Date: Tue, 3 Sep 1996 17:02:43 +0100 (BST)
From: Paul Sutton <paul@ukweb.com>
To: new-httpd@hyperreal.com
Subject: Re: Negotiation updates part II
In-Reply-To: 
 <Pine.HPP.3.95.960830231245.12071B-100000@ace.nueva.pvt.k12.ca.us>
Message-ID: <Pine.LNX.3.92.960903151809.194E-100000@star>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-new-httpd@apache.org
Precedence: bulk
Reply-To: new-httpd@hyperreal.com

On Fri, 30 Aug 1996, Alexei Kosut wrote:
> On Thu, 29 Aug 1996, Paul Sutton wrote:
> > I don't really like the idea of having some variants with languages, and
> > others without. This implies that the variants vary on language, but what
> > language do we consider foo.txt to have? For each of the other dimensions,
> > a variant with no value in that dimension is ok (no content-type means use
> > the server DefaultType, no charset means use ISO-8859-1, no encoding
> > means, er, no encoding). Language is special because there is no default.
>
> Hmm. Point. On the other hand, it's possible. Let's say you have a
> page that basically says "under construction". You might have an image
> that has a person digging a little ditch. You might also have three
> different versions of the text "Under construction", in different
> langauges. foo.gif, foo.txt.en, foo.txt.fr, foo.txt.es. I think the
> code deals with this fine, but it's an example of some variants with
> languages and some without (a picture is worth a thousand words,
> remember).

Yes, good point.

> > Well, I think Roy actaully suggested that if Accept-Langauge: en-US was
> > received, then Apache should assume en; q=0.5. This might not be the best
> > idea, since the en-US might have a lower q value itself, or their might be
> > other languages with q's less than 0.5. I would prefer to give the
> > 'assumed' en a q of 0.0001 to guarantee that any explicitly listed
> > languages with any value q are going to be preferred. This is set in the
> > patch.
>
> Right. I would actually perfer somewhat that it use 0.01 or 0.001. The
> HTTP spec defines q-values as being to the thousanth place, and if the
> values ever need to be sent over the wire, they shouldn't get rounded
> to 0.

Ok, I've replaced 0.0001 with 0.001 (it was also used as the q value for
variants with no language when negotiating on language).

> Nah, that's okay. This seems to work. +1 on
> mod_negotiation.patch2 plus this patch, with #undef HOLTMAN and
> #define PREFER_NO_ENCODING (though I'm willing to waffle on that last
> one).

Right, hopefully final version uploaded to ftp.hyperreal.com. This is the
same as the previous patch, with three changes:

   HOLTMAN undefined
   Default q values of 0.001 instead of 0.0001 for encoding and language
   PREFER_NO_ENCODING code included, #ifdef..#endif removed

It is in /httpd/incoming/mod_negotiation.patch4. This is a patch against
the current CVS source (3 Sept).

As a reminder, this patch updates three files: mod_negotiation.c (most of
the work), httpd.h (300 and 506 statuses) and http_protocol.c (300 and 506
status messages, and handing 300 and 506 responses via
send_error_response).

Paul
UK Web Ltd