Received: by taz.hyperreal.com (8.8.4/V2.0) id AAA13048;
 Tue, 18 Feb 1997 00:23:19 -0800 (PST)
Received: from paris.ics.uci.edu by taz.hyperreal.com (8.8.4/V2.0) with SMTP
 id AAA13040; Tue, 18 Feb 1997 00:23:16 -0800 (PST)
Received: from kiwi.ics.uci.edu by paris.ics.uci.edu id aa13707;
          18 Feb 97 0:21 PST
To: new-httpd@hyperreal.com
Subject: Re: lingering_close and performance 
In-reply-to: Your message of "Mon, 17 Feb 1997 10:27:42 PST."
             <Pine.SGI.3.95.970215112704.11309A-200000@fully.organic.com>
Date: Tue, 18 Feb 1997 00:21:09 -0800
From: "Roy T. Fielding" <fielding@kiwi.ICS.UCI.EDU>
Message-ID: <9702180021.aa13707@paris.ics.uci.edu>
Sender: new-httpd-owner@apache.org
Precedence: bulk
Reply-To: new-httpd@hyperreal.com

>Also, earlier I said I'd work on a generalized way to avoid multiple calls
>to mod_browser's parse_headers_browser_module -- I believe the attatched
>patch does so.  Other modules should be able to make use of the same
>functionality as described in the comments; this should also work when
>Apache runs as an inetd daemon.  I've tested it an it appears to work

Global variables are not reliable.  I prefer Doug's method of determining
if the request is the initial one.  Using my code, it would be

int is_initial_req(request_rec *r)
{
    return ((r->main == NULL) && (r->prev == NULL));
}

Is there some condition we are missing?

.....Roy