git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: Patrick Steinhardt <ps@pks.im>
Cc: git@vger.kernel.org, "Jean-Rémy Falleri" <jr.falleri@gmail.com>,
	"David Aguilar" <davvid@gmail.com>
Subject: Re: [PATCH] git-difftool--helper: honor `--trust-exit-code` with `--dir-diff`
Date: Fri, 16 Feb 2024 10:12:32 -0800	[thread overview]
Message-ID: <xmqqttm8i8hb.fsf@gitster.g> (raw)
In-Reply-To: <c071e44c52171b9b19a04d91666be48d762d19bf.1708072576.git.ps@pks.im> (Patrick Steinhardt's message of "Fri, 16 Feb 2024 09:39:18 +0100")

Patrick Steinhardt <ps@pks.im> writes:

> The `--trust-exit-code` option for git-diff-tool(1) was introduced via
> 2b52123fcf (difftool: add support for --trust-exit-code, 2014-10-26).
> When set, it makes us return the exit code of the invoked diff tool when
> diffing multiple files. This patch didn't change the code path where
> `--dir-diff` was passed because we already returned the exit code of the
> diff tool unconditionally in that case.
>
> This was changed a month later via c41d3fedd8 (difftool--helper: add
> explicit exit statement, 2014-11-20), where an explicit `exit 0` was
> added to the end of git-difftool--helper.sh. While the stated intent of
> that commit was merely a cleanup, it had the consequence that we now
> to ignore the exit code of the diff tool when `--dir-diff` was set. This
> change in behaviour is thus very likely an unintended side effect of
> this patch.
>
> Now there are two ways to fix this:
>
>   - We can either restore the original behaviour, which unconditionally
>     returned the exit code of the diffing tool when `--dir-diff` is
>     passed.
>
>   - Or we can make the `--dir-diff` case respect the `--trust-exit-code`
>     flag.
>
> The fact that we have been ignoring exit codes for 7 years by now makes
> me rather lean towards the latter option. Furthermore, respecting the
> flag in one case but not the other would needlessly make the user
> interface more complex.
>
> Fix the bug so that we also honor `--trust-exit-code` for dir diffs and
> adjust the documentation accordingly.
>
> Reported-by: Jean-Rémy Falleri <jr.falleri@gmail.com>
> Signed-off-by: Patrick Steinhardt <ps@pks.im>
> ---

Sounds sensible.

The last time David was on list seems to be in April 2023; just in 
case let's CC him for an Ack (or something else).

>  Documentation/git-difftool.txt |  1 -
>  git-difftool--helper.sh        | 14 +++++
>  t/t7800-difftool.sh            | 99 ++++++++++++++++++----------------
>  3 files changed, 68 insertions(+), 46 deletions(-)
>
> diff --git a/Documentation/git-difftool.txt b/Documentation/git-difftool.txt
> index c05f97aca9..a616f8b2e6 100644
> --- a/Documentation/git-difftool.txt
> +++ b/Documentation/git-difftool.txt
> @@ -105,7 +105,6 @@ instead.  `--no-symlinks` is the default on Windows.
>  	`merge.tool` until a tool is found.
>  
>  --[no-]trust-exit-code::
> -	'git-difftool' invokes a diff tool individually on each file.
>  	Errors reported by the diff tool are ignored by default.
>  	Use `--trust-exit-code` to make 'git-difftool' exit when an
>  	invoked diff tool returns a non-zero exit code.
> diff --git a/git-difftool--helper.sh b/git-difftool--helper.sh
> index e4e820e680..09d8542917 100755
> --- a/git-difftool--helper.sh
> +++ b/git-difftool--helper.sh
> @@ -91,6 +91,20 @@ then
>  	# ignore the error from the above --- run_merge_tool
>  	# will diagnose unusable tool by itself
>  	run_merge_tool "$merge_tool" false
> +
> +	status=$?
> +	if test $status -ge 126
> +	then
> +		# Command not found (127), not executable (126) or
> +		# exited via a signal (>= 128).
> +		exit $status
> +	fi

So these errors spawning the tool backend are always reported,
regardless of the trust-exit-code settings.  OK.

> +	if test "$status" != 0 &&
> +		test "$GIT_DIFFTOOL_TRUST_EXIT_CODE" = true
> +	then
> +		exit $status
> +	fi

I found this somehow harder to reason about than necessary.  Just

	if test "$GIT_DIFFTOOL_TRUST_EXIT_CODE" = true
	then
		exit $status
	fi

would have been a more straight-forward expression of what we want
to happen here, i.e. "if we are told to report the tool's exit
status, we do so, regardless of what the exit status is".

Not that the construct in your patch is wrong---we will exit with 0
at the end even when "trust-exit-code" thing is true and the tool
returned success.

  reply	other threads:[~2024-02-16 18:12 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-02-15 11:09 Git difftool: interaction between --dir-diff and --trust-exit-code Jean-Rémy Falleri
2024-02-16  8:39 ` [PATCH] git-difftool--helper: honor `--trust-exit-code` with `--dir-diff` Patrick Steinhardt
2024-02-16 18:12   ` Junio C Hamano [this message]
2024-02-20  9:56     ` Patrick Steinhardt
2024-02-28  1:52     ` David Aguilar
2024-02-28  2:15       ` Junio C Hamano
2024-02-20 10:08 ` [PATCH v2] " Patrick Steinhardt
2024-03-08 22:12   ` SZEDER Gábor
2024-03-08 22:36     ` Junio C Hamano
2024-03-21 13:50       ` Patrick Steinhardt

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=xmqqttm8i8hb.fsf@gitster.g \
    --to=gitster@pobox.com \
    --cc=davvid@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=jr.falleri@gmail.com \
    --cc=ps@pks.im \
    /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).