From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on dcvr.yhbt.net X-Spam-Level: X-Spam-Status: No, score=-3.6 required=3.0 tests=AWL,BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SPF_HELO_PASS, SPF_PASS shortcircuit=no autolearn=ham autolearn_force=no version=3.4.2 Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by dcvr.yhbt.net (Postfix) with ESMTP id 500A41F66E for ; Wed, 19 Aug 2020 07:18:38 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726702AbgHSHSf (ORCPT ); Wed, 19 Aug 2020 03:18:35 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:40036 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726735AbgHSHSb (ORCPT ); Wed, 19 Aug 2020 03:18:31 -0400 Received: from mail-il1-x135.google.com (mail-il1-x135.google.com [IPv6:2607:f8b0:4864:20::135]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 0182FC061389 for ; Wed, 19 Aug 2020 00:18:31 -0700 (PDT) Received: by mail-il1-x135.google.com with SMTP id t4so19839112iln.1 for ; Wed, 19 Aug 2020 00:18:30 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=Q01TrnIuItPagqZIwECtj6t0GQEUr/aKj/jthKN1Bik=; b=HeqtVzRLQltsrLkBwaOt93qN4s4GSz4e3l8fy762ZzkhWHsAYyPax6ZKdFQiNod24k xVLbp5/mXnvv4S3yArB+9KCqiQ8rN1p1dwxhSIutbyGObnihE0IjSb6PDLAakUT5W8el weI+HnPRwJr+eQsNc575rnhyQHh/9sPs042FEmtVoU3791EYqnjyvu27Bs3nvEeubUQ1 r/RhNlVlE/pI9EmntuPdHKmrjrqgDOOk+nKELva+rFwc6sWy84/aQK9lcUuGv/7+vUud 43IELrpT3sJ1OMJnCGSfQLDUyRD78aSXmx4hpeWoUtRByhzF9xw2QDWcfUOFsUUwgjgQ sDnQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=Q01TrnIuItPagqZIwECtj6t0GQEUr/aKj/jthKN1Bik=; b=fHPREquU+yw8mv40oeNfX6IRkgdbu1NUb08zMfXexxVFRjKE1V2mXEjUSiuMzr/yBd j1Ix8bV2kjmx5RFLAwDxNiRHgtgX08aiAbIutzxQo8FCvLicTzTRjX7mMPKUT2oMYOv/ HUzkmNtwWnsBHXzoZiFpUzvaj5N75IuTCn7azzFzvCBvpU9dWJKGz1oJvfI3zDc4BmE6 bA61eZky5BAqWU8pS+S28wQ8JNEcUxDMKPMpgBR/HphvrxrSC0QB26nJNViyIy1NT2rp y4CHHQnaWLrCv2fLLPtNuBX2xQS2q7F6XNvdyjBauQpOPXsbv5mm48GBC1TeauqOc4V9 KX2w== X-Gm-Message-State: AOAM532CskfRCspNquzlLKCz/ZSeasanlCbrhNVn/i9iItACfECSHlXt rXszX+ypTq6sr07SMlzWjNBfdQxq3dV4tK5HM+Zkc+V00Ss= X-Google-Smtp-Source: ABdhPJxyJspQGeMweJ2R6MamUMLQfe+UGij1jjBbav6QnPiYRrkgTT8aMxeOzGAJvadJ880Ao+Ed59wDvx+O1PCED88= X-Received: by 2002:a05:6e02:14cf:: with SMTP id o15mr21666475ilk.239.1597821508960; Wed, 19 Aug 2020 00:18:28 -0700 (PDT) MIME-Version: 1.0 References: In-Reply-To: From: Gopal Yadav Date: Wed, 19 Aug 2020 12:48:17 +0530 Message-ID: Subject: Re: [NEW] Git To: git@vger.kernel.org Cc: Johannes.Schindelin@gmx.de, sunshine@sunshineco.com Content-Type: text/plain; charset="UTF-8" Sender: git-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org On Tue, Aug 18, 2020 at 10:22 PM Eric Sunshine wrote: > > On Tue, Aug 18, 2020 at 8:23 AM Gopal Yadav wrote: > No. test_lazy_prereq() is merely the function which defines how a lazy > prerequisites should be determined when the answer about the > prerequisite is actually needed. What #353 is saying is to not perform > the actual determination if a test which requires it isn't going to be > run. > > Fixing the issue might be as simple as moving the test_verify_prereq() > call (and related "export") inside the 'if ! test_skip "$@"' > conditional in the test_expect_success() and test_expect_failure() > functions. I see that test_verify_prereq() (and related export) being used inside 3 different functions namely test_expect_failure(), test_expect_success() and test_external(). So the solution here is to move these into the if ! test_skip block. Thanks, will submit a patch. > > I know the issue talks about something related to chains but I am > > thinking of starting by resolving the lazy prereq task first. > > I took a look at the &&-chain logic and, as far as I can tell by both > direct inspection and by experimentation, the detection of broken > &&-chains is _not_ performed for tests which are being skipped. So, I > think that portion of #353 is just wrong.