Re: In which good intentions are punished, take 2
От | Andrew Dunstan |
---|---|
Тема | Re: In which good intentions are punished, take 2 |
Дата | |
Msg-id | 530D54BD.4040105@dunslane.net обсуждение исходный текст |
Ответ на | Re: In which good intentions are punished, take 2 (Alvaro Herrera <alvherre@2ndquadrant.com>) |
Список | pgsql-hackers |
On 02/25/2014 08:53 PM, Alvaro Herrera wrote: > Tom Lane wrote: > >> On the >> other hand, we evidently have got precious little other buildfarm >> coverage of the convert() family of functions, so maybe removing >> this test altogether wouldn't be the best thing either. > We do have precious little testing on encodings and conversions, yes. > The problem is how to test these things without having the tests fail > when any particular encoding is not installed in the test system. > > Maybe we can use the Perl test rig for this too: Peter said that if a > test requires something not installed, the test is skipped without > causing a failure. It seems to me that we could take advantage that so > that each member tests whatever involves only the encodings it has > installed; while each individual member would skip a large percentage of > tests, the buildfarm as a whole would be testing a sizable portion, if > not all of it. > It should be easy (at least on *nix) for the buildfarm client to check what encodings are installed on the machine and run tests accordingly. I haven't been following closely, but if someone provides me with a simple spec I'll try to code it up. We're about due for a release anyway. cheers andrew
В списке pgsql-hackers по дате отправления: