Re: [PATCH] t: make many tests depend less on the refs being files

From: Stefan Beller <sbeller@google.com>
Date: May 21, 2018, 19:34 UTC
Message-ID: <CAGZ79kbdVm-U16rMjtgnGL+TOVRqkZz+J5W_eqo=dh42xRgKBQ@mail.gmail.com>
In-Reply-To: <20180521055143.14701-1-chriscool@tuxfamily.org>

On Sun, May 20, 2018 at 10:51 PM, Christian Couder <christian.couder@gmail.com> wrote:

> From: David Turner <dturner@twopensource.com>
>
> So that they work under alternate ref storage backends.

Sometimes I have a disconnect between the subject and the commit message, (e.g. in an email reader the subject is not displayed accurately on top of the message).

So I would prefer if the first part of the body message is an actual sentence, and not a continuum from the subject.

Maybe elaborate a bit more:

  The current tests are very focused on the file system representation of
  the loose and packed refs code.  As there are plans to implement other
  ref storage systems, migrate most tests to a form that test the intent of the
  refs storage system instead of it internals. The internals of the loose and
  packed refs are tested in <TODO>, whereas the tests in this patch focus
  on testing other aspects.

>
> This will be really needed when such alternate ref storage backends are
> developed. But this could already help by making clear to readers that
> some tests do not depend on which ref backend is used.

Ah, this is what I picked up already in the suggested edit above. :/

Show 5 quoted lines

> This patch just takes care of many low hanging fruits. It does not try
> to completely solves the issue.
>
> Signed-off-by: David Turner <dturner@twopensource.com>
> Signed-off-by: Christian Couder <chriscool@tuxfamily.org>

Show 7 quoted lines

> ---
>
> Thanks for all the great feedback regarding implementing reftable [1].
>
> Looking at David Turner's tests in [2], it seems that they could indeed
> be already valuable, so let's start by extracting most of the simple
> improvements they make.

Thanks for tackling refstables!

Stefan