W3C home > Mailing lists > Public > ietf-http-wg@w3.org > January to March 2014

Re: [HPACK]Looking for huffman frequency tables

From: Ashok Kumar <ashokkumar.j@gmail.com>
Date: Fri, 7 Mar 2014 11:13:27 +0530
Message-ID: <CAOeYYReC4dP6R_rk5ZgcSDEiv2kDnjfyCm0O6dv08xzU7Pn+tA@mail.gmail.com>
To: Ilari Liusvaara <ilari.liusvaara@elisanet.fi>
Cc: "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
Thanks, encoding part was easy with the codes, but i was unsure about the
decoding side.


On Fri, Mar 7, 2014 at 10:30 AM, Ilari Liusvaara <
ilari.liusvaara@elisanet.fi> wrote:

> On Fri, Mar 07, 2014 at 10:03:29AM +0530, Ashok Kumar wrote:
> > Hi, I'm unable to find the huffman frequency tables for any of the
> drafts.
> > Is this available somewhere?
>
> Some very old expired drafts about header compression (not HPACK) might
> have
> that data (assuming the tables haven't been tweaked).
>
> Other than that, I don't think the frequencies are documented anywhere.
>
> > I'm not entirely clear on how to generate the
> > huffman tree with the codes already given.
>
> You don't really need to:
>
> Each entry explicitly gives the bit sequence coding each symbol.
> The string encoding is concatenation of those, followed by however
> many set bits (it is specified to pad with EOS, but first 7 bits of
> EOS are currently all ones) required to reach next octet boundary.
>
> E.g. ABC is:
>
> 1101111 11101011 11101100 1
> => DF D7 D9
>
> 1 padding bit.
>
> And abc is:
>
> 01000 101111 01001  (2 octets)
> => 45 E9
>
> The string happens to end on octet boundary, so no padding.
>
> -Ilari
>



-- 
.- ... .... --- -.-
Received on Friday, 7 March 2014 05:43:54 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:14:24 UTC