Re: [RFC/PATCH] Implement git-cp.

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Feb 10, 2008, 12:33 UTC
Message-ID: <alpine.LSU.1.00.0802101230160.11591@racer.site>
In-Reply-To: <7vzlu9uu7c.fsf@gitster.siamese.dyndns.org>

Hi,

On Sat, 9 Feb 2008, Junio C Hamano wrote:

Show 10 quoted lines

> Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:
> 
> > If you touch builtin-mv.c already, why not just move cmd_cp() in 
> > there?  It's not like it would be the first cmd_*() function living in 
> > the same file as other cmd_*() functions.
> 
> Why do we even want "git-cp", especially when git-mv and git-rm are 
> already pretty much redundant commands?
> 
> Especially, why do we even encourage copy-and-paste?

Heh.  I quote:

	If you want to shoot yourself in the foot Git will point you to 
	the gun rack and show you how to load the bullets. (Masukomi)

But yeah, I did not really think about it. I had the impression originally that it would make sense to copy a few tracked files around, in order to modify them.

Think "kernel module". You'd not start from scratch, but find a similar one, and then change the heck out of it.

But your comment made me think again: It's better to "cp" those files/directories, and then "git add" the relevant ones. It even avoids stupid mistakes where you commit something you "cp"ed, but did not need.

Ciao, Dscho