|
Hi All,
I have a Student table which has 3 fileds:-
ID Name PartNo
----------------------
1 Smith 3
1 Adam 1
1 Samuel 3
2 Gary 1
2 Johnson 2
3 Immanuel 4
3 Ryan 2
3 Irene 1
3 Pete 5
My problem is i need to group students based on IDs and the Name need to be a concatenation of names based on the PartNo in ascending order. The O/p need to be like:-
ID Name
---------------------------
1 Adam Smith
2 Gary Johnson
3 Irene Ryan Immanuel Pete
I am working with SQL 2000. Hope somebody can help me sort this out.
Thanks In Advance.
Payal
-- modified at 22:57 Wednesday 17th October, 2007
|
|
|
|
|
Any particular reason you didn't go with the more traditional FirstName, MiddleNames, Surname format?
|
|
|
|
|
Hi Colin,
I was being given this table(Built by somebodyelse) to get the specified result.
Thanks.
Payal
|
|
|
|
|
Your database design is really weird. Having separate forename & surname fields would make your query a lot easier.
Paul Marfleet
|
|
|
|
|
I have one solution but dont know the syntax. So telling the logic only.
execute the below tsql stmt
select * from codeproj order by id,partno asc
Result will be like this:
1 Adam 1
1 Samuel 2
1 Smith 3
2 John 1
2 king 2
Now create a cursor, fetch the data and concatinate in a variable also check for the condition ID.
You need to write stored procedure for all these.
|
|
|
|
|
Thanks for the reply.
I will try it out.
|
|
|
|
|
hi all ,
How is it possible to Install SSL for Sql Server in Workgroup environment?
Thanks in advance .
|
|
|
|
|
If you look up SSL in the index in SQL Server Books Online, one of the results is the topic "How to: Enable Encrypted Connections to the Database Engine (SQL Server Configuration Manager)" (online here[^]).
If you don't want to use a public certificate (i.e. one that you have to pay a trusted root authority to produce for you), you can set up your own stand-alone root authority to generate certificates. See here[^] for how to install certificate services. In order for the server to use the certificate you generate, and for the client to trust that certificate, the root certificate (just the certificate, not the private key associated with it - you do not distribute this) needs to be installed as a trusted root on both server and client.
You can then generate a Server Authentication certificate by going to the CA's certsrv website, clicking Request a certificate, advanced certificate request, create and submit a request to this CA. Fill in the identifying information, change Type of Certicate Needed to Server Authentication Certificate, check the Mark Private Keys As Exportable box (if you're doing this on a computer other than the target server), check Store certificate in the local computer certificate store if you're doing this on the SQL Server computer itself, [UPDATE: this is wrong, the certificate needs to go into the Personal store for the account that SQL Server is running under - use the Local Computer store only if using LocalSystem, which isn't recommended] and enter the fully-qualified domain name of the SQL Server computer in the Friendly Name field. Then click Submit.
Stand-alone CAs do not automatically approve requests. In the Certification Authority console, go to Pending Requests, right-click the request and select Issue. You can now go back to the website (don't hit Refresh, this generates another request) and click View Status. Click the request, then click Install This Certificate.
Now you can go through SQL Server Configuration Manager as described in the article and select the certificate you just generated.
You should be careful to keep your Certification Authority computer isolated. I have ours as a virtual machine on our network, and I only start it up when needing to create a new certificate.
-- modified at 12:34 Wednesday 17th October, 2007
DoEvents: Generating unexpected recursion since 1991
|
|
|
|
|
Thanks , If I generate the Certificate , Is it possible to deploy it with my .Net app setup project and Import it during setup and set SQL Server certificate to this?How?
Thanks again ,
|
|
|
|
|
Hi,
I need to check if a Value exists in a Table. I want to create 1 SP for al datachecks for different fields and tables.
I created an SP:
CREATE PROCEDURE DBO.SP_CHECK_BASEVALUE
@CodeInGrid Varchar(255),
@FieldInTable Varchar(255),
@Table Varchar(255)
AS
DECLARE @sql nvarchar(400)
SET @sql = 'SELECT COUNT(*) FROM ' + @Table + ' WHERE '+ @FieldInTable + ' = ' + @CodeInGrid
EXEC sp_executesql @sql
GO
Problem is @FieldInTable. Is says as Error that Column 'Column' does not exist.
How to create an SP that sees @FieldInTable as ColummName as stead of a varchar value.
|
|
|
|
|
You are using dynamic SQL where you are concatenating a string together to make a full statement. You have not sanitised the values @Table, @FieldInTable and @CodeInGrid. This means your application is susceptable to a SQL Injection Attack. Please read this[^]
All column and table names are nvarchar(128)
If you must do this then please check that @Table and @FieldInTable (actually, they are called columns in SQL Server) are valid first
This might be a starting point for the checking that the table and column name are valid:
SELECT * FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_NAME = @Table
SELECT * FROM INFORMATION_SCHEMA.COLUMNS WHERE COLUMN_NAME = @fieldInTable AND TABLE_NAME = @Table
Also, table and column names should be wrapped in square brackets in case they contain obscure characters or whitespace.
