Help with cascading SELECT

Hello
I have the following scenario that I need to address with which I would
appreciate some help if anyone has the time:
(Simply) I have an asp.net application:
I have 2 tables:
Table 1 is called Categories and is used to hold hierarchical data (Tiers) -
these Tiers can be altered by the users. There may be 10 Categories in Tier
1, Tier 2 will hold 1 or more sub-categories for those 10, Tier 3 will have
sub-categories of Tier 2 etc. etc. through to Tier 5
Table 2 is called Jobs and holds information pertaining to a specific task
that is categorised by Table 1, so this would have single entries that you
should be able to back-track through to Tier 1 of Table 1
This structure is to simply organise data in a highly retrievable structure.
So when a user clicks on a Category in Tier 1 I want to retrieve all Jobs
that are under that category regardless of the sub-category structure
underneath, a click on a category in Tier 2 will filter these, a click on a
sub category in Tier 3 will make the list smaller etc. etc. (I'm sure you ge
t
the picture)
When a category or a job is defined and entered by the user a unique numeric
reference is automatically assigned to that record - it also records the
unique reference of the category above in a field called "Parent_Ref". I
should therefore be able to take the parent record from a Job (which is the
end of the line) and track this back to a sub-category in Tier 3, 4, or 5,
this category record will also have a parent which goes back to Tier 2 and
this will have a recod that tracks back to Tier 1
When the user initially clicks on a Tier 1 Category I need to do a search
for all Table 1 records that have the unique reference of that Tier 1
category in their "Parent_Ref" field.... for all returned records I need to
then do a search through Table 2 for all Jobs that are associated directly
with that category. This will give me 0 to x returns from Jobs. I then need
to requery and find all records from Tier 3 in Table 1 that have one of the
unique references from the Tier 2 records that were just returned in their
"Parent_Ref" field then retrieve all Jobs that relate to these retrieved
records and loop through all the Categories in Table 1 until no more returns
are possible
I hope this is clear !?! I'm sure this is a standard scenario, I have
researched the Select options in transact SQL but can not see the light. I
appreciate that this will probably need to be a stored procedure that will b
e
triggered by my asp.net application where I simply pass the unique ref in to
SQL and let it do the search, but I do not know how to cascade the queries i
n
the above fashion
Thanks
StuartStuart
Can you post DDL+ sample data + expected result?
Like
CREATE TABLE jobs
(
...
...
)
"Stuart" <Stuart@.discussions.microsoft.com> wrote in message
news:133CA7C6-2A90-46C9-94A3-07C51F277741@.microsoft.com...
> Hello
> I have the following scenario that I need to address with which I would
> appreciate some help if anyone has the time:
> (Simply) I have an asp.net application:
> I have 2 tables:
> Table 1 is called Categories and is used to hold hierarchical data
> (Tiers) -
> these Tiers can be altered by the users. There may be 10 Categories in
> Tier
> 1, Tier 2 will hold 1 or more sub-categories for those 10, Tier 3 will
> have
> sub-categories of Tier 2 etc. etc. through to Tier 5
> Table 2 is called Jobs and holds information pertaining to a specific task
> that is categorised by Table 1, so this would have single entries that you
> should be able to back-track through to Tier 1 of Table 1
> This structure is to simply organise data in a highly retrievable
> structure.
> So when a user clicks on a Category in Tier 1 I want to retrieve all Jobs
> that are under that category regardless of the sub-category structure
> underneath, a click on a category in Tier 2 will filter these, a click on
> a
> sub category in Tier 3 will make the list smaller etc. etc. (I'm sure you
> get
> the picture)
> When a category or a job is defined and entered by the user a unique
> numeric
> reference is automatically assigned to that record - it also records the
> unique reference of the category above in a field called "Parent_Ref". I
> should therefore be able to take the parent record from a Job (which is
> the
> end of the line) and track this back to a sub-category in Tier 3, 4, or 5,
> this category record will also have a parent which goes back to Tier 2 and
> this will have a recod that tracks back to Tier 1
> When the user initially clicks on a Tier 1 Category I need to do a search
> for all Table 1 records that have the unique reference of that Tier 1
> category in their "Parent_Ref" field.... for all returned records I need
> to
> then do a search through Table 2 for all Jobs that are associated directly
> with that category. This will give me 0 to x returns from Jobs. I then
> need
> to requery and find all records from Tier 3 in Table 1 that have one of
> the
> unique references from the Tier 2 records that were just returned in their
> "Parent_Ref" field then retrieve all Jobs that relate to these retrieved
> records and loop through all the Categories in Table 1 until no more
> returns
> are possible
> I hope this is clear !?! I'm sure this is a standard scenario, I have
> researched the Select options in transact SQL but can not see the light. I
> appreciate that this will probably need to be a stored procedure that will
> be
> triggered by my asp.net application where I simply pass the unique ref in
> to
> SQL and let it do the search, but I do not know how to cascade the queries
> in
> the above fashion
> Thanks
> Stuart|||Look at this example:
http://milambda.blogspot.com/2005/0...or-monkeys.html
ML
http://milambda.blogspot.com/|||
"Uri Dimant" wrote:

