<!-- doc/src/sgml/release-9.4.sgml -->

<!-- See header comment in release.sgml about typical markup -->

<sect1 id="release-9-4-23">

<title>Release 9.4.23</title>

<formalpara>

<title>Release date:</title>

<para>2019-06-20</para>

</formalpara>

<para>

This release contains a variety of fixes from 9.4.22.

For information about new features in the 9.4 major release, see

<xref linkend="release-9-4">.

</para>

<para>

The <productname>PostgreSQL</productname> community will stop

releasing updates for the 9.4.X release series in February 2020.

Users are encouraged to update to a newer release branch soon.

</para>

<sect2>

<title>Migration to Version 9.4.23</title>

<para>

A dump/restore is not required for those running 9.4.X.

</para>

<para>

However, if you are upgrading from a version earlier than 9.4.18,

see <xref linkend="release-9-4-18">.

</para>

</sect2>

<sect2>

<title>Changes</title>

<itemizedlist>

<listitem>

<!--

Author: Tom Lane <tgl@sss.pgh.pa.us>

Branch: master [e76de8861] 2019-06-12 12:29:39 -0400

Branch: REL_11_STABLE [0b6edb9fb] 2019-06-12 12:29:41 -0400

Branch: REL_10_STABLE [2981e5a61] 2019-06-12 12:29:42 -0400

Branch: REL9_6_STABLE [77d45b790] 2019-06-12 12:29:43 -0400

Branch: REL9_5_STABLE [f7e99de95] 2019-06-12 12:29:45 -0400

Branch: REL9_4_STABLE [9695ecf36] 2019-06-12 12:29:25 -0400

-->

<para>

Fix failure of <command>ALTER TABLE ... ALTER COLUMN TYPE</command>

when the table has a partial exclusion constraint (Tom Lane)

</para>

</listitem>

<listitem>

<!--

Author: Tom Lane <tgl@sss.pgh.pa.us>

Branch: master [3d99a8139] 2019-06-12 19:43:08 -0400

Branch: REL_11_STABLE [f95d8f810] 2019-06-12 19:43:09 -0400

Branch: REL_10_STABLE [30d3df0a7] 2019-06-12 19:43:10 -0400

Branch: REL9_6_STABLE [ad3e61b28] 2019-06-12 19:42:39 -0400

Branch: REL9_5_STABLE [bf612fd3a] 2019-06-12 19:42:39 -0400

Branch: REL9_4_STABLE [8ace51a6b] 2019-06-12 19:42:39 -0400

-->

<para>

Fix incorrect printing of queries with duplicate join names

(Philip Dub&eacute;)

</para>

<para>

This oversight caused a dump/restore failure for views containing

such queries.

</para>

</listitem>

<listitem>

<!--

Author: Tom Lane <tgl@sss.pgh.pa.us>

Branch: master [8a29ed053] 2019-05-12 18:53:38 -0400

Branch: REL_11_STABLE [72ce7acaf] 2019-05-12 18:53:40 -0400

Branch: REL_10_STABLE [940f64792] 2019-05-12 18:53:41 -0400

Branch: REL9_6_STABLE [2a3dbc151] 2019-05-12 18:53:42 -0400

Branch: REL9_5_STABLE [d6de52636] 2019-05-12 18:53:43 -0400

Branch: REL9_4_STABLE [b1ca28e57] 2019-05-12 18:53:45 -0400

-->

<para>

Fix misoptimization of <literal>{1,1}</literal> quantifiers in

regular expressions (Tom Lane)

</para>

<para>

Such quantifiers were treated as no-ops and optimized away;

but the documentation specifies that they impose greediness, or

non-greediness in the case of the non-greedy

variant <literal>{1,1}?</literal>, on the subexpression they're

attached to, and this did not happen. The misbehavior occurred

only if the subexpression contained capturing parentheses or a

back-reference.

</para>

</listitem>

<listitem>

<!--

Author: Tom Lane <tgl@sss.pgh.pa.us>

Branch: master [610747d86] 2019-05-10 14:56:41 -0400

Branch: REL_11_STABLE [803f90ab7] 2019-05-10 14:56:41 -0400

Branch: REL_10_STABLE [3dcf45af5] 2019-05-10 14:56:41 -0400

Branch: REL9_6_STABLE [a73c8caea] 2019-05-10 14:56:41 -0400

Branch: REL9_5_STABLE [91a05390c] 2019-05-10 14:56:41 -0400

Branch: REL9_4_STABLE [b1cde67a4] 2019-05-10 14:56:41 -0400

-->

<para>

Fix race condition in check to see whether a pre-existing shared

memory segment is still in use by a conflicting postmaster (Tom Lane)

</para>

</listitem>

<listitem>

<!--

Author: Andres Freund <andres@anarazel.de>

Branch: master [fff2a7d7b] 2019-06-10 23:34:50 -0700

Branch: REL_11_STABLE [c01556017] 2019-06-10 23:35:38 -0700

Branch: REL_10_STABLE [52ad5fc0a] 2019-06-10 23:36:48 -0700

Branch: REL9_6_STABLE [b8607e15e] 2019-06-10 23:36:55 -0700

Branch: REL9_5_STABLE [483b2d7ae] 2019-06-10 23:36:59 -0700

Branch: REL9_4_STABLE [69f32206d] 2019-06-10 23:37:02 -0700

-->

<para>

Avoid attempting to do database accesses for parameter checking in

processes that are not connected to a specific database (Vignesh C,

Andres Freund)

</para>

<para>

This error could result in failures like <quote>cannot read pg_class

without having selected a database</quote>.

</para>

</listitem>

<listitem>

<!--

Author: Tom Lane <tgl@sss.pgh.pa.us>

Branch: master [23bd3cec6] 2018-09-13 12:36:21 -0400

Branch: REL_11_STABLE [995b4fe0b] 2019-06-14 11:25:13 -0400

Branch: REL_10_STABLE [8de574aa8] 2019-06-14 11:25:13 -0400

Branch: REL9_6_STABLE [75b0f21e1] 2019-06-14 11:25:13 -0400

Branch: REL9_5_STABLE [77dc741a1] 2019-06-14 11:25:13 -0400

Branch: REL9_4_STABLE [37011bcb3] 2019-06-14 11:25:13 -0400

Author: Andrew Gierth <rhodiumtoad@postgresql.org>

Branch: master [e3846a00c] 2019-06-15 18:15:23 +0100

Branch: REL_11_STABLE [7f28fc8e9] 2019-06-15 18:16:43 +0100

Branch: REL_10_STABLE [2913a892e] 2019-06-15 18:18:03 +0100

Branch: REL9_6_STABLE [5f5b6667e] 2019-06-15 18:19:30 +0100

Branch: REL9_5_STABLE [6335f8a99] 2019-06-15 18:21:41 +0100

Branch: REL9_4_STABLE [6c66865f3] 2019-06-15 18:23:16 +0100

-->

<para>

Improve <application>initdb</application>'s handling of multiple

equivalent names for the system time zone (Tom Lane, Andrew Gierth)

</para>

<para>

Make <application>initdb</application> examine

the <filename>/etc/localtime</filename> symbolic link, if that

exists, to break ties between equivalent names for the system time

zone. This makes <application>initdb</application> more likely to

select the time zone name that the user would expect when multiple

identical time zones exist. It will not change the behavior

if <filename>/etc/localtime</filename> is not a symlink to a zone

data file, nor if the time zone is determined from

the <envar>TZ</envar> environment variable.

</para>

<para>

Separately, prefer <literal>UTC</literal> over other spellings of

that time zone, when neither <envar>TZ</envar>

nor <filename>/etc/localtime</filename> provide a hint. This fixes

an annoyance introduced by <application>tzdata</application> 2019a's

change to make the <literal>UCT</literal> and <literal>UTC</literal>

zone names equivalent: <application>initdb</application> was then

preferring <literal>UCT</literal>, which almost nobody wants.

