Re: Unbreak mbregression test

From: Tatsuo Ishii <ishii(at)postgresql(dot)org>
To: tgl(at)sss(dot)pgh(dot)pa(dot)us
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Unbreak mbregression test
Date: 2016-01-25 23:09:10
Message-ID: 20160126.080910.2203023754563641852.t-ishii@sraoss.co.jp
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

> Tatsuo Ishii <ishii(at)postgresql(dot)org> writes:
>> The multi-byte regression tests (src/test/regress/mb) have been broken
>> for sometime due to a warning message regarding hash index usage "hash
>> indexes are not WAL-logged and their use is discouraged" (the messages
>> are different from version to version).
>
>> Attached is a patch to fix the problem for master branch and
>> REL9_5_STABLE.
>
> Isn't this simply a reversion of efc1610b64b04e7cf08cc1d6c608ede8b7d5ff07?
> What led you to make that change in the first place?

Ouch, I forgot about the commit efc1610b64b04e7cf08cc1d6c608ede8b7d5ff07.
In short, the commit was wrong as Tom Lane pointed out later on:
----------------------------------------------------------------
Subject: Re: [COMMITTERS] pgsql: Fix broken multibyte regression tests.
From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Tatsuo Ishii <ishii(at)postgresql(dot)org>
cc: pgsql-committers(at)postgresql(dot)org
Date: Sat, 28 Nov 2015 13:55:09 -0500
Comments: In-reply-to Tatsuo Ishii <ishii(at)postgresql(dot)org> message dated "Sun, 09 Aug 2015 02:09:25 -0000"

Tatsuo Ishii <ishii(at)postgresql(dot)org> writes:
> Fix broken multibyte regression tests.
> commit 9043Fe390f4f0b4586cfe59cbd22314b9c3e2957 broke multibyte
> regression tests because the commit removes the warning message when
> temporary hash indexes is created, which has been added by commit
> 07af523870bcfe930134054febd3a6a114942e5b.

> Back patched to 9.5 stable tree.

AFAICT this patch was incorrect and should be reverted, because the
src/test/mb tests all fail for me, in both HEAD and 9.5, as a consequence
of getting hash-index WARNINGs that are not in the expected-files. The
commit you mention (which is 9043ef390f4f0b4586cfe59cbd22314b9c3e2957 not
what's cited in this commit message) disabled the WARNING for temporary
and unlogged hash indexes, but the indexes created by the test scripts
are neither.

Please recheck it.

regards, tom lane
----------------------------------------------------------------
He is absolutely right here and I should have noticed earlier. So,

> Isn't this simply a reversion of efc1610b64b04e7cf08cc1d6c608ede8b7d5ff07?
> What led you to make that change in the first place?

Yes, the patch is for reverting efc1610b64b04e7cf08cc1d6c608ede8b7d5ff07.

My appolgy for the wrong commit.

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

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tatsuo Ishii 2016-01-26 00:10:34 Re: Unbreak mbregression test
Previous Message Pavel Stehule 2016-01-25 21:51:57 Re: custom function for converting human readable sizes to bytes