Date: Tue, 7 Feb 2023 09:25:42 -0600
From: Justin Pryzby <pryzby@telsasoft.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Andres Freund <andres@anarazel.de>, Noah Misch <noah@leadboat.com>,
	Andrew Dunstan <andrew@dunslane.net>,
	Jelte Fennema <postgres@jeltef.nl>, Peter Geoghegan <pg@bowt.ie>,
	Bruce Momjian <bruce@momjian.us>,
	Magnus Hagander <magnus@hagander.net>,
	Alvaro Herrera <alvherre@2ndquadrant.com>,
	Stephen Frost <sfrost@snowman.net>, Jesse Zhang <sbjesse@gmail.com>,
	pgsql-hackers@postgresql.org
Subject: Re: run pgindent on a regular basis / scripted manner
Message-ID: <20230207152542.GH1653@telsasoft.com>
References: <CAGECzQRYjHdNQ-Bq2yrBb-8s3Pjx7pfvYPDdNv9tcvDcscDhbg@mail.gmail.com>
 <20230122223806.3sgew6qswhsczae7@awork3.anarazel.de>
 <1489276.1674427657@sss.pgh.pa.us>
 <de0a7034-0186-1f91-b408-eb67f2a854dc@dunslane.net>
 <1708030.1675446770@sss.pgh.pa.us>
 <20230204041848.GC1896331@rfd.leadboat.com>
 <20230204113453.vmtnayjvhyg5qqqy@alap3.anarazel.de>
 <1923921.1675526879@sss.pgh.pa.us>
 <20230204171101.GF1653@telsasoft.com>
 <2009472.1675532231@sss.pgh.pa.us>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <2009472.1675532231@sss.pgh.pa.us>
User-Agent: Mutt/1.9.4 (2018-02-28)
Archived-At: <https://www.postgresql.org/message-id/20230207152542.GH1653%40telsasoft.com>
Precedence: bulk

On Sat, Feb 04, 2023 at 12:37:11PM -0500, Tom Lane wrote:
> Justin Pryzby <pryzby@telsasoft.com> writes:

> Hmmm ... inserting all of those as the default options would likely
> make it impossible to update pg_bsd_indent itself with anything like
> its current indent style (not that it's terribly consistent about
> that).  I could see inventing a --postgresql shortcut switch perhaps.

Or you could add ./.indent.pro, or ./src/tools/indent.profile for it to
read.

> > Would you want to make those the default options of the in-tree indent ?
> > Or provide a shortcut like --postgresql ?
> 
> But it's not clear to me why you're allergic to the perl wrapper?

My allergy is to the totality of the process, not to the perl component.
It's a bit weird to enforce a coding style that no upstream indent tool
supports.  But what's weirder is that, *having forked the indent tool*,
it still doesn't implement the desired style, and the perl wrapper tries
to work around that.

It would be more reasonable if the forked C program knew how to handle
the stuff for which the perl script currently has kludges to munge the
source code before indenting and then un-munging afterwards.

Or if the indentation were handled by the (or a) perl script itself.

Or if the perl script handled everything that an unpatched "ident"
didn't handle, rather than some things but not others, demanding use of
a patched indent as well as a perl wrapper (not just for looping around
files and fancy high-level shortcuts like indenting staged files).

On the one hand, "indent" ought to handle all the source-munging stuff.
On the other hand, it'd be better to use an unpatched indent tool.  The
current way is the worst of both worlds.

Currently, the perl wrapper supports the "/* -"-style comments that
postgres wants to use (why?) by munging the source code.  That could be
supported in pg-bsd-indent with a one line change.  I think an even
better option would be to change postgres' C files to use "/*-" without
a space, which requires neither perl munging nor patching indent.

On a less critical note, I wonder if it's a good idea to import
pgbsdindent as a git "submodule".

-- 
Justin