Re: vacuumdb and new VACUUM options

On Thu, May 9, 2019 at 1:39 PM Kyotaro HORIGUCHI
<horiguchi.kyotaro@lab.ntt.co.jp> wrote:
>
> At Thu, 9 May 2019 20:14:51 +0900, Masahiko Sawada <sawada.mshk@gmail.com> wrote in
<CAD21AoBmA9H3ZRuQFF+9io9PKhP+ePS=D+ThZ6ohRMdBm2x8Pw@mail.gmail.com>
> > On Thu, May 9, 2019 at 10:01 AM Michael Paquier <michael@paquier.xyz> wrote:
> > >
> > > On Wed, May 08, 2019 at 06:21:09PM -0300, Euler Taveira wrote:
> > > > Em qua, 8 de mai de 2019 às 14:19, Fujii Masao <masao.fujii@gmail.com> escreveu:
> > > >> The question is; we should support vacuumdb option for (1), i.e.,,
> > > >> something like --index-cleanup option is added?
> > > >> Or for (2), i.e., something like --disable-index-cleanup option is added
> > > >> as your patch does? Or for both?
> > > >
> > > > --index-cleanup=BOOL
> > >
> > > I agree with Euler's suggestion to have a 1-1 mapping between the
> > > option of vacuumdb and the VACUUM parameter
> >
> > +1. Attached the draft version patches for both options.
>
> +       printf(_("      --index-cleanup=BOOLEAN     do or do not index vacuuming and index cleanup\n"));
> +       printf(_("      --truncate=BOOLEAN          do or do not truncate off empty pages at the end of the
table\n"));
>
> I *feel* that force/inhibit is suitable than true/false for the
> options.

Indeed.

+         If not specify this option
+         the behavior depends on <literal>vacuum_index_cleanup</literal> option
+         for the table to be vacuumed.

+         If not specify this option
+         the behavior depends on <literal>vacuum_truncate</literal> option
+         for the table to be vacuumed.

Those sentences should be rephrased to something like "If this option
is not specified, the bahvior...".