Date: Fri, 11 Feb 2022 16:43:23 -0600
From: Justin Pryzby <pryzby@telsasoft.com>
To: Alvaro Herrera <alvherre@alvh.no-ip.org>
Cc: pgsql-hackers@lists.postgresql.org,
	Simon Riggs <simon.riggs@enterprisedb.com>,
	Tomas Vondra <tomas.vondra@enterprisedb.com>,
	Zhihong Yu <zyu@yugabyte.com>,
	Daniel Westermann <dwe@dbi-services.com>,
	Amit Langote <amitlangote09@gmail.com>,
	Japin Li <japinli@hotmail.com>, Erik Rijkers <er@xs4all.nl>,
	Jaime Casanova <jcasanov@systemguards.com.ec>
Subject: Re: support for MERGE
Message-ID: <20220211224323.GH31460@telsasoft.com>
References: <20220211185149.GG31460@telsasoft.com>
 <202202112025.zbeoj2u5tclj@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <202202112025.zbeoj2u5tclj@alvherre.pgsql>
User-Agent: Mutt/1.9.4 (2018-02-28)
Archived-At: 
 <https://www.postgresql.org/message-id/20220211224323.GH31460%40telsasoft.com>
Precedence: bulk

On Fri, Feb 11, 2022 at 05:25:49PM -0300, Alvaro Herrera wrote:
> > I'm not sure git diff --cherry-pick is widely known/used, but I think
> > using that relative to master may be good enough.  
> 
> I had never heard of git diff --cherry-pick, and the manpages I found
> don't document it, so frankly I doubt it's known.  I still have no idea
> what does it do.

See git-log(1)

       --cherry-pick
           Omit any commit that introduces the same change as another commit on the “other side” when the set of commits are limited with symmetric difference.

The "symmetric difference" is the triple-dot notation.

The last few years I've used this to check for missing bits in the draft
release notes.  (Actually, I tend to start my own list of features before
that).  It's doing a generic version of what git_changelog does.

https://www.postgresql.org/message-id/20210510144045.GC27406@telsasoft.com

> I suppose there is an obvious reason why using git diff with
> $(git merge-base ...) as one of the arguments doesn't work for these purposes.
> 
> > Andres thinks that does the wrong thing if CI is run manually (not by CFBOT)
> > for patches against backbranches.
> 
> I wonder if it's sufficient to handle these things (coverage
> specifically) for branch master only.

Or default to master, and maybe try to parse the commit message and pull out
Backpatch-through: NN.  It's intended to be machine-readable, after all.

Let's talk about it more on the/another thread :)

-- 
Justin