RE: #41: Header Block field name length

Yes, 8 bits should be sufficient.

>From my stats, the longest header name I found is 32 character long (access-control-allow-credentials): this means 6 bits !

However, I would defer the exact decision to the definition of the header encoding format: 1 byte could be used to carry a 1 bit flag and the length of the header name.

Hervé.

> -----Original Message-----
> From: Amos Jeffries [mailto:squid3@treenet.co.nz]
> Sent: mardi 26 février 2013 05:32
> To: ietf-http-wg@w3.org
> Subject: Re: #41: Header Block field name length
> 
> On 26/02/2013 12:31 p.m., Mark Nottingham wrote:
> > Yeah, personally I'd agree that 32 bits is a bit much...
> >     http://http2.github.com/http2-spec/#HeaderBlock
> >
> > Say, 8 bits?
> >
> > (opening as <https://github.com/http2/http2-spec/issues/41>)
> 
> +1 on 8 bits.
> 
> smaller would seem to be better, but 8-bit alignment is reasonable.
> 
> 
> > On 26/02/2013, at 10:26 AM, James M Snell wrote:
> >
> >> Sigh.. ok, how about the part about limiting header field name length
> >> to <= 0xFF?
> >>
> >> On Mon, Feb 25, 2013 at 3:02 PM, Mark Nottingham wrote:
> >>> I'd be really, really wary of this. They may not be standard or common,
> but I've seen many headers that exercise the stranger characters available,
> and having them break in HTTP/2 would not be good.
> 
> Examples? and are the custom ones or
> 
> Can the definition be narrowed down in the HTTPbis drafts to remove some
> of the characters not even being used by these exceptional cases?
> 
> Amos

Received on Tuesday, 26 February 2013 09:05:48 UTC