W3C home > Mailing lists > Public > ietf-http-wg@w3.org > January to March 2013

Re: HTTP/2 Header Encoding Status Update

From: James M Snell <jasnell@gmail.com>
Date: Sun, 3 Mar 2013 15:46:38 -0800
Message-ID: <CABP7RbdQkz-Sh2P4Eduxun5G734Yu-qD9L94AL81mkmVNHH2Yg@mail.gmail.com>
To: "Adrien W. de Croy" <adrien@qbik.com>
Cc: James Cloos <cloos@jhcloos.com>, "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
The LSB-ordered scheme used by protobufs is really drop dead simple to
implement. Extremely difficult to get wrong unless you forget to account
for signs. That's really the reason I chose it for the prototype.


On Sun, Mar 3, 2013 at 1:13 AM, Adrien W. de Croy <adrien@qbik.com> wrote:

>
>
> ------ Original Message ------
> From: "James Cloos" <cloos@jhcloos.com>
> To: "James M Snell" <jasnell@gmail.com>
> Cc: "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
> Sent: 3/03/2013 11:03:18 a.m.
> Subject: Re: HTTP/2 Header Encoding Status Update
>
>>   "JMS" == James M Snell <jasnell@gmail.com> writes:
>>>>>>>
>>>>>>
>> JMS> LSB.. I'm currently following the same scheme as protobufs but
>> that's not
>> JMS> set in stone.
>>
>> IP itself is Big Endian. No protocol over it should choose otherwise.
>>
>
> what proportion of hardware processing HTTP is big-endian?
>
>
>
>
>> There is too much room for confusion otherwise.
>>
>
> It's not actually that hard to get right.
> you use library functions to get the information from TCP and lower anyway.
>
> IME you're more likely to get errors when using big-endian data, when you
> omit a htonl / ntohl.
>
> Adrien
>
>
>> If protobufs does so, that is a bug.
>>
>> -JimC
>> --
>> James Cloos <cloos@jhcloos.com> OpenPGP: 1024D/ED7DAEA6
>>
>>
>
Received on Sunday, 3 March 2013 23:47:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 3 March 2013 23:47:30 GMT