I've revised `net/sendmail', with one noticeable change where it now
encodes the sender/subject/recipient strings if they contain non-ascii
characters. (This is fine for existing uses that already do that
before calling sendmail, since the quoted string will not have such
characters.)
But this raises two related questions:
1. The quoting (which is done by `net/unihead') uses a
quoted-printable latin-1 encoding unless there are characters
higher than 255, and in that case it switches to a base64 encoding
with utf-8.
It seems to me that always using utf-8 is better, with
quoted-printable to make it possible to decipher some of the text
when viewed in raw form. Any objections to doing that?
(IIUC, this change would affect only sirmail.)
2. There is still an issue with non-ascii content in the body.
Digging around, it seems to me that the easiest way to make it work
is to always add a "Content-Transfer-Encoding: 8bit" header.
Does anyone know if this is a correct way to do this? (I'm not
clear about the differences between the different "content"
headers.)
--
((lambda (x) (x x)) (lambda (x) (x x))) Eli Barzilay:
http://barzilay.org/ Maze is Life!
_________________________________________________
For list-related administrative tasks:
http://lists.racket-lang.org/listinfo/dev
