[Mailman-Developers] [ mailman-Patches-498766 ] See through MIME for list prefix

[noreply@sourceforge.net]

Patches item #498766, was opened at 2002-01-02 18:36
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=498766&group_id=103

Category: None
Group: None
>Status: Closed
>Resolution: Out of Date
Priority: 5
Submitted By: Mikhail Zabaluev (mzabaluev)
Assigned to: Nobody/Anonymous (nobody)
Summary: See through MIME for list prefix

Initial Comment:
In mailing lists with high percentage of non-English
messages, the subject is often encoded using MIME
=?charset?[qb]?...?= sequences. Some user agents encode
_every_ word or even the entire subject, including the
list label that might have been reproduced in a reply.
Therefore, it's useful to decode subject into Unicode
representation, to make sure the label isn't hidden
there so we can slap the label before the (encoded)
original subject without being redundant. The patch
achieves just that.

----------------------------------------------------------------------

>Comment By: Barry A. Warsaw (bwarsaw)
Date: 2002-08-09 19:48

Message:
Logged In: YES 
user_id=12800

I think this patch is no longer necessary for MM2.1b3. 
CookHeaders now checks for the prefix in each decoded part
of the subject header.

This might miss prefixes with spaces in them where each word
is encoded in a separate chunk, although I think in practice
that's unlikely.  

I'm rejecting this patch, but if you think that what's in
MM2.1b3 isn't sufficient, please do work up another patch
against cvs and I'll consider it. 

Thanks.

----------------------------------------------------------------------

Comment By: Mikhail Zabaluev (mzabaluev)
Date: 2002-04-04 16:07

Message:
Logged In: YES 
user_id=313104

Oh -- I was under the impression that mimelib has been
merged into Python completely and is developed in its CVS
trunk. Now that I look into email.Header.decode_header,
things become tricky. To be 100% correct, we have to
reassemble the Subject line into a Unicode string. Which may
raise issues with codecs. But, since the "quick and dirty"
solution has been submitted in the patch 528031, we can
afford being more thorough here. The update is underway.

----------------------------------------------------------------------

Comment By: Mikhail Zabaluev (mzabaluev)
Date: 2002-04-04 15:31

Message:
Logged In: YES 
user_id=313104

The patched file compiles dandy with Python 2.2, using its
bundled email package. Note that I didn't modify imports, as
it is now in CVS:
import email.Utils


----------------------------------------------------------------------

Comment By: Martin v. Löwis (loewis)
Date: 2002-04-04 07:45

Message:
Logged In: YES 
user_id=21627

It's deprecate in the code that is in

http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/mimelib/mimelib/email/Utils.py.diff?r1=1.12&r2=1.13

Not sure what to do about this, though, since email.Header
is not available in Python. Since mailman requires the
mimelib version anyway, it appears better to forward-port
it, in the long run.

----------------------------------------------------------------------

Comment By: Mikhail Zabaluev (mzabaluev)
Date: 2002-04-03 17:17

Message:
Logged In: YES 
user_id=313104

Since when email.Utils.decode (in the Python distribution
since 2.2) is deprecated? I cannot find anything about the
deprecation, and it's still in the docs.

----------------------------------------------------------------------

Comment By: Martin v. Löwis (loewis)
Date: 2002-04-02 02:44

Message:
Logged In: YES 
user_id=21627

This patch uses Utils.decode, which is deprecated.

Care to revise this patch? Please also take into account

https://sourceforge.net/tracker/index.php?func=detail&aid=528031&group_id=103&atid=300103

Which of the patch do you consider better/more correct?

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=498766&group_id=103