Re: CVS update of gnus/lisp (ChangeLog mm-bodies.el)

Re: CVS update of gnus/lisp (ChangeLog mm-bodies.el)

[Mark Thomas]

"Simon Josefsson" <simon@josefsson.org> writes:

>     Date: Monday, October 27, 2003 @ 21:22:47
>   Author: cvs
>     Path: /usr/local/cvsroot/gnus/lisp
>
> Modified: ChangeLog mm-bodies.el
>
> (mm-body-encoding): Don't use QP when message body
> only consists of short lines and ASCII, when
> mm-use-ultra-safe-encoding.  Refer to 'About foo' thread in
> gnus-bug, e.g. <ilullrg4k7p.fsf@extundo.com>, for more discussion.

I can't find the gnus-bug list, so I don't know the
discussion behind this change.

From this change, will you get new bugs regarding messages
where signature checks fail because lines that begin with
From have not been properly encoded?

-Mark

Re: CVS update of gnus/lisp (ChangeLog mm-bodies.el)

[Simon Josefsson]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mark Thomas <swoon@bellatlantic.net> writes:

> "Simon Josefsson" <simon@josefsson.org> writes:
>
>>     Date: Monday, October 27, 2003 @ 21:22:47
>>   Author: cvs
>>     Path: /usr/local/cvsroot/gnus/lisp
>>
>> Modified: ChangeLog mm-bodies.el
>>
>> (mm-body-encoding): Don't use QP when message body
>> only consists of short lines and ASCII, when
>> mm-use-ultra-safe-encoding.  Refer to 'About foo' thread in
>> gnus-bug, e.g. <ilullrg4k7p.fsf@extundo.com>, for more discussion.
>
> I can't find the gnus-bug list, so I don't know the
> discussion behind this change.

There is a news server on quimby.gnus.org.

> From this change, will you get new bugs regarding messages
> where signature checks fail because lines that begin with
> From have not been properly encoded?

I don't know, let's try:

- From foo

Looks dash-escaped here.  Which means it will not work without
pre-processing for PGP 2.x users, if I understand 1991 vs 2440
correctly.  This preprocessing amounts to s/^- -/-/g within the
message boundary, before running pgp.  The alternative is to use QP
for this message, but that is problematic because it is not clear if
you apply QP before or after PGP.  I suspect there are MUAs that do
both variants.  Since the s/^- -/-/g prize is paid by an ever
decreasing number of people, that do have an upgrade path AND a simple
workaround if they refuse to upgrade, I think it is worth the prize in
order to avoid QP.  Opinions?

We could also enable QP if there are '^From ' or '^-' in the body.
But the old behaviour of always QP encode ASCII appear simply wrong.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.9.2-cvs (GNU/Linux)

iQC1AwUBP528Te2iHpS1ZXFvAQIbiwUAgNIdFrMY4/fJ5CBxCbcCfJhzUrBDhhLa
5+DT53j6cAl+x5FP4sf1Ywcevy2PyaGLzsi69YNwl2qnaP4hqXqgUpJxkI/0clc9
nSoD8XXJ+PffGBQM+5uqH3/+nXk/ClSvUCSbUqNijStL5a6elqDiKKNjxZuXbm3n
sWHOcyFXkitg8I7WNo/8ZnV5njkI2m0Mm0/GHQ2KgqzCc0tOjfwSvg==
=5sAI
-----END PGP SIGNATURE-----