|
If you dont want this in Select query.. inside your stored procedure.. rather you want to place a logic use
If - Else instead. 
|
|
|
|
|
Am I the only person in the world that uses Google when I'm trying to find information? 2 seconds on Google will give you approximately 14 million search results about this including some very good explanations of what CASE is and what it does and how to use it. Just type "sql server case" and away you go.
|
|
|
|
|
Some people seem to find it easier to type in a question rather than trawl through 14 million results!  Ah I crack me up. Just because they can find excellent examples in the first 2-3 results is no reson not to be spoon fed. Send me the codz.
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
Mycroft Holmes wrote: Send me the codz.
When you get them can you forward them to me? My homework is soooo important! 
I don't speak Idiot - please talk slowly and clearly
'This space for rent'
Driven to the arms of Heineken by the wife
|
|
|
|
|
Hi, don't know what is ur actual doubt because the problem stated
sarang_k wrote: How to use the Case in stored procedure
is a normal thing.
If you are looking for syntax there is plenty in google with examples too.
However, I have given 2 sample for the same program
Sample 1:
Create procedure TestCaseStatement
as
Begin
declare @tbl table(id int)
insert into @tbl
select 0 union all select 1 union all
select 2 union all select 1 union all
select 1
select
id
,case when id = 1 then 'One' else 'Other Number' end as Value
from @tbl
End
Sample 2:
Create procedure TestCaseStatement
as
Begin
declare @tbl table(id int)
insert into @tbl
select 0 union all select 1 union all
select 2 union all select 1 union all
select 1
select
id
,Value = case when id = 1 then 'One' else 'Other Number' end
from @tbl
End
The output
id Value
0 Other Number
1 One
2 Other Number
1 One
1 One
As you can see that I have shown 2 ways of holding the case result into a pseudo column.
And as Mr. Mycroft stated, if you have any specific doubt, please don't hesitate to share it here. We all would love to help you.
Hope you understand and ur doubt as per the question title has been answered.
Have a nice day
Niladri Biswas
|
|
|
|
|
Hello All,
Sorry to bother you but I am hitting up against a wall trying to figure something out so I am turning to you for assistance (if possible.).
I have a table of let's say widgets. And for each entry there is a Quantity column. What I am needing to do is to write a Store Procedure for SQL Server that will pull all widgets for a certain customer and create a record in another table for each widget. No problem there. But I need to create a record for each widget based on it's quantity column.
So for example customer A brought 3 widget items: Widget1 5 pieces, Widget2 1 piece, and Widget3 3 pieces.
What I need to end up with is a new table with 9 entries. 1 for each widget piece.
Again I am sorry to bother you guys, but I am really at a lost.
I thank you in advance for any help you can direct me with.
Fred
|
|
|
|
|
So break it up into smaller bits.
open management studio and connect to the database - this is where you should be doing your test/creating
select *
From Widgets
Where CustomerID = @CustID
That should get you the list of widgets for the customer
Now you want to count the record for each widget type
select count(*),WidgetTypeID
From Widgets
Where CustomerID = @CustID
GroupBy WidgetTypeID
Now add the insert statement around the select
Insert OtherTable(CustID,WidgetTypeID, Qty)
select CustID, WidgetTypeID, count(*)
From Widgets
Where CustomerID = @CustID
GroupBy CustID, WidgetTypeID
Here endeth the lesson
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
According to the OP, this part of it is not a problem. There seems to be some sort of additional requirement which is the sticking point. I'm not entirely sure I understand the problem, but it seems to involve creating more rows in the target table than exist in the source table. 1 customer with 3 widgets should give 9 rows. That's the bit I don't quite get.
|
|
|
|
|
Seems like we are all struggling with this convoluted logic. I'll wait till the OP comes back with some clarification
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
Are you looking for this
declare @tblCustomers table(CustID int,CustomerName varchar(50))
declare @tblWidgets table(CustID int,WidgetTypeID int, Qty int)
declare @tblTarget table(CustID int,WidgetTypeID int, Qty int)
insert into @tblCustomers values(101,'Cust1'),(102,'Cust2'),(103,'Cust3')
insert into @tblWidgets values(101,1,5),(101,2,1),(101,3,3),(102,10,5),(103,4,1)
--Program starts
insert into @tblTarget
select CustID,WidgetTypeID,Qty
from
(
select w.CustID,w.WidgetTypeID,w.Qty from @tblWidgets w
inner join @tblCustomers c
on c.CustID = w.CustID
)X where X.CustID =101
select * from @tblTarget
The output being
CustID WidgetTypeID Qty
101 1 5
101 2 1
101 3 3
Niladri Biswas
|
|
|
|
|
ffowler wrote: So for example customer A brought 3 widget items: Widget1 5 pieces, Widget2 1 piece, and Widget3 3 pieces.
What I need to end up with is a new table with 9 entries. 1 for each widget piece.
9 entries? How do you get 9 entries from this? I can't quite follow that logic.
|
|
|
|
|
Flash of insight. The number of rows to create in the target table depends on the quantity in the source table.
So, customer A buys 3 widget items, Widget1 5 pieces, Widget2 1 piece and Widget3 3 pieces, what you want in the final result table is:
CustId WidgetID
A W1
A W1
A W1
A W1
A W1
A W2
A W3
A W3
A W3
In other words, 9 rows. Is that right?
|
|
|
|
|
That is correct David. I need a row created based on what is in the Quantity field of the original table. You have outlined it perfectly. Sorry I didn't make it clearer to you others originially.
