Re: Query not producing expected result

From: Francisco Olarte <folarte(at)peoplecall(dot)com>
To: Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>
Cc: Chuck Martin <clmartin(at)theombudsman(dot)com>, pgsql-general <pgsql-general(at)postgresql(dot)org>
Subject: Re: Query not producing expected result
Date: 2019-05-01 18:04:48
Message-ID: CA+bJJbzPv+Wa1CqhOm8ZtV8ZN6z2=OAF1wR=bB8xm0wXyR9QrA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Adrian:

On Wed, May 1, 2019 at 7:57 PM Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com> wrote:
> You will have to explain further as I am not seeing it:
> test_(postgres)# select '2019-05-01 9:52' <= '2019-05-01 24:00'::timestamp;
> ?column?
> ----------
> t
>
> test_(postgres)# select '2019-05-01 24:00' <= '2019-05-01 24:00'::timestamp;
> ?column?
> ----------
> t

Because you are using two selected examples. The one with 9:52 is ok.

The last one is misleading because you are using a constant for a
particular timestamp in MAY THE SECOND wich can be written to look
like it is in MAY THE FIRST.

Rewrite it as
select '2019-05-02'::timestamp <= '2019-05-01 24:00'::timestamp;

And you'll see and out of range date selected.

This is why <= AND 24:00 are bad and misleading.

You may not have problems with 00:00:00 times, but work a bit billing
phone calls and you'll find about one in 86400 hit it ( more in my
case as traffic distribution is skewed ). Use that kind of condition
and you end up chasing why the monthly report has a dozen less calls
than the sum of the daily ones the billing guys made using excel.

Francisco Olarte.

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Francisco Olarte 2019-05-01 18:09:52 Re: Query not producing expected result
Previous Message Adrian Klaver 2019-05-01 18:04:44 Re: Query not producing expected result