httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alexei Kosut <ako...@organic.com>
Subject Re: arrrrgh!
Date Fri, 02 Aug 1996 16:59:04 GMT
On Fri, 2 Aug 1996, Ben Laurie wrote:

[...]

> Isn't this exactly what the BrowserMatch stuff is for? (Or whatever it was
> called).

Sorta. It doesn't strike me as the most elegant solution, though.  Several
reasons, but especially necause the Host: header is a header that actually
modifies the request. Connection:, which was the Netscape problem, has not
relevance on the actual request; it just modifies how the data is
transfered.

What I'm getting at is that if we modify all the entities based on
the User-Agent string, and allow caching of them (unless you wanted to
deny caching of *all* documents the server sent...), then proxy servers
can really screw up the whole concoction.

It seems a better idea, then, to either change the semantics of how Apache
handles the Host: header, even though it's not as per spec (*sigh*...
Roy?), or get MSIE to change.

Which might be possible; it's still in beta, no? I know MSIE claims to
like standards... Hmm. I filled out a bug report, and sent it in. We'll
see.

-- Alexei Kosut <akosut@organic.com>            The Apache HTTP Server 
   http://www.nueva.pvt.k12.ca.us/~akosut/      http://www.apache.org/



Mime
View raw message