Closed Bug 1465417 Opened 7 years ago Closed 7 years ago

Thunderbird renders all email in as per the charset header and ignores folder override

Categories

(Thunderbird :: Mail Window Front End, defect)

52 Branch
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED WONTFIX

People

(Reporter: t20, Unassigned)

Details

User Agent: Mozilla/5.0 (Windows NT 6.3; WOW64; rv:60.0) Gecko/20100101 Firefox/60.0 Build ID: 20180516032328 Steps to reproduce: Select a message in incoming mail list or right-click on subject and choose "Open message in new window". Actual results: Message using UTF-8 characters is displayed in Western encoding, producing weird characters. Expected results: Incoming messages, regardless of their encoding claim, should be displayed in the user-chosen default, UTF-8 in this case. Note that a clumsy workaround is available: selecting Menu > View > Text Encoding > Unicode. However, this workaround has to be done manually for each message, each time it is viewed.
Compare with similar Bug 185523, which was fixed years ago. The current bug is currently active.
My Fallback encoding, specified under Folder > Properties, is also set to UTF-8.
I don't understand the report. If you open a message in a tab or a new window, it should be displayed in the encoding of the message. So UTF-8 messages are displayed in UTF-8. ISO-2022-JP messages are displayed in ISO-2022-JP, etc. You're saying that everything is displayed in Western (windows-1252)? Can you please attach a message that's incorrectly displayed.
Messages ARE displaying in their declared encoding, I think. This produces incorrect characters because the actual encoding of the message is UTF-8 (it is declared as iso-8859-1). I am trying to tell Thunderbird to use UTF-8 for all messages, so they will appear correctly. I will find my example message when I have time.
(In reply to David Spector from comment #4) > Messages ARE displaying in their declared encoding, I think. This produces > incorrect characters because the actual encoding of the message is UTF-8 (it > is declared as iso-8859-1). I am trying to tell Thunderbird to use UTF-8 for > all messages, so they will appear correctly. That is not a feature we support. You should contact the sender to declare the charset correctly. The folder override is only for message without declared charset.
Status: UNCONFIRMED → RESOLVED
Closed: 7 years ago
Resolution: --- → INVALID
Summary: Thunderbird renders all email in Western encoding instead of UTF-8 → Thunderbird renders all email in as per the charset header and ignores folder override
But the TB menu allows declaring the charset. Selecting UTF-8 in the menu overrides whatever was specified in the message header, even if a charset was specified. All I'm asking for is that this overriding behavior be made automatic. If you don't like that, then what about scanning the message and detecting the charset as UTF-8 based on internal evidence (nonsense characters)? I just want to be able to read even junky messages from GMail, Yahoo, AOL, or whatever. I'm not asking for adherence to standards but for "quirks mode" adaptation to messages that violate standards. Thunderbird should do this.
The |View > Text Encoding| menu is for messages that really incorrectly declare their decoding. I get one of those once in a blue moon. I don't understand what you mean by making "this overriding behaviour be made automatic". The override is for the message you're viewing and any forward or reply generated from it. If you go to the next message, it's forgotten. Scanning the message to detect the best charset would be an enhancement request. Given that that's not very useful since most message declare the charset correctly, and also hard or impossible to implement, unless you simple want to check for UTF-8, it would be very low priority. You can read "junky" messages, you just need to set the charset. Seriously, do you get so many messages which mis-declare their charset?
No, I get very few messages that have this problem. But forcing all messages to UTF-8 would work better for me (as an option). Note that TB has a built-in config editor, so it should be easy to make any such change be controlled by a boolean option that is false by default, so only people who want to force the charset to UTF-8 would be able to do just that.
There are too many (undocumented) options already. We're not implementing this. Feel free to propose a patch, but the TB module owner is in general opposed to "yet another option".
Resolution: INVALID → WONTFIX
I'm disappointed, but not surprised. I feel sure this decision will be revisited. UTF-8 is a neat solution to I18N and the logic of supporting it fully everywhere will only grow.
You need to log in before you can comment on or make changes to this bug.