528791 - Thunderbird generates utf-8 headers that are not conformant with RFC2047 (in case of long headers)

Status: REOPENED

(Thunderbird :: General, defect)

Description

[grzech]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 6.0; en-GB; rv:1.8.1.20) Gecko/20081217 Firefox/2.0.0.20
Build Identifier: version 2.0.0.23 (20090812)

If a Subject header (and probably any other) contains non US-ASCII characters and "UTF-8" is selected as a charset, and a header is too long to be encoded in single word (75 characters limit according to RFC2047), then the second and following words generated by Thunderbird are likely to exceed maximum lenght allowed for encoded word in a header. 

Reproducible: Always

Steps to Reproduce:
1. Create a mail containing following Subject: "∞, ó, ô, õ, ń, ň, ó, ô, ő, λ, ε, έ, Ø, Þ, Ð, ð, ń, ň, ó, ô, ő, ć, ś, ę, ł, ‰, •, ™, ©, ®, █, ▓, ▒, ░, ≤, ≥, ≈, °."
2. Choose "UTF-8" character set for outgoing e-mail
3. Press Send
4. View "Message Source" for sent e-mail to check the "Subject: " header
Actual Results:  
=?UTF-8?B?4oieLCDDsywgw7QsIMO1LCDFhCwgxYgsIMOzLCDDtCwgxZEsIM67LCA=?=
 =?UTF-8?B?zrUsIM6tLCDDmCwgw54sIMOQLCDDsCwgxYQsIMWILCDDsywgw7QsIMWRLCDEhyw=?=
 =?UTF-8?B?IMWbLCDEmSwgxYIsIOKAsCwg4oCiLCDihKIsIMKpLCDCriwg4paILCDilpMsIOKWkg==?=
 =?UTF-8?B?LCDilpEsIOKJpCwg4omlLCDiiYgsIMKwLiA=?=

Expected Results:  
=?UTF-8?B?4oieLCDDsywgw7QsIMO1LCDFhCwgxYgsIMOzLCDDtCwgxZEsIM67LCDOtSwg?=
 =?UTF-8?B?zq0sIMOYLCDDniwgw5AsIMOwLCDFhCwgxYgsIMOzLCDDtCwgxZEsIMSHLCA=?=
 =?UTF-8?B?xZssIMSZLCDFgiwg4oCwLCDigKIsIOKEoiwgwqksIMKuLCDilogsIOKWkywg?=
 =?UTF-8?B?4paSLCDilpEsIOKJpCwg4omlLCDiiYgsIMKwLiA=?=


Expected result is an example of a valid header, but it doesn't need to be exactly such one.

If the subject in "Steps to reproduce" due to lack of utf-8 suuport in bugzilla, please decode either "Actual Results" or "Expected Results" to obtain the original subject.

Comment 1

[grzech]

"If the subject in "Steps to reproduce" due to lack of utf-8 suuport in
bugzilla..." should read: "If the subject in "Steps to reproduce" is unreadable due to lack of utf-8 suuport in bugzilla..."

Anyway it is readable. :)

Comment 2

[Nikolay Shopik]

In latest trunk I have little bit different results, but still can reproduce bug. It appears what symbol ▒ causing problem, alone it doesn't produce problem space must go before it. Otherwise I will got perfectly standard valid header
Subject: =?UTF-8?B?4oieLCDDsywgw7QsIMO1LCDFhCwgxYgsIMOzLCDDtCwgxZEsIM67LCw=?=
 =?UTF-8?B?zrUsIM6tLCDDmCwgw54sIMOQLCDDsCwgxYQsIMWILCDDsywgw7QsIMWRLCDEhyw=?=
 =?UTF-8?B?IMWbLCDEmSwgxYIsIOKAsCwg4oCiLCDihKIsIMKpLCDCriwg4paILOKWkyzilpI=?=
 =?UTF-8?B?LOKWkSziiaQg4omlLCziiYgsIMKwLg==?=

Also I would like to say I did try experiment with Cyrillic symbols and can't reproduce bug using it.
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:2.0b9pre) Gecko/20101217 Thunderbird/3.3a2pre

Comment 4

[grzech]

Well, it's not really a duplicate of bug 493544. Bug 493544 is about incorrect decoding, this one is about incorrect encoding. Incorrect only in sense of too long encoded chunks with respect to RFC2047 - they should still decode correctly if decoder is fine with longer than 75 characters encoded words (chunks).