Re: draft-reschke-rfc2231-in-http-02

Daniel Stenberg wrote:
> Hello,
> 
> (While I know Julian Reschke's work on the revised RFC2231 for HTTP 
> isn't exactly a HTTPbis topic, I hope you bear with me for abusing this 
> list a little since it doesn't really have any other home afaik.)
> 
> As I mentioned on the HTTPbis meeting at IETF75 in Stockholm, I recently 
> came to think about the Content-Disposition header in HTTP multipart 
> upload streams [RFC1867 - Form-based File Upload in HTML] and even 
> though I guess those headers are more strictly MIME than the 
> Content-Disposition in a HTTP response, it wouldn't surprise me the 
> least if browsers deal with them identically (and possibly receivers of 
> RFC1867 will also parse them HTTP-style and not MIME-style). Some 
> feedback on that would of course be very interesting.
> ...

Hi Daniel,

thanks for the feedback. I just spent a few minutes investigating 
further, and what I found out is:

- RFC1867 is obsoleted by RFC 2854

- the latter refers to RFC 2388 for the multipart MIME type (confirmed 
in <http://www.iana.org/assignments/media-types/multipart/>)

- RFC 2388 seems to be a bit confused about what encoding to use, it 
mentions both RFC 2047 (which I think is a bug), RFC 2184 and RFC 2231 
(2231 obsoleted 2184) -- I think this should be addressed with an 
erratum (Larry, are you reading this...?)

Implementations:

- the Apache commons fileupload library doesn't seem to attempt to 
handle either RFC 2047 nor RFC 2231 encoding

- IE appears to send non-ASCII characters as raw ISO, at least if the 
source page used that encoding,

- Firefox appears to escape with &#ddd; notation.

So this seems to be a mess.

BR, Julian




> Apart from that, I believe the Julian's draft works perfectly well for 
> all purposes I can think of. I have no complaints or remarks.
>

Received on Tuesday, 4 August 2009 11:48:39 UTC