All of lore.kernel.org
 help / color / mirror / Atom feed
From: Michael J Gruber <git@drmicha.warpmail.net>
To: Andreas Ericsson <ae@op5.se>
Cc: Ondrej Certik <ondrej@certik.cz>,
	Git Mailing List <git@vger.kernel.org>,
	Jakub Narebski <jnareb@gmail.com>
Subject: Re: default aliases (ci, di, st, co)
Date: Thu, 09 Jul 2009 10:50:13 +0200	[thread overview]
Message-ID: <4A55AF45.4040700@drmicha.warpmail.net> (raw)
In-Reply-To: <4A55958E.1050401@op5.se>

Andreas Ericsson venit, vidit, dixit 09.07.2009 09:00:
> Ondrej Certik wrote:
>> Hi,
>>
>> coming to git from mercurial and svn, here is the alias part of my .gitconfig:
>>
>> [alias]
>>     ci = commit
>>     di = diff --color-words
>>     st = status
>>     co = checkout
>>
>>
>> And all is fine until I share commands to checkout my branch (for
>> example) with other people, then basically I have to write those
>> commands in full (e.g. commit, checkout, ...), since I cannot assume
>> they have their .gitconfig setup the same way I do. Especially for
>> people who are new to git.
>>
>> What is the view on this in the git community?
> 
> 
> Personally I think "plain" aliases like you use above are evil.
> git is not svn or mercurial. If you use it like svn you'll be
> surprised sooner or later and your workflow will feel awkward or
> just plain wrong. hg and git are very similar, but the fact that
> hg assumed the svn shorthands implies to me that they've tried
> to retain a compatibility that does not, in fact, exist. DVCS
> systems are enormously different from centralized ones. Borrowing
> from centralized ones to a DVCS one feels utterly backwards.
> 
> 
>> Do most of you write
>>
>> git checkout -b branch ...
>>
>> (possibly with TAB completion) or do most of you write
>>
>> git co -b branch ...
>>
>> (like I do all the time, except when sharing my commands with other people)?
>>
> 
> I have no simple aliases for any of the commands. There's just no reason
> for them since tab completion works so well.
> 
> I do have one alias, which is "wsfix". It fixes whitespace fsckups I've
> added to the index but not yet committed to the worktree. It's not a
> particularly complex one, but not exactly simple either.
> 
> 
>> Could in principle those aliases be even made default by git? Or is
>> this not a good idea.
>>
> 
> I see no reason to add default aliases. It will make a mess of things
> when trying to explain workflows to people who have older versions of
> git where those default aliases aren't available. It'll add complexity
> to explanations, which is never a good thing.
> 
> If I have a vote, I vote no.
> 

Adding my no-vote to that, too.

Aliases are great, but default aliases are a problem. We saw this when
we introduced "git stage", and I don't think we're happy in retrorespect.

Having said that, collecting useful aliases (e.g. one, who, ...) in
contrib/ or in the wiki would be helpful.

Michael

  reply	other threads:[~2009-07-09  8:50 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-07-08 23:49 default aliases (ci, di, st, co) Ondrej Certik
2009-07-09  0:13 ` Jakub Narebski
2009-07-09  8:58   ` Graeme Geldenhuys
2009-07-09 16:43   ` Junio C Hamano
2009-07-09 23:08     ` Paolo Bonzini
2009-07-13  6:40       ` Jeff King
2009-07-13  6:43         ` Jeff King
2009-07-13  9:15         ` Paolo Bonzini
2009-07-09  7:00 ` Andreas Ericsson
2009-07-09  8:50   ` Michael J Gruber [this message]
2009-07-09  9:04     ` Graeme Geldenhuys
2009-07-09  9:40     ` Jakub Narebski
2009-07-09  9:48       ` Michael J Gruber
2009-07-09  8:58   ` Luciano Miguel Ferreira Rocha
2009-07-12  5:07     ` Ondrej Certik
     [not found]   ` <20090709160249.GA12830@cthulhu>
2009-07-09 20:54     ` wsfix alias Andreas Ericsson
2009-07-09 21:30       ` Larry D'Anna
2009-07-09  9:01 ` default aliases (ci, di, st, co) Graeme Geldenhuys
2009-07-09  9:12   ` Matthieu Moy

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=4A55AF45.4040700@drmicha.warpmail.net \
    --to=git@drmicha.warpmail.net \
    --cc=ae@op5.se \
    --cc=git@vger.kernel.org \
    --cc=jnareb@gmail.com \
    --cc=ondrej@certik.cz \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.