Re: [HACKERS] Implementation of SQLCODE and SQLERRM variables

Pavel Stehule wrote:
> On Tue, 19 Apr 2005, Tom Lane wrote:
>
> > Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> > > Guys, is this patch ready for application? I think so, but am not 100%
> > > sure.
> >
> > I haven't read the code yet, but in any case I still object to choosing
> > Oracle-like names for Oracle-incompatible functionality. We need to
> > settle on better names.
> >
>
> I changed names on SQLSTATE and SQLERRM. The behave our SQLSTATE and db2
> SQLSTATE is very similar. But Oracle support SQLSTATE too (in ANSI mode).
> Only Oracle support SQLERRM.
>
> http://developer.mimer.com/howto/howto_25.htm
>
> SQLSTATE, a 5-character return value specified by the SQL standard,
> replaced SQLCODE years ago (SQL-92?). Perhaps both Oracle and DB2/2
> have support for it.
>
> But I didn't find good informations about error trapping in ANSI SQL. All
> informations are about Oracle or DB2 :-(.

Oh, it sounds like we are OK then with the names you used.

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073