Date: Thu, 26 Mar 2020 19:49:51 -0300
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Justin Pryzby <pryzby@telsasoft.com>
Cc: Masahiko Sawada <masahiko.sawada@2ndquadrant.com>,
	Amit Kapila <amit.kapila16@gmail.com>,
	Andres Freund <andres@anarazel.de>,
	Michael Paquier <michael@paquier.xyz>, pgsql-hackers@postgresql.org
Subject: Re: error context for vacuum to include block number
Message-ID: <20200326224951.GA20085@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20200326221752.GR17431@telsasoft.com>
User-Agent: Mutt/1.10.1 (2018-07-13)
Precedence: bulk

On 2020-Mar-26, Justin Pryzby wrote:

> On Thu, Mar 26, 2020 at 10:04:57AM -0500, Justin Pryzby wrote:

> > And I think you're right: we only save state when the calling function has a
> > indname=NULL, so we never "put back" a non-NULL indname.  We go from having a
> > indname=NULL at lazy_scan_heap to not not-NULL at lazy_vacuum_index, and never
> > the other way around.
> 
> I removed the free_oldindname argument.

Hah, I was wondering about that free_oldindname business this morning as
well.

> > ... So once we've "reverted back", 1) the pointer is null; and, 2)
> > the callback function doesn't access it for the previous/reverted
> > phase anyway.

BTW I'm pretty sure this "revert back" phrasing is not good English --
you should just use "revert".  Maybe get some native speaker's opinion
on it.

And speaking of language, I find the particle "cbarg" rather very ugly,
and it's *everywhere* -- function name, function argument, local
variable, enum values, enum name.  It even spread to the typedefs.list
file!  Is this a new virus???  Put some soap in it!  Can't we use "info"
or "state" or something similar, less infectious, instead?

Thanks

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services