From: Johannes Sixt <j.sixt@viscovery.net>
Date: Wed, 10 Oct 2012 14:21:43 GMT
Subject: Re: [PATCH v2 2/2] grep: stop looking at random places for .gitattributes
Message-ID: <50758477.3030304@viscovery.net>
In-Reply-To: <1349877544-17648-3-git-send-email-pclouds@gmail.com>

Am 10/10/2012 15:59, schrieb Nguyễn Thái Ngọc Duy:
> This patch passes real paths down to grep_source_load_driver(). Except
> grepping a blob, all other cases should have right paths down to

... grepping a blob or tree object...

> grep_source_load_driver(). In other words, .gitattributes are still
> respected.
...
> +test_expect_success 'grep --cached respects binary diff attribute (2)' '
> +	git add .gitattributes &&
> +	rm .gitattributes &&
> +	git grep --cached text t >actual &&
> +	test_cmp expect actual &&
> +	git checkout .gitattributes &&
> +	git rm --cached .gitattributes
> +'

This should perhaps be test_when_finished "git rm --cached .gitattributes".

> +
> +test_expect_success 'grep tree respects binary diff attribute' '

I was confused by the word "tree" here. Isn't "pathspec" more correct?

> +	git commit -m new &&
> +	echo "Binary file HEAD:t matches" >expect &&
> +	git grep text HEAD -- t >actual &&
> +	test_cmp expect actual &&
> +	git reset HEAD^
> +'

And in yet another test, should

	git grep text HEAD:t

/not/ respect the binary attribute?

At any rate, I don't observe the warnings anymore with this series.

Thanks,
-- Hannes