2 $PostgreSQL: pgsql/doc/src/sgml/release.sgml,v 1.277 2004/08/04 21:33:40 tgl Exp $
5 <appendix id="release">
6 <title>Release Notes</title>
8 <sect1 id="release-7-5">
9 <title>Release 8.0</title>
12 <title>Release date</title>
13 <simpara>2004-??-??, current as of 2004-07-22</simpara>
17 <title>Overview</title>
20 Major changes in this release:
30 This is the first PostgreSQL release to natively support Microsoft
31 Windows as a server. It can run as a service and a separate
32 installer project has been created to ease installation, <ulink
33 url="http://pgfoundry.org/projects/pginstaller">
34 http://pgfoundry.org/projects/pginstaller</ulink>. This release
35 supports NT-based Windows releases like NT4, Win2k, XP, Win2003.
36 Older releases like Windows 95, 98, and ME are not supported because
37 these operating systems do not have the infrastructure to
41 Previous releases required the Unix emulation toolkit Cygwin for
42 Win32 support. PostgreSQL has always supported clients on Win32.
49 Tablespaces Simplify Disk Layout
54 This release introduces tablespaces which allows administrators
55 to control the file system used for storage of databases,
56 schemas, tables, or indexes. This improves performance and
57 control over disk space usage. Prior releases used the inferior
58 <literal>initlocation</> and custom symlinks for such tasks.
65 Savepoints Improve Transaction Control
70 Savepoints allow specific parts of a transaction to be aborted
71 without affecting the remainder of the transaction. Prior
72 release had no such capability; there was no way to recover from
73 a statement failure in a transation. This feature is valuable
74 for application writers that required complex transaction control.
81 Point-In-Time Recovery Increases Reliability
86 Though PostgreSQL is very reliable, in previous releases there
87 was no way to recover from disk drive failure except to restore
88 from a previous backup or use a standby replication server.
89 Point-in-time recovery allows continuous backup of the server.
90 You can recover either to the point of failure or to some
91 transaction in the past.
98 Improved Buffer Management, CHECKPOINT, VACUUM
103 This release has a more intelligent buffer replacement strategy,
104 which will make better use of available shared buffers and
105 improve performance. The performance impact of vacuum and
106 checkpoints is also improved.
118 A column's data type can now be changed with ALTER TABLE.
125 New Perl Server-Side Language
130 A new version of the Perl server-side language now supports a
131 persistent, shared storage area, triggers, returning records and
132 arrays of records, and SPI calls.
139 COPY Handles Comma-Separated-Value Files
144 COPY can now read and write comma-separate-value (CSV) files. It
145 has the flexibility to interpret non-standard quoting and
146 separation characters too.
156 <title>Migration to version 8.0</title>
159 A dump/restore using <application>pg_dump</application> is
160 required for those wishing to migrate data from any previous
165 Observe the following incompatibilities:
170 Server configuration parameters SortMem and VacuumMem have been
171 renamed to work_mem and maintenance_work_mem to better reflect
172 their use. The original names still supported.
178 Server configuration parameters log_pid, log_timestamp, and
179 log_source_port have been removed now that a more flexible
180 log_line_prefix has been added.
186 Server configuration parameters virtual_host and tcpip_socket
187 have been replaced with a more general listen_addresses. Also,
188 the server now listens on localhost by default, which eliminates
189 the need for the -i postmaster switch in many scenarios.
195 Server configuration parameter syslog has been removed and
196 replaced with a more logical log_destination variable to control
197 the log output destination.
203 Server configuration parameter log_statement has been changed so
204 it can restrict logging of just database modification or data
205 definition statements.
211 Server configuration parameter max_expr_depth parameter has been
212 replaced with max_stack_depth which measures the stack size
213 rather than the number of stack levels used. This helps prevent
214 session termination due to stack overflow caused by recursive
221 EXECUTE now returns a completion tag that matches the executed
228 The length() function no longer measures trailing spaces in
235 The server now warns of empty strings passed to oid/float4/float8
236 data types. In the next major release, doing this will generate an
243 Function to_char() now returns the proper year for BC dates. It
244 previously returned one less than the current year. The function
245 now also returns the proper millennium and century.
251 psql \copy now reads from the query stdin/stdout, rather than
252 psql's stdin/stdout. The previous behavior can be accessed via new
253 pstdin/pstdout parameters.
258 <para> The JDBC client interface has been removed and is now
259 hosted at <ulink url=
260 "http://jdbc.postgresql.org">http://jdbc.postgresql.org</ulink>.
261 The TCL client interface has also been removed. There are several
262 now hosted at <ulink url=
263 "http://gborg.postgresql.org">http://gborg.postgresql.org</ulink>
269 The server now uses its own time zone database, rather than the
270 one supplied by the operating system. This should not change
271 the time zone behavior.
277 The configure threading option no longer requires users to run
278 tests or edit configuration files; threading options are now detected
285 Now that tablespaces have been implemented, initlocation has been
294 <title>Changes</title>
297 Below you will find a detailed account of the changes between
298 release 8.0 and the previous major release.
302 <title>Performance Improvements</title>
307 Support cross-data-type index usage (Tom)
310 Before this change some queries would not use an index if the data
311 types did not match exactly. This improvement makes index usage more
312 intuitive and consistent.
318 New buffer replacement strategy that improves caching (Jan)
321 Prior releases use a least-recently-used (LRU) cache to keep
322 recently referenced pages in the cache. Unfortunately, the cache
323 did not consider the number of times a specific cache entry was
324 accessed. Large table scans could force out useful cache pages.
325 The new cache uses four separate cache areas to track most
326 recently used and most frequently used cache pages to dynamically
327 optimize their replacement based on the work load. This should
328 lead to much more efficient use of the shared buffer cache.
329 Administrators who have tested shared buffer sizes in the past
330 should retest with this new cache replacement policy.
336 Add subprocess to write dirty buffers periodically to reduce checkpoint writes (Jan)
339 In previous releases, the checkpoint process, which runs every few
340 minutes, would write all dirty buffers to the operating system
341 buffer cache then flush all operating system dirty buffers to
342 disk. This often resulted in a periodic spike in disk usage that
343 hurt performance. The new code uses a background writer to trickle
344 disk writes at a steady pace so checkpoints have far fewer dirty
345 pages to write to disk. This should improve performance and
346 minimize performance degradation during checkpoints.
352 Improve btree index performance for duplicate keys (Dmitry Tkach, Tom)
355 This improves the way indexes are scanned when many duplicate
356 values exist in the index.
362 Improved index usage with OR clauses (Tom)
365 This allows the optimizer to use indexes in statements with
366 many OR clauses that were not possible in the past. It can also
367 use multi-column indexes where the first column is specified and
368 the second column is part of an OR clause.
374 Improve matching of partial index clauses (Tom)
377 The server is now smarter about using partial indexes in queries
378 involving complex WHERE clauses.
384 Improve load speed for C functions (Tom)
387 This release now uses a hash to lookup information for externally
388 loaded C functions. This improves their speed so they perform as
389 quickly as native functions that are part of the server backend.
395 Improve performance of the GEQO optimizer (Tom)
398 The GEQO optimizer is used for joining many tables (default
399 twelve). This release speeds up the way queries are analyzed to
400 decrease time spent in optimization.
406 Add ability to prolong vacuum to reduce performance impact (Jan)
409 On busy systems, VACUUM performs many I/O request which can hurt
410 performance for other users. This release allows you to slow down
411 VACUUM so it uses fewer resources though this increases the
418 Add type-specific ANALYZE statistics capability (Mark Cave-Ayland)
421 This adjustment allows more flexibility in generating statistics
422 for non-standard data types.
428 Allow collection of ANALYZE statistics for expression indexes (Tom)
431 Expression indexes (also called functional indexes) allow users to
432 index not just columns but the result of expressions and function
433 calls. With this release, expression indexes can collect analyze
434 statistics used by the optimizer in choosing the best execution
441 Sort duplicate btree items by file offset during creation (Manfred Koizar)
444 To improve performance, this release sorts duplicate index entries
445 in base table order so duplicate row lookups happen in sequential
446 order. This ordering is not maintained during table modification.
452 New two-stage sampling method for ANALYZE (Manfred Koizar)
455 This gives better statistics for asymmetric data distributions.
461 Optimize prepared queries only when first executed so constants can be used
462 for statistics (unnamed?) (Oliver Jowett)
465 Prepared statements optimize queries once and execute them many
466 times. While prepared queries run quickly, they benefit from
467 knowing all the values used in the query. This release allows
468 unnamed prepared queries to use the constants supplied by the
469 first query invocation as optimization hints.
475 Fix hash joins and aggregates of INET and CIDR data types (Tom)
478 Previous releases did not adjust to the fact that INET and CIDR
479 types have slightly different internal representations.
488 <title>Server Changes</title>
493 Add tablespace so admins can control disk layout (Gavin)
499 Add Point-in-time recovery (Simon Riggs)
505 Add new read-only server configuration parameter to query server
506 compile-time setting func_max_args, index_max_keys, namedatalen,
507 blcksz, have_int64_timestamp (Joe)
513 Make quoting of "sameuser", "samegroup", and "all" remove
514 special meaning of these terms in pg_hba.conf (Andrew)
520 Use clearer IPv6 name ::1/128 for localhost (Andrew)
526 Rename server configuration parameters SortMem and VacuumMem to
527 work_mem and maintenance_work_mem (Old names still supported)
531 This change was made to clarify that index creation uses
532 maintenance_work_mem and work_mem is for memory used for query
539 Allow logging of session disconnections using server configuration
540 log_disconnections (Andrew)
546 Add new server configuration parameter log_line_prefix to report useful
547 session information at the start of each log line (Andrew)
550 Information includes user name, database name, remote IP address,
551 and session start time.
557 Remove server configuration parameter log_pid, log_timestamp,
558 log_source_port; functionality superseded by log_line_prefix
565 Replace the virtual_host and tcpip_socket parameters with a unified
566 listen_addresses parameter (Andrew, Tom)
572 Listen on localhost by default, which eliminates the need for the -i
573 postmaster switch in many scenarios (Andrew)
576 Listening on localhost (127.0.0.1) opens no new security holes but
577 allows configurations like Win32 and JDBC, which do not support
578 local sockets, to work.
584 Remove 'syslog' server configuration parameter, and add more
585 logical 'log_destination' variable to control log output location
592 Change server configuration parameter log_statement to take values
593 "all, mod, ddl, none" which controls the queries output (Bruce)
596 This allows administrators to log only data definition changes or
597 date modification statements.
603 Allow backends to be safely terminated with SIGTERM (Bruce)
609 Allow configuration files to be placed outside the data directory using
610 server configuration parameters (mlw)
613 By default, configuration files sit in the top server directory.
614 With this addition, configuration files can be place outside the
615 data directory, easing administration.
624 <title>Query Changes</title>
629 Add savepoints (and nested transactions) (Alvaro)
635 Unsupported isolation levels are now accepted and promoted to the
636 spec-compliant level supported (Peter)
639 The SQL specification states that if a database doesn't support a
640 specific isolation level, it should use a more restrictive level.
641 This change adds that capability.
647 Allow BEGIN WORK to specify transaction isolation levels like START
654 More flexible rule/view permission checking (?) (Tom)
660 Implement dollar quoting to simplify single-quote usage (Andrew, Tom,
664 In previous releases, because single quotes had to be used to
665 quote a function's contents, the use of single quotes inside the
666 function required use of two single quotes or other error-prone
667 mechanisms. With this release we add the ability to use "dollar
668 quoting" to quote a block of text that eliminates the awkwardness
669 of single quotes used inside the function. Dollar quoting can
670 actually be used anywhere quoted text is used.
676 Make CASE val WHEN compval1 THEN evaluate 'val' only once (Tom)
679 This prevents CASE from re-evaluating the test expression multiple
680 times. This has benefits when the expression is complex or is
687 Replace max_expr_depth parameter with max_stack_depth parameter,
688 measured in kilobytes of stack size (Tom)
691 This gives us a fairly bulletproof defense against crashing due to
692 runaway recursive functions. Instead of measure the number of stack
693 calls, it measures the size of the stack.
699 Allow LIKE/ILIKE to appear in more places in a query (?) (Fabien Coelho)
705 Change EXECUTE to return a completion tag matching the executed statement
709 Previous releases return an EXECUTE tag for any EXECUTE call. In
710 this release, the tag returned will reflect the command executed.
716 Disallow NATURAL CROSS JOIN (Tom)
719 Such a clause makes no logical sense, but was not disabled in the
726 Allow arbitrary row expressions (Tom)
729 This allows columns to contain arbitrary composite types like rows
730 from other tables. It also allows functions to more easily take
731 rows as arguments and return row values.
740 <title>Object Manipulation Changes</title>
745 Add COMMENT ON casts, conversions, languages, operator classes, and
746 large objects (Christopher)
752 Add new server configuration parameter default_with_oids to
753 control the oid default during table creation (Neil)
756 This allows administrators to default all CREATE TABLE commands to
757 create tables without oid columns.
763 Add WITH / WITHOID OIDS clause to CREATE TABLE AS (Neil)
769 Allow ALTER TABLE DROP COLUMN to drop an OID column
770 (ALTER TABLE SET WITHOUT OIDS still works) (Tom)
776 Allowing composite types as table columns (Tom)
782 Allow ALTER ... ADD COLUMN with defaults and NOT NULL constraints works per SQL
786 This release will supply the appropriate defaults for columns
793 Add ALTER COLUMN TYPE to change column's type (Rod)
799 Allow multiple ALTER actions in a single ALTER TABLE command (Rod)
802 This is particularly useful for ALTER commands that rewrite the
803 table. By grouping ALTER commands together, the table can be
810 Allow ALTER DATABASE ... OWNER (Euler Taveira de Oliveira)
813 Previously this required modifying the system tables.
819 Allow temporary object creation to be limited to SECURITY DEFINER
820 functions (Sean Chittenden)
826 Add ALTER TABLE ... SET WITHOUT CLUSTER (Christopher)
829 Prior to this release, there was no way to clear an auto-cluster
830 specification except to modify the system tables.
836 Constraint/Index/SERIAL names are now table_column_type with numbers
837 appended to guarantee uniqueness within the schema (Tom)
840 The SQL specification states that such names should be unique
847 Allow ALTER TABLE to add SERIAL columns (Tom)
850 This is related to the new capability of adding default for new
857 Add pg_get_serial_sequence() to return the serial columns
858 sequence name(Christopher)
861 This allows automated scripts to reliabily find the serial
868 Allow changing the owners of aggregates, conversions, functions,
869 operators, operator classes, schemas, types, and tablespaces
876 Warn of primary/foreign key mismatch requires costly lookup
884 <title>Utility Command Changes</title>
889 Allow CREATE SCHEMA to create triggers, indexes, and sequences (Neil)
895 Add ALSO keyword to CREATE RULE (Fabien Coelho)
898 This allows ALSO to be added to rule creation to contrast it with
905 Add NOWAIT option to LOCK command (Tatsuo)
908 This allows the LOCK command to fail if they would have to wait for
915 Generate error if the COPY delimiter and NULL string conflict (Bruce)
921 Allow COPY to read comma-separated-value (CSV) files (Andrew, Bruce)
927 Erase MD5 user passwords when a user is renamed (Bruce)
930 PostgreSQL uses the user name as salt when encrypting passwords
931 via MD5. When a user name is changed, their salt no longer matches
932 the stored MD5 password so a notice is generated and the password
933 is cleared. A new password must then be assigned.
939 New pg_ctl 'kill' option for Win32 (Andrew)
942 Win32 does not have a 'kill' command to send signals to backends
943 so this capability was added to pg_ctl.
949 Information schema improvements
955 Add --pwfile option to initdb so the passwords can be set by GUI tools
962 Detect locale/encoding mismatch in initdb (Peter)
971 <title>Data Type and Function Changes</title>
976 Accept YYYY-monthname-DD as a date string (Tom)
982 Make netmask() and hostmask() functions return maximum-length
989 Change factorial function to NUMERIC (Gavin)
992 The NUMERIC data type more accurately represents the return value
999 to_char/to_date date conversion improvements (Kurt Roeckx)
1005 Make length() disregard trailing spaces in CHAR() (Gavin)
1008 This change was made for SQL specification compliance. (?)
1014 Warn of empty string being passed to oid/float4/float8 data types; 8.1
1015 will throw an error instead (Neil)
1021 Add "week" to date_trunc(Robert Creager)
1027 Allow int2/int4/int8/float4/float8 input routines to have leading
1028 or trailing whitespace (Neil)
1034 Fix to_char for 1 BC (previously it returned 1 AD) (Bruce)
1040 Fix to_char(year) for BC dates (previously it returned one less than
1041 the current year) (Bruce)
1047 Fix date_part() to return the proper millennium and century (With
1048 the previous version, the centuries and millennium had a wrong
1049 number and started the wrong year. Moreover century number 0,
1050 which does not exist in reality, lasted 200 years. Also,
1051 millennium number 0 lasted 2000 years) Fabien Coelho)
1057 Add ceiling() as an alias for ceil(), and power() as an alias for pow()
1058 for standards compliance (Neil)
1064 Change ln(), log(), power(), and sqrt() to emit the correct SQLSTATE
1065 error codes for certain error conditions, as specified by SQL2003
1072 Fix upper/lower() to work with multibyte encodings (Tom)
1078 Add boolean and bitwise integer AND/OR aggregates (Fabien Coelho)
1084 New inet_* functions to return network addresses for client
1085 and server (Sean Chittenden)
1091 Add function to determine the area of a closed path (Sean Chittenden)
1097 Add function to send cancel and terminate to other backends (Magnus)
1103 Allow interval plus datetime expressions (Tom)
1106 The reverse ordering, datetime plus interval, was already supported.
1115 <title>Server-Side Language Changes</title>
1120 Allow plpgsql parameter names to be referenced inside the function body (Tom)
1123 This basically creates an auto-reference to the names of the
1124 number parameters inside the function body.
1130 Do minimal syntax checking of plpgsql functions at creation time (Tom)
1133 This allows us to catch simple function errors sooner. (?)
1139 New plperl server-side language (Command Prompt, Andrew Dunstan)
1148 <title>psql Changes</title>
1153 Improve psql information display about database objects (Christopher)
1159 Allow psql to display group membership in \du and \dg (Markus Bertheau)
1165 Prevent psql \dn from showing temporary schemas (Bruce)
1171 Allow psql to handle tilde user expansion for file names (Zach Irmen)
1177 Allow psql to display fancy prompts, including color, via readline
1178 (Reece Hart, Chet Ramey)
1184 Have psql \copy match COPY syntax (Tom)
1190 Allow psql to show the location of syntax errors (Fabien Coelho, Tom)
1196 Add CLUSTER information to psql \d display (Bruce)
1202 Change psql \copy stdin/stdout to read from command input/output (Bruce)
1208 Add pstdin/pstdout to read from psql's stdin/stdout (Mark Feit)
1214 Add global psql config file, psqlrc.sample (Bruce)
1217 This allows a central file where global psql startup commands can
1224 Have psql \d+ indicate if the table has an oid column (Neil)
1230 Use binary file mode in psql when reading files so control-z is not seen
1237 Have \dn+ show permissions and description for schemas (Dennis
1247 <title>pg_dump Changes</title>
1252 Use dependency information to improve the reliability of pg_dump (Tom)
1258 Have pg_dump output objects in alphabetical order if possible (Tom)
1261 This should make it easier to identify changes between
1268 Allow pg_restore to ignore some SQL errors (Fabien Coelho)
1274 New start/stop of dump markers in pg_dump (Bruce)
1280 Add start/stop times for pg_dump/pg_dumpall in verbose mode (Bruce)
1286 Allow most pg_dump options in pg_dumpall (Christopher)
1292 Have pg_dump use ALTER OWNER rather than SET SESSION AUTHORIZATION
1293 by default (Christopher)
1302 <title>libpq Changes</title>
1307 Make libpq SIGPIPE thread-safe (Bruce)
1313 Add PQmbdsplen() which returns the "display length" of a character (Tatsuo)
1319 Add thread locking to SSL and Kerberos connections (Manfred Spraul)
1325 Allow PQoidValue(), PQcmdTuples(), and PQoidStatus() to work
1326 on EXECUTE commands (Neil)
1335 <title>Source Code Changes</title>
1340 Allow the database server to run natively on Win32 (Claudio, Magnus, Andrew)
1346 Shell script commands converted to C versions for Win32 support (Andrew)
1352 Use --with-docdir to choose installation location of documentation; also
1353 allow --infodir (Peter)
