On Thu, 19 Dec 2002, Vivek Khera wrote:
> >>>>> "sm" == scott marlowe <scott.marlowe> writes:
>
> sm> Please be clear here. The problem is with RT, not Postgresql. I know
> sm> that doesn't comfort you personally, but it's an important point. If RT
> sm> wants to be portable, it's the kind of thing they need to fix, as well as
> sm> probably many more things in their SQL code.
>
> Yes, RT is at fault, but just making the change without a transition
> period is harmful. All I ask is to make it easier to transition
> broken apps.
Actually, I agree with you on this. I believe that when something like
this is changed it should have an accompanying GUC var that is defaulted
to false that the DBA has to turn on, and that var can then be removed in
the next major version.
I've looked through the RT source code, and there are only about two dozen
instances of updates using '' in an insert, so it looks like it wouldn't
be all that hard to fix.