Return-Path: Delivered-To: new-httpd-archive@hyperreal.org Received: (qmail 19729 invoked by uid 6000); 20 Apr 1998 21:08:31 -0000 Received: (qmail 19720 invoked from network); 20 Apr 1998 21:08:29 -0000 Received: from devsys.jagunet.com (206.156.208.6) by taz.hyperreal.org with SMTP; 20 Apr 1998 21:08:29 -0000 Received: (from jim@localhost) by devsys.jaguNET.com (8.8.8/jag-2.4) id RAA16251 for new-httpd@apache.org; Mon, 20 Apr 1998 17:08:25 -0400 (EDT) From: Jim Jagielski Message-Id: <199804202108.RAA16251@devsys.jaguNET.com> Subject: Re: SERVER_VERSION with OS To: new-httpd@apache.org Date: Mon, 20 Apr 1998 17:08:24 -0400 (EDT) In-Reply-To: from "Dean Gaudet" at Apr 20, 98 01:52:44 pm X-Mailer: ELM [version 2.4 PL25] Content-Type: text Sender: new-httpd-owner@apache.org Precedence: bulk Reply-To: new-httpd@apache.org Dean Gaudet wrote: > > BTW Jim, I'm not adverse to doing extra work of copying strings around at > config time... it's only request time work that really concerns me. > I agree... request-time turnaround is the main factor we should worry about. Init-time is not in the same league, but I think you also agree with we simply don't just shrug our shoulders and say "Ahhh so what, doesn't make any difference" :) If there's any time to sling strings around, it _is_ config time. If we can do it faster/better, then even better. Just in case, I'm happy with Ken's solution. I'm not proposing not using pools. But just because it's config-time doesn't mean that we no longer worry about performance. The fact that using pstrcat allow us to dynamically allocate the size of the string returned as ap_get_server_version is it's main advantage and makes the most sense. -- =========================================================================== Jim Jagielski ||| jim@jaguNET.com ||| http://www.jaguNET.com/ "That's no ordinary rabbit... that's the most foul, cruel and bad-tempered rodent you ever laid eyes on"