Re: [PATCH v2 4/6] Refactor mechanics of testing in a sub test-lib

[Jeff King <peff@peff.net>]

On Wed, Sep 19, 2012 at 06:15:13PM +0100, Adam Spiers wrote:

> This will allow us to test the test framework more thoroughly
> without disrupting the top-level test metrics.

I see this is prep for the next patch, and the parts pulling out the
test-runs into functions make sense. But this hunk confuses me:

> @@ -166,7 +176,7 @@ test_expect_success 'tests clean up even on failures' "
>  	test_must_fail ./failing-cleanup.sh >out 2>err &&
>  	! test -s err &&
>  	! test -f \"trash directory.failing-cleanup/clean-after-failure\" &&
> -	sed -e 's/Z$//' -e 's/^> //' >expect <<-\\EOF &&
> +	sed -e 's/Z$//' -e 's/^> //' >expect <<-EOF &&
>  	> not ok 1 - tests clean up even after a failure
>  	> #	Z
>  	> #	touch clean-after-failure &&

Is it just that you are dropping the '\' in all of the here-docs because
they are not needed? I think our usual style is not to interpolate, and
to do so only when we explicitly want it, which can prevent accidental
errors due to missing quoting.

Also, why is this one not converted into a check_sub... invocation?

-Peff