1359 Add --without-docdir to prevent installation of documentation (Peter)
1365 Upgrade to DocBook V4.2 SGML (Peter)
1371 New "PostgreSQL" CVS tag (Marc)
1374 This was done to make it easier for organizations with their own
1375 PostgreSQL CVS repositories.
1381 Clarify locking code (Manfred Koizar)
1387 Buffer manager cleanup (Neil)
1393 Remove pg_id client program (Peter)
1399 Decouple platform tests from cpu spinlock code (Bruce, Tom)
1405 Add inlined test-and-set code on PA-RISC for gcc (ViSolve,Tom)
1411 Improved i386 spinlock code (Manfred Spraul)
1417 Remove JDBC from source tree; now a separate project
1423 More accurately estimate memory and file descriptor usage (Tom)
1429 Improvements to the MAC OS-X startup scripts (Ray A.)
1435 New fsync test program (Bruce)
1441 Major documentation improvements (Neil, Peter)
1447 Remove the libpgtcl client interface
1453 Remove pg_encoding; not needed anymore
1459 Auto detect thread flags (no more manual testing) (Bruce)
1465 Use Olson's public domain timezone library (Magnus)
1471 Allow relocatable installs (Bruce)
1477 With threading enabled, use thread flags on Unixware for
1478 backend executables too (Bruce)
1481 Unixware can not mix threaded and non-threaded object files in the
1482 same executable, so everything must be compiled as threaded.
1488 psql now uses a lexical analyzer to process command strings
1494 New linked list data structure implementation (Neil)
1497 This improved performance by allowing list append operations to be
1504 Allow external interfaces to create their own server configuration
1505 parameters (Thomas Hallgren)
1511 Remove sync() call during checkpoint and do separate fsyncs() (Tom)
1517 New Brazilian version of FAQ (Euler Taveira de Oliveira)
1523 Remove alternate database locations using initlocation (Tom)
1529 New pgevent for Win32 logging
1535 Make libpq and ecpg build as proper OS X shard libraries (Tom)
1541 Add French FAQ (Guillaume Lelarge)
1550 <title>Contrib Changes</title>
1555 Many ecpg fixes, including SET DESCRIPTOR (Michael)
1561 Overhaul of /contrib/dblink (Joe)
1567 /contrib/dbmirror improvements
1573 New /contrib/xml2 (John Gray, Torchbox)
1579 Update /contrib/mysql
1585 New version of /contrib/btree_gist (Teodor)
1591 New /contrib/trgm, trigram matching for PostgreSQL (Teodor)
1597 Many /contrib/tsearch2 improvements (Teodor)
1603 Add double metaphone to /contrib/fuzzystrmatch (Andrew)
1613 <sect1 id="release-7-4-3">
1614 <title>Release 7.4.3</title>
1617 <title>Release date</title>
1618 <simpara>2004-06-14</simpara>
1622 This release contains a variety of fixes from 7.4.2.
1627 <title>Migration to version 7.4.3</title>
1630 A dump/restore is not required for those running 7.4.X.
1635 <title>Changes</title>
1640 <listitem><para>Fix temporary memory leak when using non-hashed aggregates (Tom)</para></listitem>
1641 <listitem><para>ECPG fixes, including some for Informix compatibility (Michael)</para></listitem>
1642 <listitem><para>Fixes for compiling with thread-safety, particularly Solaris (Bruce)</para></listitem>
1643 <listitem><para>Fix error in COPY IN termination when using the old network protocol (ljb)</para></listitem>
1644 <listitem><para>Several important fixes in pg_autovacuum, including fixes for
1645 large tables, unsigned oids, stability, temp tables, and debug mode
1646 (Matthew T. O'Connor)</para></listitem>
1647 <listitem><para>Fix problem with reading tar-format dumps on NetBSD and BSD/OS (Bruce)</para></listitem>
1648 <listitem><para>Several JDBC fixes</para></listitem>
1649 <listitem><para>Fix ALTER SEQUENCE RESTART where last_value equals the restart value (Tom)</para></listitem>
1650 <listitem><para>Repair failure to recalculate nested sub-selects (Tom)</para></listitem>
1651 <listitem><para>Fix problems with non-constant expressions in LIMIT/OFFSET</para></listitem>
1652 <listitem><para>Support FULL JOIN with no join clause, such as X FULL JOIN Y ON TRUE (Tom)</para></listitem>
1653 <listitem><para>Fix another zero-column table bug (Tom)</para></listitem>
1654 <listitem><para>Improve handling of non-qualified identifiers in GROUP BY clauses in sub-selects (Tom)</para>
1656 Select-list aliases within the sub-select will now take precedence over
1657 names from outer query levels.
1659 <listitem><para>Do not generate <quote>NATURAL CROSS JOIN</> when decompiling rules (Tom)</para></listitem>
1660 <listitem><para>Add checks for invalid field length in binary COPY (Tom)</para>
1662 This fixes a difficult-to-exploit security hole.
1664 <listitem><para>Avoid locking conflict between ANALYZE and LISTEN/NOTIFY</para></listitem>
1665 <listitem><para>Numerous translation updates (various contributors)</para></listitem>
1671 <sect1 id="release-7-4-2">
1672 <title>Release 7.4.2</title>
1675 <title>Release date</title>
1676 <simpara>2004-03-08</simpara>
1680 This release contains a variety of fixes from 7.4.1.
1685 <title>Migration to version 7.4.2</title>
1688 A dump/restore is not required for those running 7.4.X. However,
1689 it may be advisable as the easiest method of incorporating fixes for
1690 two errors that have been found in the initial contents of 7.4.X system
1691 catalogs. A dump/initdb/reload sequence using 7.4.2's initdb will
1692 automatically correct these problems.
1696 The more severe of the two errors is that data type <type>anyarray</>
1697 has the wrong alignment label; this is a problem because the
1698 <structname>pg_statistic</> system catalog uses <type>anyarray</>
1699 columns. The mislabeling can cause planner misestimations and even
1700 crashes when planning queries that involve <literal>WHERE</> clauses on
1701 double-aligned columns (such as <type>float8</> and <type>timestamp</>).
1702 It is strongly recommended that all installations repair this error,
1703 either by initdb or by following the manual repair procedure given
1708 The lesser error is that the system view <structname>pg_settings</>
1709 ought to be marked as having public update access, to allow
1710 <literal>UPDATE pg_settings</> to be used as a substitute for
1711 <command>SET</>. This can also be fixed either by initdb or manually,
1712 but it is not necessary to fix unless you want to use <literal>UPDATE
1717 If you wish not to do an initdb, the following procedure will work
1718 for fixing <structname>pg_statistic</>. As the database superuser,
1722 -- clear out old data in pg_statistic:
1723 DELETE FROM pg_statistic;
1724 VACUUM pg_statistic;
1725 -- this should update 1 row:
1726 UPDATE pg_type SET typalign = 'd' WHERE oid = 2277;
1727 -- this should update 6 rows:
1728 UPDATE pg_attribute SET attalign = 'd' WHERE atttypid = 2277;
1730 -- At this point you MUST start a fresh backend to avoid a crash!
1732 -- repopulate pg_statistic:
1736 This can be done in a live database, but beware that all backends
1737 running in the altered database must be restarted before it is safe to
1738 repopulate <structname>pg_statistic</>.
1742 To repair the <structname>pg_settings</> error, simply do:
1744 GRANT SELECT, UPDATE ON pg_settings TO PUBLIC;
1749 The above procedures must be carried out in <emphasis>each</> database
1750 of an installation, including <literal>template1</>, and ideally
1751 including <literal>template0</> as well. If you do not fix the
1752 template databases then any subsequently created databases will contain
1753 the same errors. <literal>template1</> can be fixed in the same way
1754 as any other database, but fixing <literal>template0</> requires
1755 additional steps. First, from any database issue
1757 UPDATE pg_database SET datallowconn = true WHERE datname = 'template0';
1759 Next connect to <literal>template0</> and perform the above repair
1760 procedures. Finally, do
1762 -- re-freeze template0:
1764 -- and protect it against future alterations:
1765 UPDATE pg_database SET datallowconn = false WHERE datname = 'template0';
1771 <title>Changes</title>
1774 Release 7.4.2 incorporates all the fixes included in release 7.3.6,
1775 plus the following fixes:
1779 <listitem><para>Fix <structname>pg_statistics</> alignment bug that could crash optimizer</para>
1780 <para>See above for details about this problem.</para></listitem>
1781 <listitem><para>Allow non-super users to update <structname>pg_settings</></para></listitem>
1782 <listitem><para>Fix several optimizer bugs, most of which led to
1783 <quote>variable not found in subplan target lists</> errors</para></listitem>
1784 <listitem><para>Avoid out-of-memory failure during startup of large multiple
1785 index scan</para></listitem>
1786 <listitem><para>Fix multibyte problem that could lead to <quote>out of
1787 memory</> error during <command>COPY IN</></para></listitem>
1788 <listitem><para>Fix problems with <command>SELECT INTO</> / <command>CREATE
1789 TABLE AS</> from tables without OIDs</para></listitem>
1790 <listitem><para>Fix problems with <filename>alter_table</> regression test
1791 during parallel testing</para></listitem>
1792 <listitem><para>Fix problems with hitting open file limit, especially on OS X (Tom)</para></listitem>
1793 <listitem><para>Partial fix for Turkish-locale issues</para>
1794 <para>initdb will succeed now in Turkish locale, but there are still some
1795 inconveniences associated with the <literal>i/I</> problem.</para></listitem>
1796 <listitem><para>Make pg_dump set client encoding on restore</para></listitem>
1797 <listitem><para>Other minor pg_dump fixes</para></listitem>
1798 <listitem><para>Allow ecpg to again use C keywords as column names (Michael)</para></listitem>
1799 <listitem><para>Added ecpg <literal>WHENEVER NOT_FOUND</> to
1800 <literal>SELECT/INSERT/UPDATE/DELETE</> (Michael)</para></listitem>
1801 <listitem><para>Fix ecpg crash for queries calling set-returning functions (Michael)</para></listitem>
1802 <listitem><para>Various other ecpg fixes (Michael)</para></listitem>
1803 <listitem><para>Fixes for Borland compiler</para></listitem>
1804 <listitem><para>Thread build improvements (Bruce)</para></listitem>
1805 <listitem><para>Various other build fixes</para></listitem>
1806 <listitem><para>Various JDBC fixes</para></listitem>
1812 <sect1 id="release-7-4-1">
1813 <title>Release 7.4.1</title>
1816 <title>Release date</title>
1817 <simpara>2003-12-22</simpara>
1821 This release contains a variety of fixes from 7.4.
1826 <title>Migration to version 7.4.1</title>
1829 A dump/restore is <emphasis>not</emphasis> required for those
1834 If you want to install the fixes in the information schema
1835 you need to reload it into the database.
1836 This is either accomplished by initializing a new cluster
1837 by running <command>initdb</command>, or by running the following
1838 sequence of SQL commands in each database (ideally including
1839 <literal>template1</literal>) as a superuser in
1840 <application>psql</application>, after installing the new release:
1842 DROP SCHEMA information_schema CASCADE;
1843 \i /usr/local/pgsql/share/information_schema.sql
1845 Substitute your installation path in the second command.
1851 <title>Changes</title>
1854 <listitem><para>Fixed bug in <command>CREATE SCHEMA</command> parsing in ECPG (Michael)</para></listitem>
1855 <listitem><para>Fix compile error when <option>--enable-thread-safety</option> and <option>--with-perl</option> are used together (Peter)</para></listitem>
1856 <listitem><para>Fix for subqueries that used hash joins (Tom)</para>
1858 Certain subqueries that used hash joins would crash because of
1859 improperly shared structures.
1861 <listitem><para>Fix free space map compaction bug (Tom)</para>
1863 This fixes a bug where compaction of the free space map could lead
1864 to a database server shutdown.
1867 <listitem><para>Fix for Borland compiler build of libpq (Bruce)</para></listitem>
1868 <listitem><para>Fix <function>netmask()</function> and <function>hostmask()</function> to return the maximum-length masklen (Tom)</para>
1870 Fix these functions to return values consistent with pre-7.4
1874 <listitem><para>Several <filename>contrib/pg_autovacuum</filename> fixes</para>
1876 Fixes include improper variable initialization, missing vacuum after
1877 <command>TRUNCATE</command>, and duration computation overflow for long vacuums.
1880 <listitem><para>Allow compile of <filename>contrib/cube</filename> under Cygwin (Jason Tishler)</para></listitem>
1881 <listitem><para>Fix Solaris use of password file when no passwords are defined (Tom)</para>
1883 Fix crash on Solaris caused by use of any type of password
1884 authentication when no passwords were defined.
1887 <listitem><para>JDBC fix for thread problems, other fixes</para></listitem>
1888 <listitem><para>Fix for <type>bytea</type> index lookups (Joe)</para></listitem>
1889 <listitem><para>Fix information schema for bit data types (Peter)</para></listitem>
1890 <listitem><para>Force zero_damaged_pages to be on during recovery from WAL</para></listitem>
1891 <listitem><para>Prevent some obscure cases of <quote>variable not in subplan target lists</quote></para></listitem>
1892 <listitem><para>Make <function>PQescapeBytea</function> and <function>byteaout</function> consistent with each other (Joe)</para></listitem>
1893 <listitem><para>Escape <type>bytea</type> output for bytes > 0x7e(Joe)</para>
1895 If different client encodings are used for <type>bytea</type> output and input, it
1896 is possible for <type>bytea</type> values to be corrupted by the differing
1897 encodings. This fix escapes all bytes that might be affected.
1900 <listitem><para>Added missing <function>SPI_finish()</function> calls to dblink's <function>get_tuple_of_interest()</function> (Joe)</para></listitem>
1901 <listitem><para>New Czech FAQ</para></listitem>
1902 <listitem><para>Fix information schema view <literal>constraint_column_usage</literal> for foreign keys (Peter)</para></listitem>
1903 <listitem><para>ECPG fixes (Michael)</para></listitem>
1904 <listitem><para>Fix bug with multiple <literal>IN</literal> subqueries and joins in the subqueries (Tom)</para></listitem>
1905 <listitem><para>Alllow <literal>COUNT('x')</literal> to work (Tom)</para></listitem>
1906 <listitem><para>Install ECPG include files for Informix compatibility into separate directory (Peter)</para>
1908 Some names of ECPG include files for Informix compatibility conflicted with operating system include files.
1909 By installing them in their own directory, name conflicts have been reduced.
1912 <listitem><para>Fix SSL memory leak (Neil)</para>
1914 This release fixes a bug in 7.4 where SSL didn't free all memory it allocated.
1917 <listitem><para>Prevent <filename>pg_service.conf</filename> from using service name as default dbname (Bruce)</para></listitem>
1918 <listitem><para>Fix local ident authentication on FreeBSD (Tom)</para></listitem>
1924 <sect1 id="release-7-4">
1925 <title>Release 7.4</title>
1928 <title>Release date</title>
1929 <simpara>2003-11-17</simpara>
1933 <title>Overview</title>
1936 Major changes in this release:
1941 <literal>IN</literal> / <literal>NOT IN</literal> subqueries are
1942 now much more efficient
1947 In previous releases, <literal>IN</literal>/<literal>NOT
1948 IN</literal> subqueries were joined to the upper query by
1949 sequentially scanning the subquery looking for a match. The
1950 7.4 code uses the same sophisticated techniques used by
1951 ordinary joins and so is much faster. An
1952 <literal>IN</literal> will now usually be as fast as or faster
1953 than an equivalent <literal>EXISTS</literal> subquery; this
1954 reverses the conventional wisdom that applied to previous
1962 Improved <literal>GROUP BY</literal> processing by using hash buckets
1967 In previous releases, rows to be grouped had to be sorted
1968 first. The 7.4 code can do <literal>GROUP BY</literal>
1969 without sorting, by accumulating results into a hash table
1970 with one entry per group. It will still use the sort
1971 technique, however, if the hash table is estimated to be too
1972 large to fit in <varname>sort_mem</>.
1979 New multikey hash join capability
1984 In previous releases, hash joins could only occur on single
1985 keys. This release allows multicolumn hash joins.
1992 Queries using the explicit <literal>JOIN</literal> syntax are
1993 now better optimized
1998 Prior releases evaluated queries using the explicit
1999 <literal>JOIN</literal> syntax only in the order implied by
2000 the syntax. 7.4 allows full optimization of these queries,
2001 meaning the optimizer considers all possible join orderings
2002 and chooses the most efficient. Outer joins, however, must
2003 still follow the declared ordering.
2010 Faster and more powerful regular expression code
2015 The entire regular expression module has been replaced with a
2016 new version by Henry Spencer, originally written for Tcl. The
2017 code greatly improves performance and supports several flavors
2018 of regular expressions.
2025 Function-inlining for simple SQL functions
2030 Simple SQL functions can now be inlined by including their SQL
2031 in the main query. This improves performance by eliminating
2032 per-call overhead. That means simple SQL functions now
2040 Full support for IPv6 connections and IPv6 address data types
2045 Previous releases allowed only IPv4 connections, and the IP
2046 data types only supported IPv4 addresses. This release adds
2047 full IPv6 support in both of these areas.
2054 Major improvements in SSL performance and reliability
2059 Several people very familiar with the SSL API have overhauled
2060 our SSL code to improve SSL key negotiation and error
2068 Make free space map efficiently reuse empty index pages,
2069 and other free space management improvements
2074 In previous releases, B-tree index pages that were left empty
2075 because of deleted rows could only be reused by rows with
2076 index values similar to the rows originally indexed on that
2077 page. In 7.4, <command>VACUUM</command> records empty index
2078 pages and allows them to be reused for any future index rows.
2085 SQL-standard information schema
2090 The information schema provides a standardized and stable way
2091 to access information about the schema objects defined in a
2099 Cursors conform more closely to the SQL standard
2104 The commands <command>FETCH</command> and
2105 <command>MOVE</command> have been overhauled to conform more
2106 closely to the SQL standard.
2113 Cursors can exist outside transactions
2118 These cursors are also called holdable cursors.
2125 New client-to-server protocol
2130 The new protocol adds error codes, more status information,
2131 faster startup, better support for binary data transmission,
2132 parameter values separated from SQL commands, prepared
2133 statements available at the protocol level, and cleaner
2134 recovery from <command>COPY</command> failures. The older
2135 protocol is still supported by both server and clients.
2142 <application>libpq</application> and
2143 <application>ECPG</application> applications are now fully
2149 While previous <application>libpq</application> releases
2150 already supported threads, this release improves thread safety
2151 by fixing some non-thread-safe code that was used during
2152 database connection startup. The <command>configure</command>
2153 option <option>--enable-thread-safety</option> must be used to
2154 enable this feature.
2161 New version of full-text indexing
2166 A new full-text indexing suite is available in
2167 <filename>contrib/tsearch2</filename>.
