| From: | Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com> |
|---|---|
| To: | i(dot)sho(dot)0628(at)gmail(dot)com |
| Cc: | assam258(at)gmail(dot)com, pgsql-hackers(at)lists(dot)postgresql(dot)org |
| Subject: | Re: [PATCH v1] Preserve implicit composite casts in ROW deparsing |
| Date: | 2026-11-20 08:22:10 |
| Message-ID: | 20261120.172210.2020202608736055339.horikyota.ntt@gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hello,
At Thu, 19 Nov 2026 23:41:09 -0500, Sho Ito <i(dot)sho(dot)0628(at)gmail(dot)com> wrote in
> Attached v2 adds the `VARIADIC` case, including view recreation with
> pretty-printing both enabled and disabled, and the column-default test
> to cover `showimplicit = false`. The default still deparses as
> `ROW(1, 2)`. The implementation is unchanged from v1.
The change itself looks good to me, and the tests seem sufficient.
However, I think the comment might be more useful if it focused on
explaining why RECORDOID can be excluded even when showimplicit is
true.
For example:
> A bare ROW() expression has type RECORD, so omitting the cast to
> RECORD even when showimplicit is true does not affect overload
> resolution on reparse.
Regards,
--
Kyotaro Horiguchi
NTT Open Source Software Center
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alex Masterov | 2026-11-20 08:25:49 | Re: Snapshot related assert failure on skink |
| Previous Message | Chao Li | 2026-11-20 08:21:05 | Re: Avoid copying MinimalTuple twice in ExecHashSkewTableInsert() |