Feedback-ID: ia2694551:Fastmail
User-Agent: Cyrus-JMAP/3.7.0-alpha0-692-gb287c361f5-fm-20220603.003-gb287c361
Mime-Version: 1.0
Message-Id: <cbddf6bf-aa06-47fa-ab1d-d73a58d4e89f@www.fastmail.com>
In-Reply-To: <20220610.152544.1597540405935332165.horikyota.ntt@gmail.com>
References: 
 <CA+mi_8bvD0_CW3sumgwPvWdNzXY32itoG_16tDYRu_1S2gV2iw@mail.gmail.com>
 <202206081508.pwq4ne3df2si@alvherre.pgsql>
 <20220610.152544.1597540405935332165.horikyota.ntt@gmail.com>
Date: Mon, 13 Jun 2022 13:09:48 +0200
From: =?UTF-8?Q?=C3=81lvaro_Herrera?= <alvherre@alvh.no-ip.org>
To: "Kyotaro Horiguchi" <horikyota.ntt@gmail.com>
Cc: daniele.varrazzo@gmail.com,
 "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>
Subject: Re: Using PQexecQuery in pipeline mode produces unexpected Close
 messages
Content-Type: multipart/alternative;
 boundary=f5f15fe2dc004050bc54073a72f0c057
Archived-At: 
 <https://www.postgresql.org/message-id/cbddf6bf-aa06-47fa-ab1d-d73a58d4e89f%40www.fastmail.com>
Precedence: bulk

--f5f15fe2dc004050bc54073a72f0c057
Content-Type: text/plain

On Fri, Jun 10, 2022, at 8:25 AM, Kyotaro Horiguchi wrote:
> 
> The current implement of PQsendQueryInternal looks like the result of
> a misunderstanding of the doc.  In the regression tests, that path is
> excercised only for an error case, where no CloseComplete comes.
> 
> The attached adds a test for the normal-path of pipelined
> PQsendQuery() to simple_pipeline test then modifies that function not
> to send Close message. Without the fix, the test fails by "unexpected
> notice" even if the trace matches the "expected" content.

Hah, the patch I wrote is almost identical to yours, down to the notice processor counting the number of notices received.  The only difference is that I put my test in pipeline_abort.

Sadly, it looks like I won't be able to get this patched pushed for 14.4.


--f5f15fe2dc004050bc54073a72f0c057
Content-Type: text/html
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE html><html><head><title></title><style type=3D"text/css">p.Mso=
Normal,p.MsoNoSpacing{margin:0}</style></head><body><div>On Fri, Jun 10,=
 2022, at 8:25 AM, Kyotaro Horiguchi wrote:<br></div><blockquote type=3D=
"cite" id=3D"qt" style=3D""><div><br></div><div>The current implement of=
 PQsendQueryInternal looks like the result of<br></div><div>a misunderst=
anding of the doc.&nbsp; In the regression tests, that path is<br></div>=
<div>excercised only for an error case, where no CloseComplete comes.<br=
></div><div><br></div><div>The attached adds a test for the normal-path =
of pipelined<br></div><div>PQsendQuery() to simple_pipeline test then mo=
difies that function not<br></div><div>to send Close message. Without th=
e fix, the test fails by "unexpected<br></div><div>notice" even if the t=
race matches the "expected" content.<br></div></blockquote><div><br></di=
v><div><div>Hah, the patch I wrote is almost identical to yours, down to=
 the notice processor counting the number of notices received.&nbsp; The=
 only difference is that I put my test in pipeline_abort.<br></div><div>=
<br></div><div>Sadly, it looks like I won't be able to get this patched =
pushed for 14.4.</div><div><br></div></div><div><br></div></body></html>
--f5f15fe2dc004050bc54073a72f0c057--