[imp] Encoding when answering

Jānis je at ktf.rtu.lv
Tue Oct 18 17:54:13 UTC 2011


Citējot Michael M Slusarz <slusarz at horde.org>
Tue, 18 Oct 2011 11:45:52 -0600:

> Quoting Jānis <je at ktf.rtu.lv>:
>
>> Citējot Michael M Slusarz <slusarz at horde.org>
>> Tue, 18 Oct 2011 11:17:56 -0600:
>>
>>> Quoting azurIt <azurit at pobox.sk>:
>>>
>>>> It's hapenning on my installation and unicode chars are being  
>>>> corrupted cos of this. When i'm answering to e-mail with UTF-8  
>>>> charset, everything is fine, my e-mail has UTF-8 in headers. But  
>>>> when i'm answering to e-mail with, for example, ISO-8859-1,  
>>>> unicode chars are corrupted and charset ISO-8859-1 is set in  
>>>> headers (i can see this on message in Sent folder).
>>>>
>>>> How can i set preferred outgoing charset ?
>>>
>>> You can't in this situation.  If a sender is sending you messages  
>>> in ISO-8859-1, we can't assume they support UTF-8 so we can only  
>>> return messages that use the ISO-8859-1 charset.  If there are  
>>> unicode characters in there, you are out of luck.
>>>
>>> We have discussed removing outgoing charset composition selection  
>>> completely and sending messages either in US-ASCII or UTF-8.  But  
>>> this would be a major change and won't happen before 5.1.
>>
>> you can edit compose.php in a way that everything going out has  
>> UTF-8 encoding (more on that you can find in the lists archives for  
>> August/September or at bugs.horde.org).
>>
>> I know Michael will protest, but by such bad behaviuor I solved my  
>> own problems (and it seems that does not make any problem with my  
>> correspondents) as none protests about undecipherable e-mails  
>> anymore.
>
> I am not protesting.  We just can't do this in the mainline code in  
> a non-point release because it is such a huge change.
>
> The issue, which you completely gloss over, is that some recipients  
> won't be able to read the outgoing messages since they don't support  
> UTF-8 (they should be able to view the ASCII characters fine, but  
> all non-ASCII characters will be garbled.  This is no better than  
> what happens in the current situation).  So it is debatable whether  
> this is any better than the current situation.

I do not argue, may be such exists somewhere - i just do not know  
anyone from my many hundreads of correspondents (mailing lists and  
personal) who cant read UTF-8. The only thing i know is my ancient  
Siemens S10 mobile phone which shows garbage insteead of message.



More information about the imp mailing list