From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rene Scharfe Subject: Re: [PATCH 4/3] Fold get_merge_bases_clean() into get_merge_bases() Date: Sun, 02 Jul 2006 19:40:45 +0200 Message-ID: <44A8051D.6040605@lsrfire.ath.cx> References: <7vy7vedntn.fsf@assigned-by-dhcp.cox.net> <20060701150926.GA25800@lsrfire.ath.cx> <7vfyhldvd2.fsf@assigned-by-dhcp.cox.net> <44A6CD1D.2000600@lsrfire.ath.cx> <7vveqhccnk.fsf@assigned-by-dhcp.cox.net> <7vpsgpccak.fsf@assigned-by-dhcp.cox.net> <20060702094938.GA10944@lsrfire.ath.cx> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Junio C Hamano , git@vger.kernel.org, Johannes Schindelin X-From: git-owner@vger.kernel.org Sun Jul 02 19:41:12 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 1Fx5wV-0002l9-7C for gcvg-git@gmane.org; Sun, 02 Jul 2006 19:41:11 +0200 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751608AbWGBRlB convert rfc822-to-quoted-printable (ORCPT ); Sun, 2 Jul 2006 13:41:01 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751607AbWGBRlB (ORCPT ); Sun, 2 Jul 2006 13:41:01 -0400 Received: from static-ip-217-172-187-230.inaddr.intergenia.de ([217.172.187.230]:17051 "EHLO neapel230.server4you.de") by vger.kernel.org with ESMTP id S1751621AbWGBRk7 (ORCPT ); Sun, 2 Jul 2006 13:40:59 -0400 Received: from [10.0.1.3] (p508E49B6.dip.t-dialin.net [80.142.73.182]) by neapel230.server4you.de (Postfix) with ESMTP id 8C0E91229C; Sun, 2 Jul 2006 19:40:57 +0200 (CEST) User-Agent: Thunderbird 1.5.0.4 (Windows/20060516) To: Linus Torvalds In-Reply-To: X-Enigmail-Version: 0.94.0.0 Sender: git-owner@vger.kernel.org Precedence: bulk X-Mailing-List: git@vger.kernel.org Archived-At: Linus Torvalds schrieb: >=20 > On Sun, 2 Jul 2006, Rene Scharfe wrote: >> Due to popular request ;-), change get_merge_bases() to be able to >> clean up after itself if needed by adding a cleanup parameter. >=20 > Btw, I think the symmetric thing is still wrong. >=20 > Try this: >=20 > git rev-list ^HEAD~1 HEAD...HEAD~2 >=20 > which _should_ return just HEAD ("HEAD...HEAD~2" should basically exp= and=20 > into "HEAD HEAD~2 ^HEAD~2") >=20 > I haven't actually tested your patch, but I think it returns HEAD and= =20 > HEAD~1. >=20 > The reason? You clear UNINTERESTING as part of clear_commit_marks(), = so=20 > HEAD~1, which was marked thus by the user, gets cleared of its mark a= s=20 > part of the get_merge_bases() invocation. >=20 > I suspect the only way to fix that is to make "get_merge_bases()" not= use=20 > UNINTERESTING at all, but instead just explicitly use something like No and yes. Patch 1 in the 3+1 series changes the flags used in commit.c to not conflict with the ones in revision.h[*]. So we have tw= o different UNINTERESTINGs, and get_merge_bases() doesn't mess up the show/no-show markings. Ren=E9 [*] That fix was the one which reportedly made Dscho break a table. Sorry for that by the way. :-P