Re: Sections 3.3.2 and 3.3.3 allow bogus Content-Length?

Yes, you can do this in a U-A, but in a proxy we have to be strict :)

I guess those sites don't work behind proxies.

Adrien


------ Original Message ------
From: "Daniel Stenberg" <daniel@haxx.se>
To: "Matthew Kerwin" <matthew@kerwin.net.au>
Cc: "Adrien de Croy" <adrien@qbik.com>; "HTTP Working Group" 
<ietf-http-wg@w3.org>
Sent: 15/02/2017 11:05:12 AM
Subject: Re: Sections 3.3.2 and 3.3.3 allow bogus Content-Length?

>On Wed, 15 Feb 2017, Matthew Kerwin wrote:
>
>>That, along with everything else in 3.3.2 & 3.3.3, and 3.4 makes me 
>>wonder how anyone could find ambiguity there. If C-L is present, it 
>>has to be correct.
>
>... but we still see a significant amount of content on the web today 
>that serve bodies with sizes that aren't matching the C-L value.
>
>I went several rounds against that, trying to make Firefox's C-L checks 
>stricter, but finally lost in the end and had to resort to keeping 
>funky heuristics and guessworks.
>
>The web is a harsh place! =)
>
>--
>  / daniel.haxx.se
>

Received on Tuesday, 14 February 2017 22:15:52 UTC