I've got a user defined table type called StringDictionaryTVP:
CREATE TYPE [dbo].[StringDictionaryTVP] AS TABLE( [key] [varchar](500) NULL, [value] [varchar](500) NULL )
Ideally I would like to be able to have a # of columns and directions in this table like so:
DECLARE @OrderByClause StringDictionaryTVP
INSERT INTO @OrderByClause([key], [value]) values('gender','desc') INSERT INTO @OrderByClause([key], [value]) values('name','asc')
Since our database can be a bit sizable, I'd also like to use Common Table Expressions so I can page through them fairly easy.So my standard cte is something like this:
DECLARE @PageIndex INT = 0 DECLARE @PageSize INT = 20 ;WITH results_cte AS ( SELECT U.*, ROW_NUMBER() over ( ORDER BY name ) RowNum from Users U) SELECT * FROM results_cte WHERE RowNum > @Offset AND RowNum <= @Offset + @PageSize
So where 'ORDER BY name' is I'd like to use the @OrderByClause in some sort of dynamic way. I've tried all kinds of stuff but even something like this doesn't get the actual column name I need
;WITH results_cte AS ( SELECT U.*, ROW_NUMBER() over ( ORDER BY (select top 1 [key] +' '+ [value] from @OrderByClause) ) RowNum from Users U)
I may be chasing the wrong stick, but outside of dynamic sql, is something like this possible?
Implement time interval type in the form of a user defined type in SS2k8r2? Specifically an interval type described in the book Temporal Data and the Relational Model by C. J. Date at all. As an example, an interval is below:
1/4/2006:1/10/2006
which would mean the time period from 1/4 to 1/10.
what are the advantages of using user-defined data types in SQL Server?
for example, you may have Customer Number Cust_Num varchar(10) you can create a user-defined data type like this: udt_CustNum
so now your table creation script for Custome table become: Cust_Num udt_CustNum
however, once the user-defined data type is referred by other tables. it cannot be changed/deleted.
So, i wonder what are the good reason of using udt instead of fundamental data types. (built-in datatypes)
the only good thing i think is about: the abstract naming of the data type. you may have other key fields like Supplier Number which is varchar(12) instead of u have to remember varchar(10) for customer and varchar(12) for supplier u just to remember udt_CustNum for customer and udt_SuppNum for supplier.
A common request for enhancement to applications is to "make thisfield bigger". I know I've been caught with increasing a field size,and then spending hours debugging because another stored proc has avariable or temp table that uses the field defined as the originalvarchar size. SQL Server just truncates to fit the data into thesmaller varchar, and so there are no errors raised.An option suggested by a colleague is to no longer use varchars, butuse User Defined Types instead. To make this work effectively,though, they suggest we would need to make it a rule that we do notuse varchars anywhere except to define user defined types.Though there will be one point of changes I can't help thinking thisisn't a very good idea ! Any thoughts ?Thanks.
Is there a syntax to create indexes on user-defined type's fields and methods? Can I index UDT-fields?
I tried but only get syntax error.
Code Snippet
-- put an index on the picture luminance
CREATE INDEX myIdx ON myTbl(picMetaData.Brightness) -- !! error
GO
Error message:
Code Snippet
Msg 102, Level 15, State 1, Line 1
Incorrect syntax near '.'.
According to books online (BOL), section User-defined types requirements says In the SQL Server 2005 RTM version, CLR UDTs with user-defined serialization were allowed to have their fields indexed as part of non-persisted computed columns or views. In such situations, non-deterministic UDT serialization/deserialization could lead to index corruption, and therefore has been removed from SQL Server 2005 SP1. In SQL Server 2005 SP1, UDT fields must use native serialization or be persisted in order to be indexed. Any existing indexes on UDT fields should continue to function as before.
What are BOL trying to say about index on UDT fields?
Hi all. We have a mix of informix and mssql server and I want to know if something we do in informix has an analogous feature in MSSQL. We can define a "row type" in informix, like so:
create row type name_1(fname char(20),lname char(20));
The when we create any table that includes a first and last name, we do so using this row type like so:
create table sometable(name name_1, some column,...etc)
This allows us to set a standard for certain common fields and avoids having different developers build the same type of field in more than one way, different lengths, etc.
I have defined a data type "OC_BUN_NAME" in database "CVPRO". I want to create a table tempdb.dbo.CVPRO in SQL SERVER 2005 system Database tempdb. But SQL SERVER 2005 DBMS gives a Error Messages:"Can not find the data type OC_BUN_NAME".
How can I do? How to use data types in the other database?