Now, can it be done, or do I have do something externally?
|
|
|
|
|
Others may have a better suggestion, but I would select customer, widget, quantity and then use a cursor to fetch each row in turn. Then, for each row I would have a loop which repeats q times (where q is the quantity value for that row) and do an insert into the target table inside the loop. Does that make sense?
I'm not sure how you are planning to make each of the new rows unique in the target table, but at the simplest you could stick an identity column on there to handle the uniqueness. It seems a slightly odd requirement but I guess there is a reason behind it.
Sorry, no code because I don't use cursors that often, so I'm a bit rusty on the exact syntax. But Google is your friend when it comes to things like that.
|
|
|
|
|
David,
I am required to create a report whereby each item quantity has it's own line (for jotting notes or whatever next to each. I love my Sales Dept!!!)
In any event, I see where you are going with that, but I won't use a cursor. There is a niffty loop that I found that takes the place of using cursors which I have found is alot lighter.
I think that is a plan and I'll give it a go.
Thanks to all of you for your time with my issue, which is certainly not yours. I appreciate it.
|
|
|
|
|
Good Day all
i have the Following Query
DECLARE @CurrentTime DATETIME
SET @CurrentTime = CURRENT_TIMESTAMP
select tr.Descr [Room], tb.Purpose [Purpose], tb.Description [Description],
convert(varchar,datepart(hour,tb.starttime))+':'+convert(varchar,datepart(minute,tb.starttime)) [Start Time],
convert(varchar,datepart(hour,tb.endtime))+':'+convert(varchar,datepart(minute,tb.endtime)) [End Time],
tu.name [Requested by] from tbl_booking tb inner join tbl_resource tr
on tb.resources = tr.id
inner join tbl_user tu on tu.id = tb.RequestedByUser
where (day(startdate) = day(@CurrentTime))and(month(startdate)=month(@CurrentTime))and(year(startdate)=year(@CurrentTime))and(tb.status=1)
order by [Room],[Start Time]
and in the [Start Time]and [End Time] it gives me time that is not Complete
it Gives this
14:0
instead of
14:00
Thanks
Vuyiswa Maseko,
Spoted in Daniweb-- Sorry to rant. I hate websites. They are just wierd. They don't behave like normal code.
C#/VB.NET/ASP.NET/SQL7/2000/2005/2008
http://www.vuyiswamaseko.com
vuyiswa@its.co.za
http://www.itsabacus.co.za/itsabacus/
|
|
|
|
|
Why in the world would you be formatting your time in SQL? SQL Is for extracting data (and it seems your decord is already of type DateTime) and a client application should be responsible for formatting that data.
As an aside, using spaces in fields or field alias' is generally bad form.
|
|
|
|
|
Convert with style 108 will return the format as hh:mm:ss. If you don't want seconds, just strip it off.
I know the language. I've read a book. - _Madmatt
|
|
|
|
|
Damn, pipped to the post again! Note to self: must type faster.
|
|
|
|
|
Maybe you need to upgrade the 14.4 modem to broadband 
I know the language. I've read a book. - _Madmatt
|
|
|
|
|
The reason you are seeing this is because the minute part of 14:00 is 0. So, you end up getting 14 for the hour part and 0 for the minute part. Hence a result of 14 .
I agree with the previous poster, SQL is not the best place to reformat time, but sometimes the requirement is there and you have to do it. You should look into the format options of CONVERT. I think one option for you would be to convert starttime and endtime to strings using format 108 and then trim off the seconds if you don't want them.
http://msdn.microsoft.com/en-us/library/ms187928.aspx[^]
I'm also not convinced by your WHERE clause at the end. That looks to me like a potential performance killer, converting dates to do a selection like that. I may be wrong but I doubt if it will use an index for that so you will end up scanning over every row in the table.
|
|
|
|
|
Good Day
in have Resolved it by adding
right('0'+convert(varchar(2),datepart(hour,tb.starttime)),2) + ':' + right('0'+convert(varchar(2),datepart(minute,tb.starttime)),2) AS[Start Time],
right('0'+convert(varchar(2),datepart(hour,tb.endtime)),2) + ':' + right('0'+convert(varchar(2),datepart(minute,tb.endtime)),2) AS [End Time],
Thanks
Vuyiswa Maseko,
Spoted in Daniweb-- Sorry to rant. I hate websites. They are just wierd. They don't behave like normal code.
C#/VB.NET/ASP.NET/SQL7/2000/2005/2008
http://www.vuyiswamaseko.com
vuyiswa@its.co.za
http://www.itsabacus.co.za/itsabacus/
|
|
|
|
|
Hi everybody.
SQL Server 2005
I've got a table "Center (ID,Name)", it has some records registered.
I need to change the ID field, and make it IDENTITY = true.
I have been searching, and only found information about how to add a new identity column to an existing table, but I have to turn an existing field on a table to Identity.
I cannot create a new identity field and then delete the old one, because the registered IDs are not sequential...
Any ideas? Thanks a lot.
Time to come clean...
Vive y deja vivir / Live and let live
Javier
|
|
|
|
|
Follow the steps
1)Go to the design view of your table
2)Select ID Column
3) In the down, you will find Column Properties
4)In that you will find Identity Specification whose first property is Identity Specification. The default will be set to "No".
Change that to yes.
5)Save the table.
Hope this helps
Niladri Biswas
|
|
|
|
|
Thanks Niladri, but I meant how to do it in SQL language...
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
