The following code is supposed to print all rows contained in the temporary table #table_A:
create table #table_A
(
ID int NULL ,
foo int NULL
)
go
insert into #table_A values (1, 2)
insert into #table_A values (2, 3)
go
declare c_table_A cursor
for select *
from #table_A
order
by 1
open c_table_A
fetch c_table_A
while @@sqlstatus = 0
begin
print '%1!', c_table_A
fetch c_table_A
end
close c_table_A
go
However, it results in the following error message:
DECLARE CURSOR must be the only statement in a query batch.
How can I print all rows contained in a (temporary) table?
Here’s another way of putting my question:
I’m trying to do something like that:
open c_table_A
fetch c_table_A into @record_variable
while @@sqlstatus = 0
begin
print '%1!', @record_variable
fetch c_table_A into @record_variable
end
close c_table_A
Is there a way to declare a variable containing a whole row of a table in sybase?
P.S.: just using “select * from …” doesn’t work for me. I need to do some stuff with each of the rows before printing the row. (My question is supposed to focus on the essential part, which is why I didn’t go into any further details regarding other things I need to do with each row)
Advertisement
Answer
Thanks for the clarification.
In a SQL batch, rather than a stored procedure the cursor declaration must be separate from the batch that uses it, hence there needs to be a go between the declare cursor and the subsequent batch.
There is no way to define a “row variable” in Sybase ASE, sorry. Each column returned into a variable must have a variable declared for it. In the example below @id and @foo are declared as the same types as the columns id and foo in the table. Other RDBMS do have “record data types”, but unfortunately not Sybase ASE.
Before committing yourself to using a cursor, which on a large table would be relatively slow, you might be able to perform your other processing in a select statement. If there is conditional logic case ... when ... then ... else ... end could prove useful, though calling a stored procedure directly from within a select statement is not possible, you can call a SQL user defined function. That’s probably a separate question if you need help.
I’ve added a deallocate cursor statement as well, it’s part of the syntax and frees up internal workspaces associated with your connection.
You may want to execute set nocount on before running the batch, it removes the sometimes annoying (1 row affected) message.
set nocount on
go
create table #table_A
(
ID int NULL ,
foo int NULL
)
go
insert into #table_A values (1, 2)
insert into #table_A values (2, 3)
go
declare c_table_A cursor
for select *
from #table_A
order
by 1
go
declare
@id int,
@foo int
open c_table_A
fetch c_table_A into @id, @foo
while @@sqlstatus = 0
begin
print 'id: %1! foo: %2!', @id, @foo
fetch c_table_A into @id, @foo
end
close c_table_A
go
deallocate cursor c_table_A
go