Hi GuysWonder if you could help me.Basically I produce an accounts package that uses a SQL 2000 DB as theRDBMS. I always instruct users to login as 'sa' and the relevantpassword when doing an update to my program, as sometimes I need to dodatabase changes for new stuff.Found that one of my users has not only logged in with their loginname (in this case Edward), but have also made this login a 'db owner'so that when I created 2 new user-defined data types they belong toEdward rather than dbo.This must have happened a long time ago, but now that they want tomove Edward round the roles and/or delete him from a copy of thedatabase that they have, they can't because he's the owner of theseuser-defined types.This brings me to the reason for my post, how can I change the ownerfrom Edward to dbo for these data types? I found an article ontechnet of how to do this, but when it suggests changing myuser-defined type to standard format it doesn't seem to work.Any ideas?RgdsRobbie
Im trying to use a .NET IPAddress in my UDT. i create the assembly, but i get this error when i try to create the type in SQL Server:
Type "IPAddress.IPAddressUDTType" is marked for native serialization, but field "address" of type "IPAddress.IPAddressUDTType" is not valid for native serialization.
From what i understand about the IPAddress object in .NET it is serializable. what am i doing wrong
( note, im just doing this for research purposes and not planing to use the UDT)
I created a user-defined datatype in Microsoft SQL server using the Enterprise Manager. But, I am not able to find options to edit this data type. There are options to delete but not for editing an existing user defined data type.
Can any one help me how to edit this user defined datatype ?
Also, are there any better ways to create and manage user defined data types in MS SQL ?
What is 'COLLATE SQL_Latin1_General_CP1_CI_AS'? I am new to SQLServer, and couldn't find much information on the Web. Also, I am trying to understand user defined data types. For example,
In the following example, what is '[Price_DT]' data type? and how would it be referenced at the time of 'INSERT'.
CREATE TABLE [dbo].[Dist_Orders_Master_Index] ( [SubTotal] [Price_DT] NOT NULL , [Tax] [Price_DT] NOT NULL ) ON [PRIMARY]
I have several stored procedures with parameters that are defined withuser defined data types. The time it takes to run the procedures cantake 10 - 50 seconds depending on the procedure.If I change the parameter data types to the actual data type such asvarchar(10), etc., the stored procedure takes less that a second toreturn records. The user defined types are mostly varchar, but someothers such as int. They are all input type parameters.Any ideas on why the stored procedure would run much faster if notusing user defined types?Using SQL Server 2000.Thanks,DW
Hi all, In my SQL Server Management Studio Express, I have the following Database "ChemAveRpd", Table "dbo.LabTests", and Content of the Table: dbo.LabTests Column_name Type Length Prec Scale AnalyteID int 4 10 0 (Primary Key) AnalyteName nvarch 510 CasNumber nvarch 510 Result numeric 5 8 2 Unit nvarch 510 SampleID int 4 10 0 (Foreign Key)
Hello, I cannot find out to create enum data type in SQL Server 2005 Express. Can I easily create the enum type just like the MySQL does.(please the MySQL example below)
CREATE TABLE myTable ( myid INT UNSIGNED NOT NULL, myclass ENUM('FIRST','SECOND','THIRD') DEFAULT 'FIRST' PRIMARY KEY(myid) );
Does anyone know where to find or how to write a quick user defined fucntionthat will return a table object when passed the string name of the tableobject. The reason why I want dynamicallly set the table name in a storedprocudue WITHOUT using concatination and exec a SQL String.HenceIf @small_int_parameter_previous = 1 then@vchar_tablename = "sales_previous"else@vchar_tablename = "sales"Endselect * from udf_TableLookup(@vchar_tablename )So if I pass 1, that means I want all records from "sales_previous"otherwise give me all records from "sales" (Sales_Previous would last yearssales data for example).udf_TableLookup would I guess lookup in sysobjects for the table name andreturn the table object? I don't know how to do this.I want to do this to avoid having 2 stored procedures..one for current andone for previous year.Please respond to group so others may benfiit from you knowledge.ThanksErik
I have a question about creating a user defined type: I'd like to create a table of employee objects which include objects of the type employee_t. I used this structure before in Oracle 9i and would like to know how it can be done with MS SQL Server 2000 or 2005, preferably with Enteprise Manager/Management Studio. Below is an example.
CREATE TYPE employee_t AS OBJECT ( name VARCHAR(10) jobDesc VARCHAR(15) ... )
I'm trying to use a UDF that returns a table, but I'm not sure of the syntax to invoke it. I've found examples in BOL and on-line like the following:
SELECT * FROM dbo.fn_MyTableFunc( 123.09, 'MyID' )
But I need the input parameter to be obtained from another table. For a very simplistic example, I've got 4 tables (and yes, I know that I can get the results I want for this example without using a UDF, but humor me):
CREATE TABLE tUser (UserID int PRIMARY KEY, UserName varchar(50)) CREATE TABLE tAcctGroup (AcctGroupID int PRIMARY KEY, AcctGroupName varchar(50)) CREATE TABLE tAcct (AcctID int PRIMARY KEY, AcctGroupID int, AcctName varchar(50)) CREATE TABLE tMapUserToGroup (UserID int, AcctGroupID int) GO
INSERT INTO tUser VALUES (111, 'Me')
INSERT INTO tAcctGroup VALUES (1, 'NY') INSERT INTO tAcct VALUES (11, 1, 'New York City') INSERT INTO tAcct VALUES (12, 1, 'Syracuse')
INSERT INTO tAcctGroup VALUES (2, 'GA') INSERT INTO tAcct VALUES (21, 2, 'Atlanta') INSERT INTO tAcct VALUES (22, 2, 'Savannah') INSERT INTO tAcct VALUES (23, 2, 'Augusta')
INSERT INTO tAcctGroup VALUES (3, 'TX') INSERT INTO tAcct VALUES (31, 3, 'Dallas') INSERT INTO tAcct VALUES (32, 3, 'Houston') INSERT INTO tAcct VALUES (33, 3, 'El Paso') INSERT INTO tAcct VALUES (34, 3, 'San Antonio')
INSERT INTO tAcctGroup VALUES (4, 'CA') INSERT INTO tAcct VALUES (41, 4, 'Los Angeles') INSERT INTO tAcct VALUES (42, 4, 'San Francisco')
INSERT INTO tMapUserToGroup VALUES (111,2) INSERT INTO tMapUserToGroup VALUES (111,4) GO
CREATE FUNCTION dbo.ufnGetAcctList(@AcctGroupID int) RETURNS @tAcct table (AcctID int, AcctName varchar(50)) AS BEGIN INSERT INTO @tAcct SELECT AcctID, AcctName FROM tAcct WHERE AcctGroupID = @AcctGroupID RETURN END GO
I know that I can do: SELECT * FROM TestDB.dbo.ufnGetAcctList(4)
But I want the equivalent of: SELECT AcctID, AcctName FROM tAcct WHERE AcctGroupID IN (SELECT AcctGroupID FROM tMapUserToGroup WHERE UserID = 111)
Which uses tMapUserToGroup to obtain the AcctGroupID to pass into the function. The results would be: AcctID AcctName ----------------------------- 21 Atlanta 22 Savannah 23 Augusta 41 Los Angeles 42 San Francisco
Any thoughts? Thanks in advance for your help. Cat
CREATE TYPE dbo.MyTableType AS TABLE ( Name varchar(10) NOT NULL, ValueDate date NOT NULL, TenorSize smallint NOT NULL, TenorUnit char(1) NOT NULL, Rate float NOT NULL PRIMARY KEY (Name, ValueDate, TenorSize, TenorUnit) );
and I would like to create a table of this type. From this answer [URL] .... the suggestion was to try
CREATE TABLE dbo.MyNewTable AS dbo.MyTableType
which produced the following error message in my SQL Server Express 2012:
> Incorrect syntax near the keyword 'OF'.
Is this not supported by SQL Server Express? If so, could I create it some other way, for example using `DECLARE`?
I am a SSIS novice, and need help to do the following:
Every day a Table is generated by the system of format XR+present Year + present month + present day. My task is pretty simple. I am to devise a SSIS package that will import this table on to a different server.
I have defined the variable as user defined variable. I am also able to specify the variable in OLE db source editor -> Data Access Mode : Table name or view name variable. When I am click the 'Columns' feature, I get the following error
TITLE: Microsoft Visual Studio ------------------------------ Error at Data Flow Task [OLE DB Source [1]]: SSIS Error Code DTS_E_OLEDBERROR. An OLE DB error has occurred. Error code: 0x80040E37. Error at Data Flow Task [OLE DB Source [1]]: Opening a rowset for "xr+ convert(char(8),getdate(), 112)" failed. Check that the object exists in the database. ------------------------------ ADDITIONAL INFORMATION: Exception from HRESULT: 0xC02020E8 (Microsoft.SqlServer.DTSPipelineWrap) ------------------------------ BUTTONS: OK ------------------------------
In short can any one tell me a correct way of using a table name as variable in Data Access Mode : Table name or view name variable ?
One other question: can I make use of a dynamic SQL in Execute SQL Task? If yes, how do I write the query
I'm trying to create a simple function that will do a count on a table. I want to pass the table name in form of a parameter to the variable and this function will return the count as an int. See my function below...