</para>

</listitem>

<listitem>

<!--

Author: Michael Paquier <michael@paquier.xyz>

Branch: master [e51bad8fb] 2019-05-11 13:00:54 +0900

Branch: REL_11_STABLE [e16ab408f] 2019-05-11 13:01:07 +0900

Branch: REL_10_STABLE [c6354e943] 2019-05-11 13:01:12 +0900

Branch: REL9_6_STABLE [0c132d458] 2019-05-11 13:01:19 +0900

Branch: REL9_5_STABLE [e17fba8d8] 2019-05-11 13:01:24 +0900

Branch: REL9_4_STABLE [82ed20e15] 2019-05-11 13:01:29 +0900

-->

<para>

Fix misleading error reports

from <application>reindexdb</application> (Julien Rouhaud)

</para>

</listitem>

<listitem>

<!--

Author: Etsuro Fujita <efujita@postgresql.org>

Branch: master [8b6da83d1] 2019-06-13 17:59:09 +0900

Branch: REL_11_STABLE [214460182] 2019-06-13 17:59:11 +0900

Branch: REL_10_STABLE [0f2b23426] 2019-06-13 17:59:12 +0900

Branch: REL9_6_STABLE [1ade21e7f] 2019-06-13 17:59:13 +0900

Branch: REL9_5_STABLE [eddb79705] 2019-06-13 17:59:15 +0900

Branch: REL9_4_STABLE [1a3d9f62c] 2019-06-13 17:59:17 +0900

-->

<para>

In <filename>contrib/postgres_fdw</filename>, account for possible

data modifications by local <literal>BEFORE ROW UPDATE</literal>

triggers (Shohei Mochizuki)

</para>

<para>

If a trigger modified a column that was otherwise not changed by the

<command>UPDATE</command>, the new value was not transmitted to the

remote server.

</para>

</listitem>

<listitem>

<!--

Author: Noah Misch <noah@leadboat.com>

Branch: master [d02768ddd] 2019-05-12 10:33:05 -0700

Branch: REL_11_STABLE [4ec14e5aa] 2019-05-12 10:33:08 -0700

Branch: REL_10_STABLE [409f5303c] 2019-05-12 10:33:08 -0700

Branch: REL9_6_STABLE [ff7555c4f] 2019-05-12 10:33:36 -0700

Branch: REL9_5_STABLE [9dc65bcf9] 2019-05-12 10:34:22 -0700

Branch: REL9_4_STABLE [08e3fd96f] 2019-05-12 10:34:26 -0700

-->

<para>

On Windows, avoid failure when the database encoding is set to

SQL_ASCII and we attempt to log a non-ASCII string (Noah Misch)

</para>

<para>

The code had been assuming that such strings must be in UTF-8, and

would throw an error if they didn't appear to be validly encoded.

Now, just transmit the untranslated bytes to the log.

</para>

</listitem>

<listitem>

<!--

Author: Tom Lane <tgl@sss.pgh.pa.us>

Branch: master [3f61b3205] 2019-05-31 12:34:54 -0400

Branch: REL_11_STABLE [312017fcc] 2019-05-31 12:34:54 -0400

Branch: REL_10_STABLE [683c17b30] 2019-05-31 12:34:54 -0400

Branch: REL9_6_STABLE [c332c94ad] 2019-05-31 12:34:55 -0400

Branch: REL9_5_STABLE [da53e4f93] 2019-05-31 12:34:55 -0400

Branch: REL9_4_STABLE [db5d99916] 2019-05-31 12:34:55 -0400

-->

<para>

Make <application>PL/pgSQL</application>'s header files C++-safe

(George Tarasov)

</para>

</listitem>

</itemizedlist>

</sect2>

</sect1>

<sect1 id="release-9-4-22">

<title>Release 9.4.22</title>