@CodeInGrid I presume is an integer as you don't wrap it in apostrophes in your dynamic SQL. If so pass it as an integer and NOT as any sort of string type. This can easily be cast to a string for concatenation with the rest of the satement.
|
|
|
|
|
Hi,
MY SP is now:
CREATE PROCEDURE DBO.SP_CHECK_BASEVALUE
@CodeInGrid Varchar(255),
@FieldInTable Varchar(128),
@Table Varchar(128)
AS
IF EXISTS(
SELECT * FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_NAME = @Table
)
BEGIN
IF EXISTS (SELECT * FROM INFORMATION_SCHEMA.COLUMNS WHERE COLUMN_NAME = @FieldInTable AND TABLE_NAME = @Table)
BEGIN
DECLARE @sql nvarchar(400)
SET @sql = 'SELECT COUNT(*) FROM ' + @Table + ' WHERE '+ @FieldInTable + ' = ' + @CodeInGrid
EXEC sp_executesql @sql
END
END
GO
still the same error. How to convert?
AND @CodeInGrid is a String Value (CountryISOCode, CurrencyISOCODE, etc)
|
|
|
|
|
spooky manus wrote: still the same error. How to convert?
AND @CodeInGrid is a String Value (CountryISOCode, CurrencyISOCODE, etc)
Then it is being interpreted as a column name because you've not enclosed it in apostrophes. Print out @sql and see if it actually makes sense. You'll see that it doesn't because it is missing the apostrophes around @CodeInGrid
And you still haven't sanitised @CodeInGrid. Does @CodeInGrid really need 255 characters for an ISO code of a few characters?
@FieldInTable and @Table are still the wrong type. They should be NVARCHAR(128).
|
|
|
|
|
spooky manus wrote: How to add the ' to the @CodeInGrid parameter?
You don't add it to the parameter. You add it to the SQL.
There are two ways to do this.
The unsafe, but quick and dirty way: You write two apostrphes next to each other.
The safe, but slower and cleaner way: use sp_executesql[^] stored procedure.
I recommend using sp_exeuctesql as it doesn't require that you do anything funky to apostrophes at all.
|
|
|
|
|
Thanks for your help. Especially the good article about sql injection. Gat about 150 SP's to alter now (seigh). but WILL do that certainly
|
|
|
|
|
I have 2 tables AnalysisHeadings and ProductAnalysisValues.
AnalysisHeadings consist of columns :-
HeadingId
Name
ProductAnalysisValues consist of columns :-
HeadingId
ItemId
Value
What I need to do is select ALL rows from AnalysisHeadings and only the rows from ProductAnalysisValue where the ItemId = 1
The 2 tables are linked by HeadingId to HeadingId.
I stuck on how to limit the the selection of rows from ProductAnalysisValues
I appreciate your help.
Steve Jowett
-------------------------
Sometimes a man who deserves to be looked down upon because he is a fool, is only despised only because he is an 'I.T. Consultant'
|
|
|
|
|
Steven J Jowett wrote: What I need to do is select ALL rows from AnalysisHeadings and only the rows from ProductAnalysisValue where the ItemId = 1
Use an OUTER JOIN to select EVERYTHING from one table, and only matches from the other
SELECT *
FROM AnalysisHeadings AS ah
LEFT OUTER JOIN ProductAnalysisValues AS pav
ON ah.HeadingId = pav.HeadingId
WHERE ItemId = 1
|
|
|
|
|
Actually,
SELECT
*
FROM
AnalysisHeadings AS ah
LEFT JOIN
ProductAnalysisValues AS pav
ON (ah.HeadingId = pav.HeadingId)
WHERE
ItemId = 1
will not get you very far since ItemId will be NULL where there is no matching ProductAnalysisValues row. Therefore, you will get the same result as if you had used an INNER JOIN.
You will need to specify the qualification inside of the ON clause of the LEFT JOIN:
SELECT
*
FROM
AnalysisHeadings AS ah
LEFT JOIN
ProductAnalysisValues AS pav
ON (ah.HeadingId = pav.HeadingId AND
pav.ItemId = 1)
|
|
|
|
|
Well, I'm glad someone spotted my deliberate mistake. Well done! 
|
|
|
|
|
I only spotted it because I committed the same foul on one of my production servers about a month ago. It will be a least a year before I make the same mistake again 
|
|
|
|
|
Michael
Thanks for help with my select query. I now have one that does extactly what's required thanks to you and Colin.
Best regards
Steve Jowett
-------------------------
Sometimes a man who deserves to be looked down upon because he is a fool, is only despised only because he is an 'I.T. Consultant'
|
|
|
|
|
What is the data type for NULL ? I heard that NULL is untyped. But some article's say it's character type. If it is character type, then what about NULL value contained in a numeric column ?
Other article say's NULL's data type is the data type of the column which it belongs to ?
Which is correct ? Any ideas ?
|
|
|
|
|
NULL does not have a type. It can be used as a non-value in a column of any datatype, where that column's definition allows nulls. It is implemented (in SQL Server at least) as an additional bit per nullable column within the row so it doesn't restrict the range of values that can be represented.
In the .NET Framework, NULL's type is DBNull.
Do be aware that different databases implement NULL differently, and not all are completely SQL-92 conformant.
DoEvents: Generating unexpected recursion since 1991
|
|
|
|
|
Mike Dimmick wrote: It is implemented (in SQL Server at least) as an additional bit per nullable column within the row so it doesn't restrict the range of values that can be represented.
I am not cleared on this. Can you please explain it more ?
|
|
|
|
|
It is just an internal storage requirement. Nothing for anyone writing SQL to actually worry about. How NULLs are represented internally has no effect on your SQL statements.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
