Re: proposal: psql: show current user in prompt
От | Jelte Fennema |
---|---|
Тема | Re: proposal: psql: show current user in prompt |
Дата | |
Msg-id | CAGECzQSD11wX7vNZw1ZapxLQrg3HW8pOWmNueRYZQkWyUvp61Q@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: proposal: psql: show current user in prompt (Pavel Stehule <pavel.stehule@gmail.com>) |
Ответы |
Re: proposal: psql: show current user in prompt
|
Список | pgsql-hackers |
On Fri, 8 Sept 2023 at 21:08, Pavel Stehule <pavel.stehule@gmail.com> wrote: > ok changed - there is minor problem - almost all PQ functions are of int type, but ProtocolVersion is uint Your current implementation requires using the PG_PROTOCOL macros to compare versions. But clients cannot currently use those macros since they are not exported from libpq-fe.h, only from pqcomm.h which is then imported by libpq-int.h. (psql/command.c imports pcomm.h directly, but I don't think we should expect clients to do that). We could ofcourse export these macros from libpq-fe.h too. But then we'd need to document those macros too. > Using different mapping to int can be problematic - can be messy if we cannot use PG_PROTOCOL macro. I see no big problems returning an unsigned or long from the new function (even if existing functions all returned int). But I don't even think that is necessary. Returning the following as an int from PQprotocolVersionFull would work fine afaict: return PG_PROTOCOL_MAJOR(version) * 1000 + PG_PROTOCOL_MINOR(version) This would give us one thousand minor versions for each major version. This seems fine for all practical purposes. Since postgres only releases a version once every year, we'd need a protocol bump every year for one thousand years for that to ever cause any problems. So I'd prefer this approach over making the PG_PROTOCOL macros a public interface.
В списке pgsql-hackers по дате отправления: