<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<HTML
><HEAD
><TITLE
>Release 9.1</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.79"><LINK
REV="MADE"
HREF="mailto:pgsql-docs@postgresql.org"><LINK
REL="HOME"
TITLE="PostgreSQL 9.3.9 Documentation"
HREF="index.html"><LINK
REL="UP"
TITLE="Release Notes"
HREF="release.html"><LINK
REL="PREVIOUS"
TITLE="Release 9.1.1"
HREF="release-9-1-1.html"><LINK
REL="NEXT"
TITLE="Release 9.0.22"
HREF="release-9-0-22.html"><LINK
REL="STYLESHEET"
TYPE="text/css"
HREF="stylesheet.css"><META
HTTP-EQUIV="Content-Type"
CONTENT="text/html; charset=ISO-8859-1"><META
NAME="creation"
CONTENT="2015-06-13T20:07:22"></HEAD
><BODY
CLASS="SECT1"
><DIV
CLASS="NAVHEADER"
><TABLE
SUMMARY="Header navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TH
COLSPAN="5"
ALIGN="center"
VALIGN="bottom"
><A
HREF="index.html"
>PostgreSQL 9.3.9 Documentation</A
></TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
TITLE="Release 9.1.1"
HREF="release-9-1-1.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="release.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="60%"
ALIGN="center"
VALIGN="bottom"
>Appendix E. Release Notes</TD
><TD
WIDTH="20%"
ALIGN="right"
VALIGN="top"
><A
TITLE="Release 9.0.22"
HREF="release-9-0-22.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="RELEASE-9-1"
>E.43. Release 9.1</A
></H1
><DIV
CLASS="NOTE"
><BLOCKQUOTE
CLASS="NOTE"
><P
><B
>Release Date: </B
>2011-09-12</P
></BLOCKQUOTE
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN123122"
>E.43.1. Overview</A
></H2
><P
> This release shows <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> moving beyond the
traditional relational-database feature set with new, ground-breaking
functionality that is unique to <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>.
The streaming replication feature introduced in release 9.0 is
significantly enhanced by adding a synchronous-replication option,
streaming backups, and monitoring improvements.
Major enhancements include:
</P
><P
></P
><UL
><LI
><P
> Allow <A
HREF="warm-standby.html#SYNCHRONOUS-REPLICATION"
>synchronous
replication</A
>
</P
></LI
><LI
><P
> Add support for <A
HREF="sql-createforeigntable.html"
>foreign
tables</A
>
</P
></LI
><LI
><P
> Add per-column <A
HREF="collation.html"
>collation</A
> support
</P
></LI
><LI
><P
> Add <A
HREF="extend-extensions.html"
>extensions</A
> which
simplify packaging of additions to <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>
</P
></LI
><LI
><P
> Add a true <A
HREF="transaction-iso.html#XACT-SERIALIZABLE"
>serializable isolation level</A
>
</P
></LI
><LI
><P
> Support unlogged tables using the <TT
CLASS="LITERAL"
>UNLOGGED</TT
>
option in <A
HREF="sql-createtable.html"
><TT
CLASS="COMMAND"
>CREATE
TABLE</TT
></A
>
</P
></LI
><LI
><P
> Allow data-modification commands
(<TT
CLASS="COMMAND"
>INSERT</TT
>/<TT
CLASS="COMMAND"
>UPDATE</TT
>/<TT
CLASS="COMMAND"
>DELETE</TT
>) in
<A
HREF="queries-with.html"
><TT
CLASS="LITERAL"
>WITH</TT
></A
> clauses
</P
></LI
><LI
><P
> Add nearest-neighbor (order-by-operator) searching to <A
HREF="gist.html"
><ACRONYM
CLASS="ACRONYM"
>GiST</ACRONYM
> indexes</A
>
</P
></LI
><LI
><P
> Add a <A
HREF="sql-security-label.html"
><TT
CLASS="COMMAND"
>SECURITY
LABEL</TT
></A
> command and support for
<A
HREF="sepgsql.html"
><ACRONYM
CLASS="ACRONYM"
>SELinux</ACRONYM
> permissions control</A
>
</P
></LI
><LI
><P
> Update the <A
HREF="plpython.html"
>PL/Python</A
> server-side
language
</P
></LI
></UL
><P
> The above items are explained in more detail in the sections below.
</P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN123170"
>E.43.2. Migration to Version 9.1</A
></H2
><P
> A dump/restore using <SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
>,
or use of <SPAN
CLASS="APPLICATION"
>pg_upgrade</SPAN
>, is required
for those wishing to migrate data from any previous
release.
</P
><P
> Version 9.1 contains a number of changes that may affect compatibility
with previous releases. Observe the following incompatibilities:
</P
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123176"
>E.43.2.1. Strings</A
></H3
><P
></P
><UL
><LI
><P
> Change the default value of <A
HREF="runtime-config-compatible.html#GUC-STANDARD-CONFORMING-STRINGS"
><TT
CLASS="VARNAME"
>standard_conforming_strings</TT
></A
>
to on (Robert Haas)
</P
><P
> By default, backslashes are now ordinary characters in string literals,
not escape characters. This change removes a long-standing
incompatibility with the SQL standard. <A
HREF="runtime-config-compatible.html#GUC-ESCAPE-STRING-WARNING"
><TT
CLASS="VARNAME"
>escape_string_warning</TT
></A
>
has produced warnings about this usage for years. <TT
CLASS="LITERAL"
>E''</TT
>
strings are the proper way to embed backslash escapes in strings and are
unaffected by this change.
</P
><DIV
CLASS="WARNING"
><P
></P
><TABLE
CLASS="WARNING"
BORDER="1"
WIDTH="90%"
><TR
><TD
ALIGN="CENTER"
><B
>Warning</B
></TD
></TR
><TR
><TD
ALIGN="LEFT"
><P
> This change can break applications that are not expecting it and
do their own string escaping according to the old rules. The
consequences could be as severe as introducing SQL-injection security
holes. Be sure to test applications that are exposed to untrusted
input, to ensure that they correctly handle single quotes and
backslashes in text strings.
</P
></TD
></TR
></TABLE
></DIV
></LI
></UL
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123189"
>E.43.2.2. Casting</A
></H3
><P
></P
><UL
><LI
><P
> Disallow function-style and attribute-style data type casts for
composite types (Tom Lane)
</P
><P
> For example, disallow
<TT
CLASS="LITERAL"
><TT
CLASS="REPLACEABLE"
><I
>composite_value</I
></TT
>.text</TT
> and
<TT
CLASS="LITERAL"
>text(<TT
CLASS="REPLACEABLE"
><I
>composite_value</I
></TT
>)</TT
>.
Unintentional uses of this syntax have frequently resulted in bug
reports; although it was not a bug, it seems better to go back to
rejecting such expressions.
The <TT
CLASS="LITERAL"
>CAST</TT
> and <TT
CLASS="LITERAL"
>::</TT
> syntaxes are still available
for use when a cast of an entire composite value is actually intended.
</P
></LI
><LI
><P
> Tighten casting checks for domains based on arrays (Tom Lane)
</P
><P
> When a domain is based on an array type, it is allowed to <SPAN
CLASS="QUOTE"
>"look
through"</SPAN
> the domain type to access the array elements, including
subscripting the domain value to fetch or assign an element.
Assignment to an element of such a domain value, for instance via
<TT
CLASS="LITERAL"
>UPDATE ... SET domaincol[5] = ...</TT
>, will now result in
rechecking the domain type's constraints, whereas before the checks
were skipped.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123206"
>E.43.2.3. Arrays</A
></H3
><P
></P
><UL
><LI
><P
> Change <A
HREF="functions-array.html#ARRAY-FUNCTIONS-TABLE"
><CODE
CLASS="FUNCTION"
>string_to_array()</CODE
></A
>
to return an empty array for a zero-length string (Pavel
Stehule)
</P
><P
> Previously this returned a null value.
</P
></LI
><LI
><P
> Change <A
HREF="functions-array.html#ARRAY-FUNCTIONS-TABLE"
><CODE
CLASS="FUNCTION"
>string_to_array()</CODE
></A
>
so a <TT
CLASS="LITERAL"
>NULL</TT
> separator splits the string into characters
(Pavel Stehule)
</P
><P
> Previously this returned a null value.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123220"
>E.43.2.4. Object Modification</A
></H3
><P
></P
><UL
><LI
><P
> Fix improper checks for before/after triggers (Tom Lane)
</P
><P
> Triggers can now be fired in three cases: <TT
CLASS="LITERAL"
>BEFORE</TT
>,
<TT
CLASS="LITERAL"
>AFTER</TT
>, or <TT
CLASS="LITERAL"
>INSTEAD OF</TT
> some action.
Trigger function authors should verify that their logic behaves
sanely in all three cases.
</P
></LI
><LI
><P
> Require superuser or <TT
CLASS="LITERAL"
>CREATEROLE</TT
> permissions in order to
set comments on roles (Tom Lane)
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123232"
>E.43.2.5. Server Settings</A
></H3
><P
></P
><UL
><LI
><P
> Change <A
HREF="functions-admin.html#FUNCTIONS-RECOVERY-INFO-TABLE"
><CODE
CLASS="FUNCTION"
>pg_last_xlog_receive_location()</CODE
></A
>
so it never moves backwards (Fujii Masao)
</P
><P
> Previously, the value of <CODE
CLASS="FUNCTION"
>pg_last_xlog_receive_location()</CODE
>
could move backward when streaming replication is restarted.
</P
></LI
><LI
><P
> Have logging of replication connections honor <A
HREF="runtime-config-logging.html#GUC-LOG-CONNECTIONS"
><TT
CLASS="VARNAME"
>log_connections</TT
></A
>
(Magnus Hagander)
</P
><P
> Previously, replication connections were always logged.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123246"
>E.43.2.6. <A
HREF="plpgsql.html"
>PL/pgSQL</A
> Server-Side Language</A
></H3
><P
></P
><UL
><LI
><P
> Change PL/pgSQL's <TT
CLASS="LITERAL"
>RAISE</TT
> command without parameters
to be catchable by the attached exception block (Piyush Newe)
</P
><P
> Previously <TT
CLASS="LITERAL"
>RAISE</TT
> in a code block was always scoped to
an attached exception block, so it was uncatchable at the same
scope.
</P
></LI
><LI
><P
> Adjust PL/pgSQL's error line numbering code to be consistent
with other PLs (Pavel Stehule)
</P
><P
> Previously, PL/pgSQL would ignore (not count) an empty line at the
start of the function body. Since this was inconsistent with all
other languages, the special case was removed.
</P
></LI
><LI
><P
> Make PL/pgSQL complain about conflicting IN and OUT parameter names
(Tom Lane)
</P
><P
> Formerly, the collision was not detected, and the name would just
silently refer to only the OUT parameter.
</P
></LI
><LI
><P
> Type modifiers of PL/pgSQL variables are now visible to the SQL parser
(Tom Lane)
</P
><P
> A type modifier (such as a varchar length limit) attached to a PL/pgSQL
variable was formerly enforced during assignments, but was ignored for
all other purposes. Such variables will now behave more like table
columns declared with the same modifier. This is not expected to make
any visible difference in most cases, but it could result in subtle
changes for some SQL commands issued by PL/pgSQL functions.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123264"
>E.43.2.7. Contrib</A
></H3
><P
></P
><UL
><LI
><P
> All contrib modules are now installed with <A
HREF="sql-createextension.html"
><TT
CLASS="COMMAND"
>CREATE EXTENSION</TT
></A
>
rather than by manually invoking their SQL scripts
(Dimitri Fontaine, Tom Lane)
</P
><P
> To update an existing database containing the 9.0 version of a contrib
module, use <TT
CLASS="LITERAL"
>CREATE EXTENSION ... FROM unpackaged</TT
>
to wrap the existing contrib module's objects into an extension. When
updating from a pre-9.0 version, drop the contrib module's objects
using its old uninstall script, then use <TT
CLASS="LITERAL"
>CREATE EXTENSION</TT
>.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123274"
>E.43.2.8. Other Incompatibilities</A
></H3
><P
></P
><UL
><LI
><P
> Make <A
HREF="monitoring-stats.html#MONITORING-STATS-FUNCS-TABLE"
><CODE
CLASS="FUNCTION"
>pg_stat_reset()</CODE
></A
>
reset all database-level statistics (Tomas Vondra)
</P
><P
> Some <TT
CLASS="STRUCTNAME"
>pg_stat_database</TT
> counters were not being reset.
</P
></LI
><LI
><P
> Fix some <A
HREF="infoschema-triggers.html"
><TT
CLASS="STRUCTNAME"
>information_schema.triggers</TT
></A
>
column names to match the new SQL-standard names (Dean Rasheed)
</P
></LI
><LI
><P
> Treat <SPAN
CLASS="APPLICATION"
>ECPG</SPAN
> cursor names as case-insensitive
(Zoltan Boszormenyi)
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN123290"
>E.43.3. Changes</A
></H2
><P
> Below you will find a detailed account of the changes between
<SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> 9.1 and the previous major
release.
</P
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123294"
>E.43.3.1. Server</A
></H3
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123296"
>E.43.3.1.1. Performance</A
></H4
><P
></P
><UL
><LI
><P
> Support unlogged tables using the <TT
CLASS="LITERAL"
>UNLOGGED</TT
>
option in <A
HREF="sql-createtable.html"
><TT
CLASS="COMMAND"
>CREATE
TABLE</TT
></A
> (Robert Haas)
</P
><P
> Such tables provide better update performance than regular tables,
but are not crash-safe: their contents are automatically cleared in
case of a server crash. Their contents do not propagate to
replication slaves, either.
</P
></LI
><LI
><P
> Allow <TT
CLASS="LITERAL"
>FULL OUTER JOIN</TT
> to be implemented as a
hash join, and allow either side of a <TT
CLASS="LITERAL"
>LEFT OUTER JOIN</TT
>
or <TT
CLASS="LITERAL"
>RIGHT OUTER JOIN</TT
> to be hashed (Tom Lane)
</P
><P
> Previously <TT
CLASS="LITERAL"
>FULL OUTER JOIN</TT
> could only be
implemented as a merge join, and <TT
CLASS="LITERAL"
>LEFT OUTER JOIN</TT
>
and <TT
CLASS="LITERAL"
>RIGHT OUTER JOIN</TT
> could hash only the nullable
side of the join. These changes provide additional query optimization
possibilities.
</P
></LI
><LI
><P
> Merge duplicate fsync requests (Robert Haas, Greg Smith)
</P
><P
> This greatly improves performance under heavy write loads.
</P
></LI
><LI
><P
> Improve performance of <A
HREF="runtime-config-wal.html#GUC-COMMIT-SIBLINGS"
><TT
CLASS="VARNAME"
>commit_siblings</TT
></A
>
(Greg Smith)
</P
><P
> This allows the use of <TT
CLASS="VARNAME"
>commit_siblings</TT
> with
less overhead.
</P
></LI
><LI
><P
> Reduce the memory requirement for large ispell dictionaries
(Pavel Stehule, Tom Lane)
</P
></LI
><LI
><P
> Avoid leaving data files open after <SPAN
CLASS="QUOTE"
>"blind writes"</SPAN
>
(Alvaro Herrera)
</P
><P
> This fixes scenarios in which backends might hold files open long
after they were deleted, preventing the kernel from reclaiming
disk space.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123329"
>E.43.3.1.2. Optimizer</A
></H4
><P
></P
><UL
><LI
><P
> Allow inheritance table scans to return meaningfully-sorted
results (Greg Stark, Hans-Jurgen Schonig, Robert Haas, Tom Lane)
</P
><P
> This allows better optimization of queries that use <TT
CLASS="LITERAL"
>ORDER
BY</TT
>, <TT
CLASS="LITERAL"
>LIMIT</TT
>, or <TT
CLASS="LITERAL"
>MIN</TT
>/<TT
CLASS="LITERAL"
>MAX</TT
> with
inherited tables.
</P
></LI
><LI
><P
> Improve GIN index scan cost estimation (Teodor Sigaev)
</P
></LI
><LI
><P
> Improve cost estimation for aggregates and window functions (Tom Lane)
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123343"
>E.43.3.1.3. Authentication</A
></H4
><P
></P
><UL
><LI
><P
> Support host names and host suffixes
(e.g. <TT
CLASS="LITERAL"
>.example.com</TT
>) in <A
HREF="auth-pg-hba-conf.html"
><TT
CLASS="FILENAME"
>pg_hba.conf</TT
></A
>
(Peter Eisentraut)
</P
><P
> Previously only host <ACRONYM
CLASS="ACRONYM"
>IP</ACRONYM
> addresses and <ACRONYM
CLASS="ACRONYM"
>CIDR</ACRONYM
>
values were supported.
</P
></LI
><LI
><P
> Support the key word <TT
CLASS="LITERAL"
>all</TT
> in the host column of <A
HREF="auth-pg-hba-conf.html"
><TT
CLASS="FILENAME"
>pg_hba.conf</TT
></A
>
(Peter Eisentraut)
</P
><P
> Previously people used <TT
CLASS="LITERAL"
>0.0.0.0/0</TT
> or <TT
CLASS="LITERAL"
>::/0</TT
>
for this.
</P
></LI
><LI
><P
> Reject <TT
CLASS="LITERAL"
>local</TT
> lines in <A
HREF="auth-pg-hba-conf.html"
><TT
CLASS="FILENAME"
>pg_hba.conf</TT
></A
>
on platforms that don't support Unix-socket connections
(Magnus Hagander)
</P
><P
> Formerly, such lines were silently ignored, which could be surprising.
This makes the behavior more like other unsupported cases.
</P
></LI
><LI
><P
> Allow <A
HREF="auth-methods.html#GSSAPI-AUTH"
><ACRONYM
CLASS="ACRONYM"
>GSSAPI</ACRONYM
></A
>
to be used to authenticate to servers via <A
HREF="auth-methods.html#SSPI-AUTH"
><ACRONYM
CLASS="ACRONYM"
>SSPI</ACRONYM
></A
> (Christian Ullrich)
</P
><P
> Specifically this allows Unix-based <ACRONYM
CLASS="ACRONYM"
>GSSAPI</ACRONYM
> clients
to do <ACRONYM
CLASS="ACRONYM"
>SSPI</ACRONYM
> authentication with Windows servers.
</P
></LI
><LI
><P
> <A
HREF="auth-methods.html#AUTH-IDENT"
><TT
CLASS="LITERAL"
>ident</TT
></A
>
authentication over local sockets is now known as
<A
HREF="auth-methods.html#AUTH-PEER"
><TT
CLASS="LITERAL"
>peer</TT
></A
>
(Magnus Hagander)
</P
><P
> The old term is still accepted for backward compatibility, but since
the two methods are fundamentally different, it seemed better to adopt
different names for them.
</P
></LI
><LI
><P
> Rewrite <A
HREF="auth-methods.html#AUTH-PEER"
><ACRONYM
CLASS="ACRONYM"
>peer</ACRONYM
></A
>
authentication to avoid use of credential control messages (Tom Lane)
</P
><P
> This change makes the peer authentication code simpler and
better-performing. However, it requires the platform to provide the
<CODE
CLASS="FUNCTION"
>getpeereid</CODE
> function or an equivalent socket operation.
So far as is known, the only platform for which peer authentication
worked before and now will not is pre-5.0 NetBSD.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123390"
>E.43.3.1.4. Monitoring</A
></H4
><P
></P
><UL
><LI
><P
> Add details to the logging of restartpoints and checkpoints,
which is controlled by <A
HREF="runtime-config-logging.html#GUC-LOG-CHECKPOINTS"
><TT
CLASS="VARNAME"
>log_checkpoints</TT
></A
>
(Fujii Masao, Greg Smith)
</P
><P
> New details include <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> file and sync activity.
</P
></LI
><LI
><P
> Add <A
HREF="runtime-config-logging.html#GUC-LOG-FILE-MODE"
><TT
CLASS="VARNAME"
>log_file_mode</TT
></A
>
which controls the permissions on log files created by the
logging collector (Martin Pihlak)
</P
></LI
><LI
><P
> Reduce the default maximum line length for <SPAN
CLASS="APPLICATION"
>syslog</SPAN
>
logging to 900 bytes plus prefixes (Noah Misch)
</P
><P
> This avoids truncation of long log lines on syslog implementations
that have a 1KB length limit, rather than the more common 2KB.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123407"
>E.43.3.1.5. Statistical Views</A
></H4
><P
></P
><UL
><LI
><P
> Add <TT
CLASS="STRUCTFIELD"
>client_hostname</TT
> column to <A
HREF="monitoring-stats.html#MONITORING-STATS-VIEWS-TABLE"
><TT
CLASS="STRUCTNAME"
>pg_stat_activity</TT
></A
>
(Peter Eisentraut)
</P
><P
> Previously only the client address was reported.
</P
></LI
><LI
><P
> Add <A
HREF="monitoring-stats.html#MONITORING-STATS-VIEWS-TABLE"
><TT
CLASS="STRUCTNAME"
>pg_stat_xact_*</TT
></A
>
statistics functions and views (Joel Jacobson)
</P
><P
> These are like the database-wide statistics counter views, but
reflect counts for only the current transaction.
</P
></LI
><LI
><P
> Add time of last reset in database-level and background writer
statistics views (Tomas Vondra)
</P
></LI
><LI
><P
> Add columns showing the number of vacuum and analyze operations
in <A
HREF="monitoring-stats.html#MONITORING-STATS-VIEWS-TABLE"
><TT
CLASS="STRUCTNAME"
>pg_stat_*_tables</TT
></A
>
views (Magnus Hagander)
</P
></LI
><LI
><P
> Add <TT
CLASS="STRUCTFIELD"
>buffers_backend_fsync</TT
> column to <A
HREF="monitoring-stats.html#MONITORING-STATS-VIEWS-TABLE"
><TT
CLASS="STRUCTNAME"
>pg_stat_bgwriter</TT
></A
>
(Greg Smith)
</P
><P
> This new column counts the number of times a backend fsyncs a
buffer.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123433"
>E.43.3.1.6. Server Settings</A
></H4
><P
></P
><UL
><LI
><P
> Provide auto-tuning of <A
HREF="runtime-config-wal.html#GUC-WAL-BUFFERS"
><TT
CLASS="VARNAME"
>wal_buffers</TT
></A
> (Greg
Smith)
</P
><P
> By default, the value of <TT
CLASS="VARNAME"
>wal_buffers</TT
> is now chosen
automatically based on the value of <TT
CLASS="VARNAME"
>shared_buffers</TT
>.
</P
></LI
><LI
><P
> Increase the maximum values for
<A
HREF="runtime-config-locks.html#GUC-DEADLOCK-TIMEOUT"
><TT
CLASS="VARNAME"
>deadlock_timeout</TT
></A
>,
<A
HREF="runtime-config-logging.html#GUC-LOG-MIN-DURATION-STATEMENT"
><TT
CLASS="VARNAME"
>log_min_duration_statement</TT
></A
>, and
<A
HREF="runtime-config-autovacuum.html#GUC-LOG-AUTOVACUUM-MIN-DURATION"
><TT
CLASS="VARNAME"
>log_autovacuum_min_duration</TT
></A
>
(Peter Eisentraut)
</P
><P
> The maximum value for each of these parameters was previously
only about 35 minutes. Much larger values are now allowed.
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123452"
>E.43.3.2. Replication and Recovery</A
></H3
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123454"
>E.43.3.2.1. Streaming Replication and Continuous Archiving</A
></H4
><P
></P
><UL
><LI
><P
> Allow <A
HREF="warm-standby.html#SYNCHRONOUS-REPLICATION"
>synchronous
replication</A
> (Simon Riggs, Fujii Masao)
</P
><P
> This allows the primary server to wait for a standby to write a
transaction's information to disk before acknowledging the commit.
One standby at a time can take the role of the synchronous standby,
as controlled by the
<A
HREF="runtime-config-replication.html#GUC-SYNCHRONOUS-STANDBY-NAMES"
><TT
CLASS="VARNAME"
>synchronous_standby_names</TT
></A
>
setting. Synchronous replication can be enabled or disabled on a
per-transaction basis using the
<A
HREF="runtime-config-wal.html#GUC-SYNCHRONOUS-COMMIT"
><TT
CLASS="VARNAME"
>synchronous_commit</TT
></A
>
setting.
</P
></LI
><LI
><P
> Add protocol support for sending file system backups to standby servers
using the streaming replication network connection (Magnus Hagander,
Heikki Linnakangas)
</P
><P
> This avoids the requirement of manually transferring a file
system backup when setting up a standby server.
</P
></LI
><LI
><P
> Add
<TT
CLASS="VARNAME"
>replication_timeout</TT
>
setting (Fujii Masao, Heikki Linnakangas)
</P
><P
> Replication connections that are idle for more than the
<TT
CLASS="VARNAME"
>replication_timeout</TT
> interval will be terminated
automatically. Formerly, a failed connection was typically not
detected until the TCP timeout elapsed, which is inconveniently
long in many situations.
</P
></LI
><LI
><P
> Add command-line tool <A
HREF="app-pgbasebackup.html"
><SPAN
CLASS="APPLICATION"
>pg_basebackup</SPAN
></A
>
for creating a new standby server or database backup (Magnus
Hagander)
</P
></LI
><LI
><P
> Add a <A
HREF="sql-createrole.html"
>replication permission</A
>
for roles (Magnus Hagander)
</P
><P
> This is a read-only permission used for streaming replication.
It allows a non-superuser role to be used for replication connections.
Previously only superusers could initiate replication
connections; superusers still have this permission by default.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123481"
>E.43.3.2.2. Replication Monitoring</A
></H4
><P
></P
><UL
><LI
><P
> Add system view <A
HREF="monitoring-stats.html#MONITORING-STATS-VIEWS-TABLE"
><TT
CLASS="STRUCTNAME"
>pg_stat_replication</TT
></A
>
which displays activity of <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> sender processes (Itagaki
Takahiro, Simon Riggs)
</P
><P
> This reports the status of all connected standby servers.
</P
></LI
><LI
><P
> Add monitoring function <A
HREF="functions-admin.html#FUNCTIONS-RECOVERY-INFO-TABLE"
><CODE
CLASS="FUNCTION"
>pg_last_xact_replay_timestamp()</CODE
></A
>
(Fujii Masao)
</P
><P
> This returns the time at which the primary generated the most
recent commit or abort record applied on the standby.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123495"
>E.43.3.2.3. Hot Standby</A
></H4
><P
></P
><UL
><LI
><P
> Add configuration parameter <A
HREF="runtime-config-replication.html#GUC-HOT-STANDBY-FEEDBACK"
><TT
CLASS="VARNAME"
>hot_standby_feedback</TT
></A
>
to enable standbys to postpone cleanup of old row versions on the
primary (Simon Riggs)
</P
><P
> This helps avoid canceling long-running queries on the standby.
</P
></LI
><LI
><P
> Add the <A
HREF="monitoring-stats.html#MONITORING-STATS-VIEWS-TABLE"
><TT
CLASS="STRUCTNAME"
>pg_stat_database_conflicts</TT
></A
>
system view to show queries that have been canceled and the
reason (Magnus Hagander)
</P
><P
> Cancellations can occur because of dropped tablespaces, lock
timeouts, old snapshots, pinned buffers, and deadlocks.
</P
></LI
><LI
><P
> Add a <TT
CLASS="STRUCTFIELD"
>conflicts</TT
> count to <A
HREF="monitoring-stats.html#MONITORING-STATS-VIEWS-TABLE"
><TT
CLASS="STRUCTNAME"
>pg_stat_database</TT
></A
>
(Magnus Hagander)
</P
><P
> This is the number of conflicts that occurred in the database.
</P
></LI
><LI
><P
> Increase the maximum values for
<A
HREF="runtime-config-replication.html#GUC-MAX-STANDBY-ARCHIVE-DELAY"
><TT
CLASS="VARNAME"
>max_standby_archive_delay</TT
></A
> and
<A
HREF="runtime-config-replication.html#GUC-MAX-STANDBY-STREAMING-DELAY"
><TT
CLASS="VARNAME"
>max_standby_streaming_delay</TT
></A
>
</P
><P
> The maximum value for each of these parameters was previously
only about 35 minutes. Much larger values are now allowed.
</P
></LI
><LI
><P
> Add <A
HREF="errcodes-appendix.html#ERRCODES-TABLE"
><TT
CLASS="LITERAL"
>ERRCODE_T_R_DATABASE_DROPPED</TT
></A
>
error code to report recovery conflicts due to dropped databases
(Tatsuo Ishii)
</P
><P
> This is useful for connection pooling software.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123526"
>E.43.3.2.4. Recovery Control</A
></H4
><P
></P
><UL
><LI
><P
> Add functions to control streaming replication replay (Simon Riggs)
</P
><P
> The new functions are <A
HREF="functions-admin.html#FUNCTIONS-RECOVERY-CONTROL-TABLE"
><CODE
CLASS="FUNCTION"
>pg_xlog_replay_pause()</CODE
></A
>,
<A
HREF="functions-admin.html#FUNCTIONS-RECOVERY-CONTROL-TABLE"
><CODE
CLASS="FUNCTION"
>pg_xlog_replay_resume()</CODE
></A
>,
and the status function <A
HREF="functions-admin.html#FUNCTIONS-RECOVERY-CONTROL-TABLE"
><CODE
CLASS="FUNCTION"
>pg_is_xlog_replay_paused()</CODE
></A
>.
</P
></LI
><LI
><P
> Add <TT
CLASS="FILENAME"
>recovery.conf</TT
> setting
<TT
CLASS="VARNAME"
>pause_at_recovery_target</TT
>
to pause recovery at target (Simon Riggs)
</P
><P
> This allows a recovery server to be queried to check whether
the recovery point is the one desired.
</P
></LI
><LI
><P
> Add the ability to create named restore points using <A
HREF="functions-admin.html#FUNCTIONS-ADMIN-BACKUP-TABLE"
><CODE
CLASS="FUNCTION"
>pg_create_restore_point()</CODE
></A
>
(Jaime Casanova)
</P
><P
> These named restore points can be specified as recovery
targets using the new <TT
CLASS="FILENAME"
>recovery.conf</TT
> setting
<A
HREF="recovery-target-settings.html#RECOVERY-TARGET-NAME"
><TT
CLASS="VARNAME"
>recovery_target_name</TT
></A
>.
</P
></LI
><LI
><P
> Allow standby recovery to switch to a new timeline automatically
(Heikki Linnakangas)
</P
><P
> Now standby servers scan the archive directory for new
timelines periodically.
</P
></LI
><LI
><P
> Add <A
HREF="runtime-config-error-handling.html#GUC-RESTART-AFTER-CRASH"
><TT
CLASS="VARNAME"
>restart_after_crash</TT
></A
>
setting which disables automatic server restart after a backend
crash (Robert Haas)
</P
><P
> This allows external cluster management software to control
whether the database server restarts or not.
</P
></LI
><LI
><P
> Allow <A
HREF="recovery-config.html"
><TT
CLASS="FILENAME"
>recovery.conf</TT
></A
>
to use the same quoting behavior as <TT
CLASS="FILENAME"
>postgresql.conf</TT
>
(Dimitri Fontaine)
</P
><P
> Previously all values had to be quoted.
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123565"
>E.43.3.3. Queries</A
></H3
><P
></P
><UL
><LI
><P
> Add a true <A
HREF="transaction-iso.html#XACT-SERIALIZABLE"
>serializable isolation level</A
>
(Kevin Grittner, Dan Ports)
</P
><P
> Previously, asking for serializable isolation guaranteed only that a
single MVCC snapshot would be used for the entire transaction, which
allowed certain documented anomalies. The old snapshot isolation
behavior is still available by requesting the <A
HREF="transaction-iso.html#XACT-REPEATABLE-READ"
><TT
CLASS="LITERAL"
>REPEATABLE READ</TT
></A
>
isolation level.
</P
></LI
><LI
><P
> Allow data-modification commands
(<TT
CLASS="COMMAND"
>INSERT</TT
>/<TT
CLASS="COMMAND"
>UPDATE</TT
>/<TT
CLASS="COMMAND"
>DELETE</TT
>) in
<A
HREF="queries-with.html"
><TT
CLASS="LITERAL"
>WITH</TT
></A
> clauses
(Marko Tiikkaja, Hitoshi Harada)
</P
><P
> These commands can use <TT
CLASS="LITERAL"
>RETURNING</TT
> to pass data up to the
containing query.
</P
></LI
><LI
><P
> Allow <A
HREF="queries-with.html"
><TT
CLASS="LITERAL"
>WITH</TT
></A
>
clauses to be attached to <TT
CLASS="COMMAND"
>INSERT</TT
>, <TT
CLASS="COMMAND"
>UPDATE</TT
>,
<TT
CLASS="COMMAND"
>DELETE</TT
> statements (Marko Tiikkaja, Hitoshi Harada)
</P
></LI
><LI
><P
> Allow non-<A
HREF="queries-table-expressions.html#QUERIES-GROUP"
><TT
CLASS="LITERAL"
>GROUP
BY</TT
></A
> columns in the query target list when the primary
key is specified in the <TT
CLASS="LITERAL"
>GROUP BY</TT
> clause (Peter
Eisentraut)
</P
><P
> The SQL standard allows this behavior, and
because of the primary key, the result is unambiguous.
</P
></LI
><LI
><P
> Allow use of the key word <TT
CLASS="LITERAL"
>DISTINCT</TT
> in <A
HREF="queries-union.html"
><TT
CLASS="LITERAL"
>UNION</TT
>/<TT
CLASS="LITERAL"
>INTERSECT</TT
>/<TT
CLASS="LITERAL"
>EXCEPT</TT
></A
>
clauses (Tom Lane)
</P
><P
> <TT
CLASS="LITERAL"
>DISTINCT</TT
> is the default behavior so use of this
key word is redundant, but the SQL standard allows it.
</P
></LI
><LI
><P
> Fix ordinary queries with rules to use the same snapshot behavior
as <TT
CLASS="COMMAND"
>EXPLAIN ANALYZE</TT
> (Marko Tiikkaja)
</P
><P
> Previously <TT
CLASS="COMMAND"
>EXPLAIN ANALYZE</TT
> used slightly different
snapshot timing for queries involving rules. The
<TT
CLASS="COMMAND"
>EXPLAIN ANALYZE</TT
> behavior was judged to be more logical.
</P
></LI
></UL
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123611"
>E.43.3.3.1. Strings</A
></H4
><P
></P
><UL
><LI
><P
> Add per-column <A
HREF="collation.html"
>collation</A
> support
(Peter Eisentraut, Tom Lane)
</P
><P
> Previously collation (the sort ordering of text strings) could only be
chosen at database creation.
Collation can now be set per column, domain, index, or
expression, via the SQL-standard <TT
CLASS="LITERAL"
>COLLATE</TT
> clause.
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123619"
>E.43.3.4. Object Manipulation</A
></H3
><P
></P
><UL
><LI
><P
> Add <A
HREF="extend-extensions.html"
>extensions</A
> which
simplify packaging of additions to <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>
(Dimitri Fontaine, Tom Lane)
</P
><P
> Extensions are controlled by the new <A
HREF="sql-createextension.html"
><TT
CLASS="COMMAND"
>CREATE</TT
></A
>/<A
HREF="sql-alterextension.html"
><TT
CLASS="COMMAND"
>ALTER</TT
></A
>/<A
HREF="sql-dropextension.html"
><TT
CLASS="COMMAND"
>DROP EXTENSION</TT
></A
>
commands. This replaces ad-hoc methods of grouping objects that
are added to a <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> installation.
</P
></LI
><LI
><P
> Add support for <A
HREF="sql-createforeigntable.html"
>foreign
tables</A
> (Shigeru Hanada, Robert Haas, Jan Urbanski,
Heikki Linnakangas)
</P
><P
> This allows data stored outside the database to be used like
native <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>-stored data. Foreign tables
are currently read-only, however.
</P
></LI
><LI
><P
> Allow new values to be added to an existing enum type via
<A
HREF="sql-altertype.html"
><TT
CLASS="COMMAND"
>ALTER TYPE</TT
></A
> (Andrew
Dunstan)
</P
></LI
><LI
><P
> Add <A
HREF="sql-altertype.html"
><TT
CLASS="COMMAND"
>ALTER TYPE ...
ADD/DROP/ALTER/RENAME ATTRIBUTE</TT
></A
> (Peter Eisentraut)
</P
><P
> This allows modification of composite types.
</P
></LI
></UL
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123648"
>E.43.3.4.1. <TT
CLASS="COMMAND"
>ALTER</TT
> Object</A
></H4
><P
></P
><UL
><LI
><P
> Add <TT
CLASS="LITERAL"
>RESTRICT</TT
>/<TT
CLASS="LITERAL"
>CASCADE</TT
> to <A
HREF="sql-altertype.html"
><TT
CLASS="COMMAND"
>ALTER TYPE</TT
></A
> operations
on typed tables (Peter Eisentraut)
</P
><P
> This controls
<TT
CLASS="LITERAL"
>ADD</TT
>/<TT
CLASS="LITERAL"
>DROP</TT
>/<TT
CLASS="LITERAL"
>ALTER</TT
>/<TT
CLASS="LITERAL"
>RENAME
ATTRIBUTE</TT
> cascading behavior.
</P
></LI
><LI
><P
> Support <TT
CLASS="LITERAL"
>ALTER TABLE <TT
CLASS="REPLACEABLE"
><I
>name</I
></TT
> {OF | NOT OF}
<TT
CLASS="REPLACEABLE"
><I
>type</I
></TT
></TT
>
(Noah Misch)
</P
><P
> This syntax allows a standalone table to be made into a typed table,
or a typed table to be made standalone.
</P
></LI
><LI
><P
> Add support for more object types in <TT
CLASS="COMMAND"
>ALTER ... SET
SCHEMA</TT
> commands (Dimitri Fontaine)
</P
><P
> This command is now supported for conversions, operators, operator
classes, operator families, text search configurations, text search
dictionaries, text search parsers, and text search templates.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123673"
>E.43.3.4.2. <A
HREF="sql-createtable.html"
><TT
CLASS="COMMAND"
>CREATE/ALTER TABLE</TT
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Add <TT
CLASS="COMMAND"
>ALTER TABLE ...
ADD UNIQUE/PRIMARY KEY USING INDEX</TT
>
(Gurjeet Singh)
</P
><P
> This allows a primary key or unique constraint to be defined using an
existing unique index, including a concurrently created unique index.
</P
></LI
><LI
><P
> Allow <TT
CLASS="COMMAND"
>ALTER TABLE</TT
>
to add foreign keys without validation (Simon Riggs)
</P
><P
> The new option is called <TT
CLASS="LITERAL"
>NOT VALID</TT
>. The constraint's
state can later be modified to <TT
CLASS="LITERAL"
>VALIDATED</TT
> and validation
checks performed. Together these allow you to add a foreign key
with minimal impact on read and write operations.
</P
></LI
><LI
><P
> Allow <A
HREF="sql-altertable.html"
><TT
CLASS="COMMAND"
>ALTER TABLE
... SET DATA TYPE</TT
></A
> to avoid table rewrites in
appropriate cases (Noah Misch, Robert Haas)
</P
><P
> For example, converting a <TT
CLASS="TYPE"
>varchar</TT
> column to
<TT
CLASS="TYPE"
>text</TT
> no longer requires a rewrite of the table.
However, increasing the length constraint on a
<TT
CLASS="TYPE"
>varchar</TT
> column still requires a table rewrite.
</P
></LI
><LI
><P
> Add <A
HREF="sql-createtable.html"
><TT
CLASS="COMMAND"
>CREATE TABLE IF
NOT EXISTS</TT
></A
> syntax (Robert Haas)
</P
><P
> This allows table creation without causing an error if the
table already exists.
</P
></LI
><LI
><P
> Fix possible <SPAN
CLASS="QUOTE"
>"tuple concurrently updated"</SPAN
> error
when two backends attempt to add an inheritance
child to the same table at the same time (Robert Haas)
</P
><P
> <A
HREF="sql-altertable.html"
><TT
CLASS="COMMAND"
>ALTER TABLE</TT
></A
>
now takes a stronger lock on the parent table, so that the sessions
cannot try to update it simultaneously.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123707"
>E.43.3.4.3. Object Permissions</A
></H4
><P
></P
><UL
><LI
><P
> Add a <A
HREF="sql-security-label.html"
><TT
CLASS="COMMAND"
>SECURITY
LABEL</TT
></A
> command (KaiGai Kohei)
</P
><P
> This allows security labels to be assigned to objects.
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123715"
>E.43.3.5. Utility Operations</A
></H3
><P
></P
><UL
><LI
><P
> Add transaction-level <A
HREF="explicit-locking.html#ADVISORY-LOCKS"
>advisory
locks</A
> (Marko Tiikkaja)
</P
><P
> These are similar to the existing session-level advisory locks,
but such locks are automatically released at transaction end.
</P
></LI
><LI
><P
> Make <A
HREF="sql-truncate.html"
><TT
CLASS="COMMAND"
>TRUNCATE ... RESTART
IDENTITY</TT
></A
> restart sequences transactionally (Steve
Singer)
</P
><P
> Previously the counter could have been left out of sync if a
backend crashed between the on-commit truncation activity and
commit completion.
</P
></LI
></UL
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123727"
>E.43.3.5.1. <A
HREF="sql-copy.html"
><TT
CLASS="COMMAND"
>COPY</TT
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Add <TT
CLASS="LITERAL"
>ENCODING</TT
> option to <A
HREF="sql-copy.html"
><TT
CLASS="COMMAND"
>COPY TO/FROM</TT
></A
> (Hitoshi
Harada, Itagaki Takahiro)
</P
><P
> This allows the encoding of the <TT
CLASS="COMMAND"
>COPY</TT
> file to be
specified separately from client encoding.
</P
></LI
><LI
><P
> Add bidirectional <A
HREF="sql-copy.html"
><TT
CLASS="COMMAND"
>COPY</TT
></A
>
protocol support (Fujii Masao)
</P
><P
> This is currently only used by streaming replication.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123744"
>E.43.3.5.2. <A
HREF="sql-explain.html"
><TT
CLASS="COMMAND"
>EXPLAIN</TT
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Make <TT
CLASS="COMMAND"
>EXPLAIN VERBOSE</TT
> show the function call expression
in a <TT
CLASS="LITERAL"
>FunctionScan</TT
> node (Tom Lane)
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123753"
>E.43.3.5.3. <A
HREF="sql-vacuum.html"
><TT
CLASS="COMMAND"
>VACUUM</TT
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Add additional details to the output of <A
HREF="sql-vacuum.html"
><TT
CLASS="COMMAND"
>VACUUM FULL VERBOSE</TT
></A
>
and <A
HREF="sql-cluster.html"
><TT
CLASS="COMMAND"
>CLUSTER VERBOSE</TT
></A
>
(Itagaki Takahiro)
</P
><P
> New information includes the live and dead tuple count and
whether <TT
CLASS="COMMAND"
>CLUSTER</TT
> is using an index to rebuild.
</P
></LI
><LI
><P
> Prevent <A
HREF="routine-vacuuming.html#AUTOVACUUM"
>autovacuum</A
> from
waiting if it cannot acquire a table lock (Robert Haas)
</P
><P
> It will try to vacuum that table later.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123770"
>E.43.3.5.4. <A
HREF="sql-cluster.html"
><TT
CLASS="COMMAND"
>CLUSTER</TT
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Allow <TT
CLASS="COMMAND"
>CLUSTER</TT
> to sort the table rather than scanning
the index when it seems likely to be cheaper (Leonardo Francalanci)
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123778"
>E.43.3.5.5. Indexes</A
></H4
><P
></P
><UL
><LI
><P
> Add nearest-neighbor (order-by-operator) searching to <A
HREF="gist.html"
><ACRONYM
CLASS="ACRONYM"
>GiST</ACRONYM
> indexes</A
> (Teodor Sigaev, Tom Lane)
</P
><P
> This allows <ACRONYM
CLASS="ACRONYM"
>GiST</ACRONYM
> indexes to quickly return the
<TT
CLASS="REPLACEABLE"
><I
>N</I
></TT
> closest values in a query with <TT
CLASS="LITERAL"
>LIMIT</TT
>.
For example
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT * FROM places ORDER BY location <-> point '(101,456)' LIMIT 10;</PRE
><P>
finds the ten places closest to a given target point.
</P
></LI
><LI
><P
> Allow <A
HREF="gin.html"
><ACRONYM
CLASS="ACRONYM"
>GIN</ACRONYM
> indexes</A
> to index null
and empty values (Tom Lane)
</P
><P
> This allows full <ACRONYM
CLASS="ACRONYM"
>GIN</ACRONYM
> index scans, and fixes various
corner cases in which GIN scans would fail.
</P
></LI
><LI
><P
> Allow <A
HREF="gin.html"
><ACRONYM
CLASS="ACRONYM"
>GIN</ACRONYM
> indexes</A
> to
better recognize duplicate search entries (Tom Lane)
</P
><P
> This reduces the cost of index scans, especially in cases where
it avoids unnecessary full index scans.
</P
></LI
><LI
><P
> Fix <A
HREF="gist.html"
><ACRONYM
CLASS="ACRONYM"
>GiST</ACRONYM
> indexes</A
> to be fully
crash-safe (Heikki Linnakangas)
</P
><P
> Previously there were rare cases where a <TT
CLASS="COMMAND"
>REINDEX</TT
>
would be required (you would be informed).
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123807"
>E.43.3.6. Data Types</A
></H3
><P
></P
><UL
><LI
><P
> Allow <TT
CLASS="TYPE"
>numeric</TT
> to use a more compact, two-byte header
in common cases (Robert Haas)
</P
><P
> Previously all <TT
CLASS="TYPE"
>numeric</TT
> values had four-byte headers;
this change saves on disk storage.
</P
></LI
><LI
><P
> Add support for dividing <TT
CLASS="TYPE"
>money</TT
> by <TT
CLASS="TYPE"
>money</TT
>
(Andy Balholm)
</P
></LI
><LI
><P
> Allow binary I/O on type <TT
CLASS="TYPE"
>void</TT
> (Radoslaw Smogura)
</P
></LI
><LI
><P
> Improve hypotenuse calculations for geometric operators (Paul Matthews)
</P
><P
> This avoids unnecessary overflows, and may also be more accurate.
</P
></LI
><LI
><P
> Support hashing array values (Tom Lane)
</P
><P
> This provides additional query optimization possibilities.
</P
></LI
><LI
><P
> Don't treat a composite type as sortable unless all its column types
are sortable (Tom Lane)
</P
><P
> This avoids possible <SPAN
CLASS="QUOTE"
>"could not identify a comparison function"</SPAN
>
failures at runtime, if it is possible to implement the query without
sorting. Also, <TT
CLASS="COMMAND"
>ANALYZE</TT
> won't try to use inappropriate
statistics-gathering methods for columns of such composite types.
</P
></LI
></UL
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123833"
>E.43.3.6.1. Casting</A
></H4
><P
></P
><UL
><LI
><P
> Add support for casting between <TT
CLASS="TYPE"
>money</TT
> and <TT
CLASS="TYPE"
>numeric</TT
>
(Andy Balholm)
</P
></LI
><LI
><P
> Add support for casting from <TT
CLASS="TYPE"
>int4</TT
> and <TT
CLASS="TYPE"
>int8</TT
>
to <TT
CLASS="TYPE"
>money</TT
> (Joey Adams)
</P
></LI
><LI
><P
> Allow casting a table's row type to the table's supertype if
it's a typed table (Peter Eisentraut)
</P
><P
> This is analogous to the existing facility that allows casting a row
type to a supertable's row type.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123848"
>E.43.3.6.2. <A
HREF="functions-xml.html"
><ACRONYM
CLASS="ACRONYM"
>XML</ACRONYM
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Add <ACRONYM
CLASS="ACRONYM"
>XML</ACRONYM
> function <A
HREF="functions-xml.html#XML-EXISTS"
><TT
CLASS="LITERAL"
>XMLEXISTS</TT
></A
> and <A
HREF="functions-xml.html#XML-EXISTS"
><CODE
CLASS="FUNCTION"
>xpath_exists()</CODE
></A
>
functions (Mike Fowler)
</P
><P
> These are used for XPath matching.
</P
></LI
><LI
><P
> Add <ACRONYM
CLASS="ACRONYM"
>XML</ACRONYM
> functions <A
HREF="functions-xml.html#XML-IS-WELL-FORMED"
><CODE
CLASS="FUNCTION"
>xml_is_well_formed()</CODE
></A
>,
<A
HREF="functions-xml.html#XML-IS-WELL-FORMED"
><CODE
CLASS="FUNCTION"
>xml_is_well_formed_document()</CODE
></A
>,
<A
HREF="functions-xml.html#XML-IS-WELL-FORMED"
><CODE
CLASS="FUNCTION"
>xml_is_well_formed_content()</CODE
></A
>
(Mike Fowler)
</P
><P
> These check whether the input is properly-formed <ACRONYM
CLASS="ACRONYM"
>XML</ACRONYM
>.
They provide functionality that was previously available only in
the deprecated <TT
CLASS="FILENAME"
>contrib/xml2</TT
> module.
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123873"
>E.43.3.7. Functions</A
></H3
><P
></P
><UL
><LI
><P
> Add SQL function <A
HREF="functions-string.html#FORMAT"
><CODE
CLASS="FUNCTION"
>format(text, ...)</CODE
></A
>, which
behaves analogously to C's <CODE
CLASS="FUNCTION"
>printf()</CODE
> (Pavel Stehule,
Robert Haas)
</P
><P
> It currently supports formats for strings, SQL literals, and
SQL identifiers.
</P
></LI
><LI
><P
> Add string functions <A
HREF="functions-string.html#FUNCTIONS-STRING-OTHER"
><CODE
CLASS="FUNCTION"
>concat()</CODE
></A
>,
<A
HREF="functions-string.html#FUNCTIONS-STRING-OTHER"
><CODE
CLASS="FUNCTION"
>concat_ws()</CODE
></A
>,
<A
HREF="functions-string.html#FUNCTIONS-STRING-OTHER"
><CODE
CLASS="FUNCTION"
>left()</CODE
></A
>,
<A
HREF="functions-string.html#FUNCTIONS-STRING-OTHER"
><CODE
CLASS="FUNCTION"
>right()</CODE
></A
>,
and <A
HREF="functions-string.html#FUNCTIONS-STRING-OTHER"
><CODE
CLASS="FUNCTION"
>reverse()</CODE
></A
>
(Pavel Stehule)
</P
><P
> These improve compatibility with other database products.
</P
></LI
><LI
><P
> Add function <A
HREF="functions-admin.html#FUNCTIONS-ADMIN-GENFILE"
><CODE
CLASS="FUNCTION"
>pg_read_binary_file()</CODE
></A
>
to read binary files (Dimitri Fontaine, Itagaki Takahiro)
</P
></LI
><LI
><P
> Add a single-parameter version of function <A
HREF="functions-admin.html#FUNCTIONS-ADMIN-GENFILE"
><CODE
CLASS="FUNCTION"
>pg_read_file()</CODE
></A
>
to read an entire file (Dimitri Fontaine, Itagaki Takahiro)
</P
></LI
><LI
><P
> Add three-parameter forms of <A
HREF="functions-array.html#ARRAY-FUNCTIONS-TABLE"
><CODE
CLASS="FUNCTION"
>array_to_string()</CODE
></A
>
and <A
HREF="functions-array.html#ARRAY-FUNCTIONS-TABLE"
><CODE
CLASS="FUNCTION"
>string_to_array()</CODE
></A
>
for null value processing control (Pavel Stehule)
</P
></LI
></UL
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123909"
>E.43.3.7.1. Object Information Functions</A
></H4
><P
></P
><UL
><LI
><P
> Add the <A
HREF="functions-info.html#FUNCTIONS-INFO-CATALOG-TABLE"
><CODE
CLASS="FUNCTION"
>pg_describe_object()</CODE
></A
>
function (Alvaro Herrera)
</P
><P
> This function is used to obtain a human-readable string describing
an object, based on the <A
HREF="catalog-pg-class.html"
><TT
CLASS="STRUCTNAME"
>pg_class</TT
></A
>
OID, object OID, and sub-object ID. It can be used to help
interpret the contents of <A
HREF="catalog-pg-depend.html"
><TT
CLASS="STRUCTNAME"
>pg_depend</TT
></A
>.
</P
></LI
><LI
><P
> Update comments for built-in operators and their underlying
functions (Tom Lane)
</P
><P
> Functions that are meant to be used via an associated operator
are now commented as such.
</P
></LI
><LI
><P
> Add variable <A
HREF="runtime-config-compatible.html#GUC-QUOTE-ALL-IDENTIFIERS"
><TT
CLASS="VARNAME"
>quote_all_identifiers</TT
></A
>
to force the quoting of all identifiers in <TT
CLASS="COMMAND"
>EXPLAIN</TT
>
and in system catalog functions like <A
HREF="functions-info.html#FUNCTIONS-INFO-CATALOG-TABLE"
><CODE
CLASS="FUNCTION"
>pg_get_viewdef()</CODE
></A
>
(Robert Haas)
</P
><P
> This makes exporting schemas to tools and other databases with
different quoting rules easier.
</P
></LI
><LI
><P
> Add columns to the <A
HREF="infoschema-sequences.html"
><TT
CLASS="STRUCTNAME"
>information_schema.sequences</TT
></A
>
system view (Peter Eisentraut)
</P
><P
> Previously, though the view existed, the columns about the
sequence parameters were unimplemented.
</P
></LI
><LI
><P
> Allow <TT
CLASS="LITERAL"
>public</TT
> as a pseudo-role name in <A
HREF="functions-info.html#FUNCTIONS-INFO-ACCESS-TABLE"
><CODE
CLASS="FUNCTION"
>has_table_privilege()</CODE
></A
>
and related functions (Alvaro Herrera)
</P
><P
> This allows checking for public permissions.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123943"
>E.43.3.7.2. Function and Trigger Creation</A
></H4
><P
></P
><UL
><LI
><P
> Support <A
HREF="sql-createtrigger.html"
><TT
CLASS="LITERAL"
>INSTEAD
OF</TT
></A
> triggers on views (Dean Rasheed)
</P
><P
> This feature can be used to implement fully updatable views.
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN123951"
>E.43.3.8. Server-Side Languages</A
></H3
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123953"
>E.43.3.8.1. <A
HREF="plpgsql.html"
>PL/pgSQL</A
> Server-Side Language</A
></H4
><P
></P
><UL
><LI
><P
> Add <A
HREF="plpgsql-control-structures.html#PLPGSQL-FOREACH-ARRAY"
><TT
CLASS="COMMAND"
>FOREACH IN
ARRAY</TT
></A
> to PL/pgSQL
(Pavel Stehule)
</P
><P
> This is more efficient and readable than previous methods of
iterating through the elements of an array value.
</P
></LI
><LI
><P
> Allow <TT
CLASS="COMMAND"
>RAISE</TT
> without parameters to be caught in
the same places that could catch a <TT
CLASS="COMMAND"
>RAISE ERROR</TT
>
from the same location (Piyush Newe)
</P
><P
> The previous coding threw the error
from the block containing the active exception handler.
The new behavior is more consistent with other DBMS products.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123967"
>E.43.3.8.2. <A
HREF="plperl.html"
>PL/Perl</A
> Server-Side Language</A
></H4
><P
></P
><UL
><LI
><P
> Allow generic record arguments to PL/Perl functions (Andrew
Dunstan)
</P
><P
> PL/Perl functions can now be declared to accept type <TT
CLASS="TYPE"
>record</TT
>.
The behavior is the same as for any named composite type.
</P
></LI
><LI
><P
> Convert PL/Perl array arguments to Perl arrays (Alexey Klyukin,
Alex Hunsaker)
</P
><P
> String representations are still available.
</P
></LI
><LI
><P
> Convert PL/Perl composite-type arguments to Perl hashes
(Alexey Klyukin, Alex Hunsaker)
</P
><P
> String representations are still available.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN123981"
>E.43.3.8.3. <A
HREF="plpython.html"
>PL/Python</A
> Server-Side Language</A
></H4
><P
></P
><UL
><LI
><P
> Add table function support for PL/Python (Jan Urbanski)
</P
><P
> PL/Python can now return multiple <TT
CLASS="LITERAL"
>OUT</TT
> parameters
and record sets.
</P
></LI
><LI
><P
> Add a validator to PL/Python (Jan Urbanski)
</P
><P
> This allows PL/Python functions to be syntax-checked at function
creation time.
</P
></LI
><LI
><P
> Allow exceptions for SQL queries in PL/Python (Jan Urbanski)
</P
><P
> This allows access to SQL-generated exception error codes from
PL/Python exception blocks.
</P
></LI
><LI
><P
> Add explicit subtransactions to PL/Python (Jan Urbanski)
</P
></LI
><LI
><P
> Add PL/Python functions for quoting strings (Jan Urbanski)
</P
><P
> These functions are <A
HREF="plpython-util.html"
><TT
CLASS="LITERAL"
>plpy.quote_ident</TT
></A
>,
<A
HREF="plpython-util.html"
><TT
CLASS="LITERAL"
>plpy.quote_literal</TT
></A
>,
and <A
HREF="plpython-util.html"
><TT
CLASS="LITERAL"
>plpy.quote_nullable</TT
></A
>.
</P
></LI
><LI
><P
> Add traceback information to PL/Python errors (Jan Urbanski)
</P
></LI
><LI
><P
> Report PL/Python errors from iterators with <TT
CLASS="LITERAL"
>PLy_elog</TT
> (Jan
Urbanski)
</P
></LI
><LI
><P
> Fix exception handling with Python 3 (Jan Urbanski)
</P
><P
> Exception classes were previously not available in
<TT
CLASS="LITERAL"
>plpy</TT
> under Python 3.
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN124015"
>E.43.3.9. Client Applications</A
></H3
><P
></P
><UL
><LI
><P
> Mark <A
HREF="app-createlang.html"
><SPAN
CLASS="APPLICATION"
>createlang</SPAN
></A
>
and <A
HREF="app-droplang.html"
><SPAN
CLASS="APPLICATION"
>droplang</SPAN
></A
>
as deprecated now that they just invoke extension commands (Tom
Lane)
</P
></LI
></UL
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124024"
>E.43.3.9.1. <A
HREF="app-psql.html"
><SPAN
CLASS="APPLICATION"
>psql</SPAN
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Add <SPAN
CLASS="APPLICATION"
>psql</SPAN
> command <TT
CLASS="LITERAL"
>\conninfo</TT
>
to show current connection information (David Christensen)
</P
></LI
><LI
><P
> Add <SPAN
CLASS="APPLICATION"
>psql</SPAN
> command <TT
CLASS="LITERAL"
>\sf</TT
> to
show a function's definition (Pavel Stehule)
</P
></LI
><LI
><P
> Add <SPAN
CLASS="APPLICATION"
>psql</SPAN
> command <TT
CLASS="LITERAL"
>\dL</TT
> to list
languages (Fernando Ike)
</P
></LI
><LI
><P
> Add the <TT
CLASS="OPTION"
>S</TT
> (<SPAN
CLASS="QUOTE"
>"system"</SPAN
>) option to <SPAN
CLASS="APPLICATION"
>psql</SPAN
>'s
<TT
CLASS="LITERAL"
>\dn</TT
> (list schemas) command (Tom Lane)
</P
><P
> <TT
CLASS="LITERAL"
>\dn</TT
> without <TT
CLASS="LITERAL"
>S</TT
> now suppresses system
schemas.
</P
></LI
><LI
><P
> Allow <SPAN
CLASS="APPLICATION"
>psql</SPAN
>'s <TT
CLASS="LITERAL"
>\e</TT
> and <TT
CLASS="LITERAL"
>\ef</TT
>
commands to accept a line number to be used to position the
cursor in the editor (Pavel Stehule)
</P
><P
> This is passed to the editor according to the
<TT
CLASS="ENVAR"
>PSQL_EDITOR_LINENUMBER_ARG</TT
> environment variable.
</P
></LI
><LI
><P
> Have <SPAN
CLASS="APPLICATION"
>psql</SPAN
> set the client encoding from the
operating system locale by default (Heikki Linnakangas)
</P
><P
> This only happens if the <TT
CLASS="ENVAR"
>PGCLIENTENCODING</TT
> environment
variable is not set.
</P
></LI
><LI
><P
> Make <TT
CLASS="LITERAL"
>\d</TT
> distinguish between unique
indexes and unique constraints (Josh Kupershmidt)
</P
></LI
><LI
><P
> Make <TT
CLASS="LITERAL"
>\dt+</TT
> report <CODE
CLASS="FUNCTION"
>pg_table_size</CODE
>
instead of <CODE
CLASS="FUNCTION"
>pg_relation_size</CODE
> when talking to 9.0 or
later servers (Bernd Helmle)
</P
><P
> This is a more useful measure of table size, but note that it is
not identical to what was previously reported in the same display.
</P
></LI
><LI
><P
> Additional tab completion support (Itagaki Takahiro, Pavel Stehule,
Andrey Popp, Christoph Berg, David Fetter, Josh Kupershmidt)
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124073"
>E.43.3.9.2. <A
HREF="app-pgdump.html"
><SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Add <A
HREF="app-pgdump.html"
><SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
></A
>
and <A
HREF="app-pg-dumpall.html"
><SPAN
CLASS="APPLICATION"
>pg_dumpall</SPAN
></A
>
option <TT
CLASS="OPTION"
>--quote-all-identifiers</TT
> to force quoting
of all identifiers (Robert Haas)
</P
></LI
><LI
><P
> Add <TT
CLASS="LITERAL"
>directory</TT
> format to <SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
>
(Joachim Wieland, Heikki Linnakangas)
</P
><P
> This is internally similar to the <TT
CLASS="LITERAL"
>tar</TT
>
<SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
> format.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124092"
>E.43.3.9.3. <A
HREF="app-pg-ctl.html"
><SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Fix <SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
>
so it no longer incorrectly reports that the server is not
running (Bruce Momjian)
</P
><P
> Previously this could happen if the server was running but
<SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
> could not authenticate.
</P
></LI
><LI
><P
> Improve <SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
> start's <SPAN
CLASS="QUOTE"
>"wait"</SPAN
>
(<TT
CLASS="OPTION"
>-w</TT
>) option (Bruce Momjian, Tom Lane)
</P
><P
> The wait mode is now significantly more robust. It will not get
confused by non-default postmaster port numbers, non-default
Unix-domain socket locations, permission problems, or stale
postmaster lock files.
</P
></LI
><LI
><P
> Add <TT
CLASS="LITERAL"
>promote</TT
> option to <SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
> to
switch a standby server to primary (Fujii Masao)
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN124112"
>E.43.3.10. <SPAN
CLASS="APPLICATION"
>Development Tools</SPAN
></A
></H3
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124115"
>E.43.3.10.1. <A
HREF="libpq.html"
><SPAN
CLASS="APPLICATION"
>libpq</SPAN
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Add a libpq connection option <A
HREF="libpq-connect.html#LIBPQ-CONNECT-CLIENT-ENCODING"
><TT
CLASS="LITERAL"
>client_encoding</TT
></A
>
which behaves like the <TT
CLASS="ENVAR"
>PGCLIENTENCODING</TT
> environment
variable (Heikki Linnakangas)
</P
><P
> The value <TT
CLASS="LITERAL"
>auto</TT
> sets the client encoding based on
the operating system locale.
</P
></LI
><LI
><P
> Add <A
HREF="libpq-misc.html#LIBPQ-PQLIBVERSION"
><CODE
CLASS="FUNCTION"
>PQlibVersion()</CODE
></A
>
function which returns the libpq library version (Magnus
Hagander)
</P
><P
> libpq already had <CODE
CLASS="FUNCTION"
>PQserverVersion()</CODE
> which returns
the server version.
</P
></LI
><LI
><P
> Allow libpq-using clients to
check the user name of the server process
when connecting via Unix-domain sockets, with the new <A
HREF="libpq-connect.html#LIBPQ-CONNECT-REQUIREPEER"
><TT
CLASS="LITERAL"
>requirepeer</TT
></A
>
connection option
(Peter Eisentraut)
</P
><P
> <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> already allowed servers to check
the client user name when connecting via Unix-domain sockets.
</P
></LI
><LI
><P
> Add <A
HREF="libpq-connect.html#LIBPQ-PQPING"
><CODE
CLASS="FUNCTION"
>PQping()</CODE
></A
>
and <A
HREF="libpq-connect.html#LIBPQ-PQPINGPARAMS"
><CODE
CLASS="FUNCTION"
>PQpingParams()</CODE
></A
>
to libpq (Bruce Momjian, Tom Lane)
</P
><P
> These functions allow detection of the server's status without
trying to open a new session.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124146"
>E.43.3.10.2. <A
HREF="ecpg.html"
><SPAN
CLASS="APPLICATION"
>ECPG</SPAN
></A
></A
></H4
><P
></P
><UL
><LI
><P
> Allow ECPG to accept dynamic cursor names even in
<TT
CLASS="LITERAL"
>WHERE CURRENT OF</TT
> clauses
(Zoltan Boszormenyi)
</P
></LI
><LI
><P
> Make <SPAN
CLASS="APPLICATION"
>ecpglib</SPAN
> write <TT
CLASS="TYPE"
>double</TT
> values with a
precision of 15 digits, not 14 as formerly (Akira Kurosawa)
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN124158"
>E.43.3.11. Build Options</A
></H3
><P
></P
><UL
><LI
><P
> Use <TT
CLASS="LITERAL"
>+Olibmerrno</TT
> compile flag with HP-UX C compilers
that accept it (Ibrar Ahmed)
</P
><P
> This avoids possible misbehavior of math library calls on recent
HP platforms.
</P
></LI
></UL
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124165"
>E.43.3.11.1. Makefiles</A
></H4
><P
></P
><UL
><LI
><P
> Improved parallel make support (Peter Eisentraut)
</P
><P
> This allows for faster compiles. Also, <TT
CLASS="LITERAL"
>make -k</TT
>
now works more consistently.
</P
></LI
><LI
><P
> Require <ACRONYM
CLASS="ACRONYM"
>GNU</ACRONYM
> <A
HREF="install-requirements.html"
><SPAN
CLASS="APPLICATION"
>make</SPAN
></A
>
3.80 or newer (Peter Eisentraut)
</P
><P
> This is necessary because of the parallel-make improvements.
</P
></LI
><LI
><P
> Add <TT
CLASS="LITERAL"
>make maintainer-check</TT
> target
(Peter Eisentraut)
</P
><P
> This target performs various source code checks that are not
appropriate for either the build or the regression tests. Currently:
duplicate_oids, SGML syntax and tabs check, NLS syntax check.
</P
></LI
><LI
><P
> Support <TT
CLASS="LITERAL"
>make check</TT
> in <TT
CLASS="FILENAME"
>contrib</TT
>
(Peter Eisentraut)
</P
><P
> Formerly only <TT
CLASS="LITERAL"
>make installcheck</TT
> worked, but now
there is support for testing in a temporary installation.
The top-level <TT
CLASS="LITERAL"
>make check-world</TT
> target now includes
testing <TT
CLASS="FILENAME"
>contrib</TT
> this way.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124190"
>E.43.3.11.2. Windows</A
></H4
><P
></P
><UL
><LI
><P
> On Windows, allow <A
HREF="app-pg-ctl.html"
><SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
></A
> to register
the service as auto-start or start-on-demand (Quan Zongliang)
</P
></LI
><LI
><P
> Add support for collecting <A
HREF="installation-platform-notes.html#WINDOWS-CRASH-DUMPS"
>crash
dumps</A
> on Windows (Craig Ringer, Magnus Hagander)
</P
><P
> <SPAN
CLASS="PRODUCTNAME"
>minidumps</SPAN
> can now be generated by non-debug
Windows binaries and analyzed by standard debugging tools.
</P
></LI
><LI
><P
> Enable building with the MinGW64 compiler (Andrew Dunstan)
</P
><P
> This allows building 64-bit Windows binaries even on non-Windows
platforms via cross-compiling.
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN124205"
>E.43.3.12. Source Code</A
></H3
><P
></P
><UL
><LI
><P
> Revise the API for GUC variable assign hooks (Tom Lane)
</P
><P
> The previous functions of assign hooks are now split between check
hooks and assign hooks, where the former can fail but the latter
shouldn't. This change will impact add-on modules that define custom
GUC parameters.
</P
></LI
><LI
><P
> Add latches to the source code to support waiting for events (Heikki
Linnakangas)
</P
></LI
><LI
><P
> Centralize data modification permissions-checking logic
(KaiGai Kohei)
</P
></LI
><LI
><P
> Add missing <CODE
CLASS="FUNCTION"
>get_<TT
CLASS="REPLACEABLE"
><I
>object</I
></TT
>_oid()</CODE
> functions, for consistency
(Robert Haas)
</P
></LI
><LI
><P
> Improve ability to use C++ compilers for <A
HREF="xfunc-c.html"
>compiling add-on modules</A
> by removing
conflicting key words (Tom Lane)
</P
></LI
><LI
><P
> Add support for DragonFly <ACRONYM
CLASS="ACRONYM"
>BSD</ACRONYM
> (Rumko)
</P
></LI
><LI
><P
> Expose <CODE
CLASS="FUNCTION"
>quote_literal_cstr()</CODE
> for backend use
(Robert Haas)
</P
></LI
><LI
><P
> Run <A
HREF="install-procedure.html#BUILD"
>regression tests</A
> in the
default encoding (Peter Eisentraut)
</P
><P
> Regression tests were previously always run with
<TT
CLASS="LITERAL"
>SQL_ASCII</TT
> encoding.
</P
></LI
><LI
><P
> Add <SPAN
CLASS="APPLICATION"
>src/tools/git_changelog</SPAN
> to replace
<SPAN
CLASS="APPLICATION"
>cvs2cl</SPAN
> and <SPAN
CLASS="APPLICATION"
>pgcvslog</SPAN
> (Robert
Haas, Tom Lane)
</P
></LI
><LI
><P
> Add <SPAN
CLASS="APPLICATION"
>git-external-diff</SPAN
> script to
<TT
CLASS="FILENAME"
>src/tools</TT
> (Bruce Momjian)
</P
><P
> This is used to generate context diffs from git.
</P
></LI
><LI
><P
> Improve support for building with
<SPAN
CLASS="APPLICATION"
>Clang</SPAN
> (Peter Eisentraut)
</P
></LI
></UL
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124246"
>E.43.3.12.1. Server Hooks</A
></H4
><P
></P
><UL
><LI
><P
> Add source code hooks to check permissions (Robert Haas,
Stephen Frost)
</P
></LI
><LI
><P
> Add post-object-creation function hooks for use by security
frameworks (KaiGai Kohei)
</P
></LI
><LI
><P
> Add a client authentication hook (KaiGai Kohei)
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN124255"
>E.43.3.13. Contrib</A
></H3
><P
></P
><UL
><LI
><P
> Modify <TT
CLASS="FILENAME"
>contrib</TT
> modules and procedural
languages to install via the new <A
HREF="extend-extensions.html"
>extension</A
> mechanism (Tom Lane,
Dimitri Fontaine)
</P
></LI
><LI
><P
> Add <A
HREF="file-fdw.html"
><TT
CLASS="FILENAME"
>contrib/file_fdw</TT
></A
>
foreign-data wrapper (Shigeru Hanada)
</P
><P
> Foreign tables using this foreign data wrapper can read flat files
in a manner very similar to <TT
CLASS="COMMAND"
>COPY</TT
>.
</P
></LI
><LI
><P
> Add nearest-neighbor search support to <A
HREF="pgtrgm.html"
><TT
CLASS="FILENAME"
>contrib/pg_trgm</TT
></A
> and <A
HREF="btree-gist.html"
><TT
CLASS="FILENAME"
>contrib/btree_gist</TT
></A
>
(Teodor Sigaev)
</P
></LI
><LI
><P
> Add <A
HREF="btree-gist.html"
><TT
CLASS="FILENAME"
>contrib/btree_gist</TT
></A
>
support for searching on not-equals (Jeff Davis)
</P
></LI
><LI
><P
> Fix <A
HREF="fuzzystrmatch.html"
><TT
CLASS="FILENAME"
>contrib/fuzzystrmatch</TT
></A
>'s
<CODE
CLASS="FUNCTION"
>levenshtein()</CODE
> function to handle multibyte characters
(Alexander Korotkov)
</P
></LI
><LI
><P
> Add <CODE
CLASS="FUNCTION"
>ssl_cipher()</CODE
> and <CODE
CLASS="FUNCTION"
>ssl_version()</CODE
>
functions to <A
HREF="sslinfo.html"
><TT
CLASS="FILENAME"
>contrib/sslinfo</TT
></A
> (Robert
Haas)
</P
></LI
><LI
><P
> Fix <A
HREF="intarray.html"
><TT
CLASS="FILENAME"
>contrib/intarray</TT
></A
>
and <A
HREF="hstore.html"
><TT
CLASS="FILENAME"
>contrib/hstore</TT
></A
>
to give consistent results with indexed empty arrays (Tom Lane)
</P
><P
> Previously an empty-array query that used an index might return
different results from one that used a sequential scan.
</P
></LI
><LI
><P
> Allow <A
HREF="intarray.html"
><TT
CLASS="FILENAME"
>contrib/intarray</TT
></A
>
to work properly on multidimensional arrays (Tom Lane)
</P
></LI
><LI
><P
> In
<A
HREF="intarray.html"
><TT
CLASS="FILENAME"
>contrib/intarray</TT
></A
>,
avoid errors complaining about the presence of nulls in cases where no
nulls are actually present (Tom Lane)
</P
></LI
><LI
><P
> In
<A
HREF="intarray.html"
><TT
CLASS="FILENAME"
>contrib/intarray</TT
></A
>,
fix behavior of containment operators with respect to empty arrays
(Tom Lane)
</P
><P
> Empty arrays are now correctly considered to be contained in any other
array.
</P
></LI
><LI
><P
> Remove <A
HREF="xml2.html"
><TT
CLASS="FILENAME"
>contrib/xml2</TT
></A
>'s
arbitrary limit on the number of
<TT
CLASS="REPLACEABLE"
><I
>parameter</I
></TT
>=<TT
CLASS="REPLACEABLE"
><I
>value</I
></TT
> pairs that can be
handled by <CODE
CLASS="FUNCTION"
>xslt_process()</CODE
> (Pavel Stehule)
</P
><P
> The previous limit was 10.
</P
></LI
><LI
><P
> In <A
HREF="pageinspect.html"
><TT
CLASS="FILENAME"
>contrib/pageinspect</TT
></A
>,
fix heap_page_item to return infomasks as 32-bit values (Alvaro Herrera)
</P
><P
> This avoids returning negative values, which was confusing. The
underlying value is a 16-bit unsigned integer.
</P
></LI
></UL
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124322"
>E.43.3.13.1. Security</A
></H4
><P
></P
><UL
><LI
><P
> Add <A
HREF="sepgsql.html"
><TT
CLASS="FILENAME"
>contrib/sepgsql</TT
></A
>
to interface permission checks with <ACRONYM
CLASS="ACRONYM"
>SELinux</ACRONYM
> (KaiGai Kohei)
</P
><P
> This uses the new <A
HREF="sql-security-label.html"
><TT
CLASS="COMMAND"
>SECURITY LABEL</TT
></A
>
facility.
</P
></LI
><LI
><P
> Add contrib module <A
HREF="auth-delay.html"
><TT
CLASS="FILENAME"
>auth_delay</TT
></A
> (KaiGai
Kohei)
</P
><P
> This causes the server to pause before returning authentication
failure; it is designed to make brute force password attacks
more difficult.
</P
></LI
><LI
><P
> Add <TT
CLASS="FILENAME"
>dummy_seclabel</TT
>
contrib module (KaiGai Kohei)
</P
><P
> This is used for permission regression testing.
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124342"
>E.43.3.13.2. Performance</A
></H4
><P
></P
><UL
><LI
><P
> Add support for <TT
CLASS="LITERAL"
>LIKE</TT
> and <TT
CLASS="LITERAL"
>ILIKE</TT
> index
searches to <A
HREF="pgtrgm.html"
><TT
CLASS="FILENAME"
>contrib/pg_trgm</TT
></A
> (Alexander
Korotkov)
</P
></LI
><LI
><P
> Add <CODE
CLASS="FUNCTION"
>levenshtein_less_equal()</CODE
> function to <A
HREF="fuzzystrmatch.html"
><TT
CLASS="FILENAME"
>contrib/fuzzystrmatch</TT
></A
>,
which is optimized for small distances (Alexander Korotkov)
</P
></LI
><LI
><P
> Improve performance of index lookups on <A
HREF="seg.html"
><TT
CLASS="FILENAME"
>contrib/seg</TT
></A
> columns (Alexander
Korotkov)
</P
></LI
><LI
><P
> Improve performance of <A
HREF="pgupgrade.html"
><SPAN
CLASS="APPLICATION"
>pg_upgrade</SPAN
></A
> for
databases with many relations (Bruce Momjian)
</P
></LI
><LI
><P
> Add flag to <A
HREF="pgbench.html"
><TT
CLASS="FILENAME"
>contrib/pgbench</TT
></A
> to
report per-statement latencies (Florian Pflug)
</P
></LI
></UL
></DIV
><DIV
CLASS="SECT4"
><H4
CLASS="SECT4"
><A
NAME="AEN124368"
>E.43.3.13.3. Fsync Testing</A
></H4
><P
></P
><UL
><LI
><P
> Move <TT
CLASS="FILENAME"
>src/tools/test_fsync</TT
> to <A
HREF="pgtestfsync.html"
><TT
CLASS="FILENAME"
>contrib/pg_test_fsync</TT
></A
>
(Bruce Momjian, Tom Lane)
</P
></LI
><LI
><P
> Add <TT
CLASS="LITERAL"
>O_DIRECT</TT
> support to <A
HREF="pgtestfsync.html"
><TT
CLASS="FILENAME"
>contrib/pg_test_fsync</TT
></A
>
(Bruce Momjian)
</P
><P
> This matches the use of <TT
CLASS="LITERAL"
>O_DIRECT</TT
> by <A
HREF="runtime-config-wal.html#GUC-WAL-SYNC-METHOD"
><TT
CLASS="VARNAME"
>wal_sync_method</TT
></A
>.
</P
></LI
><LI
><P
> Add new tests to <A
HREF="pgtestfsync.html"
><TT
CLASS="FILENAME"
>contrib/pg_test_fsync</TT
></A
>
(Bruce Momjian)
</P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN124389"
>E.43.3.14. Documentation</A
></H3
><P
></P
><UL
><LI
><P
> Extensive <A
HREF="ecpg.html"
><SPAN
CLASS="APPLICATION"
>ECPG</SPAN
></A
>
documentation improvements (Satoshi Nagayasu)
</P
></LI
><LI
><P
> Extensive proofreading and documentation improvements
(Thom Brown, Josh Kupershmidt, Susanne Ebrecht)
</P
></LI
><LI
><P
> Add documentation for <A
HREF="runtime-config-error-handling.html#GUC-EXIT-ON-ERROR"
><TT
CLASS="VARNAME"
>exit_on_error</TT
></A
>
(Robert Haas)
</P
><P
> This parameter causes sessions to exit on any error.
</P
></LI
><LI
><P
> Add documentation for <A
HREF="functions-info.html#FUNCTIONS-INFO-CATALOG-TABLE"
><CODE
CLASS="FUNCTION"
>pg_options_to_table()</CODE
></A
>
(Josh Berkus)
</P
><P
> This function shows table storage options in a readable form.
</P
></LI
><LI
><P
> Document that it is possible to access all composite type
fields using <A
HREF="sql-expressions.html#FIELD-SELECTION"
><TT
CLASS="LITERAL"
>(compositeval).*</TT
></A
>
syntax (Peter Eisentraut)
</P
></LI
><LI
><P
> Document that <A
HREF="functions-string.html#FUNCTIONS-STRING-OTHER"
><CODE
CLASS="FUNCTION"
>translate()</CODE
></A
>
removes characters in <TT
CLASS="LITERAL"
>from</TT
> that don't have a
corresponding <TT
CLASS="LITERAL"
>to</TT
> character (Josh Kupershmidt)
</P
></LI
><LI
><P
> Merge documentation for <TT
CLASS="COMMAND"
>CREATE CONSTRAINT TRIGGER</TT
> and <A
HREF="sql-createtrigger.html"
><TT
CLASS="COMMAND"
>CREATE TRIGGER</TT
></A
>
(Alvaro Herrera)
</P
></LI
><LI
><P
> Centralize <A
HREF="ddl-priv.html"
>permission</A
> and <A
HREF="upgrading.html"
>upgrade</A
> documentation (Bruce Momjian)
</P
></LI
><LI
><P
> Add <A
HREF="kernel-resources.html#SYSVIPC-PARAMETERS"
>kernel tuning
documentation</A
> for Solaris 10 (Josh Berkus)
</P
><P
> Previously only Solaris 9 kernel tuning was documented.
</P
></LI
><LI
><P
> Handle non-ASCII characters consistently in <TT
CLASS="FILENAME"
>HISTORY</TT
> file
(Peter Eisentraut)
</P
><P
> While the <TT
CLASS="FILENAME"
>HISTORY</TT
> file is in English, we do have to deal
with non-ASCII letters in contributor names. These are now
transliterated so that they are reasonably legible without assumptions
about character set.
</P
></LI
></UL
></DIV
></DIV
></DIV
><DIV
CLASS="NAVFOOTER"
><HR
ALIGN="LEFT"
WIDTH="100%"><TABLE
SUMMARY="Footer navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
><A
HREF="release-9-1-1.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="index.html"
ACCESSKEY="H"
>Home</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
><A
HREF="release-9-0-22.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>Release 9.1.1</TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="release.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>Release 9.0.22</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>
