| From: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> |
|---|---|
| To: | Vivek Khera <khera(at)kcilink(dot)com> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: trouble caused by change in 7.3 handling of '' in |
| Date: | 2002-12-19 17:01:21 |
| Message-ID: | 200212191701.gBJH1LI26908@candle.pha.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Vivek Khera wrote:
> >>>>> "PE" == Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
>
> PE> It's one thing to have warts for allowing the market-leading GUI frontend
> PE> for databases (MS Access) to work, but if every, with all due respect,
> PE> random client wants to have its own break-the-spec-for-me switch we'll go
> PE> mad.
>
> PE> Explain to me why it's so hard to say 0 when you mean 0.
>
> It is not hard. What is hard is when you change behavior abruptly
> between releases without a transition period in which you warn about
> an upcoming change when that deprecated feature is used. Ideally, it
> would have logged this error for the 7.3 release, and disallowed it
> for 7.4, something akin to the LIMIT transition.
>
> I guess nobody realized how some popular applications used this
> non-compliant feature before the change was implemented. Perhaps we
> learn from this experience and don't do this type of change again.
> Perhaps not.
Yes, this was really it --- we didn't realize how many apps used this.
Also, we expected to hit them before final release, when we could have
reverted the change.
The prompting of the change was that '' -> 0 didn't make sense, and
it was masking some COPY file format errors.
--
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
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Larry Rosenman | 2002-12-19 17:06:08 | Re: trouble caused by change in 7.3 handling of '' in |
| Previous Message | Csaba Nagy | 2002-12-19 16:58:26 | Replacement for Oracle's "ROWNUM" in Postgres |