pgnus 0.98 bug: double re-encoding of forwarded messages

pgnus 0.98 bug: double re-encoding of forwarded messages

[Vladimir Volovich]

-- forwarded message --
From: vvv@vsu.ru
Subject: This is a test. Это тест.
Date: 13 Nov 1999 21:55:07 +0300
To: nobody

Hi,

The present "forwarded message" which contains 8-bit chars in koi8-r
encoding, attached not as MIME part, but between lines
containing "-- forwarded message --" and "-- end of forwarded message --"

So, it is a bug in some gnus-* function which treats these messages as
MIME attachments.

Это тест. Это тест. Это тест. Это тест.
This is a test. This is a test. This is a test.
Это тест. Это тест. Это тест. Это тест.
This is a test. This is a test. This is a test.
Это тест. Это тест. Это тест. Это тест.
This is a test. This is a test. This is a test.
Это тест. Это тест. Это тест. Это тест.
This is a test. This is a test. This is a test.
-- end of forwarded message --

-- 
Vladimir.

Re: pgnus 0.98 bug: double re-encoding of forwarded messages

[Shenghuo ZHU]

>>>>> "VV" == Vladimir Volovich <vvv@vvv.vsu.ru> writes:

VV> From: vvv@vsu.ru
VV> Subject: This is a test. Это тест.
VV> To: nobody
VV> Date: 13 Nov 1999 21:55:07 +0300

VV> Hi,

VV> The present "forwarded message" which contains 8-bit chars in koi8-r
VV> encoding, attached not as MIME part, but between lines
VV> containing "-- forwarded message --" and "-- end of forwarded message --"

VV> So, it is a bug in some gnus-* function which treats these messages as
VV> MIME attachments.

VV> Это тест. Это тест. Это тест. Это тест.
VV> This is a test. This is a test. This is a test.
VV> Это тест. Это тест. Это тест. Это тест.
VV> This is a test. This is a test. This is a test.
VV> Это тест. Это тест. Это тест. Это тест.
VV> This is a test. This is a test. This is a test.
VV> Это тест. Это тест. Это тест. Это тест.
VV> This is a test. This is a test. This is a test.
VV> ----------


VV> -- 
VV> Vladimir.

It is a long story. I have been fixing this problem in mm-uu for
several months. Now it should work for most cases.

The fix has been checked into CVS.

-- 
Shenghuo ZHU

Re: pgnus 0.98 bug: double re-encoding of forwarded messages

[Shenghuo ZHU]

>>>>> "VV" == Vladimir Volovich <vvv@vvv.vsu.ru> writes:

VV> "ZSH" == Shenghuo ZHU writes:
ZSH> It is a long story. I have been fixing this problem in mm-uu for
ZSH> several months. Now it should work for most cases.

ZSH> The fix has been checked into CVS.

VV> Thanks a lot (i did not yet check, -- i'll wait for a patch at
VV> ding-patches list :). I have one more "bug" in gnus which you probably
VV> might be interested to know about.

VV> When an article contains 8-bit chars (e.g. cyrillic text in koi8-r
VV> encoding), then gnus-article-de-quoted-unreadable (W q) does not work:
VV> it washes the broken quoted-unreadable text, but all the other
VV> non-ascii (8-bit) text becomes unreadable (again because of double
VV> re-encoding). Could gnus-article-de-quoted-unreadable be fixed to not
VV> break non-ascii chars?

VV> 	Best regards, -- Vladimir.

The function article-de-quoted-unreable does two steps, decoding
quoted-printable, then decoding charset. Decoding charset generally
decodes the whole buffer, therefore all the other non-ascii texts are
double-decoded. It is not easy to fix, unless restricting to each
quoted-printable region to decode charset.

But you can try `C-u g W q'. However, this command will not decode
headers.

-- 
Shenghuo