git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jonathan Nieder <jrnieder@gmail.com>
To: Matthieu Moy <Matthieu.Moy@imag.fr>
Cc: git@vger.kernel.org, gitster@pobox.com
Subject: Re: [PATCH v2] test-lib: user-friendly alternatives to test [!] [-d|-f]
Date: Fri, 6 Aug 2010 17:57:05 -0500	[thread overview]
Message-ID: <20100806225705.GA2534@burratino> (raw)
In-Reply-To: <1281027831-22739-1-git-send-email-Matthieu.Moy@imag.fr>

Matthieu Moy wrote:

> The helper functions are implemented, documented, and used in a few
> places to validate them

When I first read this, I thought you were saying these helpers
already existed.  This is where the rationale goes, anyway, so maybe:

	Add new test_file_must_not_exist et al helpers for
	use by tests to more loudly diagnose failures that
	manifest themselves by the existence or nonexistence
	of a file or directory.

	So now you can use

		test_file_must_exist foo "so there"

	from your test, and when it fails due to foo being
	absent or being a symlink instead, instead of silence
	you will get (if debugging with "-v") the helpful message

		file foo does not exist. so there.

> +++ b/t/README
> @@ -467,6 +467,14 @@ library for your script to use.
>     <expected> file.  This behaves like "cmp" but produces more
>     helpful output when the test is run with "-v" option.
>  
> + - test_file_must_exist <file> [<diagnosis>]
> +   test_file_must_not_exist <file> [<diagnosis>]
> +   test_dir_must_exist <dir> [<diagnosis>]
> +   test_dir_must_not_exist <dir> [<diagnosis>]
> +
> +   check whether a file/directory exists or doesn't. <diagnosis> will
> +   be displayed if the test fails.

Maybe:

	- test_file_exists <name> [<diagnosis>]
	- test_dir_exists <name> [<diagnosis>]

	  Check that <name> exists and is a file or directory,
	  printing a diagnostic if it does not.  The <diagnosis>
	  if present will be used to give some added context to
	  the diagnostic.

	- test_does_not_exist <name> [<diagnosis>]

	  Check that <name> does not exist, printing a
	  diagnostic if it does.  The <diagnosis> will be
	  printed on failure as added context if present.

I think the ..._must_exist names put the emphasis in the
wrong place, and they look funny in "if" statements.

> +++ b/t/t3404-rebase-interactive.sh
> +++ b/t/t3407-rebase-abort.sh
[examples]

Makes sense.

> +++ b/t/test-lib.sh
> @@ -541,6 +541,38 @@ test_external_without_stderr () {
>  	fi
>  }
>  
> +# debugging-friendly alternatives to "test [!] [-f|-d]"
> +# The commands test the existence or non-existance of $1. $2 can be
> +# given to provide a more precise diagnosis.
> +test_file_must_exist () {
> +	if ! [ -f "$1" ]; then
> +		echo "file $1 doesn't exist. $*"
> +		false
> +	fi
> +}

Style nitpick: if statementss in the test-lib have tended to look like

 if [ foo ]
 then
	bar
 fi

so far.  Here the whole function is a glorified "test -f", so I wonder
if

	[ -f "$1" ] ||
	{
		echo >&2 "file $1 doesn't exist. $*"
		false
	}

would not be clearer.  I dunno.

> +test_file_must_not_exist () {
> +	if [ -f "$1" ]; then
> +		echo "file $1 exists. $*"
> +		false
> +	fi
> +}

What should happen if $1 exists and is not a file?

I have often run into silent test failures of the sort your patch
is designed to avoid.  Thanks for tackling it.

  reply	other threads:[~2010-08-06 22:58 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-05 13:00 [PATCH] rebase -i: add exec command to launch a shell command Matthieu Moy
2010-08-05 13:31 ` Ævar Arnfjörð Bjarmason
2010-08-05 16:47   ` Matthieu Moy
2010-08-05 16:54     ` [PATCH 1/2] " Matthieu Moy
2010-08-05 16:54     ` [PATCH 2/2] test-lib: user-friendly alternatives to test [!] [-d|-f] Matthieu Moy
2010-08-05 17:03       ` [PATCH v2] " Matthieu Moy
2010-08-06 22:57         ` Jonathan Nieder [this message]
2010-08-07  0:21           ` Ævar Arnfjörð Bjarmason
2010-08-09 16:25       ` [PATCH 2/2] " Junio C Hamano
2010-08-10 11:00         ` [PATCH] test-lib: user-friendly alternatives to test [-d|-f|-e] Matthieu Moy
2010-08-10 11:11           ` Joshua Juran
2010-08-10 11:18             ` [PATCH v3] " Matthieu Moy
2010-08-05 18:24     ` [PATCH] rebase -i: add exec command to launch a shell command Erik Faye-Lund
2010-08-05 18:37     ` Jacob Helwig
2010-08-05 20:16       ` Junio C Hamano

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20100806225705.GA2534@burratino \
    --to=jrnieder@gmail.com \
    --cc=Matthieu.Moy@imag.fr \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).