Horton

Re: Bad canonicalization for dateranges with 'infinity' bounds

Thomas Munro <thomas.munro@gmail.com>

View thread View on postgr.es
From: Thomas Munro <thomas.munro@gmail.com>
To: Laurenz Albe <laurenz.albe@cybertec.at>
Cc: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2019-07-14T03:27:47Z
Lists: pgsql-hackers 
On Sun, Jul 14, 2019 at 12:44 AM Thomas Munro <thomas.munro@gmail.com> wrote:
> Even though !(X || Y) is equivalent to !X && !Y, by my reading of
> range_in(), lower.value can be uninitialised when lower.infinite is
> true, and it's also a bit hard to read IMHO, so I'd probably write
> that as !upper.infinite && !DATE_NOT_FINITE(upper.val) &&
> upper.inclusive.  I don't think it can affect the result but it might
> upset Valgrind or similar.

I take back the bit about reading an uninitialised value (X || Y
doesn't access Y if X is true... duh), but I still think the other way
of putting it is a bit easier to read.  YMMV.

Generally, +1 for this patch.  I'll wait a couple of days for more
feedback to appear.

-- 
Thomas Munro
https://enterprisedb.com

Commits

  1. Fix daterange canonicalization for +/- infinity.