From mboxrd@z Thu Jan 1 00:00:00 1970 From: Luben Tuikov Subject: Re: [PATCH] gitweb.cgi: Customization Date: Wed, 2 Aug 2006 13:30:03 -0700 (PDT) Message-ID: <20060802203003.61929.qmail@web31802.mail.mud.yahoo.com> References: <1154535801.19994.15.camel@cashmere.sps.mot.com> Reply-To: ltuikov@yahoo.com Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Cc: Git List , Martin Waitz , Matthias Lederhofer X-From: git-owner@vger.kernel.org Wed Aug 02 22:30:50 2006 Return-path: Envelope-to: gcvg-git@gmane.org Received: from vger.kernel.org ([209.132.176.167]) by ciao.gmane.org with esmtp (Exim 4.43) id 1G8NM5-0006DP-Dl for gcvg-git@gmane.org; Wed, 02 Aug 2006 22:30:14 +0200 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932226AbWHBUaG (ORCPT ); Wed, 2 Aug 2006 16:30:06 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932228AbWHBUaF (ORCPT ); Wed, 2 Aug 2006 16:30:05 -0400 Received: from web31802.mail.mud.yahoo.com ([68.142.207.65]:51613 "HELO web31802.mail.mud.yahoo.com") by vger.kernel.org with SMTP id S932226AbWHBUaE (ORCPT ); Wed, 2 Aug 2006 16:30:04 -0400 Received: (qmail 61931 invoked by uid 60001); 2 Aug 2006 20:30:03 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:Received:Date:From:Reply-To:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=ViFbmafpGB0Gdu6SpiRcVUZgQ2HEqCZ3R1oB556csnR+JoXAXCiDdt8Ox/5BgKxuj1ww0PTJiiw4Gm2Sh04H7kRa8nEtA2/+dc+LRXb6aKoUzIIIfnYQhdKpFwnvUeiUaJ93/uxKRJkogK+GPuWUlruYwf/wAKpC4gIIoJi5zAk= ; Received: from [64.215.88.90] by web31802.mail.mud.yahoo.com via HTTP; Wed, 02 Aug 2006 13:30:03 PDT To: Jon Loeliger , Junio C Hamano In-Reply-To: <1154535801.19994.15.camel@cashmere.sps.mot.com> Sender: git-owner@vger.kernel.org Precedence: bulk X-Mailing-List: git@vger.kernel.org Archived-At: --- Jon Loeliger wrote: > On Tue, 2006-08-01 at 18:54, Junio C Hamano wrote: > > > In any case, I think tweaking gitweb.cgi from Makefile like > > Martin Waitz did is as easy and clean for people who want to > > customize; it should just be the matter of defining the > > necessary params in config.mak. > > I disagree. I run multiple virtual web servers on one > physical machine. Several of them run different gitweb > instances, each with different configurations. > > With this "params in config.mk" approach, I have to > run it multiple times, once for each web server I run. That's exactly my sentiments. > I _really_ would prefer an "include from ." feature > where I can place the specific gitweb_config.pm parts > in the same directory where gitweb.{pl,cgi} is installed. Yes, this approach is simple and straightforward enough. When upgrading, one only needs to do a "cp". > We really need to separate out these config values > from the gitweb.{pl,cgi} script itself. We _need_ to > be able to update the gitweb script independently, > and easily. That's exactly where I was going with this patch. I don't want to have to: - configure httpd, or - configure the environment of user "apache", or - have a separate forked off branch in order to "configure" gitweb at compile time for each gitweb instance I'm running, - etc, etc, etc. Configuration, especially of user space programs should be completely dynamic. Compilation for user space programs should be a separate process from configuration. > > I do not think there is much difference between any of the > > customization proposed so far (yours, Martin's and the one from > > Matthias Lederhofer) from functionality and ease-of-use point of > > view. They all try to make customization can be done in one > > place, and the difference is mostly of taste, so I'd just pick > > one from Martin. > > Let's just make sure it is a separate config file, please. I thing your and mine deployment are pretty similar, thus warranting a more flexible approach, withouth having to recompile or reconfigure completely unrelated programs and users. Luben