git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Ævar Arnfjörð Bjarmason" <avarab@gmail.com>
To: Daniel Johnson <computerdruid@gmail.com>
Cc: Junio C Hamano <gitster@pobox.com>,
	Tay Ray Chuan <rctay89@gmail.com>,
	git@vger.kernel.org
Subject: Re: [RFC/PATCH] t5525: test the tagopt variable and that it can be overridden
Date: Fri, 13 Aug 2010 20:55:35 +0000	[thread overview]
Message-ID: <AANLkTimYm+GmJ4BmZKOmcZkJf_wgeUiKhTs06qxex+0q@mail.gmail.com> (raw)
In-Reply-To: <1281730410-20009-1-git-send-email-ComputerDruid@gmail.com>

On Fri, Aug 13, 2010 at 20:13, Daniel Johnson <computerdruid@gmail.com> wrote:
> ---
>> The current behaviour seems to me a bug introduced while git-fetch was
>> rewritten in C (the original found in contrib/examples reads from the
>> config only when no --tags/--no-tags option is given from the command
>> line).
>>
>> Is this something we can protect with a test script from future breakages?
> This should test that behavior. I'd appreciate feedback on how to improve this
> test. I'm not sure if this is the right name/number either.

Thanks for tackling this.

>  t/t5525-fetch-tagopt.sh |   44 ++++++++++++++++++++++++++++++++++++++++++++
>  1 files changed, 44 insertions(+), 0 deletions(-)
>  create mode 100755 t/t5525-fetch-tagopt.sh
>
> diff --git a/t/t5525-fetch-tagopt.sh b/t/t5525-fetch-tagopt.sh

The test name looks fine.

(The t55* test names are a bit of a mess with mixed pull/fetch, but
that's not something that should be dealt with here)

> new file mode 100755
> index 0000000..17bd407
> --- /dev/null
> +++ b/t/t5525-fetch-tagopt.sh
> @@ -0,0 +1,44 @@
> +
> +#!/bin/sh

Is that an empty line before the test begins? The shebang should be on
the first line.

> +test_description='tagopt variable affects "git fetch" and is overridden by commandline.'
> +
> +. ./test-lib.sh
> +
> +setup_clone () {
> +       (git clone --mirror . $1 &&
> +       git remote add remote_$1 $1 &&
> +       cd $1 &&
> +       git tag tag_$1)
> +}

Maybe only put the "cd $1 ..." inside a subshell for clarity.

> +test_expect_success setup '
> +       echo >file original &&
> +       git add file &&
> +       git commit -a -m original &&

Maybe this can use test_commit if you don't mind it creating a tag
too.

> +       setup_clone one &&
> +       git config remote.remote_one.tagopt --no-tags &&
> +       setup_clone two &&
> +       git config remote.remote_two.tagopt --tags
> +       '
> +
> +test_expect_success "fetch with tagopt=--no-tags does not get tag" '
> +       git fetch remote_one &&
> +       ! (git show-ref tag_one)
> +       '

Doesn't need a subshell? You should also use:

    test_must_fail git show-ref ...

> +test_expect_success "fetch --tags with tagopt=--no-tags gets tag" '
> +       git fetch --tags remote_one &&
> +       (git show-ref tag_one)
> +       '

Doesn't need a subshell?

> +test_expect_success "fetch --no-tags with tagopt=--tags does not get tag" '
> +       git fetch --no-tags remote_two &&
> +       ! (git show-ref tag_two)
> +       '
> +
> +test_expect_success "fetch with tagopt=--tags gets tag" '
> +       git fetch remote_two &&
> +       (git show-ref tag_two)
> +       '
> +test_done

test_must_fail etc etc.

Otherwise it looks good.

  reply	other threads:[~2010-08-13 20:55 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-04 18:56 [PATCH/RFC] fetch: allow command line --tags to override config Daniel Johnson
2010-08-05  9:56 ` Tay Ray Chuan
2010-08-06 13:54   ` [PATCH] Documentation: changes in the behavior of tagopt Daniel Johnson
2010-08-08  2:17     ` Tay Ray Chuan
2010-08-11 22:57       ` [RFC/PATCHv2] fetch: allow command line --tags to override config Daniel Johnson
2010-08-13  1:22         ` Junio C Hamano
2010-08-13 20:13           ` [RFC/PATCH] t5525: test the tagopt variable and that it can be overridden Daniel Johnson
2010-08-13 20:55             ` Ævar Arnfjörð Bjarmason [this message]
2010-08-13 21:27               ` [RFC/PATCHv2] " Daniel Johnson
2010-08-13 21:39                 ` Ævar Arnfjörð Bjarmason
2010-08-14 21:32                   ` Daniel Johnson
2010-08-14 21:37                     ` Ævar Arnfjörð Bjarmason

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=AANLkTimYm+GmJ4BmZKOmcZkJf_wgeUiKhTs06qxex+0q@mail.gmail.com \
    --to=avarab@gmail.com \
    --cc=computerdruid@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=rctay89@gmail.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).