Horton

Re: [18] Unintentional behavior change in commit e9931bfb75

Noah Misch <noah@leadboat.com>

View thread View on postgr.es
From: Noah Misch <noah@leadboat.com>
To: Jeff Davis <pgsql@j-davis.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, pgsql-hackers@postgresql.org
Date: 2025-04-16T01:02:13Z
Lists: pgsql-hackers 
On Tue, Apr 15, 2025 at 04:08:55PM -0700, Jeff Davis wrote:
> On Mon, 2025-04-14 at 13:44 -0700, Noah Misch wrote:
> > However, I think v17's
> > concept of separate PG_REGEX_ symbols for the default-locale case is
> > still the
> > right thing for v18.  In other words, this code should change to look
> > more
> > like v17, with the removal of non-locale_t calls being the main
> > change.
> 
> I tried that in v2-0003, but I think it ended up worse. Most
> pg_wc_xyz() functions don't care if it's the default collation or not,
> so there are a lot of duplicate cases.

I'd likely adopt 0003, but I'm fine if you stop at 0002.

> + * As a special case, in the "default" collation we force ASCII letters to

I'd change s/we force/(2) and (3) force/ to make explicit that this isn't
specific to (3).  That's all I would change in v2.

Commits

  1. Another unintentional behavior change in commit e9931bfb75.

  2. Improve comment in regc_pg_locale.c.

  3. Fix unintentional behavior change in commit e9931bfb75.

  4. Remove support for null pg_locale_t most places.

  5. Fix up handling of C/POSIX collations.