
SET @SQLSTRING = 'SELECT * INTO ' + @OutputTableName + ' FROM #RESULTTABLE'
EXEC @RETVAL = sp_executeSQL @SQLSTRING
How can I use SSIS to move the complete #RESULTTABLE to Excel or to a Flat File? (e.g. as a *.csv -File)
I found a way but I think i'ts only a workaround:
1. Write the #Resulttable to DB (changed Prozedure)
2. create data flow task (ole DB Source - Data Conversion - Excel Destination)
Does anyone know a better way to transfer the #RESULTTABLE to Excel or Flat file?
Thanks for an early Answer
Chaepp
I have a bunch of SPs that all rely on a UDF that parses a commadelimitted list of numbers into a table. Everything was working fine,but now my application is growing and I'm starting to approach the 8000character limit of the varChar variable used to store the list.I would like to change the UDF only and avoid having to dig through allof my stored procedures. I was hoping to use the text datatype toallow for much larger lists, but I am unable to perform anymanipulations necessary to parse the list into a table. I have triedPATINDEX, but it alone is not enough without the text maniuplations andI don't think the sp_xml_preparedocument can be used in a UDF.Anyone with any thoughts on managing large arrays in t-sql?thanks,Matt Weiner
View 2 Replies View RelatedHello Folks,I encountered a problem with SQL server 2000 and UDFs.I have a scalar UDF and a table UDF where I would like the scalar UDFto provide the argument for the table UDF like in:SELECT*FROMtransaction_tWHEREtrxn_gu_id in (select get_trxns_for_quarter(get_current_quarter( GetDate() ) ))'get_current_quarter' returns an integer which is a GUID in a tablecontaining business quarter definitions, like start date, end date.'get_current_quarter' is a scalar UDF.'get_trxns_for_quarter' will then get all transctions that fall intothat quarter and return their GUID's in a table.'get_trxns_for_quarter' is a table UDF.This doesn't seem to work at all. Regardless whether I provide thenamespace (schema) calling the scalar UDF or not. Error message isjust different.Both functions operate correctly invoked un-nested.The whole expression does work fine if I turn 'get_trxns_for_quarter'into a scalar UDF as well, e.g. by returning just one trxn_gu_id withe.g. MAX() in a scalar datatype. But of course that's no good to me.It also works fine if I select the result of 'get_current_quarter'into a variable and pass that variable into 'get_trxns_for_quarter'.But that's no good to me either since then I cannot use the wholething embedded into other SELECT clauses.Both UDF's are non-deterministic but I couldnt see how that would havean impact anyway.Never mind the syntax on that example or anyhting, I tried all theobvious and not so obvious stuff and it really seems to come down tothe fact that one UDF is scalar and the other one is not. However, Idid not come across any type of information saying that this cannot bedone.Have you any ideas?Any help would be greatly appreciated.Carsten
View 6 Replies View RelatedI am in the process of building my first "large scale" database system (after 15+ years of developing Windows Apps and Web Apps) - so I am very VERY "Green" when it comes to Database development & SQL et al.
A little context setting: I am building a multi-tier Statistical Analysis & Reporting system where the "end product" will be Reports created in Reporting Services. There are a ton of business rules that I am implementing in a Business Logic Tier (hidden from the "end user" by a Data Access Tier) comprised of SQL in the form of UDFs (scalar) and Views.
The question: I have been reading that UDFs cause a performance hit compared to things like in-line functions. Alot of the Rules (implemented as Scalar UDFs) build on each other so that the output of UDF #1 is used as input to UDF #2.
So far I am implementing the Business Logic as a hierarchy of Views (7 Views to be exact) with each view implementing multiple Rules; each Rule basically a Scalar UDF. Below is an example of what I am doing:
Example
View #1 -> Select A, B, C, funcX1(A) as ValueX1, funcY1(B, C) as ValueY1 FROM someView
Then
View #2 -> Select A, B, C, ValueX1, ValueY1, funcX2 (ValueX1) as ValueX2, funcY2(ValueY2) as ValueY2 FROM View#1
Currently I have a hierarchy of 7 views that each use UDFs to implement the Business Rules, where the value calculated from a UDF in one View is used as input to UDF in a View further down the Hierarchy.
Is there a better way of implementing all of the Rules instead of using multiple Views with a bunch of UDFs?
The "end product" dataset is then exposed as a Stored Procedure to the reports in Reporting Services.
Any help would be GREATLY appreciated.
Thanks!
