Re: draft-ietf-httpbis-p5-range-21, "3.2 416 Requested Range Not Satisfiable"


------ Original Message ------
From: "Zhong Yu" <zhong.j.yu@gmail.com>
To: "Adrien W. de Croy" <adrien@qbik.com>
Cc: "Julian Reschke" <julian.reschke@gmx.de>;"ietf-http-wg@w3.org" 
<ietf-http-wg@w3.org>
Sent: 25/10/2012 10:00:04 a.m.
Subject: Re: draft-ietf-httpbis-p5-range-21, "3.2 416 Requested Range 
Not Satisfiable"
>On Wed, Oct 24, 2012 at 2:58 PM, Adrien W. de Croy <adrien@qbik.com> wrote:
>
>>
>>I've always considered multipart/byteranges to be less than optimal design
>>for the problem.
>>
>>A server should be able to send the byte ranges coalesced in a single
>>message body, since it advertised the ranges coming back it's possible to
>>unpick it, and doesn't then require the part separators etc.
>>
>
>
>That doesn't work for other range units though (but does anyone
>actually use non-byte units?)
>

I've always considered any other range unit to be a Really Bad Idea 
which I hope never sees the light of day.  In fact I'd love to see the 
extensibility of Range deprecated.

there is always another way to do different types of range (e.g. pages, 
frames etc) which doesn't break everything (esp caches).

>
>
>I agree multipart sucks. It was probably designed for human eyes? It's
>hard for programming, both in generating and in parsing.
>
It's horrible to parse.  MIME multipart is not a good format for 
performance.
>>
>>
>>that way you don't need to overload the Content-Type which then removes your
>>ability to transfer the actual content type (although presumably this has
>>been communicated earlier).
>>
>>Does anyone actually use multiple ranges?
>>
>
>
>It doesn't seem necessary at all. The client can always send multiple
>requests, each for a single range; the overhead of multiple requests
>is probably inconsequential compared to the bytes of the body.
>
>
>>
>>
>>
>>------ Original Message ------
>>From: "Zhong Yu" <zhong.j.yu@gmail.com>
>>To: "Julian Reschke" <julian.reschke@gmx.de>
>>Cc: "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
>>Sent: 25/10/2012 4:52:03 a.m.
>>Subject: Re: draft-ietf-httpbis-p5-range-21, "3.2 416 Requested Range Not
>>Satisfiable"
>>
>>>
>>>
>>>Wouldn't "Content-Type: multipart/byteranges" cause confusions if it's
>>>used anywhere other than in a 206 response?
>>>
>>>Suppose a representation itself has the content type of
>>>"multipart/byteranges"
>>>
>>>  Get /slivers HTTP/1.1
>>>
>>>
>>>  HTTP/1.1 200 OK
>>>  Content-Type: multipart/byteranges
>>>
>>>That's pretty confusing for observers. Even more confusingly
>>>
>>>  Get/slivers HTTP/1.1
>>>  Range: bytes=0-499
>>>
>>>
>>>  HTTP/1.1 206 Partial Content
>>>  Content-Type: multipart/byteranges
>>>  Content-Range: bytes 0-499/1234
>>>
>>>Maybe we should strongly discourage the use of multipart/byteranges in
>>>any application except in a HTTP 206 response.
>>>
>>>Zhong Yu
>>>
>>>
>>>
>>>On Wed, Oct 24, 2012 at 7:21 AM, Julian Reschke <julian.reschke@gmx.de>
>>>wrote:
>>>
>>>
>>>>
>>>>
>>>>
>>>><http://greenbytes.de/tech/webdav/draft-ietf-httpbis-p5-range-21.html#status.416>:
>>>>
>>>>"When this status code is returned for a byte-range request, the response
>>>>SHOULD include a Content-Range header field specifying the current length
>>>>of
>>>>the representation (see Section 5.2). This response MUST NOT use the
>>>>multipart/byteranges content-type. For example,"
>>>>
>>>>What is this "MUST NOT" about? Are there clients that will ignore the
>>>>status
>>>>code and assume success if they see the expected content-type?
>>>>
>>>>Best regards, Julian
>>>>
>>>>
>>>>
>

Received on Wednesday, 24 October 2012 21:07:34 UTC