[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: fine-tuning new flags - feedback needed

From: David Kastrup
Subject: Re: fine-tuning new flags - feedback needed
Date: Mon, 07 Feb 2011 16:03:45 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.0.50 (gnu/linux)

Janek Warchoł <address@hidden> writes:

> 2011/2/7 Carl Sorensen <address@hidden>:
>> On 2/6/11 4:44 PM, "Janek Warchoł" <address@hidden> wrote:
>>> As for now, i've already reset my git repository (that is i clicked
>>> "Abort changes - reset to origin" in lily-git) and made the changes
>>> again. So now i call
>>>   git status (everything looks fine, 3 files i've changed are listed)
>>>   git diff HEAD (i see something resembling patch file)
>>>   git commit -a (it asked me for message - i'm not sure if it's needed
>>> since it will be an update of existing commit, but i wrote something
>>> there and answered yes to questions that shown up...)
>> There's no such thing as "an update of an existing commit".  Every time you
>> do git commit, it's a new commit, even though it's an existing patch set.
>> If you want to make it part of the previous commit, you can do so with
>> git commit -a --amend
>> But I don't recommend doing that.
> Aha. Ok.

Rule of thumb: --amend is fine as long as long as the original commit
never got into any repository other than the one being amended.

A patchset on Rietveld does not count I should think since "commit" here
means a complete unit including author, committer, commit message, diff
and parent commit.  Applying a patchset from Rietveld does not result in
a reliable reproduction of the "commit" as identified by git since
Rietveld is fundamentally Subversion.

David Kastrup

reply via email to

[Prev in Thread] Current Thread [Next in Thread]