<!--
-$PostgreSQL: pgsql/doc/src/sgml/datatype.sgml,v 1.134 2003/12/01 20:34:53 tgl Exp $
+$PostgreSQL: pgsql/doc/src/sgml/datatype.sgml,v 1.135 2003/12/01 22:07:55 momjian Exp $
-->
<chapter id="datatype">
<para>
Object identifiers (OIDs) are used internally by
- <productname>PostgreSQL</productname> as primary keys for various system
- tables. Also, an OID system column is added to user-created tables
- (unless <literal>WITHOUT OIDS</> is specified at table creation time).
- Type <type>oid</> represents an object identifier. There are also
- several alias types for <type>oid</>: <type>regproc</>, <type>regprocedure</>,
- <type>regoper</>, <type>regoperator</>, <type>regclass</>,
- and <type>regtype</>. <xref linkend="datatype-oid-table"> shows an overview.
+ <productname>PostgreSQL</productname> as primary keys for various
+ system tables. An OID system column is also added to user-created
+ tables, unless <literal>WITHOUT OIDS</literal> is specified when
+ the table is created, or the <varname>default_with_oids</varname>
+ configuration variable is set to false. Type <type>oid</>
+ represents an object identifier. There are also several alias
+ types for <type>oid</>: <type>regproc</>, <type>regprocedure</>,
+ <type>regoper</>, <type>regoperator</>, <type>regclass</>, and
+ <type>regtype</>. <xref linkend="datatype-oid-table"> shows an
+ overview.
</para>
<para>
- The <type>oid</> type is currently implemented as an unsigned four-byte
- integer.
- Therefore, it is not large enough to provide database-wide uniqueness
- in large databases, or even in large individual tables. So, using a
- user-created table's OID column as a primary key is discouraged.
- OIDs are best used only for references to system tables.
+ The <type>oid</> type is currently implemented as an unsigned
+ four-byte integer. Therefore, it is not large enough to provide
+ database-wide uniqueness in large databases, or even in large
+ individual tables. So, using a user-created table's OID column as
+ a primary key is discouraged. OIDs are best used only for
+ references to system tables.
</para>
+ <note>
+ <para>
+ OIDs are included by default in user-created tables in
+ <productname>PostgreSQL</productname> &version;. However, this
+ behavior is likely to change in a future version of
+ <productname>PostgreSQL</productname>. Eventually, user-created
+ tables will not include an OID system column unless <literal>WITH
+ OIDS</literal> is specified when the table is created, or the
+ <varname>default_with_oids</varname> configuration variable is set
+ to true. If your application requires the presence of an OID
+ system column in a table, it should specify <literal>WITH
+ OIDS</literal> when that table is created to ensure compatibility
+ with future releases of <productname>PostgreSQL</productname>.
+ </para>
+ </note>
+
<para>
The <type>oid</> type itself has few operations beyond comparison.
It can be cast to
<!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/alter_table.sgml,v 1.63 2003/11/29 19:51:38 pgsql Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/alter_table.sgml,v 1.64 2003/12/01 22:07:58 momjian Exp $
PostgreSQL documentation
-->
of the OID are kept indefinitely. This is semantically similar
to the <literal>DROP COLUMN</literal> process.
</para>
+
+ <para>
+ Note that there is no variant of <command>ALTER TABLE</command>
+ that allows OIDs to be restored to a table once they have been
+ removed.
+ </para>
</listitem>
</varlistentry>
<!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/create_table.sgml,v 1.76 2003/11/29 19:51:38 pgsql Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/create_table.sgml,v 1.77 2003/12/01 22:07:58 momjian Exp $
PostgreSQL documentation
-->
<para>
If specified, the table is created as a temporary table.
Temporary tables are automatically dropped at the end of a
- session, or optionally at the end of the current transaction
- (see ON COMMIT below). Existing permanent tables with the same
- name are not visible to the current session while the temporary
- table exists, unless they are referenced with schema-qualified
- names. Any indexes created on a temporary table are automatically
- temporary as well.
+ session, or optionally at the end of the current transaction
+ (see <literal>ON COMMIT</literal> below). Existing permanent
+ tables with the same name are not visible to the current session
+ while the temporary table exists, unless they are referenced
+ with schema-qualified names. Any indexes created on a temporary
+ table are automatically temporary as well.
</para>
<para>
<listitem>
<para>
This optional clause specifies whether rows of the new table
- should have OIDs (object identifiers) assigned to them. The
- default is to have OIDs. (If the new table inherits from any
- tables that have OIDs, then <literal>WITH OIDS</> is forced even
- if the command says <literal>WITHOUT OIDS</>.)
+ should have OIDs (object identifiers) assigned to them. If
+ neither <literal>WITH OIDS</literal> nor <literal>WITHOUT
+ OIDS</literal> is specified, the default value depends upon the
+ <varname>default_with_oids</varname> configuration parameter. (If
+ the new table inherits from any tables that have OIDs, then
+ <literal>WITH OIDS</> is forced even if the command says
+ <literal>WITHOUT OIDS</>.)
</para>
<para>
- Specifying <literal>WITHOUT OIDS</> allows the user to suppress
- generation of OIDs for rows of a table. This may be worthwhile
- for large tables, since it will reduce OID consumption and
- thereby postpone wraparound of the 32-bit OID counter. Once the
- counter wraps around, uniqueness of OIDs can no longer be
- assumed, which considerably reduces their usefulness. Specifying
- <literal>WITHOUT OIDS</literal> also reduces the space required
- to store the table on disk by 4 bytes per row of the table,
- thereby improving performance.
+ If <literal>WITHOUT OIDS</literal> is specified or implied, this
+ means that the generation of OIDs for this table will be
+ supressed. This is generally considered worthwhile, since it
+ will reduce OID consumption and thereby postpone the wraparound
+ of the 32-bit OID counter. Once the counter wraps around, OIDs
+ can no longer be assumed to be unique, which makes them
+ considerably less useful. In addition, excluding OIDs from a
+ table reduces the space required on disk to storage the table by
+ 4 bytes per row, leading to increased performance.
+ </para>
+
+ <para>
+ To remove OIDs from a table after it has been created, use <xref
+ linkend="sql-altertable" endterm="sql-altertable-title">.
</para>
</listitem>
</varlistentry>
<itemizedlist>
<listitem>
<para>
- Whenever an application makes use of OIDs to identify specific
+ Using OIDs in new applications is not recommended: where
+ possible, using a <literal>SERIAL</literal> or other sequence
+ generator as the table's primary key is preferred. However, if
+ your application does make use of OIDs to identify specific rows
rows of a table, it is recommended to create a unique constraint
on the <structfield>oid</> column of that table, to ensure that
OIDs in the table will indeed uniquely identify rows even after
counter wraparound. Avoid assuming that OIDs are unique across
tables; if you need a database-wide unique identifier, use the
combination of <structfield>tableoid</> and row OID for the
- purpose. (It is likely that future <productname>PostgreSQL</>
- releases will use a separate OID counter for each table, so that
- it will be <emphasis>necessary</>, not optional, to include
- <structfield>tableoid</> to have a unique identifier
- database-wide.)
+ purpose.
</para>
<tip>
<!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/create_table_as.sgml,v 1.17 2003/11/29 19:51:38 pgsql Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/create_table_as.sgml,v 1.18 2003/12/01 22:07:58 momjian Exp $
PostgreSQL documentation
-->
<refsect1>
<title>Parameters</title>
-
+
+ <variablelist>
+ <varlistentry>
+ <term><literal>GLOBAL</literal> or <literal>LOCAL</literal></term>
+ <listitem>
+ <para>
+ Ignored for compatibility. Refer to <xref
+ linkend="sql-createtable" endterm="sql-createtable-title"> for
+ details.
+ </para>
+ </listitem>
+ </varlistentry>
+ </variablelist>
+
<variablelist>
<varlistentry>
<term><literal>TEMPORARY</> or <literal>TEMP</></term>
<title>Notes</title>
<para>
- This command is functionally equivalent to <xref
- linkend="sql-selectinto" endterm="sql-selectinto-title">, but it is preferred since it is less
- likely to be confused with other uses of the <command>SELECT
- ... INTO</command> syntax.
+ This command is functionally similar to <xref
+ linkend="sql-selectinto" endterm="sql-selectinto-title">, but it is
+ preferred since it is less likely to be confused with other uses of
+ the <command>SELECT INTO</command> syntax.
+ </para>
+
+ <para>
+ Prior to PostgreSQL 7.5, <command>CREATE TABLE AS</command> always
+ included OIDs in the table it produced. Furthermore, these OIDs
+ were newly generated: they were distinct from the OIDs of any of
+ the rows in the source tables of the <command>SELECT</command> or
+ <command>EXECUTE</command> statement. Therefore, if <command>CREATE
+ TABLE AS</command> was frequently executed, the OID counter would
+ be rapidly incremented. As of PostgreSQL 7.5, the inclusion of OIDs
+ in the table generated by <command>CREATE TABLE AS</command> is
+ controlled by the <varname>default_with_oids</varname> configuration
+ variable. This variable currently defaults to true, but will likely
+ default to false in a future release of <productname>PostgreSQL</>.
</para>
</refsect1>
<simplelist type="inline">
<member><xref linkend="sql-createtable" endterm="sql-createtable-title"></member>
- <member><xref linkend="sql-createview" endterm="sql-createview-title"></member>
<member><xref linkend="sql-execute" endterm="sql-execute-title"></member>
<member><xref linkend="sql-select" endterm="sql-select-title"></member>
<member><xref linkend="sql-selectinto" endterm="sql-selectinto-title"></member>
<!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/pg_dump.sgml,v 1.67 2003/11/29 19:51:39 pgsql Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/pg_dump.sgml,v 1.68 2003/12/01 22:07:58 momjian Exp $
PostgreSQL documentation
-->
</para>
<para>
- Once restored, it is wise to run <command>ANALYZE</> on each
- restored table so the optimizer has useful statistics.
+ The dump file produced by <application>pg_dump</application> does
+ not contain the statistics used by the optimizer to make query
+ planning decisions. Therefore, it is wise to run
+ <command>ANALYZE</command> after restoring from a dump file to
+ ensure good performance.
</para>
</refsect1>
<!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/prepare.sgml,v 1.8 2003/11/29 19:51:39 pgsql Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/prepare.sgml,v 1.9 2003/12/01 22:07:58 momjian Exp $
PostgreSQL documentation
-->
</para>
<para>
- Prepared statements are only stored in and for the duration of
- the current database session. When
- the session ends, the prepared statement is forgotten, and so it must be
- recreated before being used again. This also means that a single
- prepared statement cannot be used by multiple simultaneous database
- clients; however, each client can create their own prepared statement
- to use.
+ Prepared statements are only for the duration of the current
+ database session. When the session ends, the prepared statement is
+ forgotten, so it must be recreated before being used again. This
+ also means that a single prepared statement cannot be used by
+ multiple simultaneous database clients; however, each client can
+ create their own prepared statement to use.
</para>
<para>
<!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/select_into.sgml,v 1.25 2003/11/29 19:51:39 pgsql Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/select_into.sgml,v 1.26 2003/12/01 22:07:58 momjian Exp $
PostgreSQL documentation
-->
<title>Notes</title>
<para>
- <xref linkend="sql-createtableas" endterm="sql-createtableas-title">
- is functionally equivalent to <command>SELECT INTO</command>.
- <command>CREATE TABLE AS</command> is the recommended syntax, since
- this form of <command>SELECT INTO</command> is not available in
- <application>ECPG</application> or
- <application>PL/pgSQL</application>, because they interpret the
- <literal>INTO</literal> clause differently.
+ <xref linkend="sql-createtableas"
+ endterm="sql-createtableas-title"> is functionally similar to
+ <command>SELECT INTO</command>. <command>CREATE TABLE AS</command>
+ is the recommended syntax, since this form of <command>SELECT
+ INTO</command> is not available in <application>ECPG</application>
+ or <application>PL/pgSQL</application>, because they interpret the
+ <literal>INTO</literal> clause differently. Furthermore,
+ <command>CREATE TABLE AS</command> offers a superset of the
+ functionality provided by <command>SELECT INTO</command>.
+ </para>
+
+ <para>
+ Prior to PostgreSQL 7.5, the table created by <command>SELECT
+ INTO</command> always included OIDs. Furthermore, these OIDs were
+ newly generated: they were distinct from the OIDs of any of the
+ rows in the source tables of the <command>SELECT INTO</command>
+ statement. Therefore, if <command>SELECT INTO</command> was
+ frequently executed, the OID counter would be rapidly
+ incremented. As of PostgreSQL 7.5, the inclusion of OIDs in the
+ table created by <command>SELECT INTO</command> is controlled by
+ the <varname>default_with_oids</varname> configuration
+ variable. This variable currently defaults to true, but will likely
+ default to false in a future release of <productname>PostgreSQL</>.
</para>
</refsect1>
<title>Compatibility</title>
<para>
- The SQL standard uses <command>SELECT ... INTO</command> to
+ The SQL standard uses <command>SELECT INTO</command> to
represent selecting values into scalar variables of a host program,
rather than creating a new table. This indeed is the usage found
in <application>ECPG</application> (see <xref linkend="ecpg">) and
<!--
-$PostgreSQL: pgsql/doc/src/sgml/runtime.sgml,v 1.223 2003/11/29 19:51:37 pgsql Exp $
+$PostgreSQL: pgsql/doc/src/sgml/runtime.sgml,v 1.224 2003/12/01 22:07:56 momjian Exp $
-->
<Chapter Id="runtime">
</listitem>
</varlistentry>
- </variablelist>
+ <varlistentry>
+ <term><varname>default_with_oids</varname> (<type>boolean</type>)</term>
+ <listitem>
+ <para>
+ This controls whether <command>CREATE TABLE</command> will
+ include OIDs in newly-created tables, if neither <literal>WITH
+ OIDS</literal> or <literal>WITHOUT OIDS</literal> have been
+ specified. It also determines whether OIDs will be included in
+ the table generated by <command>SELECT INTO</command> and
+ <command>CREATE TABLE AS</command>. In
+ <productname>PostgreSQL</productname> &version;
+ <varname>default_with_oids</varname> defaults to true. This is
+ also the behavior of previous versions of
+ <productname>PostgreSQL</productname>. However, assuming that
+ tables will contain OIDs by default is not
+ encouraged. Therefore, this option will default to false in a
+ future release of <productname>PostgreSQL</productname>.
+ </para>
+
+ <para>
+ To ease compatibility with applications that make use of OIDs,
+ this option should left enabled. To ease compatibility with
+ future versions of <productname>PostgreSQL</productname>, this
+ option should be disabled, and applications that require OIDs
+ on certain tables should explictely specify <literal>WITH
+ OIDS</literal> when issuing the <command>CREATE
+ TABLE</command> statements for the tables in question.
+ </para>
+ </listitem>
+ </varlistentry>
+
+ </variablelist>
</sect3>
<sect3 id="runtime-config-compatible-clients">
<title>Platform and Client Compatibility</title>
<!--
-$PostgreSQL: pgsql/doc/src/sgml/spi.sgml,v 1.29 2003/11/29 19:51:37 pgsql Exp $
+$PostgreSQL: pgsql/doc/src/sgml/spi.sgml,v 1.30 2003/12/01 22:07:57 momjian Exp $
-->
<chapter id="spi">
<listitem>
<para>
if a <command>SELECT</command> (but not <command>SELECT
- ... INTO</>) was executed
+ INTO</>) was executed
</para>
</listitem>
</varlistentry>
<term><symbol>SPI_OK_SELINTO</symbol></term>
<listitem>
<para>
- if a <command>SELECT ... INTO</command> was executed
+ if a <command>SELECT INTO</command> was executed
</para>
</listitem>
</varlistentry>
*
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/backend/executor/execMain.c,v 1.222 2003/11/29 19:51:48 pgsql Exp $
+ * $PostgreSQL: pgsql/src/backend/executor/execMain.c,v 1.223 2003/12/01 22:07:58 momjian Exp $
*
*-------------------------------------------------------------------------
*/
#include "optimizer/var.h"
#include "parser/parsetree.h"
#include "utils/acl.h"
+#include "utils/guc.h"
#include "utils/lsyscache.h"
do_select_into = true;
/*
- * For now, always create OIDs in SELECT INTO; this is for
- * backwards compatibility with pre-7.3 behavior. Eventually we
- * might want to allow the user to choose.
+ * The presence of OIDs in the result set of SELECT INTO is
+ * controlled by the default_with_oids GUC parameter. The
+ * behavior in versions of PostgreSQL prior to 7.5 is to
+ * always include OIDs.
*/
- estate->es_force_oids = true;
+ estate->es_force_oids = default_with_oids;
}
/*
*
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/backend/parser/gram.y,v 2.440 2003/11/29 19:51:51 pgsql Exp $
+ * $PostgreSQL: pgsql/src/backend/parser/gram.y,v 2.441 2003/12/01 22:07:58 momjian Exp $
*
* HISTORY
* AUTHOR DATE MAJOR EVENT
#include "utils/numeric.h"
#include "utils/datetime.h"
#include "utils/date.h"
+#include "utils/guc.h"
extern List *parsetree; /* final parse result is delivered here */
OptWithOids:
WITH OIDS { $$ = TRUE; }
| WITHOUT OIDS { $$ = FALSE; }
- | /*EMPTY*/ { $$ = TRUE; }
+ /*
+ * If the user didn't explicitely specify WITH or WITHOUT
+ * OIDS, decide whether to include OIDs based on the
+ * "default_with_oids" GUC var
+ */
+ | /*EMPTY*/ { $$ = default_with_oids; }
;
OnCommitOption: ON COMMIT DROP { $$ = ONCOMMIT_DROP; }
* Written by Peter Eisentraut <peter_e@gmx.net>.
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/backend/utils/misc/guc.c,v 1.173 2003/12/01 03:55:21 momjian Exp $
+ * $PostgreSQL: pgsql/src/backend/utils/misc/guc.c,v 1.174 2003/12/01 22:08:00 momjian Exp $
*
*--------------------------------------------------------------------
*/
bool Password_encryption = true;
+bool default_with_oids = true;
+
int log_min_error_statement = PANIC;
int log_min_messages = NOTICE;
int client_min_messages = NOTICE;
/* QUERY_TUNING */
gettext_noop("Query Tuning"),
/* QUERY_TUNING_METHOD */
- gettext_noop("Query Tuning / Planner Method Enabling"),
+ gettext_noop("Query Tuning / Planner Method Configuration"),
/* QUERY_TUNING_COST */
gettext_noop("Query Tuning / Planner Cost Constants"),
/* QUERY_TUNING_GEQO */
&check_function_bodies,
true, NULL, NULL
},
+ {
+ {"default_with_oids", PGC_USERSET, COMPAT_OPTIONS_PREVIOUS,
+ gettext_noop("By default, newly-created tables should have OIDs"),
+ NULL
+ },
+ &default_with_oids,
+ true, NULL, NULL
+ },
/* End-of-list marker */
{
# QUERY TUNING
#---------------------------------------------------------------------------
-# - Planner Method Enabling -
+# - Planner Method Configuration -
#enable_hashagg = true
#enable_hashjoin = true
#add_missing_from = true
#regex_flavor = advanced # advanced, extended, or basic
#sql_inheritance = true
+#default_with_oids = true
# - Other Platforms & Clients -
* by PostgreSQL
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/bin/pg_dump/pg_dump.c,v 1.358 2003/11/29 19:52:05 pgsql Exp $
+ * $PostgreSQL: pgsql/src/bin/pg_dump/pg_dump.c,v 1.359 2003/12/01 22:08:01 momjian Exp $
*
*-------------------------------------------------------------------------
*/
appendPQExpBuffer(q, ")");
}
- if (!tbinfo->hasoids)
- appendPQExpBuffer(q, " WITHOUT OIDS");
+ appendPQExpBuffer(q, tbinfo->hasoids ? " WITH OIDS" : " WITHOUT OIDS");
appendPQExpBuffer(q, ";\n");
*
* Copyright (c) 2000-2003, PostgreSQL Global Development Group
*
- * $PostgreSQL: pgsql/src/bin/psql/tab-complete.c,v 1.94 2003/11/29 19:52:07 pgsql Exp $
+ * $PostgreSQL: pgsql/src/bin/psql/tab-complete.c,v 1.95 2003/12/01 22:08:01 momjian Exp $
*/
/*----------------------------------------------------------------------
"default_statistics_target",
"default_transaction_isolation",
"default_transaction_read_only",
+ "default_with_oids",
"dynamic_library_path",
"effective_cache_size",
"enable_hashagg",
* Copyright (c) 2000-2003, PostgreSQL Global Development Group
* Written by Peter Eisentraut <peter_e@gmx.net>.
*
- * $PostgreSQL: pgsql/src/include/utils/guc.h,v 1.42 2003/11/29 22:41:15 pgsql Exp $
+ * $PostgreSQL: pgsql/src/include/utils/guc.h,v 1.43 2003/12/01 22:08:02 momjian Exp $
*--------------------------------------------------------------------
*/
#ifndef GUC_H
extern bool SQL_inheritance;
extern bool Australian_timezones;
+extern bool default_with_oids;
+
extern int log_min_error_statement;
extern int log_min_messages;
extern int client_min_messages;
--
-- WITHOUT OID
--
-CREATE TABLE wi (i INT);
+CREATE TABLE wi (i INT) WITH OIDS;
CREATE TABLE wo (i INT) WITHOUT OIDS;
INSERT INTO wi VALUES (1); -- 1
INSERT INTO wo SELECT i FROM wi; -- 1
-- WITHOUT OID
--
-CREATE TABLE wi (i INT);
+CREATE TABLE wi (i INT) WITH OIDS;
CREATE TABLE wo (i INT) WITHOUT OIDS;
INSERT INTO wi VALUES (1); -- 1
INSERT INTO wo SELECT i FROM wi; -- 1
projects / users / bernd / postgres.git / commitdiff