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-ASN: AS53758 23.128.96.0/24 X-Spam-Status: No, score=-4.2 required=3.0 tests=AWL,BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,RCVD_IN_DNSWL_LOW, 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 D3CBF1F5AE for ; Thu, 20 May 2021 17:05:01 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234763AbhETRGU (ORCPT ); Thu, 20 May 2021 13:06:20 -0400 Received: from cloud.peff.net ([104.130.231.41]:60814 "EHLO cloud.peff.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233986AbhETRGT (ORCPT ); Thu, 20 May 2021 13:06:19 -0400 Received: (qmail 28740 invoked by uid 109); 20 May 2021 17:04:58 -0000 Received: from Unknown (HELO peff.net) (10.0.1.2) by cloud.peff.net (qpsmtpd/0.94) with ESMTP; Thu, 20 May 2021 17:04:58 +0000 Authentication-Results: cloud.peff.net; auth=none Received: (qmail 1016 invoked by uid 111); 20 May 2021 17:04:58 -0000 Received: from coredump.intra.peff.net (HELO sigill.intra.peff.net) (10.0.0.2) by peff.net (qpsmtpd/0.94) with (TLS_AES_256_GCM_SHA384 encrypted) ESMTPS; Thu, 20 May 2021 13:04:58 -0400 Authentication-Results: peff.net; auth=none Date: Thu, 20 May 2021 13:04:57 -0400 From: Jeff King To: Patrick Steinhardt Cc: git@vger.kernel.org Subject: Re: [PATCH 2/8] p5400: add perf tests for git-receive-pack(1) Message-ID: References: MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org On Wed, May 19, 2021 at 09:13:27PM +0200, Patrick Steinhardt wrote: > +while read name repo > +do > + refs=("create updated:new") This (and the other array manipulation below) is a bash-ism. Presumably you've got TEST_SHELL_PATH pointed at bash. Without that, on a system where /bin/sh is dash, the script chokes here. For your purposes here, I think you can get by with just a single string with newlines in it. Or even a file (see below). > + while read desc ref > + do > + test_expect_success "setup $name $desc" " > + test_must_fail git push --force '$repo' '$ref' \ > + --receive-pack='tee pack | git receive-pack' 2>err && > + grep 'failed in pre-receive hook' err > + " This inverts the double- and single- quotes from our usual style. So if $repo is "foo", you are creating a string that has: test_must_fail git push --force 'foo' ... in it, and then the test harness will eval that string. That will fail if $repo itself contains a single quote. Pretty unlikely, but I think it contains the absolute path. The usual style is: test_expect_success "setup $name $desc" ' test_must_fail git push --force "$repo" "$ref" \ ...etc... ' where the variables are dereferenced inside the eval'd snippet. So no quoting is necessary, no matter what's in the variables. > + test_perf "receive-pack $name $desc" " > + git receive-pack '$repo' negotiation && > + grep 'pre-receive hook declined' negotiation > + " Likewise here, but note that test_perf is tricky! It runs the snippet in a sub-process. You have to export $repo to make it visible (you can use test_export, but you don't need to; there's some discussing in t/perf/README). > + done < <(printf "%s\n" "${refs[@]}") > +done < <(printf "%s\n" "clone $TARGET_REPO_CLONE" "extrarefs $TARGET_REPO_REFS" "empty $TARGET_REPO_EMPTY") These process substitutions are also bash-isms. I guess you're trying to avoid putting the while on the right-hand side of a pipe like: printf "%s\n" ... | while read ... which is good, because they set variables and those values don't reliably make it out of the pipeline. If you stick the contents of $refs into a file, then you can just do: while read ... do ... done