CREATE FUNCTION count_rows (@tablename varchar(100) RETURNS int AS BEGIN DECLARE @emp_count AS int declare @declaration varchar(100)
[Code] ....
The errors I am getting are as follows:
Msg 102, Level 15, State 1, Procedure count_rows, Line 3 Incorrect syntax near 'RETURNS'. Msg 102, Level 15, State 1, Procedure count_rows, Line 10 Incorrect syntax near '@declaration'. Msg 178, Level 15, State 1, Procedure count_rows, Line 14
A RETURN statement with a return value cannot be used in this context.
     field ([idConteudo] [int] NULL)     to      ([idConteudo] [BigInt] NULL)
Or really should drop and create again?
CREATE TYPE [dbo].[tbProjeto] AS TABLE( [dsConteudo] [varchar](64) NOT NULL, [idConteudo] [int] NULL, [dtConteudo] [datetime] NULL DEFAULT (getdate()) ) GO
We have created several Table Valued User Defined Functions in a Production SQL Server 2005 DB that are returning large (tens of thousands of) rows obtained through a web service. Our code is based on the MSDN article Extending SQL Server Reporting Services with SQL CLR Table-Valued Functions .
What we have found in our implementations of variations of this code on three seperate servers is that as the rowset grows, the length of time required to return the rows grows exponentially. With 10 columns, we have maxed out at approximately 2 500 rows. Once our rowset hit that size, no rows were being returned and the queries were timing out.
Here is a chart comparing the time elapsed to the rows returned at that time for a sample trial i ran:
It took 570 seconds (just over 9 1/2 minutes to return 2566 rows).
The minute breakdown during my trial is as follows: 1 = 655 (+ 655) 2 = 1081 (+ 426) 3 = 1325 (+244) 4 = 1610 (+285) 5 = 1822 (+212) 6 = 1988 (+166) 7 = 2160 (+172) 8 = 2308 (+148) 9 = 2451 (+143)
As you can tell, except for a few discrepancies to the resulting row count at minutes 4 and 7 (I will attribute these to timing as the results grid in SQL Management Studio was being updated once every 5 seconds or so), as time went on, fewer and fewer rows were being returned in a given time period. This was a "successful" run as the entire rowset was returned but on more than several occasions, we have reached the limit and have had 0 new rows per minute towards the end of execution.
Allow me to explain the code in further detail:
[SqlFunction(FillRowMethodName = "FillListItem")] public static IEnumerable DiscoverListItems(...) {
ArrayList listItems = new ArrayList();
SPToSQLService service = new SPToSQLService();
[...]
DataSet itemQueryResult = service.DoItemQuery(...); // This is a synchronous call returning a DataSet from the Web Service
//Load the DS to the ArrayList
return listItems; }
public static void FillListItem(object obj, out string col1, out string col2, out string col3, ...) {
As you will notice, the web service is called, and the DataSet is loaded to an ArrayList object (containing ArrayList objects), before the main ArrayList is returned by the UDF method. There are 237 rows returned within 10 seconds, which leads me to believe that all of this has occured within 10 seconds. The method GetListItems has executed completely and the ArrayList is now being iterated through by the code calling the FillListItem method. I believe that this code is causing the result set to be returned at a decreasing rate. I know that the GetListItems code is only being executed once and that the WebService is only being called once.
Now alot of my larger queries ( > 20 000 rows) have timed out because of this behaviour, and my workaround was to customize my web service to page the data in reasonable chunks and call my UDF's in a loop using T-SQL. This means calling the Web Service up to 50 times per query in order to return the result set.
Surely someone else who has used Table Valued UDFs has come accross this problem. I would appreciate some feedback from someone in the know, as to whether I'm doing something wrong in my code, or how to optimize an SQL Server properly to allow for better performance with CLR functions.
I am working with a vendor application called Cisco Unified Attendant Console - it operates on a Windows server with a SQL express database. The CUPs function of the application needs to reference a "contact" field with only the user portion of the contact's email address - generally, the contact's User ID will match the user portion of their email address, however, for this customer it does not (they use the employee number as the User ID and firstname.lastname as user portion of the email address.
Writing a script to accomplish the following:
The dbo.Contact_Properties table of the ATTCFG database has the following fields that we can work with:  - First_Name  - Last_Name  - Email  - User_Field_2  - Contact_Unique_Ref (appears to be the field that ties all other contact tables together ?)
Is it possible to create a script that could run daily to either, combine the First_Name and Last_Name fields (with a period between) and populate the User_Field_2 field for each user, or populate the User_Field_2 field with everything before the @ symbol in the Email field for each user?
Also, by default the servers that this application is installed on does not have SQL Server Management Studio installed - is it possible to accomplish with PowerShell script triggered from the Windows Scheduler?