| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Aleksander Alekseev <aleksander(at)timescale(dot)com> |
| Cc: | pgsql-hackers(at)lists(dot)postgresql(dot)org, "Jelte Fennema-Nio" <postgres(at)jeltef(dot)nl>, "Bruce Momjian" <bruce(at)momjian(dot)us>, "Andreas Karlsson" <andreas(at)proxel(dot)se> |
| Subject: | Re: IANA timezone abbreviations versus timezone_abbreviations |
| Date: | 2025-01-10 20:19:05 |
| Message-ID: | 2601969.1736540345@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
I wrote:
> Hmph ... the cfbot doesn't like that one bit. It appears that the
> "LMT" abbrev is known in my machine's system-provided tzdata tree,
> but not when building our own timezone files. I wonder if this is
> exposing an existing bug (perhaps our copy of zic is too far out of
> date?). Will look into it tomorrow.
Found it: there is indeed pre-existing breakage in
pg_interpret_timezone_abbrev(), but it doesn't manifest when using
--with-system-tzdata on RHEL8, so I'd missed it in my local testing.
0001-0003 are the same as last time (so I kept the patch version
as v2), and 0004 adds a fix for the pre-existing problem.
regards, tom lane
| Attachment | Content-Type | Size |
|---|---|---|
| v2-0001-Seek-zone-abbreviations-in-the-IANA-data-before-t.patch | text/x-diff | 18.6 KB |
| v2-0002-Improve-DecodeTimezoneAbbrev-s-caching-logic.patch | text/x-diff | 6.0 KB |
| v2-0003-Add-zone-derived-abbreviations-to-the-pg_timezone.patch | text/x-diff | 11.9 KB |
| v2-0004-Make-pg_interpret_timezone_abbrev-check-sp-defaul.patch | text/x-diff | 3.9 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | m.litsarev | 2025-01-10 20:39:07 | Re: SQL function which allows to distinguish a server being in point in time recovery mode and an ordinary replica |
| Previous Message | Nathan Bossart | 2025-01-10 19:26:06 | Re: Proposal: add new API to stringinfo |