[Evolution] Character encoding and expunge?

[Bjørn T Johansen]

Not Zed wrote:
> On Mon, 2004-05-24 at 11:06 +0200, Bjørn T Johansen wrote:
> 
>>Jeffrey Stedfast wrote:
>>> because the subject encoding is broken and does not conform to the spec
>>> (which means this is probably spam anyway, so who cares???)
>>> 
>>> for more information, read rfc2047 and rfc0822
>>> 
>>
>>Well, it's not spam, it's actually mail from a customer... And I am 
>>guessing that it does not conform to the spec... But I was just curious 
>>why Evolution doesn't handle this since most of the other MUA does?
>>
> The short of it is that the code in question is hairy enough as it is, 
> without adding more workarounds for software who'se authors clearly 
> haven't read the spec, or tried to interoperate in any way.
> 
> Particularly with the plethora of hacked up mailers out there, 
> everything from perl web mailers to [gk]mailers, it would seem pretty 
> well a waste of time to put in a fix just because SlopMail pre-Beta 
> 0.0.95 didn't send out valid messages (as evolution has itself done in 
> the past), particularly since it might even be fixed by now.
> 

Well, that's fair enough... it's just annoying, especially since I am 
getting a lot of mail from my customer...

> 
>>> 
>>>>
>>>>And also, when using IMAP, why isn't the seen flag marked when I have
>>>>read the email? I have to eg. go to another folder en back again to make
>>>>sure the mail is marked as read..
>>> 
>>> 
>>> because it's a lot slower??? :-)
>>> 
>>> flag syncing is delayed until either:
>>> 
>>> 1. a timeout expires
>>> 2. folder change
>>> 
>>
>>It can't be that much slower but then I am running my own mailserver on 
>>my LAN and I don't have any trouble with bandwidth... :)
>>
> Yes but we can't tell that from evolution.  The delay could be 
> unspecifically long.
> 
> Sure we could do better, e.g. launch a background thread to do the flag 
> syncing while we're displaying our version of the flags.  But there's 
> only so many hours in a day and other more important problems to fix 
> too.  The code didn't have the mechamisms to manage background tasks 
> until fairly recently.  And well, as i keep saying (hinting), nobody's 
> sent us a patch either.
> 

Or just an option to tell if the flag syncing should be done immediately 
or not, but I wan't nag about this anymore... :)

BTJ