MIME-Version: 1.0
From: Jianghua Yang <yjhjstz@gmail.com>
Date: Tue, 24 Mar 2026 05:31:55 -0700
Message-ID: 
 <CAAZLFmS2OMiwe65gdm-MKgO=LnKatGMSK6JWxhycGN3TWrhbnw@mail.gmail.com>
Subject: BUG: test_bloomfilter error message reports wrong variable and wrong
 format specifier
To: pgsql-hackers@lists.postgresql.org
Content-Type: multipart/mixed; boundary="0000000000006d3a63064dc458a2"
Archived-At: 
 <https://www.postgresql.org/message-id/CAAZLFmS2OMiwe65gdm-MKgO%3DLnKatGMSK6JWxhycGN3TWrhbnw%40mail.gmail.com>
Precedence: bulk

--0000000000006d3a63064dc458a2
Content-Type: multipart/alternative; boundary="0000000000006d3a62064dc458a0"

--0000000000006d3a62064dc458a0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi,

  I found a small bug in
src/test/modules/test_bloomfilter/test_bloomfilter.c, line 128.

  The current code:

  int64   nelements =3D PG_GETARG_INT64(1);
  int     tests =3D PG_GETARG_INT32(3);

  if (tests <=3D 0)
      elog(ERROR, "invalid number of tests: %d", tests);

  if (nelements < 0)
      elog(ERROR, "invalid number of elements: %d", tests);

  The second elog has two issues:

  1. It checks nelements but prints tests. For example, with
  nelements =3D -1 and tests =3D 1, the error message would be
  "invalid number of elements: 1" =E2=80=94 which is misleading when debugg=
ing.
  2. nelements is int64, so the format specifier should be
  INT64_FORMAT rather than %d. Using %d for int64 is
  undefined behavior on 32-bit platforms.

  The fix:

  if (nelements < 0)
      elog(ERROR, "invalid number of elements: " INT64_FORMAT, nelements);

  A patch is attached.

  Regards,
  Jianghua Yang

--0000000000006d3a62064dc458a0
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hi,<br><br>=C2=A0 I found a small bug in src/test/modules/=
test_bloomfilter/test_bloomfilter.c, line 128.<br><br>=C2=A0 The current co=
de:<br><br>=C2=A0 int64 =C2=A0 nelements =3D PG_GETARG_INT64(1);<br>=C2=A0 =
int =C2=A0 =C2=A0 tests =3D PG_GETARG_INT32(3);<br><br>=C2=A0 if (tests &lt=
;=3D 0)<br>=C2=A0 =C2=A0 =C2=A0 elog(ERROR, &quot;invalid number of tests: =
%d&quot;, tests);<br><br>=C2=A0 if (nelements &lt; 0)<br>=C2=A0 =C2=A0 =C2=
=A0 elog(ERROR, &quot;invalid number of elements: %d&quot;, tests);<br><br>=
=C2=A0 The second elog has two issues:<br><br>=C2=A0 1. It checks nelements=
 but prints tests. For example, with<br>=C2=A0 nelements =3D -1 and tests =
=3D 1, the error message would be<br>=C2=A0 &quot;invalid number of element=
s: 1&quot; =E2=80=94 which is misleading when debugging.<br>=C2=A0 2. nelem=
ents is int64, so the format specifier should be<br>=C2=A0 INT64_FORMAT rat=
her than %d. Using %d for int64 is<br>=C2=A0 undefined behavior on 32-bit p=
latforms.<br><br>=C2=A0 The fix:<br><br>=C2=A0 if (nelements &lt; 0)<br>=C2=
=A0 =C2=A0 =C2=A0 elog(ERROR, &quot;invalid number of elements: &quot; INT6=
4_FORMAT, nelements);<br><br>=C2=A0 A patch is attached.<br><br>=C2=A0 Rega=
rds,<br>=C2=A0 Jianghua Yang</div>

--0000000000006d3a62064dc458a0--

--0000000000006d3a63064dc458a2
Content-Type: application/octet-stream;
	name="v1-0001-Fix-wrong-variable-and-format-specifier-in-test_b.patch"
Content-Disposition: attachment;
	filename="v1-0001-Fix-wrong-variable-and-format-specifier-in-test_b.patch"
Content-Transfer-Encoding: base64
Content-ID: <f_mn4lflxf0>
X-Attachment-Id: f_mn4lflxf0
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--0000000000006d3a63064dc458a2--