git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Ævar Arnfjörð Bjarmason" <avarab@gmail.com>
To: Elijah Newren <newren@gmail.com>
Cc: git@vger.kernel.org, gitster@pobox.com, jrnieder@gmail.com
Subject: Re: [PATCHv6 16/16] Introduce portable_unset and use it to ensure proper && chaining
Date: Mon, 4 Oct 2010 03:26:58 +0000	[thread overview]
Message-ID: <AANLkTinU_fv7pJqw_ghM4RoF6dsiV8PFM3GgCqD-hh7L@mail.gmail.com> (raw)
In-Reply-To: <1286136014-7728-17-git-send-email-newren@gmail.com>

On Sun, Oct 3, 2010 at 20:00, Elijah Newren <newren@gmail.com> wrote:
> Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
> Signed-off-by: Elijah Newren <newren@gmail.com>
> ---
>  t/README          |   13 +++++--------
>  t/t0001-init.sh   |   28 ++++++++++++++--------------
>  t/t7006-pager.sh  |   10 +++++-----
>  t/t7502-commit.sh |    4 ++--
>  t/test-lib.sh     |   11 +++++++++++
>  5 files changed, 37 insertions(+), 29 deletions(-)
>
> diff --git a/t/README b/t/README
> index ee4c0cf..e10c0ce 100644
> --- a/t/README
> +++ b/t/README
> @@ -259,14 +259,11 @@ Do:
>        test ...
>
>    That way all of the commands in your tests will succeed or fail. If
> -   you must ignore the return value of something (e.g., the return
> -   after unsetting a variable that was already unset is unportable) it's
> -   best to indicate so explicitly with a semicolon:
> -
> -       unset HLAGH;
> -       git merge hla &&
> -       git push gh &&
> -       test ...
> +   you must ignore the return value of something, consider using a
> +   helper function (e.g. use portable_unset instead of unset, in order
> +   to avoid unportable return value for unsetting a variable that was
> +   already unset), or prepending the command with test_might_fail or
> +   test_must_fail.

This should be:

    diff --git a/t/README b/t/README
    index a1eb7c8..dc4b9d1 100644
    --- a/t/README
    +++ b/t/README
    @@ -259,11 +259,10 @@ Do:
            test ...

        That way all of the commands in your tests will succeed or fail. If
    -   you must ignore the return value of something (e.g., the return
    -   after unsetting a variable that was already unset is unportable) it's
    -   best to indicate so explicitly with a semicolon:
    +   you must ignore the return value of something it's best to indicate
    +   so explicitly with a test_might_fail:

    -       unset HLAGH;
    +       test_might_fail some-command &&
            git merge hla &&
            git push gh &&
            test ...
    @@ -522,6 +521,10 @@ library for your script to use.
                    ...
            '

    + - portable_unset <variable>
    +
    +   .....
    +
     Prerequisites
     -------------

Or something like that. I don't know if we came to an agreement on
using test_might_fail for non-builtins, but the docs on do's & dont's
should mention how to do that, and portable_unset should be documented
in a new section in "Test harness library".

