git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Eric Sunshine <sunshine@sunshineco.com>
To: git@vger.kernel.org
Cc: Lars Schneider <larsxschneider@gmail.com>,
	matthew.k.gumbel@intel.com,
	Eric Sunshine <sunshine@sunshineco.com>
Subject: [PATCH 0/2] worktree: change to new worktree dir before running hook(s)
Date: Sun, 11 Feb 2018 22:15:24 -0500	[thread overview]
Message-ID: <20180212031526.40039-1-sunshine@sunshineco.com> (raw)
In-Reply-To: <20180210010132.33629-1-lars.schneider@autodesk.com>

This patch series replaces "worktree: set worktree environment in
post-checkout hook"[1] from Lars, which is a proposed bug fix for
ade546be47 (worktree: invoke post-checkout hook, 2017-12-07).

The problem that patch addresses is that "git worktree add" does not
provide proper context to the invoked 'post-checkout' hook, so the hook
doesn't know where the newly-created worktree is. Lars's approach was to
set GIT_WORK_TREE to point at the new worktree directory, however, doing
so has a few drawbacks:

1. GIT_WORK_TREE is normally assigned in conjunction with GIT_DIR; it is
   unusual and possibly problematic to set one but not the other.

2. Assigning GIT_WORK_TREE unconditionally may lead to unforeseen
   interactions and problems with end-user scripts and aliases or even
   within Git itself. It seems better to avoid unconditional assignment
   rather than risk problems such as those described and worked around
   by 86d26f240f (setup.c: re-fix d95138e (setup: set env $GIT_WORK_TREE
   when .., 2015-12-20)

3. Assigning GIT_WORK_TREE is too specialized a solution; it "fixes"
   only Git commands run by the hook, but does nothing for other
   commands ('mv', 'cp', etc.) that the hook might invoke.

The real problem with ade546be47 is that it neglects to change to the
directory of the newly-created worktree before running the hook, thus
the hook incorrectly runs in the directory in which "git worktree add"
was invoked. Rather than messing with GIT_WORK_TREE, this replacement
patch series fixes the problem by ensuring that the directory is changed
before the hook is invoked.

[1]: https://public-inbox.org/git/20180210010132.33629-1-lars.schneider@autodesk.com/

Eric Sunshine (2):
  run-command: teach 'run_hook' about alternate worktrees
  worktree: add: change to new worktree directory before running hook

 builtin/worktree.c      | 11 ++++++++---
 run-command.c           | 23 +++++++++++++++++++++--
 run-command.h           |  4 ++++
 t/t2025-worktree-add.sh | 25 ++++++++++++++++++++++---
 4 files changed, 55 insertions(+), 8 deletions(-)

-- 
2.16.1.291.g4437f3f132

  parent reply	other threads:[~2018-02-12  3:17 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-02-10  1:01 [PATCH v1] worktree: set worktree environment in post-checkout hook lars.schneider
2018-02-10  1:28 ` Lars Schneider
2018-02-12  3:15 ` Eric Sunshine [this message]
2018-02-12  3:15   ` [PATCH 1/2] run-command: teach 'run_hook' about alternate worktrees Eric Sunshine
2018-02-12 20:58     ` Lars Schneider
2018-02-12 21:49       ` Eric Sunshine
2018-02-12  3:15   ` [PATCH 2/2] worktree: add: change to new worktree directory before running hook Eric Sunshine
2018-02-12 19:37     ` Junio C Hamano
2018-02-12 20:31       ` Eric Sunshine
2018-02-12 20:01     ` Lars Schneider
2018-02-13  4:42       ` Eric Sunshine
2018-02-13  4:48         ` Eric Sunshine
2018-02-13  7:27     ` Johannes Sixt
2018-02-13  7:34       ` Eric Sunshine
2018-02-15 19:18   ` [PATCH v2] worktree: add: fix 'post-checkout' not knowing new worktree location Eric Sunshine
2018-02-15 20:52     ` Junio C Hamano
2018-02-15 21:27       ` Eric Sunshine
2018-02-15 21:36         ` Junio C Hamano
2018-02-15 23:09         ` Eric Sunshine
2018-02-15 23:09     ` [PATCH v3] " Eric Sunshine
2018-02-16 16:55       ` Lars Schneider
2018-02-16 18:27         ` Eric Sunshine
2018-02-16 18:05       ` Junio C Hamano
2018-02-12  3:27 ` [PATCH v1] worktree: set worktree environment in post-checkout hook Eric Sunshine

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

  List information: http://vger.kernel.org/majordomo-info.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180212031526.40039-1-sunshine@sunshineco.com \
    --to=sunshine@sunshineco.com \
    --cc=git@vger.kernel.org \
    --cc=larsxschneider@gmail.com \
    --cc=matthew.k.gumbel@intel.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.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

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 read-only IMAP folder(s) and NNTP newsgroup(s).