Re: A test for replay of regression tests

From: Andres Freund <andres(at)anarazel(dot)de>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Thomas Munro <thomas(dot)munro(at)gmail(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: A test for replay of regression tests
Date: 2021-04-23 17:22:49
Message-ID: 20210423172249.wxzzn3oi35e2bocb@alap3.anarazel.de
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

On 2021-04-23 13:13:15 -0400, Tom Lane wrote:
> contrib/bloom is indeed the only(?) case within contrib, but in my mind
> that's a proxy for what people will be needing to test out-of-core AMs.
> It might not be a test to run by default, but there needs to be a way
> to do it.

Hm. My experience in the past was that the best way to test an external
AM is to run the core regression tests with a different
default_table_access_method. That does require some work of ensuring the
AM is installed and the relevant extension created, which in turn
requires a different test schedule, or hacking template1. So likely a
different test script anyway?

> Also, I suspect that there are bits of GIST/GIN/SPGIST that are not
> well exercised if you don't run the contrib modules that add opclasses
> for those.

Possible - still think it'd be best to get the minimal thing in asap,
and then try to extend further afterwards... Perfect being the enemy of
good and all that.

Greetings,

Andres Freund

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Robert Haas 2021-04-23 17:28:50 Re: pg_amcheck contrib application
Previous Message Tom Lane 2021-04-23 17:13:15 Re: A test for replay of regression tests