From: SZEDER Gábor <szeder@ira.uka.de>
Date: Sun, 17 Apr 2016 09:42:53 GMT
Subject: Re: [PATCH v2 2/2] t1500-rev-parse: rewrite each test to run in isolation
Message-ID: <20160417114253.Horde.giIo57RkUzhAe6GP-RahIrw@webmail.informatik.kit.edu>
In-Reply-To: <1460823230-45692-3-git-send-email-rappazzo@gmail.com>


Quoting Michael Rappazzo <rappazzo@gmail.com>:

> +test_expect_success 'GIT_DIR=../.git, core.bare = false:  
> is-bare-repository' '
> +	mkdir work &&
> +	test_when_finished "rm -rf work" &&
> +	test_config -C "$(pwd)"/.git core.bare false &&
> +	GIT_DIR=../.git test_stdout false git -C work rev-parse  
> --is-bare-repository
> +'

Here and in the following tests as well: some shells don't cope that well
with a one-shot environmental variable set in front of a shell function.
See commit 512477b17528:

     tests: use "env" to run commands with temporary env-var settings

     Ordinarily, we would say "VAR=VAL command" to execute a tested
     command with environment variable(s) set only for that command.
     This however does not work if 'command' is a shell function (most
     notably 'test_must_fail'); the result of the assignment is retained
     and affects later commands.

     To avoid this, we used to assign and export environment variables
     and run such a test in a subshell, like so:

             (
                     VAR=VAL && export VAR &&
                     test_must_fail git command to be tested
             )

     But with "env" utility, we should be able to say:

             test_must_fail env VAR=VAL git command to be tested

     which is much shorter and easier to read.