Re: Refactor replication origin state reset helpers
| От | Chao Li |
|---|---|
| Тема | Re: Refactor replication origin state reset helpers |
| Дата | |
| Msg-id | 12089A8D-AB3C-4152-B8DF-C1073CEEE080@gmail.com обсуждение исходный текст |
| Ответ на | Re: Refactor replication origin state reset helpers (Masahiko Sawada <sawada.mshk@gmail.com>) |
| Список | pgsql-hackers |
> On Jan 8, 2026, at 09:46, Masahiko Sawada <sawada.mshk@gmail.com> wrote: > > On Wed, Jan 7, 2026 at 5:15 PM Chao Li <li.evan.chao@gmail.com> wrote: >> >> >> On Wed, Jan 7, 2026 at 4:49 PM Chao Li <li.evan.chao@gmail.com> wrote: >>> >>> >>> <v5-0001-Refactor-replication-origin-state-reset-helpers.patch><v5-0002-Consolidate-replication-origin-session-globals-in.patch> >>> >>> The CF CI failed a test case, but I don’t think that’s mine. I just did a self-review for v5 and couldn’t find any logicchange. So I would guess the CI failure was intermittent. I don’t know how to rerun the CI test other than bumping thepatch version and re-posting. >> >> >> V6 is only a rebase, nothing new. > > Thank you for updating the patches. > > I've made some cosmetic changes to both patches (comments and the > commit messages). Please review them and let me know what you think. > > Regards, > > -- > Masahiko Sawada > Amazon Web Services: https://aws.amazon.com > <v7-0002-Consolidate-replication-origin-session-globals-in.patch><v7-0001-Refactor-replication-origin-state-reset-helpers.patch> Only issue: ``` -/* external variables */ -RepOriginId replorigin_session_origin = InvalidRepOriginId; /* assumed identity */ -XLogRecPtr replorigin_session_origin_lsn = InvalidXLogRecPtr; -TimestampTz replorigin_session_origin_timestamp = 0; +/* Global variable for per-transaction replication origin state */ +RepOriginXactState replorigin_xact_state = { + .origin = InvalidRepOriginId, /* assumed identify */ + .origin_lsn = InvalidXLogRecPtr, + .origin_timestamp = 0 +}; ``` “identify" should be “identity". And a nitpick: ``` +/* Global variable for per-transaction replication origin state */ ``` Maybe change “for” to “holding”. But this is really a nitpick, it’s up to you. Otherwise v7 looks good. Commit messages are now neater. Best regards, -- Chao Li (Evan) HighGo Software Co., Ltd. https://www.highgo.com/
В списке pgsql-hackers по дате отправления: