git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Pete Wyckoff <pw@padd.com>
To: Jens Lehmann <Jens.Lehmann@web.de>
Cc: Jeff King <peff@peff.net>, Git Mailing List <git@vger.kernel.org>,
	Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH] test_interactive: interactive debugging in test scripts
Date: Mon, 16 Jan 2012 17:07:12 -0500	[thread overview]
Message-ID: <20120116220712.GA30785@padd.com> (raw)
In-Reply-To: <4F148D3B.4070206@web.de>

Jens.Lehmann@web.de wrote on Mon, 16 Jan 2012 21:48 +0100:
> Am 16.01.2012 21:11, schrieb Jeff King:
> > On Mon, Jan 16, 2012 at 09:01:21PM +0100, Jens Lehmann wrote:
> > 
> >> So I vote for your patch as it takes my initial idea even further. I
> >> really like that HOME, TERM and SHELL are honored in your version
> >> leaving the user with a fully functional shell of his choice.
> > 
> > I'm actually mildly negative on this feature, as it interferes with the
> > tests themselves. Probably TERM and SHELL don't matter. But $HOME means
> > git will read your personal .gitconfig, not any config (or lack thereof)
> > in the trash directory.
> 
> Good point, I haven't thought of that ... so yes, at least $HOME should
> go.

I, too, have stumbled over differences that are due to picking up
something in $HOME.  It takes a minute to realize what's going
on.

TERM can interfere with at least one test: c2116a1 (test-lib: fix TERM
to dumb for test repeatability, 2008-03-06).  SHELL can cause
issues when it is more feature-ful than SHELL_PATH.

On the other hand, it's frustrating to work in an environment
without my shell aliases, git aliases, readline customizations,
personal path, and assorted environment variables.

I'm comfortable with the (rare?) possibility of confusion.  But
perhaps it is unwise to support a feature with so many caveats,
even if it is only for debugging.

		-- Pete

  reply	other threads:[~2012-01-16 22:07 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-01-15 20:00 [PATCH] test-lib: add the test_bash convenience function Jens Lehmann
2012-01-15 23:24 ` Jeff King
2012-01-16 15:49   ` [PATCH] test_interactive: interactive debugging in test scripts Pete Wyckoff
2012-01-16 20:01     ` Jens Lehmann
2012-01-16 20:11       ` Jeff King
2012-01-16 20:48         ` Jens Lehmann
2012-01-16 22:07           ` Pete Wyckoff [this message]
2012-01-16 20:19     ` Jeff King
2012-01-16 22:51   ` [PATCH] test-lib: add the test_bash convenience function Junio C Hamano
2012-01-17  8:21     ` [PATCH] test-lib: add the test_pause " Jens Lehmann
2012-01-17 19:15       ` Junio C Hamano
2012-01-17 21:04         ` [PATCH v2] " Jens Lehmann

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=20120116220712.GA30785@padd.com \
    --to=pw@padd.com \
    --cc=Jens.Lehmann@web.de \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=peff@peff.net \
    /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).