2179 The new autovacuum tool in
2180 <filename>contrib/autovacuum</filename> monitors the database
2181 statistics tables for
2182 <command>INSERT</command>/<command>UPDATE</command>/<command>DELETE</command>
2183 activity and automatically vacuums tables when needed.
2190 Array handling has been improved and moved into the server core
2195 Many array limitations have been removed, and arrays behave
2196 more like fully-supported data types.
2205 <title>Migration to version 7.4</title>
2208 A dump/restore using <application>pg_dump</application> is
2209 required for those wishing to migrate data from any previous
2214 Observe the following incompatibilities:
2219 The server-side autocommit setting was removed and
2220 reimplemented in client applications and languages.
2221 server-side autocommit was causing too many problems with
2222 languages and applications that wanted to control their own
2223 autocommit behavior, so autocommit was removed from the server
2224 and added to individual client APIs as appropriate.
2230 Error message wording has changed substantially in this
2231 release. Significant effort was invested to make the messages
2232 more consistent and user-oriented. If your applications try to
2233 detect different error conditions by parsing the error message,
2234 you are strongly encouraged to use the new error code facility instead.
2240 Inner joins using the explicit <literal>JOIN</literal> syntax
2241 may behave differently because they are now better
2248 A number of server configuration parameters have been renamed
2249 for clarity, primarily those related to
2256 <literal>FETCH 0</literal> or <literal>MOVE 0</literal> now
2257 does nothing. In prior releases, <literal>FETCH 0</literal>
2258 would fetch all remaining rows, and <literal>MOVE 0</literal>
2259 would move to the end of the cursor.
2265 <command>FETCH</command> and <command>MOVE</command> now return
2266 the actual number of rows fetched/moved, or zero if at the
2267 beginning/end of the cursor. Prior releases would return the
2268 row count passed to the command, not the number of rows
2269 actually fetched or moved.
2275 <command>COPY</command> now can process files that use
2276 carriage-return or carriage-return/line-feed end-of-line
2277 sequences. Literal carriage-returns and line-feeds are no
2278 longer accepted in data values; use <literal>\r</literal> and
2279 <literal>\n</literal> instead.
2285 Trailing spaces are now trimmed when converting from type
2286 <type>char(<replaceable>n</>)</type> to
2287 <type>varchar(<replaceable>n</>)</type> or <type>text</type>.
2288 This is what most people always expected to happen anyway.
2294 The data type <type>float(<replaceable>p</>)</type> now
2295 measures <replaceable>p</> in binary digits, not decimal
2296 digits. The new behavior follows the SQL standard.
2302 Ambiguous date values now must match the ordering specified by
2303 the <varname>datestyle</varname> setting. In prior releases, a
2304 date specification of <literal>10/20/03</> was interpreted as a
2305 date in October even if <varname>datestyle</> specified that
2306 the day should be first. 7.4 will throw an error if a date
2307 specification is invalid for the current setting of
2308 <varname>datestyle</>.
2314 The functions <function>oidrand</function>,
2315 <function>oidsrand</function>, and
2316 <function>userfntest</function> have been removed. These
2317 functions were determined to be no longer useful.
2323 String literals specifying time-varying date/time values, such
2324 as <literal>'now'</literal> or <literal>'today'</literal> will
2325 no longer work as expected in column default expressions; they
2326 now cause the time of the table creation to be the default, not
2327 the time of the insertion. Functions such as
2328 <function>now()</>, <function>current_timestamp</>, or
2329 <function>current_date</function> should be used instead.
2333 In previous releases, there was special code so that strings
2334 such as <literal>'now'</literal> were interpreted at
2335 <command>INSERT</> time and not at table creation time, but
2336 this work around didn't cover all cases. Release 7.4 now
2337 requires that defaults be defined properly using functions such
2338 as <function>now()</> or <function>current_timestamp</>. These
2339 will work in all situations.
2345 The dollar sign (<literal>$</>) is no longer allowed in
2346 operator names. It can instead be a non-first character in
2347 identifiers. This was done to improve compatibility with other
2348 database systems, and to avoid syntax problems when parameter
2349 placeholders (<literal>$<replaceable>n</></>) are written
2350 adjacent to operators.
2358 <title>Changes</title>
2361 Below you will find a detailed account of the changes between
2362 release 7.4 and the previous major release.
2366 <title>Server Operation Changes</title>
2371 Allow IPv6 server connections (Nigel Kukard, Johan Jordaan,
2372 Bruce, Tom, Kurt Roeckx, Andrew Dunstan)
2378 Fix SSL to handle errors cleanly (Nathan Mueller)
2381 In prior releases, certain SSL API error reports were not
2382 handled correctly. This release fixes those problems.
2388 SSL protocol security and performance improvements (Sean Chittenden)
2391 SSL key renegotiation was happening too frequently, causing poor
2392 SSL performance. Also, initial key handling was improved.
2398 Print lock information when a deadlock is detected (Tom)
2401 This allows easier debugging of deadlock situations.
2407 Update <filename>/tmp</filename> socket modification times
2408 regularly to avoid their removal (Tom)
2411 This should help prevent <filename>/tmp</filename> directory
2412 cleaner administration scripts from removing server socket
2417 <listitem><para>Enable PAM for Mac OS X (Aaron Hillegass)</para></listitem>
2420 <para>Make B-tree indexes fully WAL-safe (Tom)</para>
2422 In prior releases, under certain rare cases, a server crash
2423 could cause B-tree indexes to become corrupt. This release
2424 removes those last few rare cases.
2428 <listitem><para>Allow B-tree index compaction and empty page reuse (Tom)</para></listitem>
2432 Fix inconsistent index lookups during split of first root page (Tom)
2435 In prior releases, when a single-page index split into two
2436 pages, there was a brief period when another database session
2437 could miss seeing an index entry. This release fixes that rare
2442 <listitem><para>Improve free space map allocation logic (Tom)</para></listitem>
2445 <para>Preserve free space information between server restarts (Tom)</para>
2447 In prior releases, the free space map was not saved when the
2448 postmaster was stopped, so newly started servers had no free
2449 space information. This release saves the free space map, and
2450 reloads it when the server is restarted.
2454 <listitem><para>Add start time to <literal>pg_stat_activity</literal> (Neil)</para></listitem>
2455 <listitem><para>New code to detect corrupt disk pages; erase with <varname>zero_damaged_pages</varname> (Tom)</para></listitem>
2456 <listitem><para>New client/server protocol: faster, no username length limit, allow clean exit from <command>COPY</command> (Tom)</para></listitem>
2457 <listitem><para>Add transaction status, table ID, column ID to client/server protocol (Tom)</para></listitem>
2458 <listitem><para>Add binary I/O to client/server protocol (Tom)</para></listitem>
2459 <listitem><para>Remove autocommit server setting; move to client applications (Tom)</para></listitem>
2460 <listitem><para>New error message wording, error codes, and three levels of error detail (Tom, Joe, Peter)</para></listitem>
2465 <title>Performance Improvements</title>
2468 <listitem><para>Add hashing for <literal>GROUP BY</literal> aggregates (Tom)</para></listitem>
2469 <listitem><para>Make nested-loop joins be smarter about multicolumn indexes (Tom)</para></listitem>
2470 <listitem><para>Allow multikey hash joins (Tom)</para></listitem>
2471 <listitem><para>Improve constant folding (Tom)</para></listitem>
2472 <listitem><para>Add ability to inline simple SQL functions (Tom)</para></listitem>
2475 <para>Reduce memory usage for queries using complex functions (Tom)</para>
2477 In prior releases, functions returning allocated memory would
2478 not free it until the query completed. This release allows the
2479 freeing of function-allocated memory when the function call
2480 completes, reducing the total memory used by functions.
2485 <para>Improve GEQO optimizer performance (Tom)</para>
2487 This release fixes several inefficiencies in the way the GEQO optimizer
2488 manages potential query paths.
2494 Allow <literal>IN</>/<literal>NOT IN</> to be handled via hash
2501 Improve <literal>NOT IN (<replaceable>subquery</>)</literal>
2508 Allow most <literal>IN</literal> subqueries to be processed as
2515 Pattern matching operations can use indexes regardless of
2519 There is no way for non-ASCII locales to use the standard
2520 indexes for <literal>LIKE</literal> comparisons. This release
2521 adds a way to create a special index for
2522 <literal>LIKE</literal>.
2527 <para>Allow the postmaster to preload libraries using <varname>preload_libraries</varname> (Joe)</para>
2529 For shared libraries that require a long time to load, this
2530 option is available so the library can be preloaded in the
2531 postmaster and inherited by all database sessions.
2537 Improve optimizer cost computations, particularly for subqueries (Tom)
2543 Avoid sort when subquery <literal>ORDER BY</literal> matches upper query (Tom)
2549 Deduce that <literal>WHERE a.x = b.y AND b.y = 42</literal> also
2550 means <literal>a.x = 42</literal> (Tom)
2556 Allow hash/merge joins on complex joins (Tom)
2562 Allow hash joins for more data types (Tom)
2568 Allow join optimization of explicit inner joins, disable with
2569 <varname>join_collapse_limit</varname> (Tom)
2575 Add parameter <varname>from_collapse_limit</varname> to control
2576 conversion of subqueries to joins (Tom)
2582 Use faster and more powerful regular expression code from Tcl
2583 (Henry Spencer, Tom)
2589 Use bit-mapped relation sets in the optimizer (Tom)
2594 <para>Improve connection startup time (Tom)</para>
2596 The new client/server protocol requires fewer network packets to
2597 start a database session.
2603 Improve trigger/constraint performance (Stephan)
2609 Improve speed of <literal>col IN (const, const, const, ...)</literal> (Tom)
2615 Fix hash indexes which were broken in rare cases (Tom)
2619 <listitem><para>Improve hash index concurrency and speed (Tom)</para>
2621 Prior releases suffered from poor hash index performance,
2622 particularly for high concurrency situations. This release fixes
2623 that, and the development group is interested in reports
2624 comparing B-tree and hash index performance.
