MIME-Version: 1.0
In-Reply-To: 
 <CAHGQGwE2dC5aNBgV5AtpMMdZNpkGn1PAa9o=whi8_Wb3m3ruAA@mail.gmail.com>
References: 
 <CAOzEurTpMTUEW8kHu-zKB0EBtuPfpvyoJ--8pxKe87p24BGrpg@mail.gmail.com>
 <CAOtHd0AY4LuM+-JqFyzs4mPHG8yeAwfP5yr2EZg21RVO0442BA@mail.gmail.com>
 <CAHGQGwE2dC5aNBgV5AtpMMdZNpkGn1PAa9o=whi8_Wb3m3ruAA@mail.gmail.com>
From: "David G. Johnston" <david.g.johnston@gmail.com>
Date: Sun, 5 Apr 2026 22:17:14 -0700
Message-ID: 
 <CAKFQuwZUSetJvoXewwofTNP5dLfhg7S31NNJhZ=G+ZbOOCZ=ug@mail.gmail.com>
Subject: Re: doc: Clarify ANALYZE VERBOSE output
To: Fujii Masao <masao.fujii@gmail.com>
Cc: "maciek@sakrejda.org" <maciek@sakrejda.org>,
 Shinya Kato <shinya11.kato@gmail.com>,
	"pgsql-docs@lists.postgresql.org" <pgsql-docs@lists.postgresql.org>
Content-Type: multipart/alternative; boundary="000000000000a6e890064ec3c7e9"
Archived-At: 
 <https://www.postgresql.org/message-id/CAKFQuwZUSetJvoXewwofTNP5dLfhg7S31NNJhZ%3DG%2BZbOOCZ%3Dug%40mail.gmail.com>
Precedence: bulk

--000000000000a6e890064ec3c7e9
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sunday, April 5, 2026, Fujii Masao <masao.fujii@gmail.com> wrote:

> On Mon, Apr 6, 2026 at 3:10=E2=80=AFAM Maciek Sakrejda <m.sakrejda@gmail.=
com>
> wrote:
> >
> > It makes sense to align these, but I think the existing VACUUM wording
> > is not great. What do you think about something like the attached?
> > Basically, I changed both option descriptions to just be
> >
> >   Prints detailed progress for each table at <literal>INFO</literal>
> level.
> >
> > I think the idea of _progress_ is important to communicate here. The
> > word "report" suggests more detailed information, that comes in a
> > batch after the action is completed.
>
> Referring to it only as "progress" seems like a step backward, doesn't it=
?
> The VERBOSE option reports per-table activity details (e.g., pages to sca=
n,
> buffer usage), not just progress.
>
> Since these details are shown for each table, they can also serve as
> progress
> indicators, but they're more than that.
>
> If that understanding is correct, the existing term "vacuum activity
> report"
> seems more appropriate to me. Thought?
>

How about something like:
=E2=80=9CEnables sending an INFO message to the client (and server log) as =
each
table is processed.  This message contains: etc=E2=80=A6=E2=80=9D

And then let=E2=80=99s tell the user what info they are getting and what it=
 means
(where necessary).

I concur being specific about when these messages arrive, and IMO where,
should be specified.  But losing the detail of =E2=80=9Creport=E2=80=9D is =
not good; but
not sure why we are being vague so suggest we just go all-in on specificity=
.

David J.

--000000000000a6e890064ec3c7e9
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sunday, April 5, 2026, Fujii Masao &lt;<a href=3D"mailto:masao.fujii@gma=
il.com">masao.fujii@gmail.com</a>&gt; wrote:<br><blockquote class=3D"gmail_=
quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1=
ex">On Mon, Apr 6, 2026 at 3:10=E2=80=AFAM Maciek Sakrejda &lt;<a href=3D"m=
ailto:m.sakrejda@gmail.com">m.sakrejda@gmail.com</a>&gt; wrote:<br>
&gt;<br>
&gt; It makes sense to align these, but I think the existing VACUUM wording=
<br>
&gt; is not great. What do you think about something like the attached?<br>
&gt; Basically, I changed both option descriptions to just be<br>
&gt;<br>
&gt;=C2=A0 =C2=A0Prints detailed progress for each table at &lt;literal&gt;=
INFO&lt;/literal&gt; level.<br>
&gt;<br>
&gt; I think the idea of _progress_ is important to communicate here. The<b=
r>
&gt; word &quot;report&quot; suggests more detailed information, that comes=
 in a<br>
&gt; batch after the action is completed.<br>
<br>
Referring to it only as &quot;progress&quot; seems like a step backward, do=
esn&#39;t it?<br>
The VERBOSE option reports per-table activity details (e.g., pages to scan,=
<br>
buffer usage), not just progress.<br>
<br>
Since these details are shown for each table, they can also serve as progre=
ss<br>
indicators, but they&#39;re more than that.<br>
<br>
If that understanding is correct, the existing term &quot;vacuum activity r=
eport&quot;<br>
seems more appropriate to me. Thought?<br>
</blockquote><div><br></div><div>How about something like:</div><div>=E2=80=
=9CEnables sending an INFO message to the client (and server log) as each t=
able is processed.=C2=A0 This message contains: etc=E2=80=A6=E2=80=9D</div>=
<div><br></div><div>And then let=E2=80=99s tell the user what info they are=
 getting and what it means (where necessary).</div><div><br></div><div>I co=
ncur being specific about when these messages arrive, and IMO where, should=
 be specified.=C2=A0 But losing the detail of =E2=80=9Creport=E2=80=9D is n=
ot good; but not sure why we are being vague so suggest we just go all-in o=
n specificity.</div><div><br></div><div>David J.</div><div><br></div>

--000000000000a6e890064ec3c7e9--