Re: Define variables in the approprieate scope - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Define variables in the approprieate scope
Date
Msg-id 15375.1585017694@sss.pgh.pa.us
Whole thread Raw
In response to Re: Define variables in the approprieate scope  (Michael Paquier <michael@paquier.xyz>)
List pgsql-hackers
Michael Paquier <michael@paquier.xyz> writes:
> On Mon, Mar 23, 2020 at 08:50:55PM -0400, Bruce Momjian wrote:
>> I am fine with either usage, frankly.  I was just pointing out what
>> might be the benefit of the current coding.

> Personal opinion here.  I tend to prefer putting variable declarations
> into the inner portions because it makes it easier to reason about the
> code, though I agree that this concept does not need to be applied all
> the time.

My vote is to not make this sort of change until there's another
reason to touch the code in question.  All changes create hazards for
back-patching, and I don't think this change is worth it on its own.
But if there are going to be diffs in the immediate vicinity anyway,
then sure.

(I'm feeling a bit sensitized to this, perhaps, because of recent
unpleasant experience with back-patching b4570d33a.  That didn't touch
very much code, and the functions in question seemed like fairly stagnant
backwaters of the code base, so it should not have been painful to
back-patch ... but it was, because of assorted often-cosmetic changes
in said code.)

            regards, tom lane



pgsql-hackers by date:

Previous
From: Fujii Masao
Date:
Subject: Re: Wait event that should be reported while waiting for WALarchiving to finish
Next
From: Thomas Munro
Date:
Subject: Re: weird hash plan cost, starting with pg10