  reply	other threads:[~2010-10-04  3:27 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-10-03 19:59 [PATCHv6 00/16] Add missing &&'s in the testsuite Elijah Newren
2010-10-03 19:59 ` [PATCHv6 01/16] test-lib: make test_expect_code a test command Elijah Newren
2010-10-04  0:54   ` Junio C Hamano
2010-10-04  3:48     ` Ævar Arnfjörð Bjarmason
2010-10-04  3:50       ` Jonathan Nieder
2010-10-04  4:04         ` Ævar Arnfjörð Bjarmason
2010-10-04  4:06           ` Jonathan Nieder
2010-10-04  6:57             ` Ævar Arnfjörð Bjarmason
2010-10-04  9:28               ` Jonathan Nieder
2010-10-04  3:37   ` Ævar Arnfjörð Bjarmason
2010-10-03 20:00 ` [PATCHv6 02/16] t3020 (ls-files-error-unmatch): remove stray '1' from end of file Elijah Newren
2010-10-04  0:56   ` Junio C Hamano
2010-10-03 20:00 ` [PATCHv6 03/16] t4017 (diff-retval): replace manual exit code check with test_expect_code Elijah Newren
2010-10-03 20:00 ` [PATCHv6 04/16] t100[12] (read-tree-m-2way, read_tree_m_u_2way): add missing && Elijah Newren
2010-10-03 20:00 ` [PATCHv6 05/16] t4002 (diff-basic): use test_might_fail for commands that might fail Elijah Newren
2010-10-03 20:00 ` [PATCHv6 06/16] t4202 (log): Replace '<git-command> || :' with test_might_fail Elijah Newren
2010-10-03 20:00 ` [PATCHv6 07/16] t3600 (rm): add lots of missing && Elijah Newren
2010-10-03 20:56   ` Jonathan Nieder
2010-10-03 23:32     ` Junio C Hamano
2010-10-03 20:00 ` [PATCHv6 08/16] t4019 (diff-wserror): " Elijah Newren
2010-10-03 20:00 ` [PATCHv6 09/16] t4026 (color): remove unneeded and unchained command Elijah Newren
2010-10-03 20:00 ` [PATCHv6 10/16] t5602 (clone-remote-exec): add missing && Elijah Newren
2010-10-03 20:00 ` [PATCHv6 11/16] t6016 (rev-list-graph-simplify-history): " Elijah Newren
2010-10-03 20:00 ` [PATCHv6 12/16] t7001 (mv): " Elijah Newren
2010-10-03 20:00 ` [PATCHv6 13/16] t7601 (merge-pull-config): " Elijah Newren
2010-10-03 20:00 ` [PATCHv6 14/16] t7800 (difftool): " Elijah Newren
2010-10-03 20:00 ` [PATCHv6 15/16] Add missing &&'s throughout the testsuite Elijah Newren
2010-10-03 20:59   ` Jonathan Nieder
2010-10-03 21:17   ` Jonathan Nieder
2010-10-31  1:46   ` [PATCH en/cascade-tests] tests: add missing && Jonathan Nieder
2010-10-31  3:31     ` Junio C Hamano
2010-10-31  7:26       ` [PATCH/RFC 00/10] " Jonathan Nieder
2010-10-31  7:30         ` [PATCH 01/10] tests: add missing &&, batch 2 Jonathan Nieder
2010-10-31  7:33         ` [RFC/PATCH 02/10] test-lib: introduce test_line_count to measure files Jonathan Nieder
2010-11-09 22:56           ` Junio C Hamano
2010-11-09 23:09           ` Ævar Arnfjörð Bjarmason
2010-10-31  7:34         ` [PATCH 03/10] t6022 (renaming merge): chain test commands with && Jonathan Nieder
2010-10-31  7:35         ` [PATCH 04/10] t1502 (rev-parse --parseopt): test exit code from "-h" Jonathan Nieder
2010-10-31  7:36         ` [PATCH 05/10] t1400 (update-ref): use test_must_fail Jonathan Nieder
2010-10-31  7:36         ` [PATCH 06/10] t3301 (notes): use test_expect_code for clarity Jonathan Nieder
2010-10-31  7:38         ` [PATCH 07/10] t3404 (rebase -i): unroll test_commit loops Jonathan Nieder
2010-10-31  7:39         ` [PATCH 08/10] t3404 (rebase -i): move comment to description Jonathan Nieder
2010-11-17 23:12           ` Junio C Hamano
2010-10-31  7:40         ` [PATCH 09/10] t3404 (rebase -i): introduce helper to check position of HEAD Jonathan Nieder
2010-11-17 17:55           ` Junio C Hamano
2010-10-31  7:41         ` [PATCH 10/10] t4124 (apply --whitespace): use test_might_fail Jonathan Nieder
2010-11-09 22:25       ` [PATCH en/cascade-tests] tests: add missing && Junio C Hamano
2010-11-05  4:57     ` Elijah Newren
2010-10-03 20:00 ` [PATCHv6 16/16] Introduce portable_unset and use it to ensure proper && chaining Elijah Newren
2010-10-04  3:26   ` Ævar Arnfjörð Bjarmason [this message]
2010-10-04  4:44     ` Jonathan Nieder
2010-10-04  7:52       ` Junio C Hamano
2010-10-04 11:38         ` Jonathan Nieder
2010-10-03 21:09 ` [PATCHv6 00/16] Add missing &&'s in the testsuite Jonathan Nieder
2010-10-04  3:11 ` Ævar Arnfjörð Bjarmason
2010-10-04  3:44   ` Jonathan Nieder
2010-10-04  3:50     ` Ævar Arnfjörð Bjarmason
2010-10-04 12:28 ` yj2133011
2010-10-06  5:31 ` [TOY PATCH] test-lib: &&-chaining tester Jonathan Nieder
2010-10-06  8:09   ` Matthieu Moy
2010-10-06  8:25     ` Johannes Sixt
2010-10-06  8:52   ` Sverre Rabbelier

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=AANLkTinU_fv7pJqw_ghM4RoF6dsiV8PFM3GgCqD-hh7L@mail.gmail.com \
    --to=avarab@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=jrnieder@gmail.com \
    --cc=newren@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).