After I started to use MacOS Sequoia 15.5, Japanese characters encoded
in iso-2022-jp and quoted-printable are sometimes garbled.  It seems
what is happening is:

- A single Japanese character in iso-2022-jp consists of two bytes.
  Say I have three Japanese characters that are encoded as %5%$%H in a
  long line.

- When a line is long, quoted-printable inserts = and break the line.

- If = is inserted at the character boundary, like %5%$= %H, it is
  decoded OK.

- However, if = is inserted in the middle of a character, like %5%= $%H,
  the first character %5 is decoded OK, but the second one is not
  considered as a single character, and the raw text %$%H becomes the
  result of decoding the second and third characters.

This phenomenon did not occur when I was using MacOS BigSur 11.6.7.
The mutt version I am using is 2.2.11 (2023-08-18) for BigSur and
2.2.14 (2025-05-03) for Sequoia.  However, today, I recompiled the old
2.2.11 on Sequoia and the same phenomenon happened.  This suggests
that mutt itself is OK and some strange thing is happening somewhere
else.  Still, do you have any guesses as to why this is happening?

Any help appreciated.  Sincerely,

-- 
Kenichi Asai

Reply via email to