here is a couple of ways

set nocount on

create table #customers (

[cid] int identity primary key

,[ac1] money not null

,[ac2] money not null

,[ac3] money not null

)--create table

-- don't leave "include actual execution...

I gave him a way out by asking him if he was pretending

He chose not to take the way out

I even proved he was stupid with working code

Which he ignored...

I always liked the idea that Microsoft deliberately keep things complicated to keep us IT folk in work

things getting more powerful, more advanced and more complex every year just works...

oh my god

you weren't joking!

given that Java catch blocks are consistently non-resumptive

and T-SQL catch blocks MAY or MAY NOT be resumptive

yes I am calling you stupid

LOL you have even proved...

The part of IT I find most challenging is not learning or understanding the technologies ...

... but discovering all the really important things that *aren't* documented

those are the niggly bits...

Funny, you are pretending to be stupid as a joke?

Lynn Pettis (7/15/2010)

---

Here is what I hear from the OP:

I have a stored procedure written where after each DML statement the value of @@ERROR is evaluated to determine the success...

First, there is NO "implicit Try..Catch" in C#. If you do NOT put a Try..Catch in the code, it will bubble the exception all the way back to the Framework,...

repeatedly adding a random offset (> 0) to an initial seed value

will give a randomised sequence of unique integers

something like:

create table #temp (

[rid] int

...

these are dreary non-arguments

in VB6 you can control whether a given block of code is resumptive or not

in T-SQL only code that is EXPLICITLY DECLARED INSIDE A LOCAL TRY BLOCK...

VB6 is a good example, because you can use

On Error Resume Next

or

On Error Goto 0

at the top of any module and it is will be PREDICTABLE and CONSISTENT regardless of...

I am having same problem

if I wrap the SP code in an

execute ('code here')

then the stored procedure takes 250 ms

otherwise it takes 8500 ms

the exe plans are completely different

the SP...

that is the output from query analyzer

spTryTestA:1

Msg 50000, Level 16, State 1, Procedure spTryTestA, Line 6

spTryTestA:2

spTryTestA:3 << codepath reaches here

spTryTestB:1

spTryTestA:1 << oh, codepath exits

spTryTestB:3

It is not incorrect - I am...

in the following code which statements are in a TRY block?

create dbo.spBlah

as

raiserror('is this?', 16, 1)

begin try

raiserror('we all agree this one is', 16, 1)

end try

begin catch

...