Question

I have a t-sql user-defined-sp (call it sp1) which gets userId as param and return

I want to create a new user-defined-sp (call it sp2) that gets a list of userIds.

sp2 executes sp1 for each userId and union all output to one big output.

Each set of outputs for userId=a will be consecutive in the big table

i.e. the big table can look like:

a  result1,1 
a  result1,2
a  result1,3
b  result2,1
b  result2,1

how can I do this?

Answer 1

Inside the main stored proc (sp2), you can create a table variable, and populate it with the data, then select all from that table variable. Something like this:

DECLARE @loop_tbl TABLE (ID int identity(1,1), UserID int)
DECLARE @return_tbl TABLE (column_a varchar(1), column_result varchar(20), column_int int)
DECLARE @min int, @max int

INSERT INTO @loop_tbl(UserID) SELECT UserID FROM Users
SELECT @min=MIN(ID), @max=MAX(ID) FROM @loop_tbl

WHILE @min <= @max
BEGIN
    INSERT INTO @return_tbl (column_a, column_result, column_int)
    EXEC sp1 @userid=(SELECT UserID FROM @loop_tbl WHERE ID=@min)
    --Proc must return a data set using SELECT with the proper columns in the proper order

    SET @min=@min+1
END

SELECT * FROM @return_tbl 

The only thing to watch out for is that INSERT EXEC will fail if sp2 is nested inside of another stored proc.