2629 <para>Align shared buffers on 32-byte boundary for copy speed improvement (Manfred Spraul)</para>
2631 Certain CPU's perform faster data copies when addresses are
2637 <para>Data type <type>numeric</type> reimplemented for better performance (Tom)</para>
2639 <type>numeric</type> used to be stored in base 100. The new code
2640 uses base 10000, for significantly better performance.
2647 <title>Server Configuration Changes</title>
2651 <para>Rename server parameter <varname>server_min_messages</> to <varname>log_min_messages</> (Bruce)</para>
2653 This was done so most parameters that control the server logs
2654 begin with <literal>log_</>.
2658 <listitem><para>Rename <varname>show_*_stats</> to <varname>log_*_stats</> (Bruce)</para></listitem>
2659 <listitem><para>Rename <varname>show_source_port</> to <varname>log_source_port</> (Bruce)</para></listitem>
2660 <listitem><para>Rename <varname>hostname_lookup</> to <varname>log_hostname</> (Bruce)</para></listitem>
2663 <para>Add <varname>checkpoint_warning</> to warn of excessive checkpointing (Bruce)</para>
2665 In prior releases, it was difficult to determine if checkpoint
2666 was happening too frequently. This feature adds a warning to the
2667 server logs when excessive checkpointing happens.
2671 <listitem><para>New read-only server parameters for localization (Tom)</para></listitem>
2675 Change debug server log messages to output as <literal>DEBUG</>
2676 rather than <literal>LOG</> (Bruce)
2681 <para>Prevent server log variables from being turned off by non-superusers (Bruce)</para>
2683 This is a security feature so non-superusers cannot disable
2684 logging that was enabled by the administrator.
2690 <varname>log_min_messages</>/<varname>client_min_messages</> now
2691 controls <varname>debug_*</> output (Bruce)
2694 This centralizes client debug information so all debug output
2695 can be sent to either the client or server logs.
2700 <para>Add Mac OS X Rendezvous server support (Chris Campbell)</para>
2702 This allows Mac OS X hosts to query the network for available
2703 <productname>PostgreSQL</productname> servers.
2709 Add ability to print only slow statements using
2710 <varname>log_min_duration_statement</varname>
2714 This is an often requested debugging feature that allows
2715 administrators to see only slow queries in their server logs.
2720 <para>Allow <filename>pg_hba.conf</filename> to accept netmasks in CIDR format (Andrew Dunstan)</para>
2722 This allows administrators to merge the host IP address and
2723 netmask fields into a single CIDR field in <filename>pg_hba.conf</filename>.
2727 <listitem><para>New read-only parameter <varname>is_superuser</varname> (Tom)</para></listitem>
2730 <para>New parameter <varname>log_error_verbosity</varname> to control error detail (Tom)</para>
2732 This works with the new error reporting feature to supply
2733 additional error information like hints, file names and line
2739 <para><literal>postgres --describe-config</literal> now dumps server config variables (Aizaz Ahmed, Peter)</para>
2741 This option is useful for administration tools that need to know
2742 the configuration variable names and their minimums, maximums,
2743 defaults, and descriptions.
2749 Add new columns in <literal>pg_settings</literal>:
2750 <literal>context</>, <literal>type</>, <literal>source</>,
2751 <literal>min_val</>, <literal>max_val</> (Joe)
2757 Make default <varname>shared_buffers</> 1000 and
2758 <varname>max_connections</> 100, if possible (Tom)
2761 Prior versions defaulted to 64 shared buffers so <productname>PostgreSQL</productname>
2762 would start on even very old systems. This release tests the
2763 amount of shared memory allowed by the platform and selects more
2764 reasonable default values if possible. Of course, users are
2765 still encouraged to evaluate their resource load and size
2766 <varname>shared_buffers</varname> accordingly.
2772 New <filename>pg_hba.conf</filename> record type
2773 <literal>hostnossl</> to prevent SSL connections (Jon
2777 In prior releases, there was no way to prevent SSL connections
2778 if both the client and server supported SSL. This option allows
2785 Remove parameter <varname>geqo_random_seed</varname>
2792 Add server parameter <varname>regex_flavor</varname> to control regular expression processing (Tom)
2798 Make <command>pg_ctl</command> better handle nonstandard ports (Greg)
2805 <title>Query Changes</title>
2808 <listitem><para>New SQL-standard information schema (Peter)</para></listitem>
2809 <listitem><para>Add read-only transactions (Peter)</para></listitem>
2810 <listitem><para>Print key name and value in foreign-key violation messages (Dmitry Tkach)</para></listitem>
2813 <para>Allow users to see their own queries in <literal>pg_stat_activity</literal> (Kevin Brown)</para>
2815 In prior releases, only the superuser could see query strings
2816 using <literal>pg_stat_activity</literal>. Now ordinary users
2817 can see their own query strings.
2822 <para>Fix aggregates in subqueries to match SQL standard (Tom)</para>
2824 The SQL standard says that an aggregate function appearing
2825 within a nested subquery belongs to the outer query if its
2826 argument contains only outer-query variables. Prior
2827 <productname>PostgreSQL</productname> releases did not handle
2828 this fine point correctly.
2833 <para>Add option to prevent auto-addition of tables referenced in query (Nigel J. Andrews)</para>
2835 By default, tables mentioned in the query are automatically
2836 added to the <literal>FROM</> clause if they are not already
2837 there. This is compatible with historic
2838 <productname>POSTGRES</productname> behavior but is contrary to
2839 the SQL standard. This option allows selecting
2840 standard-compatible behavior.
2845 <para>Allow <literal>UPDATE ... SET col = DEFAULT</literal> (Rod)</para>
2847 This allows <command>UPDATE</command> to set a column to its
2848 declared default value.
2853 <para>Allow expressions to be used in <literal>LIMIT</>/<literal>OFFSET</> (Tom)</para>
2855 In prior releases, <literal>LIMIT</>/<literal>OFFSET</> could
2856 only use constants, not expressions.
2861 <para>Implement <literal>CREATE TABLE AS EXECUTE</literal> (Neil, Peter)</para>
2867 <title>Object Manipulation Changes</title>
2871 <para>Make <command>CREATE SEQUENCE</command> grammar more conforming to SQL 2003 (Neil)</para>
2875 <para>Add statement-level triggers (Neil)</para>
2877 While this allows a trigger to fire at the end of a statement,
2878 it does not allow the trigger to access all rows modified by the
2879 statement. This capability is planned for a future release.
2884 <para>Add check constraints for domains (Rod)</para>
2886 This greatly increases the usefulness of domains by allowing
2887 them to use check constraints.
2892 <para>Add <command>ALTER DOMAIN</command> (Rod)</para>
2894 This allows manipulation of existing domains.
2899 <para>Fix several zero-column table bugs (Tom)</para>
2901 <productname>PostgreSQL</productname> supports zero-column tables. This fixes various bugs
2902 that occur when using such tables.
2907 <para>Have <literal>ALTER TABLE ... ADD PRIMARY KEY</literal> add not-null constraint (Rod)</para>
2909 In prior releases, <literal>ALTER TABLE ... ADD
2910 PRIMARY</literal> would add a unique index, but not a not-null
2911 constraint. That is fixed in this release.
2915 <listitem><para>Add <literal>ALTER TABLE ... WITHOUT OIDS</literal> (Rod)</para>
2917 This allows control over whether new and updated rows will have
2918 an OID column. This is most useful for saving storage space.
2924 Add <literal>ALTER SEQUENCE</literal> to modify minimum, maximum,
2925 increment, cache, cycle values (Rod)
2930 <para>Add <literal>ALTER TABLE ... CLUSTER ON</literal> (Alvaro Herrera)</para>
2932 This command is used by <command>pg_dump</command> to record the
2933 cluster column for each table previously clustered. This
2934 information is used by database-wide cluster to cluster all
2935 previously clustered tables.
2939 <listitem><para>Improve automatic type casting for domains (Rod, Tom)</para></listitem>
2940 <listitem><para>Allow dollar signs in identifiers, except as first character (Tom)</para></listitem>
2941 <listitem><para>Disallow dollar signs in operator names, so <literal>x=$1</> works (Tom)</para></listitem>
2945 Allow copying table schema using <literal>LIKE
2946 <replaceable>subtable</replaceable></literal>, also SQL 2003
2947 feature <literal>INCLUDING DEFAULTS</literal> (Rod)
2953 Add <literal>WITH GRANT OPTION</literal> clause to
2954 <command>GRANT</command> (Peter)
2957 This enabled <command>GRANT</command> to give other users the
2958 ability to grant privileges on a object.
2965 <title>Utility Command Changes</title>
2969 <para>Add <literal>ON COMMIT</literal> clause to <command>CREATE TABLE</command> for temporary tables (Gavin)</para>
2971 This adds the ability for a table to be dropped or all rows
2972 deleted on transaction commit.
2977 <para>Allow cursors outside transactions using <literal>WITH HOLD</literal> (Neil)</para>
2979 In previous releases, cursors were removed at the end of the
2980 transaction that created them. Cursors can now be created with
2981 the <literal>WITH HOLD</literal> option, which allows them to
2982 continue to be accessed after the creating transaction has
2988 <para><literal>FETCH 0</literal> and <literal>MOVE 0 </literal> now do nothing (Bruce)</para>
2990 In previous releases, <literal>FETCH 0</literal> fetched all
2991 remaining rows, and <literal>MOVE 0</literal> moved to the end
2998 Cause <command>FETCH</command> and <command>MOVE</command> to
2999 return the number of rows fetched/moved, or zero if at the
3000 beginning/end of cursor, per SQL standard (Bruce)
3003 In prior releases, the row count returned by
3004 <command>FETCH</command> and <command>MOVE</command> did not
3005 accurately reflect the number of rows processed.
