Odd behavior for Intel build

[Ruslan Zasukhin]

On 11/7/06 5:02 AM, "jda" <jda at his.com> wrote:

> Hi Ruslan,
> 
> I have a try-catch block set up to make sure that a record with a
> duplicated uniqueID isn't imported.
> 
>        success = false
>        do
>          try
>            theUniqueID = theUniqueID + 1
>            myCursor.LongField("uniqueID").value = theUniqueID
>            call myCursor.addRecord
>            success = true
>          catch theError5 as VException
>            if theError5.ErrorNumber =
> EVError.ConstraintUniqueViolation then //EVError4.ErrorNumber =
> 144643 then //and myDatabase.errNumber = 344 then //duplicate uniqueID
>              //just loop back to the try and, uh, try again
>            else
>              myCursor = nil
>              raise theError5
>            end if
>          end try
>        loop until success
>       
> 
> This works fine for PPC builds. But with a i386 build (RB 2006r4,
> Valentina b12), the catch code doesn't fire, and I get this error if
> the uniqueID isn't unique:
> 
> "Attempt to insert duplicate key row in object "thereferences" with
> unique index "U_thereferences_uniqueID".

So?

You insert non-unique exception. Valentina throw exception as expected.
Right?

I think you can see in the V4RB LOG for this function call
    ERROR ...
Right?

So if it is not catched then this sounds like RB problem.




-- 
Best regards,

Ruslan Zasukhin
VP Engineering and New Technology
Paradigma Software, Inc

Valentina - Joining Worlds of Information
http://www.paradigmasoft.com

[I feel the need: the need for speed]