Return-Path: Delivered-To: apmail-apr-dev-archive@www.apache.org Received: (qmail 63439 invoked from network); 5 Oct 2007 01:07:50 -0000 Received: from hermes.apache.org (HELO mail.apache.org) (140.211.11.2) by minotaur.apache.org with SMTP; 5 Oct 2007 01:07:50 -0000 Received: (qmail 94867 invoked by uid 500); 5 Oct 2007 01:07:38 -0000 Delivered-To: apmail-apr-dev-archive@apr.apache.org Received: (qmail 94818 invoked by uid 500); 5 Oct 2007 01:07:38 -0000 Mailing-List: contact dev-help@apr.apache.org; run by ezmlm Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Id: Delivered-To: mailing list dev@apr.apache.org Received: (qmail 94807 invoked by uid 99); 5 Oct 2007 01:07:38 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 04 Oct 2007 18:07:38 -0700 X-ASF-Spam-Status: No, hits=1.2 required=10.0 tests=SPF_NEUTRAL X-Spam-Check-By: apache.org Received-SPF: neutral (athena.apache.org: local policy) Received: from [64.202.165.33] (HELO smtpauth11.prod.mesa1.secureserver.net) (64.202.165.33) by apache.org (qpsmtpd/0.29) with SMTP; Fri, 05 Oct 2007 01:07:38 +0000 Received: (qmail 22469 invoked from network); 5 Oct 2007 01:07:17 -0000 Received: from unknown (24.15.193.17) by smtpauth11.prod.mesa1.secureserver.net (64.202.165.33) with ESMTP; 05 Oct 2007 01:07:16 -0000 Message-ID: <47058E44.6010607@rowe-clan.net> Date: Thu, 04 Oct 2007 20:07:16 -0500 From: "William A. Rowe, Jr." User-Agent: Thunderbird 1.5.0.12 (X11/20070719) MIME-Version: 1.0 To: "William A. Rowe, Jr." CC: dev@apr.apache.org Subject: Re: [vote] apr 1.2.x win32 compatibility References: <47058C2F.90304@rowe-clan.net> In-Reply-To: <47058C2F.90304@rowe-clan.net> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Virus-Checked: Checked by ClamAV on apache.org William A. Rowe, Jr. wrote: > I see there are only two conclusions based on Joe's veto, therefore vote > for one; > > [ ] Revert to inheriting INVALID_HANDLE_VALUE on win32 for all unspecified > stdio streams (1.2.8 behavior, retaining all other inheritance bugfixes) > and for the lifespan of 1.2.x remain incompatible with Unix. (In 1.3.0 > we still follow unix behavior of inheriting the parent's stdio streams > for unspecified channels). > > [ ] "Tweak" our versioning rules to permit Windows users to continue to use > the incompatible behavior via a new APR_NO_FILE macro flag in APR 1.2.12 > (this remains ABI-compatible since it's a macro). > > Please vote, I will be closing the debate shortly to roll out 1.2.12 in the > very near future. Essentially option #1 says we can fix platform-incompatible > in a subversion release, option #2 says we don't attempt to solve platform > inconsistencies until the next minor release. Choose your preference. Sorry - flip #1/#2 in the explanation above - Revert means we can't really solve a platform inconsistency until the minor release, while "Tweak" means we can introduce a platform specific option to preserve inconsistent behavior when it's needed by a small minority of our users, while addressing the larger incompatibility for the vast majority of APR applications.