Date: Mon, 24 Apr 2023 10:30:22 +0900 (JST)
Message-Id: <20230424.103022.1084509624041818400.horikyota.ntt@gmail.com>
To: pgsql-translators@lists.postgresql.org
Subject: Oddities in NLS Status Tables
From: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
User-Agent: Mew version 6.8 on Emacs 27.2
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Archived-At: 
 <https://www.postgresql.org/message-id/20230424.103022.1084509624041818400.horikyota.ntt%40gmail.com>
Precedence: bulk

Hello. I just realized I haven't subscribed this list yet..

I recently checked out the "NLS Status Tables" page [1] and noticed
that the status of some files appears to be off for Japanese and
seemingly a few other other languages as well.

Everything looks perfect for the 15 branch and perhaps master about
Japanese.

For the 14 branch, pg_amcheck is at 97% (only 14 has it). and
pg_waldump is shown as around 65% for all versions older than
15. However, I have both of them completed on my end.

I decided to dig into what's going on, so I downloaded one of the
"incomplete" files and found that, as expected, poedit says the
translation is complete. This leads me to believe that the workhorse
script for calculating completeness might have some issues with
certain languages, like Japanese, or it might not be compatible with
the format poedit uses.

I does't matter much, just thought I'd share the info.

regards.

[1] https://babel.postgresql.org/

-- 
Kyotaro Horiguchi
NTT Open Source Software Center