From mboxrd@z Thu Jan 1 00:00:00 1970 From: Petr Baudis Subject: Re: Understanding git.git's branch policy Date: Mon, 25 Jan 2010 01:17:32 +0100 Message-ID: <20100125001732.GD9553@machine.or.cz> References: <83pr4znklo.fsf@torus.sehlabs.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: git@vger.kernel.org To: "Steven E. Harris" X-From: git-owner@vger.kernel.org Mon Jan 25 01:17:43 2010 Return-path: Envelope-to: gcvg-git-2@lo.gmane.org Received: from vger.kernel.org ([209.132.180.67]) by lo.gmane.org with esmtp (Exim 4.50) id 1NZCeE-0007Cc-KO for gcvg-git-2@lo.gmane.org; Mon, 25 Jan 2010 01:17:42 +0100 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753759Ab0AYARg convert rfc822-to-quoted-printable (ORCPT ); Sun, 24 Jan 2010 19:17:36 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751441Ab0AYARg (ORCPT ); Sun, 24 Jan 2010 19:17:36 -0500 Received: from w241.dkm.cz ([62.24.88.241]:55722 "EHLO machine.or.cz" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751181Ab0AYARf (ORCPT ); Sun, 24 Jan 2010 19:17:35 -0500 Received: by machine.or.cz (Postfix, from userid 2001) id D56BF125A0E9; Mon, 25 Jan 2010 01:17:32 +0100 (CET) Content-Disposition: inline In-Reply-To: <83pr4znklo.fsf@torus.sehlabs.com> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: git-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org Archived-At: On Sun, Jan 24, 2010 at 06:34:11PM -0500, Steven E. Harris wrote: > Radding the maintain-git.txt document=C2=B9, there are a few points t= hat I'm > having trouble decoding. Under "The Policy", it notes >=20 > ,---- > | The tips of 'master', 'maint' and 'next' branches will always > | fast-forward, to allow people to build their own customization on t= op > | of them. > `---- >=20 > I understand that a "fast-forward merge" means that one's current HEA= D > commit is an ancestor of the evolved branch's head, so that the HEAD > pointer can move forward to "catch up" without needing to combine > disparate content. >=20 > How does this relate to the prescribed use of the "master", "maint", = and > "next" branches? What operations or patterns does it constrain agains= t? Rebases or other jumps. New tip of 'master' will always be descendant o= f old tip of 'master', never a commit from a parallel commit line. This i= s preserved over commits and merges, but not over operations that rewrite history - rebase, filter-branch and such. The term "fast-forward" is used commonly in this sense. E.g. git push will typically deny you to push out a branch that is not fastforwarding the currently pushed out branch, unless you force it to. --=20 Petr "Pasky" Baudis A lot of people have my books on their bookshelves. That's the problem, they need to read them. -- Don Knuth