Re: Removing useless \. at the end of copy in pgbench

From: Tatsuo Ishii <ishii(at)sraoss(dot)co(dot)jp>
To: tgl(at)sss(dot)pgh(dot)pa(dot)us
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Removing useless \. at the end of copy in pgbench
Date: 2018-07-28 03:03:07
Message-ID: 20180728.120307.1134923282983882586.t-ishii@sraoss.co.jp
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

> Tatsuo Ishii <ishii(at)sraoss(dot)co(dot)jp> writes:
>> While populating pgbench_account table by using COPY FROM STDIN,
>> pgbench sends out "\." at the end of the copy data. However this is
>> only necessary in the version 2 of frontend/backend protocol (i.e. the
>> version 3 protocol does not need it). I think we can safely remove the
>> code to save a few CPU cycle since we only support back to PostgreSQL
>> 9.3 and the version 3 protocol has been supported since 7.4.
>
> What exactly is the benefit of breaking compatibility with old servers?
> Saving a few cycles during initial table population seems like it could
> not be of interest to anyone.

Then we can do protocol version checking to not break backward
compatibility. I believe this is what psql does. So this will bring a
benefit to make our clients more consistent.

Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2018-07-28 03:38:57 Bizarre behavior in libpq's searching of ~/.pgpass
Previous Message David Rowley 2018-07-28 01:54:43 Re: Speeding up INSERTs and UPDATEs to partitioned tables