291325 - gb2312 encoded header display wrong

Status: NEW

(Thunderbird :: Message Reader UI, defect)

Description

[admin]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; zh-CN; rv:1.7.7) Gecko/20050414 Firefox/1.0.3
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; zh-CN; rv:1.7.7) Gecko/20050414 Firefox/1.0.3

Subject: =?gb2312?Q?Re:java=D7=F7=D2=B5_03_=3D??=
	=?gb2312?Q?gb2312?Q?281283_=3D?=
	=?gb2312?Q?E5=3DAD=3D99=3DE5=3DAE=3D87?=
	=?gb2312?Q?=3DE8=3D81??=3D_size=3D?=
Message-ID: <[e-mail]>
Date: Thu, 21 Apr 2005 13:20:49 +0800
From: =?gb2312?Q?=D0=EC=B1=A3=C3=F1?= <....>

I captured it from one of my email source
node there's many =?gb2312?xxx?= pairs in the text above
thunderbird can only display the first part of those many =?gb2312 lines.
the others are displaied wrongly,or displaie un-uncodely...

Reproducible: Always

Steps to Reproduce:
1.receive such mail
2.see


Actual Results:  
it display in subject field
?gb2312?Q?Re:java=D7=F7=D2=B5_03_=3D??= =?gb2312?Q?gb2312?Q?281283_=3D?=
E5=AD=99=E5=AE=87 =?gb2312?Q?=3DE8=3D81??=3D_size=3D?=


Expected Results:  
should display in chinese

this can often be improved by rewriting the gb2312 decoding programe

Comment 1

[Mike Cowperthwaite]

Reproduced with TB 1.0+0414, Win2K.  The header looks correctly formatted to me; 
Jungshik, is there a problem using Q encoding for 2047 with Asian charsets?  All 
the examples I have that work use B encoding.

Comment 2

[admin]

B Encoding with Asian characters also have bug
this is my another mail which subject is

Subject: =?ISO-8859-1?B?Q2hpbmFSZW7Qo9PRwrw6ufOw4NPQ0MI=?=
 =?ISO-8859-1?B?s8nUsbzTyOssv+zIpbTyuPbV0Lr0sMkh?=
and it's result is

ChinaRenУÓѼ:¹иобуидÓÐгÉÔб└¼ÓÈë,¿имÈ¥´и░¸öÕкôбуÉ!

which should be display in chinese also.

what does it actually mean?I have searched in source code,but I can't decide
where to solve this problem.I think this is a major problem with chinese
users.Thank you.

Comment 3

[Jungshik Shin]

(In reply to comment #1)
> Reproduced with TB 1.0+0414, Win2K.  The header looks correctly formatted to me; 
> Jungshik, is there a problem using Q encoding for 2047 with Asian charsets?  

EIther B or Q can be used. Usually, B encoding is more efficient for a string
with  8bit octets taking more than a certain fraction, but quite a lot of MUAs
don't check which one is more efficient for a given  string and just use Q
encodings for all cases. I'll take a look what's wrong with our code. 

I'm taking it

Comment 4

[admin]

just ask :) , how is it going ?
I want to see If there's something I can do.
:)

Comment 5

[Jeffrey Stedfast]

read the encoded words again and you'll see something funny ;-)

Subject: =?gb2312?Q?Re:java=D7=F7=D2=B5_03_=3D??=
        =?gb2312?Q?gb2312?Q?281283_=3D?=
        =?gb2312?Q?E5=3DAD=3D99=3DE5=3DAE=3D87?=
        =?gb2312?Q?=3DE8=3D81??=3D_size=3D?=

The end of the first enc-word decodes to "=?"
the beginning of the second enc-word decodes to: "gb2312?Q?..."

see the problem? :)

hint: these encoded-words decode to more encoded-words.

Comment 6

[Wayne Mery (:wsmwk)]

still reproduces afaict