[Email-SIG] fixing the current email module

[Glenn Linderman]

On approximately 10/9/2009 5:25 AM, came the following characters from 
the keyboard of Barry Warsaw:
> On Oct 8, 2009, at 7:02 PM, Glenn Linderman wrote:
>
>> Well, that is a feature of some mailing list programs.  Those that 
>> want to do that, will have to decode and re-encode.
>>
>> However, there are definitely mailing lists that don't do that.  
>> Google Groups is one example that doesn't collapse, and always 
>> prepends the headers in front of Re:.  Seems like all the Python 
>> lists do the collapsing (I wonder why! :) )  Other lists don't do 
>> prepending (I think the RFCs recommend not prepending in Subject, 
>> actually), of the others I'm subscribed to, that prepend, some 
>> collapse and some don't.
>>
>> I'm saying that there are use cases where prepending could be done 
>> without decoding; while you are positing use cases where that is 
>> insufficient, but you shouldn't have said "Except"... you should have 
>> said "There are also other use cases".
>>
>> And when you collapse Re:, do you also collapse various 
>> language-specific spellings of Re: ???  that is a hard problem.
>
> I don't disagree with any of that.  It's all firmly in the scope of 
> the application, not the email package.  The email package just has to 
> make it possible. 

Yes.  So since the application has such latitude to make such decisions, 
it seems that the email package should do minimal parsing, analysis, and 
decoding of incoming messages until such time as the application chooses 
to request particular information.

So it seems there need to be APIs to retrieve and set (using your 
terminology from another reply) wire format header values, as well as 
decoded header values.

-- 
Glenn -- http://nevcal.com/
===========================
A protocol is complete when there is nothing left to remove.
-- Stuart Cheshire, Apple Computer, regarding Zero Configuration Networking