- marty
There is something very strange going on here. Tested with ADO 2.7 andMSDE/2000. At first, things look quite sensible.You have a simple SQL query, let's sayselect * from mytab where col1 = 1234Now, let's write a simple VB program to do this query back to anMSDE/2000 database on our local machine. Effectively, we'llrs.open sSQLrs.closeand do that 1,000 times. We wont bother fetching the result set, itisn't important in this example.No problem. On my machine this takes around 1.6 seconds and modifyingthe code so that the column value in the where clause changes eachtime (i.e col1 = nnnn), doesn't make a substantial difference to thistime. Well, that all seems reasonable, so moving right along...Now we do it with a stored procedurecreate procedure proctest(@id int)asselect * from mytab where col1 = @idand we now find that executingproctest nnnn1,000 times takes around 1.6 seconds whether or not the argumentchanges. So far so good. No obvious saving, but then we wouldn'texpect any. The query is very simple, after all.Well, get to the point!Now create a table-returning UDFcreate function functest(@id int) returns table asreturn(select * from mytab where col1 = @id)try calling that 1,000 times asselect * from functest(nnnn)and we get around 5.5 seconds on my machine if the argument changes,otherwise 1.6 seconds if it remains the same for each call.Hmm, looks like the query plan is discarded if the argument changes.Well, that's fair enough I guess. UDFs might well be more expensive...gotta be careful about using them. It's odd that discarding the queryplan seems to be SO expensive, but hey, waddya expect?. (perhaps theUDF is completely rebuilt, who knows)last test, then. Create an SP that calls the UDFcreate procedure proctest1(@id int)asselect * from functest(@id)Ok, here's the $64,000 question. How long will this take if @idchanges each time. The raw UDF took 5.5 seconds, remember, so thisshould be slightly slower.But... IT IS NOT.. It takes 1.6 seconds whether or not @id changes.Somehow, the UDF becomes FOUR TIMES more efficient when wrapped in anSP.My theory, which I stress is not entirely scientific, goes somethinglike this:-I deduce that SQL Server decides to reuse the query plan in thiscircumstance but does NOT when the UDF is called directly. This iscounter-intuitive but it may be because SQL Server's query parser istuned for conventional SQL i.e it can saywell, I've gotselect * from mytab WHERE [something or other]and now I've gotselect * from mytab WHERE [something else]so I can probably re-use the query plan from last time. (I don't knowif it is this clever, but it does seem to know when twotextually-different queries have some degree of commonality)Whereas withselect * from UDF(arg1)andselect * from UDF(arg2)it goes... hmm, mebbe not.... I better not risk it.But withsp_something arg1andsp_something arg2it goes... yup, i'll just go call it... and because the SP was alreadycompiled, the internal call to the UDF already has a query plan.Anyway, that's the theory. For more complex UDFs, by the way, theperformance increase can be a lot more substantial. On a big complexUDF with a bunch of joins, I measured a tenfold increase inperformance just by wrapping it in an SP, as above.Obviously, wrapping a UDF in an SP isn't generally a good thing; theidea of UDFs is to allow the column list and where clause to filterthe rowset of the UDF, but if you are repeatedly calling the UDF withthe same where clause and column list, this will make it a *lot*faster.
View 3 Replies View RelatedWe have migrated a database (myDb2014) to a SQL Server 2014 instance. We have another on a 2008R2 instance (myDb2008). Both on the same server.
I have setup a linked server to the 2014 instance from the 2008R2 one.
I have a number of sprocs on myDb2008 that call TVFs and sprocs on myDb2014. This worked fine when they were both on the same instance.
Now however, I have an issue where I would have to update all the references in each calling query.
I looked into creating Synonyms but this only works for Tables/Views.
Is there a workaround to querying myDb2014 TVFs/sprocs without having to update each calling query in myDb2008?
Hi,
Do you know how to write stored procedures inside another stored procedure in MS SQL.
Create procedure spMyProc inputData varchar(50)
AS
----- some logical
procedure spMyProc inputInsideData varchar(10)
AS
--- some logical
--- go
-------
I am writing a set of store procedures (around 30), most of them require the same basic logic to get an ID, I was thinking to add this logic into an stored procedure.
The question is: Would calling an stored procedure from within an stored procedure affect performance? I mean, would it need to create a separate db connection? am I better off copying and pasting the logic into all the store procedures (in terms of performance)?
Thanks in advance
John
Hi all - I'm trying to optimized my stored procedures to be a bit easier to maintain, and am sure this is possible, not am very unclear on the syntax to doing this correctly. For example, I have a simple stored procedure that takes a string as a parameter, and returns its resolved index that corresponds to a record in my database. ie
exec dbo.DeriveStatusID 'Created'
returns an int value as 1
(performed by "SELECT statusID FROM statusList WHERE statusName= 'Created')
but I also have a second stored procedure that needs to make reference to this procedure first, in order to resolve an id - ie:
exec dbo.AddProduct_Insert 'widget1'
which currently performs:SET @statusID = (SELECT statusID FROM statusList WHERE statusName='Created')INSERT INTO Products (productname, statusID) VALUES (''widget1', @statusID)
I want to simply the insert to perform (in one sproc):
SET @statusID = EXEC deriveStatusID ('Created')INSERT INTO Products (productname, statusID) VALUES (''widget1', @statusID)
This works fine if I call this stored procedure in code first, then pass it to the second stored procedure, but NOT if it is reference in the second stored procedure directly (I end up with an empty value for @statusID in this example).
My actual "Insert" stored procedures are far more complicated, but I am working towards lightening the business logic in my application ( it shouldn't have to pre-vet the data prior to executing a valid insert).
Hopefully this makes some sense - it doesn't seem right to me that this is impossible, and am fairly sure I'm just missing some simple syntax - can anyone assist?