git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Phillip Wood <phillip.wood123@gmail.com>
To: Patrick Steinhardt <ps@pks.im>,
	"D. Ben Knoble" <ben.knoble+github@gmail.com>
Cc: git@vger.kernel.org,
	Johannes Schindelin <johannes.schindelin@gmx.de>,
	Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH 4/4] drop git_exec_path() from non-Git commands' PATH
Date: Wed, 21 May 2025 14:07:25 +0100	[thread overview]
Message-ID: <01301fa6-eea7-427d-a61f-5cb8dac99afd@gmail.com> (raw)
In-Reply-To: <aC2OaeLYJQAOE_S1@pks.im>

On 21/05/2025 09:27, Patrick Steinhardt wrote:
> On Tue, May 20, 2025 at 03:34:58PM -0400, D. Ben Knoble wrote:
>> We setup_path() with git_exec_path() unconditionally (8e3462837b (Modify
>> setup_path() to only add git_exec_path() to PATH, 2009-01-18)) when Git
>> starts; as a result, all child processes see Git's exec-path when run,
>> including editors and other programs that don't need it [1]. This can
>> cause confusion for such programs or shells, especially when they rely
>> on finding "git" in PATH to locate other nearby directories,

I'm skeptical that using the location of the git executable to find 
"git-jump" and "git-completion.zsh" is a good idea. The location of 
those (and whether they are packaged at all) is entirely at the 
discretion of whoever packages git - they are not installed by the 
Makefile and so there is no fixed relationship between their location in 
the filesystem and the git executable. For instance the example in Ben's 
blog post looks for "git-jump" at 
"$prefix/share/git-core/contrib/git-jump" but on my system it is located 
at "$prefix/share/git/contrib/git-jump"

> I agree with what Junio mentioned in a parallel thread, especially
> around Git hooks. The expectation there is that those may execute other
> Git commands, and that should typically be using the same execution
> environment as the original Git command that has been invoking the hook.
> So refining this patch so that the mechanism is opt-in probably makes
> sense.
> 
>      A slight tangent: I wonder whether it is even required nowadays to
>      adapt PATH at all anymore. As far as I understand this was a
>      requirement back when people still executed dashed binaries
>      directly. But nowadays scripts don't really do that anymore, but
>      instead use the git binary. And that one doesn't need PATH to be
>      adapted at all, as it knows to listen to GIT_EXEC_PATH and its
>      built-in path anyway.

But don't we still need to change PATH so that hooks, shell aliases, git 
rebase --exec, git bisect run, etc. still run the same git executable 
that started them? For example "/usr/bin/git -c alias.g=!git g 
--version" should report the version of /usr/bin/git, not 
~/.local/bin/git which comes first in my PATH if git doesn't change it.

