From: Junio C Hamano <gitster@pobox.com>
Date: Thu, 04 Aug 2011 00:43:45 GMT
Subject: Re: [PATCH v16 6/7] bisect: add tests for the --no-checkout option.
Message-ID: <7voc066m66.fsf@alter.siamese.dyndns.org>
In-Reply-To: <1312408626-8600-7-git-send-email-jon.seymour@gmail.com>

Jon Seymour <jon.seymour@gmail.com> writes:

> @@ -616,6 +628,14 @@ cat > expected.missing-tree.default <<EOF
>  fatal: unable to read tree 39f7e61a724187ab767d2e08442d9b6b9dab587d
>  EOF
>  
> +check_same()
> +{
> +    echo "Checking $1 is the same as $2" &&
> +    git rev-parse "$1" > expected.same &&
> +    git rev-parse "$2" > expected.actual &&
> +    test_cmp expected.same expected.actual

Please indent with tabs; I know some existing lines in this file needs
fixing in a separate patch, but we do not have to make it worse.

> +test_expect_success 'bisect: demonstrate identification of damage boundary' "
> +	git bisect reset &&
> +	git checkout broken &&
> +	git bisect start broken master --no-checkout &&
> +	git bisect run sh -c '
> +		GOOD=$(git for-each-ref "--format=%(objectname)" refs/bisect/good-*) &&

I have a suspicion that the dq around the format=%(...) here does not do
what you seem to think; doesn't it just step outside of the dq context of
the second parameter of the outermost test_expect_success?  Wouldn't your
shell see

	git for-each-ref --format=%(objectname)

without dq and barf on () as a consequence?