3010 <para>Properly handle <literal>SCROLL</literal> with cursors, or
3011 report an error (Neil)</para>
3013 Allowing random access (both forward and backward scrolling) to
3014 some kinds of queries cannot be done without some additional
3015 work. If <literal>SCROLL</literal> is specified when the cursor
3016 is created, this additional work will be performed. Furthermore,
3017 if the cursor has been created with <literal>NO SCROLL</literal>,
3018 no random access is allowed.
3024 Implement SQL-compatible options <literal>FIRST</>,
3025 <literal>LAST</>, <literal>ABSOLUTE <replaceable>n</></>,
3026 <literal>RELATIVE <replaceable>n</></> for
3027 <command>FETCH</command> and <command>MOVE</command> (Tom)
3032 <para>Allow <command>EXPLAIN</command> on <command>DECLARE CURSOR</command> (Tom)</para>
3036 <para>Allow <command>CLUSTER</command> to use index marked as pre-clustered by default (Alvaro Herrera)</para>
3040 <para>Allow <command>CLUSTER</command> to cluster all tables (Alvaro Herrera)</para>
3042 This allows all previously clustered tables in a database to be
3043 reclustered with a single command.
3047 <listitem><para>Prevent <command>CLUSTER</command> on partial indexes (Tom)</para></listitem>
3049 <listitem><para>Allow DOS and Mac line-endings in <command>COPY</> files (Bruce)</para></listitem>
3053 Disallow literal carriage return as a data value,
3054 backslash-carriage-return and <literal>\r</> are still allowed
3060 <para><command>COPY</> changes (binary, <literal>\.</>) (Tom)</para>
3064 <para>Recover from <command>COPY</command> failure cleanly (Tom)</para>
3068 <para>Prevent possible memory leaks in <command>COPY</command> (Tom)</para>
3072 <para>Make <command>TRUNCATE</command> transaction-safe (Rod)</para>
3074 <command>TRUNCATE</command> can now be used inside a
3075 transaction. If the transaction aborts, the changes made by the
3076 <command>TRUNCATE</command> are automatically rolled back.
3082 Allow prepare/bind of utility commands like
3083 <command>FETCH</command> and <command>EXPLAIN</command> (Tom)
3088 <para>Add <command>EXPLAIN EXECUTE</command> (Neil)</para>
3092 <para>Improve <command>VACUUM</command> performance on indexes by reducing WAL traffic (Tom)</para>
3096 <para>Functional indexes have been generalized into indexes on expressions (Tom)</para>
3098 In prior releases, functional indexes only supported a simple
3099 function applied to one or more column names. This release
3100 allows any type of scalar expression.
3106 Have <command>SHOW TRANSACTION ISOLATION</command> match input
3107 to <command>SET TRANSACTION ISOLATION</command>
3114 Have <command>COMMENT ON DATABASE</command> on nonlocal
3115 database generate a warning, rather than an error (Rod)
3119 Database comments are stored in database-local tables so
3120 comments on a database have to be stored in each database.
3126 Improve reliability of <command>LISTEN</>/<command>NOTIFY</> (Tom)
3131 <para>Allow <command>REINDEX</command> to reliably reindex nonshared system catalog indexes (Tom)</para>
3133 This allows system tables to be reindexed without the
3134 requirement of a standalone session, which was necessary in
3135 previous releases. The only tables that now require a standalone
3136 session for reindexing are the global system tables
3137 <literal>pg_database</>, <literal>pg_shadow</>, and
3138 <literal>pg_group</>.
3145 <title>Data Type and Function Changes</title>
3150 New server parameter <varname>extra_float_digits</varname> to
3151 control precision display of floating-point numbers (Pedro
3155 This controls output precision which was causing regression
3160 <listitem><para>Allow <literal>+1300</literal> as a numeric time-zone specifier, for FJST (Tom)</para></listitem>
3164 Remove rarely used functions <function>oidrand</>,
3165 <function>oidsrand</>, and <function>userfntest</> functions
3171 <para>Add <function>md5()</> function to main server, already in <filename>contrib/pgcrypto</filename> (Joe)</para>
3173 An MD5 function was frequently requested. For more complex
3174 encryption capabilities, use
3175 <filename>contrib/pgcrypto</filename>.
3179 <listitem><para>Increase date range of <type>timestamp</type> (John Cochran)</para></listitem>
3183 Change <literal>EXTRACT(EPOCH FROM timestamp)</literal> so
3184 <type>timestamp without time zone</type> is assumed to be in
3185 local time, not GMT (Tom)
3189 <listitem><para>Trap division by zero in case the operating system doesn't prevent it (Tom)</para></listitem>
3190 <listitem><para>Change the <type>numeric</type> data type internally to base 10000 (Tom)</para></listitem>
3191 <listitem><para>New <function>hostmask()</function> function (Greg Wickham)</para></listitem>
3192 <listitem><para>Fixes for <function>to_char()</function> and <function>to_timestamp()</function> (Karel)</para></listitem>
3196 Allow functions that can take any argument data type and return
3197 any data type, using <type>anyelement</type> and
3198 <type>anyarray</type> (Joe)
3201 This allows the creation of functions that can work with any
3208 Arrays may now be specified as <literal>ARRAY[1,2,3]</literal>,
3209 <literal>ARRAY[['a','b'],['c','d']]</literal>, or
3210 <literal>ARRAY[ARRAY[ARRAY[2]]]</literal> (Joe)
3216 Allow proper comparisons for arrays, including <literal>ORDER
3217 BY</literal> and <literal>DISTINCT</literal> support
3222 <listitem><para>Allow indexes on array columns (Joe)</para></listitem>
3223 <listitem><para>Allow array concatenation with <literal>||</literal> (Joe)</para></listitem>
3227 Allow <literal>WHERE</literal> qualification
3228 <literal><replaceable>expr</> <replaceable>op</> ANY/SOME/ALL
3229 (<replaceable>array_expr</>)</literal> (Joe)
3232 This allows arrays to behave like a list of values, for purposes
3233 like <literal>SELECT * FROM tab WHERE col IN
3234 (array_val)</literal>.
3240 New array functions <function>array_append</>,
3241 <function>array_cat</>, <function>array_lower</>,
3242 <function>array_prepend</>, <function>array_to_string</>,
3243 <function>array_upper</>, <function>string_to_array</> (Joe)
3247 <listitem><para>Allow user defined aggregates to use polymorphic functions (Joe)</para></listitem>
3248 <listitem><para>Allow assignments to empty arrays (Joe)</para></listitem>
3252 Allow 60 in seconds fields of <type>time</type>,
3253 <type>timestamp</type>, and <type>interval</type> input values
3257 Sixty-second values are needed for leap seconds.
3261 <listitem><para>Allow <type>cidr</type> data type to be cast to <type>text</type> (Tom)</para></listitem>
3263 <listitem><para>Disallow invalid time zone names in SET TIMEZONE</para></listitem>
3267 Trim trailing spaces when <type>char</type> is cast to
3268 <type>varchar</> or <type>text</> (Tom)
3274 Make <type>float(<replaceable>p</>)</> measure the precision
3275 <replaceable>p</> in binary digits, not decimal digits
3281 <para>Add IPv6 support to the <type>inet</type> and <type>cidr</type> data types (Michael Graff)</para>
3285 <para>Add <function>family()</function> function to report whether address is IPv4 or IPv6 (Michael Graff)</para>
3290 Have <literal>SHOW datestyle</literal> generate output similar
3291 to that used by <literal>SET datestyle</literal> (Tom)
3297 Make <literal>EXTRACT(TIMEZONE)</literal> and <literal>SET/SHOW
3298 TIME ZONE</literal> follow the SQL convention for the sign of
3299 time zone offsets, i.e., positive is east from UTC (Tom)
3304 <para>Fix <literal>date_trunc('quarter', ...)</literal> (Böjthe Zoltán)</para>
3306 Prior releases returned an incorrect value for this function call.
3311 <para>Make <function>initcap()</function> more compatible with Oracle (Mike Nolan)</para>
3313 <function>initcap()</function> now uppercases a letter appearing
3314 after any non-alphanumeric character, rather than only after
3320 <para>Allow only <varname>datestyle</varname> field order for date values not in ISO-8601 format (Greg)</para>
3325 Add new <varname>datestyle</varname> values <literal>MDY</>,
3326 <literal>DMY</>, and <literal>YMD</> to set input field order;
3327 honor <literal>US</> and <literal>European</> for backward
3334 String literals like <literal>'now'</literal> or
3335 <literal>'today'</literal> will no longer work as a column
3336 default. Use functions such as <function>now()</function>,
3337 <function>current_timestamp</function> instead. (change
3338 required for prepared statements) (Tom)
3343 <para>Treat NaN as larger than any other value in <function>min()</>/<function>max()</> (Tom)</para>
3345 NaN was already sorted after ordinary numeric values for most
3346 purposes, but <function>min()</> and <function>max()</> didn't
3352 <para>Prevent interval from suppressing <literal>:00</literal>
3353 seconds display</para>
3358 New functions <function>pg_get_triggerdef(prettyprint)</function>
3359 and <function>pg_conversion_is_visible()</function> (Christopher)
3364 <para>Allow time to be specified as <literal>040506</> or <literal>0405</> (Tom)</para>
3369 Input date order must now be <literal>YYYY-MM-DD</literal> (with 4-digit year) or
3370 match <varname>datestyle</varname>
3376 Make <function>pg_get_constraintdef</function> support
3377 unique, primary-key, and check constraints (Christopher)
3384 <title>Server-Side Language Changes</title>
3389 Prevent PL/pgSQL crash when <literal>RETURN NEXT</literal> is
3390 used on a zero-row record variable (Tom)
3396 Make PL/Python's <function>spi_execute</function> interface
3397 handle null values properly (Andrew Bosma)
3402 <para>Allow PL/pgSQL to declare variables of composite types without <literal>%ROWTYPE</literal> (Tom)</para>
3406 <para>Fix PL/Python's <function>_quote()</function> function to handle big integers</para>
3410 <para>Make PL/Python an untrusted language, now called <literal>plpythonu</literal> (Kevin Jacobs, Tom)</para>
3412 The Python language no longer supports a restricted execution
3413 environment, so the trusted version of PL/Python was removed. If
3414 this situation changes, a version of PL/Python that can be used
3415 by non-superusers will be readded.