Best Wishes

Phillip


  reply	other threads:[~2025-05-21 13:07 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-20 19:34 [PATCH 0/4] Drop git-exec-path from non-Git child programs D. Ben Knoble
2025-05-20 19:34 ` [PATCH 1/4] t7005: sanitize test environment for subsequent tests D. Ben Knoble
2025-05-21  7:56   ` Patrick Steinhardt
2025-05-21 13:23     ` D. Ben Knoble
2025-05-20 19:34 ` [PATCH 2/4] editor: use standard strvec API to receive environment for external editors D. Ben Knoble
2025-05-21  7:56   ` Patrick Steinhardt
2025-05-21 13:26     ` D. Ben Knoble
2025-05-21 13:34       ` Patrick Steinhardt
2025-05-21 15:20       ` Junio C Hamano
2025-05-21 15:10     ` Junio C Hamano
2025-05-20 19:34 ` [PATCH 3/4] run-command: prep_childenv on all platforms D. Ben Knoble
2025-05-21  7:56   ` Patrick Steinhardt
2025-05-21 13:07     ` Phillip Wood
2025-05-21 13:33       ` D. Ben Knoble
2025-05-20 19:34 ` [PATCH 4/4] drop git_exec_path() from non-Git commands' PATH D. Ben Knoble
2025-05-20 20:33   ` Junio C Hamano
2025-05-21 13:44     ` D. Ben Knoble
2025-05-21  8:27   ` Patrick Steinhardt
2025-05-21 13:07     ` Phillip Wood [this message]
2025-05-21 13:17       ` Patrick Steinhardt
2025-05-21 15:27         ` Phillip Wood
2025-05-26  6:34           ` Patrick Steinhardt
2025-05-21 15:50         ` Justin Tobler
2025-05-26  6:31           ` Patrick Steinhardt
2025-05-21 13:48     ` D. Ben Knoble
2025-05-21 14:47     ` Junio C Hamano
2025-05-22 13:21 ` [PATCH 0/4] Drop git-exec-path from non-Git child programs Phillip Wood
2025-08-05  1:41   ` D. Ben Knoble
2025-08-05  2:40 ` [PATCH 0/2] clean up some code around editors D. Ben Knoble
2025-08-06 10:07   ` Phillip Wood
2025-08-05  2:40 ` [PATCH 1/2] t7005: sanitize test environment for subsequent tests D. Ben Knoble
2025-08-05  2:40 ` [PATCH 2/2] editor: use standard strvec API to receive environment for external editors D. Ben Knoble
2025-08-10 16:03 ` [PATCH 0/3] clean up some code around editors D. Ben Knoble
2025-08-10 16:06   ` D. Ben Knoble
2025-08-10 16:34     ` Ben Knoble
2025-08-11 22:16   ` [PATCH v3 0/4] " D. Ben Knoble
2025-08-11 22:59     ` Ben Knoble
2025-08-12  0:16       ` Eric Sunshine
2025-08-12 16:42         ` D. Ben Knoble
2025-08-12 17:35         ` Junio C Hamano
2025-08-12 18:13           ` Eric Sunshine
2025-08-12 18:22             ` Junio C Hamano
2025-08-12 18:30               ` Eric Sunshine
2025-08-12 17:02     ` [PATCH v4 0/3] " D. Ben Knoble
2025-08-13 10:14       ` Phillip Wood
2025-08-13 15:41         ` Junio C Hamano
2025-08-13 17:36           ` D. Ben Knoble
2025-08-13 17:50       ` [PATCH v5 " D. Ben Knoble
2025-08-15 16:00         ` Phillip Wood
2025-08-13 17:50       ` [PATCH v5 1/3] t7005: use modern test style D. Ben Knoble
2025-08-13 17:50       ` [PATCH v5 2/3] t7005: stop abusing --exec-path D. Ben Knoble
2025-08-13 17:50       ` [PATCH v5 3/3] t7005: sanitize test environment for subsequent tests D. Ben Knoble
2025-08-12 17:02     ` [PATCH v4 1/3] t7005: use modern test style D. Ben Knoble
2025-08-12 17:02     ` [PATCH v4 2/3] t7005: stop abusing --exec-path D. Ben Knoble
2025-08-12 17:02     ` [PATCH v4 3/3] t7005: sanitize test environment for subsequent tests D. Ben Knoble
2025-08-11 22:16   ` [PATCH v3 1/4] t7005: use modern test style D. Ben Knoble
2025-08-11 22:16   ` [PATCH v3 2/4] t7005: stop abusing --exec-path D. Ben Knoble
2025-08-11 22:16   ` [PATCH v3 3/4] t7005: sanitize test environment for subsequent tests D. Ben Knoble
2025-08-11 22:34     ` Eric Sunshine
2025-08-12 16:40       ` D. Ben Knoble
2025-08-12 17:13         ` Eric Sunshine
2025-08-11 22:16   ` [PATCH v3 4/4] editor: use standard strvec API to receive environment for external editors D. Ben Knoble
2025-08-11 22:46     ` Eric Sunshine
2025-08-11 23:58       ` Junio C Hamano
2025-08-10 16:03 ` [PATCH 1/3] t7005: use modern test style D. Ben Knoble
2025-08-10 19:42   ` Phillip Wood
2025-08-11 10:04   ` Phillip Wood
2025-08-10 16:03 ` [PATCH 2/3] t7005: sanitize test environment for subsequent tests D. Ben Knoble
2025-08-10 19:44   ` Phillip Wood
2025-08-10 19:53     ` Ben Knoble
2025-08-10 20:58       ` Eric Sunshine
2025-08-11  9:59     ` Phillip Wood
2025-08-10 16:03 ` [PATCH 3/3] editor: use standard strvec API to receive environment for external editors D. Ben Knoble

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=01301fa6-eea7-427d-a61f-5cb8dac99afd@gmail.com \
    --to=phillip.wood123@gmail.com \
    --cc=ben.knoble+github@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=johannes.schindelin@gmx.de \
    --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).