Re: Converting pqsignal to void return
От | Andres Freund |
---|---|
Тема | Re: Converting pqsignal to void return |
Дата | |
Msg-id | moawtxi6zza5wsvrs7y2p5yhff46vla5eivusqhgqhr5reyvih@xu34av5xxi3s обсуждение исходный текст |
Ответ на | Re: Converting pqsignal to void return (Andres Freund <andres@anarazel.de>) |
Список | pgsql-hackers |
Hi, On 2025-01-22 11:50:30 -0500, Andres Freund wrote: > On 2025-01-22 07:57:52 -0800, Paul Ramsey wrote: > > It is possible we have been Doing It Wrong all this time, and would love some pointers on the right way to do this. > > All your interrupt handler is doing "for you" is setting > http_interrupt_requested, right? Why do you need that variable? Seems you > could just check postgres' QueryCancelPending? And perhaps ProcDiePending, if > you want to handle termination too. FWIW, the even better approach in this case probably would be to use curl in a non-blocking manner. Then you can do all the blocking in a loop around WaitLatchOrSocket (or, for slightly improved efficiency, create a wait event and wait with WaitEventSetWait()) and react to interrupts with CHECK_FOR_INTERRUPTS() like normal postgres code. Greetings, Andres Freund
В списке pgsql-hackers по дате отправления: