Re: [PATCH v2] glossary: add definition for overlay

From: Thomas Gummerer <t.gummerer@gmail.com>
To: Junio C Hamano <gitster@pobox.com>
Cc: Duy Nguyen <pclouds@gmail.com>,
	Philip Oakley <philipoakley@iee.org>,
	Git Mailing List <git@vger.kernel.org>,
	Jonathan Nieder <jrnieder@gmail.com>,
	Elijah Newren <newren@gmail.com>
Subject: Re: [PATCH v2] glossary: add definition for overlay
Date: Fri, 15 Mar 2019 23:10:54 +0000	[thread overview]
Message-ID: <20190315231054.GK16414@hank.intra.tgummerer.com> (raw)
In-Reply-To: <xmqq5zsnh7my.fsf@gitster-ct.c.googlers.com>

On 03/13, Junio C Hamano wrote:
> Thomas Gummerer <t.gummerer@gmail.com> writes:
> 
> > diff --git a/Documentation/glossary-content.txt b/Documentation/glossary-content.txt
> > index 023ca95e7c..53df6ecb0a 100644
> > --- a/Documentation/glossary-content.txt
> > +++ b/Documentation/glossary-content.txt
> > @@ -287,6 +287,13 @@ This commit is referred to as a "merge commit", or sometimes just a
> >  	origin/name-of-upstream-branch, which you can see using
> >  	`git branch -r`.
> >  
> > +[[def_overlay]]overlay::
> > +	Only update and add files to the working directory, but don't
> > +	delete them, similar to how 'cp -R' would work.  This is the
> > +	default mode in a <<def_checkout,checkout>>.  In contrast,
> > +	no-overlay mode will also delete tracked files not present in
> > +	the source, similar to 'rsync --delete'.
> > +
> 
> At least the mention of "checkout" needs to be a lot tightened to
> clarify that it is talking about "checkout <pathspec>", aka
> "checking out files out of the index or a tree-ish", as opposed to
> "checking out a branch to work on it", as checking out a branch will
> not work in the overlay fashion.  What's not in the newly checked
> out branch will disappear from the working tree.

Good point, I'll fix that in v3.

> If readers happen to be not paying close attention to the fact that
> the difference between overlay and non-overlay is about the
> destination, "similar to how 'cp -R' would work" may not click to
> their minds.  "similar to how 'cp -R' updates the contents in the
> destination directory" may avoid such a risk, albeit it might be a
> bit too verbose.  I dunno.

I feel like it's okay to be a bit more verbose in the glossary.  I
think your suggestion above is better than what I had, I'll use that
in v3, thanks!