> Stuart
> Can you post DDL+ sample data + expected result?
> Like
> CREATE TABLE jobs
> (
> ...
> ....
> )
>
>
> "Stuart" <Stuart@.discussions.microsoft.com> wrote in message
> news:133CA7C6-2A90-46C9-94A3-07C51F277741@.microsoft.com...
>
>|||Stuart
'?
"Stuart" <Stuart@.discussions.microsoft.com> wrote in message
news:147912F5-39C6-4868-B368-ED57D26CC465@.microsoft.com...
>
> "Uri Dimant" wrote:
>|||Thanks very much - seems to be just the thing ! - but in your experience
would this function in an acceptable way with a max of 1000 entries in the
table ?
"ML" wrote:

> Look at this example:
> http://milambda.blogspot.com/2005/0...or-monkeys.html
>
> ML
> --
> http://milambda.blogspot.com/|||Sorry Uri - I spent some time composing a complete response to your request
only to have this interface bomb out when I posted it ! I appreciate your
time, but the answer from ML actually answers my question...
"Uri Dimant" wrote:

> Stuart
> '?
>
>
> "Stuart" <Stuart@.discussions.microsoft.com> wrote in message
> news:147912F5-39C6-4868-B368-ED57D26CC465@.microsoft.com...
>
>|||The performance of the function depends on two facts:
1) the number of rows in the table; and
2) the average depth of hierarchies.
The best way to know would be to test it on your system. It should peform
adequately with several thousands of rows with an average depth of about ten
levels. When I say "adequatly" that does not mean "perfectly". In SQL 2005
the function should be redesigned using a recursive common table expression
(rCTE), which might improve performance, since it's built into the SQL Serve
r
engine and requires less CPU time to execute.
ML
http://milambda.blogspot.com/

Help talking to stored proc

Can someone please lend a hand.
I am a total noob at this .NET/ASP stuff. I simply need help passing and returning values to a stored procedure.
Here’s my sp.

ALTER PROCEDURE [dbo].[returnIdUser]-- Add the parameters for the stored procedure here @.sessionUservarchar(25)OUTPUTASBEGIN-- SET NOCOUNT ON added to prevent extra result sets from -- interfering with SELECT statements.SET NOCOUNT ON;-- Insert statements for procedure hereSELECT idUserFROM tblUsersWHERE domainUser = @.sessionUserEND

So when the user logs into to the default page (windows authentication) I capture their domain\login in a session variable (sessionUser). On the next page, when the page loads I need to pass this to my stored proc and get the idUser in return.
Here is what I have for the page load so far.

try{ SqlConnection cxnReturnID =new SqlConnection(ConfigurationManager.ConnectionStrings["cxnLeaveRecords"].ConnectionString); SqlCommand cmdReturnID =new SqlCommand("returnIdUser", cxnReturnID); cmdReturnID.CommandType = CommandType.StoredProcedure; cmdReturnID.Parameters.Add("@.sessionUser", SqlDbType.VarChar, 25).Direction = ParameterDirection.Output; cmdReturnID.Connection.Open(); cmdReturnID.ExecuteReader(); cmdReturnID.Connection.Close(); cmdReturnID.Connection.Dispose(); Session["sessionUserID"] = cmdReturnID.Parameters["idUser"].Value;}catch (Exception ex){ lblStatus.Text = ex.Message;}

Of course this fails cause I don’t know what I’m doing. My error label shows the following: "An SqlParameter with ParameterName 'idUser' is not contained by this SqlParameterCollection."

Can someone point me to what I’m doing wrong?

Hmm.. You mixed up every thing here..

First you need to a pass an input parameter and expect an out put paremeter. So, your stored procedure should have two paramets defined in it.. You have only one

And also, in your c# code, you need to assign the value of session variable to the input parameter, which you have not.

Anyway, try this..

