Date: Wed, 9 Dec 2015 06:27:09 -0800
From: David Fetter <david@fetter.org>
To: Michael Paquier <michael.paquier@gmail.com>
Cc: Thomas Munro <thomas.munro@enterprisedb.com>,
 Alvaro Herrera <alvherre@2ndquadrant.com>,
 Jeff Janes <jeff.janes@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>,
 Pg Hackers <pgsql-hackers@postgresql.org>
Subject: Re: Making tab-complete.c easier to maintain
Message-ID: <20151209142709.GA7050@fetter.org>
References: 
 <CAEepm=1=Hdt9TCM3BRiVE6NCnXHyDDki+YVPhfZ02_=niU=8cA@mail.gmail.com>
 <CAMkU=1yt=Ed7+K4sUisdUkt-wDhGnxdkDqs_43uJvhJYFA56gQ@mail.gmail.com>
 <CAEepm=1ns0rbo+xun_j6_PwDT6zmEPuNggy=Vaie-3peF2C44Q@mail.gmail.com>
 <CAMkU=1zRsaaSF_8qyQNYqOcDt0Xa7Awk==YdWCeTZtNhjnPNww@mail.gmail.com>
 <CAEepm=1Oaq+L_6KCZ5o_hNn-=T93_LiK=_T1C9DcPzgXs4H84g@mail.gmail.com>
 <CAEepm=1EOiP9_fKfb2B_ncp2hWYFxWEM+OsoD-=-iL_Wh=46Vg@mail.gmail.com>
 <20151116151923.GX614468@alvherre.pgsql>
 <CAB7nPqTfXV5Jwd1j-JF4wTuVgt8ggsNAy=yBByWB0nV1_3CEsA@mail.gmail.com>
 <CAEepm=0+vDukbD7St2tjvpF9Ro9Z5Gj1Q9Mh6hKg+Of2YMNxzQ@mail.gmail.com>
 <CAB7nPqRoQVqzz9y5hOYBhMwPy5_OoTrObf=5a1XP+zjkPNAAkg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CAB7nPqRoQVqzz9y5hOYBhMwPy5_OoTrObf=5a1XP+zjkPNAAkg@mail.gmail.com>
User-Agent: Mutt/1.5.24 (2015-08-30)
Precedence: bulk
Sender: pgsql-hackers-owner@postgresql.org

On Wed, Dec 09, 2015 at 08:49:22PM +0900, Michael Paquier wrote:
> On Wed, Dec 9, 2015 at 8:17 PM, Thomas Munro
> <thomas.munro@enterprisedb.com> wrote:
> > On Mon, Dec 7, 2015 at 8:41 PM, Michael Paquier
> > <michael.paquier@gmail.com> wrote:
> >> On Tue, Nov 17, 2015 at 12:19 AM, Alvaro Herrera <alvherre@2ndquadrant.com>
> >> wrote:
> >>> Thomas Munro wrote:
> >>>> New version attached, merging recent changes.
> >>>
> >>> I wonder about the TailMatches and Matches macros --- wouldn't it be
> >>> better to have a single one, renaming TailMatches to Matches and
> >>> replacing the current Matches() with an initial token that corresponds
> >>> to anchoring to start of command?  Just wondering, not terribly attached
> >>> to the idea.
> >>
> >> +       /* TODO:TM -- begin temporary, not part of the patch! */
> >> +       Assert(!word_matches(NULL, ""));
> >> + [...]
> >> +       Assert(!word_matches("foo", ""));
> >> +       /* TODO:TM -- end temporary */
> >>
> >> Be sure to not forget to remove that later.
> >
> > Thanks for looking at this Michael.  It's probably not much fun to
> > review!  Here is a new version with that bit removed.  More responses
> > inline below.
> 
> I had a hard time not sleeping when reading it... That's very mechanical.
> 
> > I agree that would probably be better but Alvaro suggested following
> > the existing logic in the first pass, which was mostly based on tails,
> > and then considering simpler head-based patterns in a future pass.
> 
> Fine with me.
> 
> So what do we do now? There is your patch, which is already quite big,
> but as well a second patch based on regexps, which is far bigger. And
> at the end they provide a similar result:
> 
> Here is for example what the regexp patch does for some complex
> checks, like ALTER TABLE RENAME:
>      /* ALTER TABLE xxx RENAME yyy */
> -    else if (pg_strcasecmp(prev4_wd, "TABLE") == 0 &&
> -             pg_strcasecmp(prev2_wd, "RENAME") == 0 &&
> -             pg_strcasecmp(prev_wd, "CONSTRAINT") != 0 &&
> -             pg_strcasecmp(prev_wd, "TO") != 0)
> +    else if (MATCH("TABLE #id RENAME !CONSTRAINT|TO"))
> 
> And what Thomas's patch does:
> +    else if (TailMatches5("ALTER", "TABLE", MatchAny, "RENAME", MatchAny) &&
> +             !TailMatches1("CONSTRAINT|TO"))
> 
> The regexp patch makes the negative checks somewhat easier to read
> (there are 19 positions in tab-complete.c doing that), still inventing
> a new langage and having a heavy refactoring just tab completion of
> psql seems a bit too much IMO, so my heart balances in favor of
> Thomas' stuff. Thoughts from others?

Agreed that the "whole new language" aspect seems like way too big a
hammer, given what it actually does.

Cheers,
David.
-- 
David Fetter <david@fetter.org> http://fetter.org/
Phone: +1 415 235 3778  AIM: dfetter666  Yahoo!: dfetter
Skype: davidfetter      XMPP: david.fetter@gmail.com

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate


-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers