Re: Wrong results from function that selects from vier after "created or replace"

On 12-Oct-2020, at 18:58, Tom Lane <tgl@sss.pgh.pa.us> wrote:

Bryn Llewellyn <bryn@yugabyte.com> writes:
> I hadn’t realized that submitting a bug as the doc at www.postgresql.org/docs/12/bug-reporting.html seems to suggest
isthe recommended method (“In general, send bug reports to the bug report mailing list at
<pgsql-bugs@lists.postgresql.org>…”)would _never_ result in a bug being filed (even if it’s acknowledged to be a bug).
Sothanks for that tip off! 

As far as that goes, the closest thing we have to an "official bug report"
is an email to the pgsql-bugs mailing list --- which is what you've done.
The only thing that webform does differently is attach a subject line with
a serial number; but the subject line has little bearing on how we treat
the report.

As David indicated, what we usually use as a formal reference is the
email's message ID, which can be used conveniently to index into the
Postgres project's email archives.  We do that in part because many
bug reports arrive via other email lists.

In this case, multiple people have opined that it isn't a bug, but you'd
have gotten the exact same responses if you'd gone via the webform.

            regards, tom lane

Thank you very much for this clarification. I’ll know what to do next time. I’ll also be sure to think very much harder
aboutmy analysis before sticking my neck out!