ALTER PROCEDURE [dbo].[returnIdUser]
-- Add the parameters for the stored procedure here

 @.sessionUservarchar(25),

 @.idOUT varchar(25) out
AS
BEGIN
-- SET NOCOUNT ON added to prevent extra result sets from
 -- interfering with SELECT statements.
SET NOCOUNT ON;
-- Insert statements for procedure here
set idOUT = (SELECT idUserFROM tblUsersWHERE domainUser = @.sessionUser)

return @.idOUT
END

and your code to

cmdReturnID.Parameters.Add("@.sessionUser", SqlDbType.VarChar, 25).Direction = ParameterDirection.Input;

command.Parameters["@.sessionUser"].Value =  Session["UserName"]

 cmdReturnID.Parameters.Add("@.idOUT", SqlDbType.VarChar, 25).Direction = ParameterDirection.Output;

SqlDataReader reader = cmdReturnID.ExecuteReader();

 reader.close();

 Session["sessionUserID"] = cmdReturnID.Parameters["idOUT"].Value;

Even though, the code looks real, its not... I typed by hand, so beware of small errors... 

|||

Hi , in your example I guess you want to pass a session user into the stored procedure and return idUser. In this case your stored procedure should be :

ALTER PROCEDURE [dbo].[returnIdUser]-- Add the parameters for the stored procedure here (@.sessionUservarchar(25))ASBEGIN-- SET NOCOUNT ON added to prevent extra result sets from -- interfering with SELECT statements.SET NOCOUNT ON;-- Insert statements for procedure hereSELECT idUserFROM tblUsersWHERE domainUser = @.sessionUserEND

And here is the modified code:

try { SqlConnection cxnReturnID =new SqlConnection(ConfigurationManager.ConnectionStrings["cxnLeaveRecords"].ConnectionString); SqlCommand cmdReturnID =new SqlCommand("returnIdUser", cxnReturnID); cmdReturnID.CommandType = CommandType.StoredProcedure; cmdReturnID.Parameters.Add("@.sessionUser", SqlDbType.VarChar, 25); cmdReturnID.Connection.Open(); SqlDataReader sqldr = cmdReturnID.ExecuteReader(); cmdReturnID.Connection.Close(); cmdReturnID.Connection.Dispose(); Session["sessionUserID"] = sqldr.GetString(0); }catch (Exception ex) { lblStatus.Text = ex.Message; }

Here I use sessionUser as an input value and read idUser with SqlDataReader.Hope this helps.|||it took some alterations, but finally got it to work. Thanks to both of you!

Help Simply the Update Statement

Hello all,

I need some help in simplyfying the following update statement -

update table <table_a> set <col_1> = NULL where <col_1> = 'N/A'
update table <table_a> set <col_2> = NULL where <col_2> = 'N/A'
update table <table_a> set <col_3> = NULL where <col_3> = 'N/A'
update table <table_a> set <col_4> = NULL where <col_4> = 'N/A'
update table <table_a> set <col_5> = NULL where <col_5> = 'N/A'
update table <table_a> set <col_6> = NULL where <col_6> = 'N/A'
update table <table_a> set <col_7> = NULL where <col_7> = 'N/A'
...and there are 73 columns

Anyway I can create a loop or array and store the column name as a parameter and then pass it to the update statement?

Thanks in advance,
SauravPut this in a SP and keep @.table as parameter

DECLARE @.COLUMN VARCHAR(100)
DECLARE @.TABLE VARCHAR(100)

CREATE TABLE #TEMP_TABLE
(
COLUMNNAME VARCHAR(100)
)

SET @.TABLE = 'PRODUCTS'

INSERT INTO #TEMP_TABLE

SELECT A.NAME FROM
SYSCOLUMNS AS A
INNER JOIN
SYSOBJECTS AS B
ON
A.ID = B.ID
WHERE
B.NAME = @.TABLE --TABLE NAME

WHILE ((SELECT COUNT(*) FROM #TEMP_TABLE) > 0)
BEGIN

SELECT TOP 1 @.COLUMN = COLUMNNAME
FROM #TEMP_TABLE

EXEC('UPDATE' + @.TABLENAME + ' SET ' + @.COLUMN + ' = NULL WHERE ' + @.COLUMN + ' = ''N/A''')

DELETE FROM #TEMP_TABLE WHERE COLUMNNAME = @.COLUMN

END

DROP #TEMP_TABLE|||There's rarely need to send table names to sprocs for dynamic sql, and this isn't one of those times in any case. It's all on the same table.

Simpler:
update table <table_a>
set <col_1> = NULLIF(<col_1>, 'N/A')
, <col_2> = NULLIF(<col_2>, 'N/A')
, <col_3> = NULLIF(<col_3>, 'N/A')
....
FROM <table_a>
WHERE <col_1> = 'N/A'
OR <col_2> = 'N/A'
OR <col_3> = 'N/A'...|||You rock, Blindman. Thanks you!!|||You rock, Blindman. Thanks you!!I'll be sure to tell him ;)

--blindman is quoted in my sig - he ain't me and he will be pretty appalled when he sees this.|||Thank you nick.cs! I also found that there is no need to update all columns but around 28 of them. But new columns may be added to the update statement and removed. Therefore, I have added another variable to your query which is a flag field to indicate which ones are needed to be updated. Your query also helped me!

Thanks all!

This forum is great!