Re: ICU locale validation / canonicalization

From: Peter Eisentraut <peter(dot)eisentraut(at)enterprisedb(dot)com>
To: Jeff Davis <pgsql(at)j-davis(dot)com>, Robert Haas <robertmhaas(at)gmail(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: ICU locale validation / canonicalization
Date: 2023-03-28 06:41:24
Message-ID: 3e45a760-a7f8-80af-e4b0-be91427359d3@enterprisedb.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

[PATCH v9 1/5] Fix error inconsistency in older ICU versions.

ok

[PATCH v9 2/5] initdb: replace check_icu_locale() with
default_icu_locale().

I would keep the #ifdef USE_ICU inside the lower-level function
default_icu_locale(), like it was before, so that the higher-level
setlocales() doesn't need to know about it.

Otherwise ok.

[PATCH v9 3/5] initdb: emit message when using default ICU locale.

ok

[PATCH v9 4/5] Validate ICU locales.

Also here, let's keep the #ifdef USE_ICU in the lower-level function and
move more logic in there. Otherwise you have to repeat various things
in DefineCollation() and createdb().

I'm not sure we need the IsBinaryUpgrade checks. Users can set
icu_validation_level on the target instance if they don't want that.

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Laurenz Albe 2023-03-28 06:50:45 Re: Add standard collation UNICODE
Previous Message Masahiko Sawada 2023-03-28 06:41:15 Re: Support logical replication of global object commands