git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Shawn O. Pearce" <spearce@spearce.org>
To: Ted Pavlic <ted@tedpavlic.com>
Cc: git <git@vger.kernel.org>, Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH] Simplest update to bash completions to prevent unbounded variable errors
Date: Tue, 13 Jan 2009 07:20:47 -0800	[thread overview]
Message-ID: <20090113152047.GO10179@spearce.org> (raw)
In-Reply-To: <496C1F5F.9020604@tedpavlic.com>

Ted Pavlic <ted@tedpavlic.com> wrote:
> Another try at fixing bash completions in "set -u" environments.

I agree with Junio; setting -u in your interactive shell is as bad
as export CDPATH.  Its crazy.

> Additionally added some comments and omitted things like Vim modelines.

These are orthogonal to the -u corrections.  They should be in a
different patch.  The comments are wecome.  The '#!bash' looks like
a good idea.  But a vim specific modeline, I don't like, for the
reasons Junio has already stated.

> +# __gitdir accepts 0 or 1 arguments (i.e., location)
> +# returns location of .git repo
>  __gitdir ()
>  {
> -	if [ -z "$1" ]; then
> +	if [ $# -eq 0 ] || [ -z "$1" ]; then

This is one of those places where [ -z "${1-}" ] is likely easier
to read then the || usage you have introduced.  We don't care if
we got no args, or we got one that is the empty string, either way
the $1 cannot be a gitdir and we need to guess it.

> @@ -111,7 +116,7 @@ __git_ps1 ()
>  			fi
>  		fi
>
> -		if [ -n "$1" ]; then
> +		if [ $# -gt 0 ] && [ -n "$1" ]; then
>  			printf "$1" "${b##refs/heads/}$r"

Eh, I'd rather see [ -n "${1-}" ] over the && test.

> -complete -o default -o nospace -F _git git
> -complete -o default -o nospace -F _gitk gitk
> +complete -o bashdefault -o default -o nospace -F _git git 2>/dev/null \
> +	|| complete -o default -o nospace -F _git git
> +complete -o bashdefault -o default -o nospace -F _gitk gitk 2>/dev/null \
> +	|| complete -o default -o nospace -F _gitk gitk

Why are we switching to bashdefault?  Is this an unrelated change
from the -u stuff and should go into its own commit, with its own
justification?

-- 
Shawn.

  reply	other threads:[~2009-01-13 15:22 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-01-13  4:58 [PATCH] Simplest update to bash completions to prevent unbounded variable errors Ted Pavlic
2009-01-13 15:20 ` Shawn O. Pearce [this message]
2009-01-13 15:30   ` Ted Pavlic
2009-01-13 15:33     ` Shawn O. Pearce

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=20090113152047.GO10179@spearce.org \
    --to=spearce@spearce.org \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=ted@tedpavlic.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).