>
> I discovered the hard way that there is no regression test for
> copy in/out (COPY table TO stdout, etc). This is not good.
> pg_dump depends on copy in/out, and pg_dump is rather a critical
> facility, wouldn't you say?
>
> I'd suggest, in fact, that there ought to be a regression test
> specifically exercising pg_dump and the resulting reload script.
>
> regards, tom lane
What I have tried in the past is to dump the completed regression
database and reload it and dump it again, and compare the output. Seems
to find problems, but did not do that for 6.3.*.
--
Bruce Momjian | 830 Blythe Avenue
maillist@candle.pha.pa.us | Drexel Hill, Pennsylvania 19026
+ If your life is a hard drive, | (610) 353-9879(w)
+ Christ can be your backup. | (610) 853-3000(h)