Re: 8.1.7 Bug ?

On 2 Nov 2001 23:48:39 GMT, Damien Salvador <damien.salvador_at_via.ecp.fr> a écrit:
>>Probably because the decode will be a bucketload slower (if they are
>>using this function not as a replacement for decode in sql)

>This is indeed not the full function, but only the part that causes the
>failure. BTW, the exact pb seems to be in the 'if' statement.

and we did quite a number of "side-tests". Checking the length of the variable instead of "is not null", or using nvl does no good.

Did not try the 'decode' in our sql statement calling the function. I will try tomorrow :-)

But still, it's only avoiding the issue.

Oh, and I did not see the message from Sybrand directly. Only the answer, and he doesn't seem to be in my kill-file :-( Cancelled post or definitively me ?

-- 
Damien