Re: [PATCH] fix GIN index search sometimes losing results
Pavel Borisov <pashkin.elfe@gmail.com>
From: Pavel Borisov <pashkin.elfe@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-hackers@lists.postgresql.org
Date: 2020-07-02T11:23:13Z
Lists: pgsql-hackers
Attachments
- gin-gist-weight-patch-v4.diff (application/octet-stream) patch v4
ср, 1 июл. 2020 г. в 23:16, Tom Lane <tgl@sss.pgh.pa.us>: > Pavel Borisov <pashkin.elfe@gmail.com> writes: > > Below is my variant how to patch Gin-Gist weights issue: > > I looked at this patch, but I'm unimpressed, because it's buggy. > Thank you, i'd noticed and made minor corrections in the patch. Now it should work correctly, As for preserving the option to use legacy bool-style calls, personally I see much value of not changing API ad hoc to fix something. This may not harm vanilla reseases but can break many possible side things like RUM index etc which I think are abundant around there. Furthermore if we leave legacy bool callback along with newly proposed and recommended for further use it will cost nothing. So I've attached a corrected patch. Also I wrote some comments to the code and added your test as a part of apatch. Again thank you for sharing your thoughts and advice. As always I'd appreciate everyone's opinion on the bugfix. -- Best regards, Pavel Borisov Postgres Professional: http://postgrespro.com <http://www.postgrespro.com>
Commits
-
Replace TS_execute's TS_EXEC_CALC_NOT flag with TS_EXEC_SKIP_NOT.
- 79d6d1a277ee 14.0 landed
- 70eca6a9a6df 13.0 landed
-
Fix assorted bugs by changing TS_execute's callback API to ternary logic.
- 92fe6895d66d 13.0 landed
- 2f2007fbb255 14.0 landed