Re: Factoring out Content-Disposition (i123), was: Content-Disposition (new issue?)

William A. Rowe, Jr. wrote:
> 
> Julian Reschke wrote:
>> William A. Rowe, Jr. wrote:
>>>
>>> Or more to the point, TEXT* is defined as RFC2047 charset-encoded 
>>> values,
>>> so defining Content-Disposition filename as TEXT* solves the 
>>> ascii/iso/uft8
>>> puzzle.
>>> ...
>>
>> Doesn't work for me. We *know* that RFC2231-encoding already is in 
>> use, and that 2 out of 4 UAs have been supporting it for a long time.
>>
>> Why invent something new? How do you deploy it?
> 
> Ok, color me confused; RFC 2616 is quite a bit older than your draft.

But RFC2616 is not clear about whether it allows RFC2047-style encoding 
in parameter values. Again, see 
<http://tools.ietf.org/wg/httpbis/trac/ticket/111> and 
<http://greenbytes.de/tech/httpbis/issue-111.xhtml>.

Furthermore, you may have missed the fact that my draft doesn't define 
anything new; it just re-states a subset of RFC2231 (which BTW does 
predate RFC2616), which already is implemented in several UAs.

Turning it around: do you have any evidence of HTTP clients supporting 
RFC2047 encoding in parameter values? If yes, which of the encoding 
methods defined in <http://tools.ietf.org/html/rfc2047#section-4>? With 
what character sets?

> ...

BR, Julian