git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Holger Hellmuth <hellmuth@ira.uka.de>
To: Massimo Manca <massimo.manca@micronengineering.it>
Cc: "Jakub Narebski" <jnareb@gmail.com>,
	"Jonathan Nieder" <jrnieder@gmail.com>, Mike <xandrani@gmail.com>,
	git@vger.kernel.org, "Ben Walton" <bwalton@artsci.utoronto.ca>,
	"Ramkumar Ramachandra" <artagnon@gmail.com>,
	"Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>
Subject: Re: Undo last commit?
Date: Tue, 28 Jun 2011 15:57:02 +0200	[thread overview]
Message-ID: <4E09DDAE.30801@ira.uka.de> (raw)
In-Reply-To: <4DFF382A.5030206@micronengineering.it>

Let me play devils advocate:
Wouldn't it be nice to have a command 'git uncommit' (in core git) that 
would be an alias to "git reset --keep HEAD^" ? And if not already done, 
it should hint at "git reset --soft HEAD^" in case of failure because of 
conflicts.

It would be a nice companion to the not-yet-realized "git unadd" ;-)

Holger.


On 20.06.2011 14:08, Massimo Manca wrote:
> Hello,
> I several times made the mistake of a wrong commit also if generally the
> error born for a wrong add expecially:
>
> git add . in a directory with some files that haven't to be managed by git.
>
> So, as wrote in some emails here, if I wrote something like:
> git commit -m "Added file.c" -a
>
> I tryed to solve with:
> git commit --amend -m "Added file.c" -a
>
> hoping to have a status like before the commit and then sending:
> git reset .
>
> hoping to have a status like that before the wrong add.
> But this is not what git status say so, my solution to solve commi
> problems is ALWAYS:
>
> git reset --soft HEAD^
>
> that for my point of view works better then all others permitting to
> redo last add and commit to solve not only a -m problem but also a wrong
> git add command.
>
> Il 19/06/2011 12.37, Jakub Narebski ha scritto:
>> On Sun, 19 Jun 2011, Jonathan Nieder wrote:
>>> Jakub Narebski wrote:
>>>> Mike<xandrani@gmail.com>  writes:
>>>>> % git reset --hard HEAD~1
>>>> Errr... here you screwed up.  This reset state of you working area to
>>>> the state at last commit, removing all your changes to tracked files.
>>> Or rather, here we screwed up.  Jakub and others gave some useful
>>> advice about how to recover, so let's consider how the UI or
>>> documentation could be improved to prevent it from happening again.
>>>
>>> * In this example if I understand correctly then the index contained
>>>    some useful information, perhaps about a larger commit intended for
>>>    later.  To preserve that, you could have used
>>>
>>> 	git reset --soft HEAD~1
>>>
>>>    which would _just_ undo the effect of "git commit", leaving the index
>>>    and worktree alone.
>> Another issue is that Mike haven't realized that `--amend' option can be
>> used *in combination* with other "git commit" options, which means that
>> the solution to his problem was using "git commit" as it should have
>> been done, but with '--amend' added.
>>
>> I'm not sure if git documentation talks about 'git reset --soft HEAD^',
>> and when to use it; from what I remember it encourages use of
>> 'git commit --amend' instead (which was I guess most often used reason
>> of using soft reset before there was '--amend').
>>
>>> * Another situation that comes up from time to time is making a change
>>>    that just turned out to be a bad idea.  After commiting it, you might
>>>    want to discard the erroneous change, like so:
>>>
>>> 	git reset --keep HEAD~1
>>>
>>>    The "--keep" option uses some safeguards to make sure that only the
>>>    committed change gets discarded, instead of clobbering local changes
>>>    at the same time.
>>>
>>> * In the early days of git, the "--keep" option did not exist.  So a lot
>>>    of old documentation recommends to do
>>>
>>> 	git reset --hard HEAD~1
>>>
>>>    which is the same if you don't have any local changes.
>> Yes, it would be good idea to examine git documentation (tutorials,
>> user's manual, manpages, perhaps "Git Community Book" and "Pro Git"
>> too) to encourage use of new safer options of hard reset, namely
>> '--keep' and '--merge' instead of '--hard'.
>>
>

  reply	other threads:[~2011-06-28 13:59 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-18 13:15 Undo last commit? Mike
2011-06-18 13:43 ` Ben Walton
2011-06-18 13:48 ` Ramkumar Ramachandra
2011-06-18 13:54 ` Jakub Narebski
2011-06-19  0:37   ` Jonathan Nieder
2011-06-19 10:37     ` Jakub Narebski
2011-06-20 12:08       ` Massimo Manca
2011-06-28 13:57         ` Holger Hellmuth [this message]
2011-06-28 14:31           ` Mike
2011-06-30  4:50             ` Ramkumar Ramachandra
2011-06-30 18:38               ` Jonathan Nieder
2011-06-30 19:48                 ` Mike
2011-06-30 17:29             ` Junio C Hamano
2012-02-11 18:19           ` Neal Kreitzinger
     [not found]             ` <CAHK-92oMc62O0S8Bxt6+uxobE+kg5wOeRDoOsHWvvenXaXmZGQ@mail.gmail.com>
2012-02-11 22:07               ` Jakub Narebski
2012-02-11 22:29                 ` Jonathan Nieder

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=4E09DDAE.30801@ira.uka.de \
    --to=hellmuth@ira.uka.de \
    --cc=artagnon@gmail.com \
    --cc=bwalton@artsci.utoronto.ca \
    --cc=git@vger.kernel.org \
    --cc=jnareb@gmail.com \
    --cc=jrnieder@gmail.com \
    --cc=massimo.manca@micronengineering.it \
    --cc=pclouds@gmail.com \
    --cc=xandrani@gmail.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).