3420 <para>Allow polymorphic PL/pgSQL functions (Joe, Tom)</para>
3424 <para>Allow polymorphic SQL functions (Joe)</para>
3429 Improved compiled function caching mechanism in PL/pgSQL with
3430 full support for polymorphism (Joe)
3436 Add new parameter <literal>$0</> in PL/pgSQL representing the
3437 function's actual return type (Joe)
3443 Allow PL/Tcl and PL/Python to use the same trigger on multiple tables (Tom)
3449 Fixed PL/Tcl's <function>spi_prepare</function> to accept fully
3450 qualified type names in the parameter type list
3458 <title>psql Changes</title>
3462 <para>Add <literal>\pset pager always</literal> to always use pager (Greg)</para>
3464 This forces the pager to be used even if the number of rows is
3465 less than the screen height. This is valuable for rows that
3466 wrap across several screen rows.
3470 <listitem><para>Improve tab completion (Rod, Ross Reedstrom, Ian Barwick)</para></listitem>
3471 <listitem><para>Reorder <literal>\?</> help into groupings (Harald Armin Massa, Bruce)</para></listitem>
3472 <listitem><para>Add backslash commands for listing schemas, casts, and conversions (Christopher)</para></listitem>
3476 <command>\encoding</> now changes based on the server parameter
3477 <varname>client_encoding</varname> (Tom)
3480 In previous versions, <command>\encoding</command> was not aware
3481 of encoding changes made using <literal>SET
3482 client_encoding</literal>.
3487 <para>Save editor buffer into readline history (Ross)</para>
3489 When <command>\e</> is used to edit a query, the result is saved
3490 in the readline history for retrieval using the up arrow.
3494 <listitem><para>Improve <command>\d</command> display (Christopher)</para></listitem>
3495 <listitem><para>Enhance HTML mode to be more standards-conforming (Greg)</para></listitem>
3498 <para>New <command>\set AUTOCOMMIT off</command> capability (Tom)</para>
3500 This takes the place of the removed server parameter <varname>autocommit</varname>.
3505 <para>New <command>\set VERBOSITY</command> to control error detail (Tom)</para>
3507 This controls the new error reporting details.
3511 <listitem><para>New prompt escape sequence <literal>%x</literal> to show transaction status (Tom)</para></listitem>
3512 <listitem><para>Long options for <application>psql</application> are now available on all platforms</para></listitem>
3517 <title>pg_dump Changes</title>
3520 <listitem><para>Multiple pg_dump fixes, including tar format and large objects</para></listitem>
3521 <listitem><para>Allow pg_dump to dump specific schemas (Neil)</para></listitem>
3524 <para>Make pg_dump preserve column storage characteristics (Christopher)</para>
3526 This preserves <literal>ALTER TABLE ... SET STORAGE</literal> information.
3530 <listitem><para>Make pg_dump preserve <command>CLUSTER</command> characteristics (Christopher)</para></listitem>
3534 Have pg_dumpall use <command>GRANT</>/<command>REVOKE</> to dump database-level privleges (Tom)
3540 Allow pg_dumpall to support the options <option>-a</>,
3541 <option>-s</>, <option>-x</> of pg_dump (Tom)
3545 <listitem><para>Prevent pg_dump from lowercasing identifiers specified on the command line (Tom)</para></listitem>
3549 pg_dump options <option>--use-set-session-authorization</option>
3550 and <option>--no-reconnect</option> now do nothing, all dumps
3551 use <command>SET SESSION AUTHORIZATION</command>
3554 pg_dump no longer reconnects to switch users, but instead always
3555 uses <command>SET SESSION AUTHORIZATION</command>. This will
3556 reduce password prompting during restores.
3561 <para>Long options for <application>pg_dump</application> are now available on all platforms</para>
3563 <productname>PostgreSQL</productname> now includes its own
3564 long-option processing routines.
3571 <title>libpq Changes</title>
3576 Add function <function>PQfreemem</function> for freeing memory on
3577 Windows, suggested for <command>NOTIFY</command> (Bruce)
3580 Windows requires that memory allocated in a library be freed by
3581 a function in the same library, hence
3582 <function>free()</function> doesn't work for freeing memory
3583 allocated by libpq. <function>PQfreemem</function> is the proper
3584 way to free libpq memory, especially on Windows, and is
3585 recommended for other platforms as well.
3590 <para>Document service capability, and add sample file (Bruce)</para>
3592 This allows clients to look up connection information in a
3593 central file on the client machine.
3599 Make <function>PQsetdbLogin</function> have the same defaults as
3600 <function>PQconnectdb</function> (Tom)
3604 <listitem><para>Allow libpq to cleanly fail when result sets are too large (Tom)</para></listitem>
3608 Improve performance of function <function>PQunescapeBytea</function> (Ben Lamb)
3614 Allow thread-safe libpq with <filename>configure</filename>
3615 option <option>--enable-thread-safety</option> (Lee Kindness,
3622 Allow function <function>pqInternalNotice</function> to accept a
3623 format string and arguments instead of just a preformatted
3624 message (Tom, Sean Chittenden)
3630 Control SSL negotiation with <literal>sslmode</literal> values
3631 <literal>disable</literal>, <literal>allow</literal>,
3632 <literal>prefer</literal>, and <literal>require</literal> (Jon
3638 <para>Allow new error codes and levels of text (Tom)</para>
3642 <para>Allow access to the underlying table and column of a query result (Tom)</para>
3644 This is helpful for query-builder applications that want to know
3645 the underlying table and column names associated with a specific
3650 <listitem><para>Allow access to the current transaction status (Tom)</para></listitem>
3651 <listitem><para>Add ability to pass binary data directly to the server (Tom)</para></listitem>
3655 Add function <function>PQexecPrepared</function> and
3656 <function>PQsendQueryPrepared</function> functions which perform
3657 bind/execute of previously prepared statements (Tom)
3664 <title>JDBC Changes</title>
3667 <listitem><para>Allow <function>setNull</function> on updateable result sets</para></listitem>
3668 <listitem><para>Allow <function>executeBatch</function> on a prepared statement (Barry)</para></listitem>
3669 <listitem><para>Support SSL connections (Barry)</para></listitem>
3670 <listitem><para>Handle schema names in result sets (Paul Sorenson)</para></listitem>
3671 <listitem><para>Add refcursor support (Nic Ferrier)</para></listitem>
3676 <title>Miscellaneous Interface Changes</title>
3680 <para>Prevent possible memory leak or core dump during libpgtcl shutdown (Tom)</para>
3683 <para>Add Informix compatibility to ECPG (Michael)</para>
3685 This allows ECPG to process embedded C programs that were
3686 written using certain Informix extensions.
3691 <para>Add type <type>decimal</type> to ECPG that is fixed length, for Informix (Michael)</para>
3696 Allow thread-safe embedded SQL programs with
3697 <filename>configure</filename> option
3698 <option>--enable-thread-safety</option> (Lee Kindness, Bruce)
3701 This allows multiple threads to access the database at the same
3707 <para>Moved Python client PyGreSQL to <ulink url="http://www.pygresql.org"></ulink> (Marc)</para>
3713 <title>Source Code Changes</title>
3716 <listitem><para>Prevent need for separate platform geometry regression result files (Tom)</para></listitem>
3717 <listitem><para>Improved PPC locking primitive (Reinhard Max)</para></listitem>
3718 <listitem><para>New function <function>palloc0</function> to allocate and clear memory (Bruce)</para></listitem>
3719 <listitem><para>Fix locking code for s390x CPU (64-bit) (Tom)</para></listitem>
3720 <listitem><para>Allow OpenBSD to use local ident credentials (William Ahern)</para></listitem>
3721 <listitem><para>Make query plan trees read-only to executor (Tom)</para></listitem>
3722 <listitem><para>Add Darwin startup scripts (David Wheeler)</para></listitem>
3723 <listitem><para>Allow libpq to compile with Borland C++ compiler (Lester Godwin, Karl Waclawek)</para></listitem>
3724 <listitem><para>Use our own version of <function>getopt_long()</function> if needed (Peter)</para></listitem>
3725 <listitem><para>Convert administration scripts to C (Peter)</para></listitem>
3726 <listitem><para> Bison >= 1.85 is now required to build the <productname>PostgreSQL</> grammar, if building from CVS</para></listitem>
3727 <listitem><para>Merge documentation into one book (Peter)</para></listitem>
3728 <listitem><para>Add Windows compatibility functions (Bruce)</para></listitem>
3729 <listitem><para>Allow client interfaces to compile under MinGW (Bruce)</para></listitem>
3730 <listitem><para>New <function>ereport()</function> function for error reporting (Tom)</para></listitem>
3731 <listitem><para>Support Intel compiler on Linux (Peter)</para></listitem>
3732 <listitem><para>Improve Linux startup scripts (Slawomir Sudnik, Darko Prenosil)</para></listitem>
3733 <listitem><para>Add support for AMD Opteron and Itanium (Jeffrey W. Baker, Bruce)</para></listitem>
3735 <para>Remove <option>--enable-recode</option> option from <command>configure</command></para>
3737 This was no longer needed now that we have <command>CREATE CONVERSION</command>.
3741 <para>Generate a compile error if spinlock code is not found (Bruce)</para>
3743 Platforms without spinlock code will now fail to compile, rather
3744 than silently using semaphores. This failure can be disabled
3745 with a new <command>configure</command> option.
3752 <title>Contrib Changes</title>
3755 <listitem><para>Change dbmirror license to BSD</para></listitem>
3756 <listitem><para>Improve earthdistance (Bruno Wolff III)</para></listitem>
3757 <listitem><para>Portability improvements to pgcrypto (Marko Kreen)</para></listitem>
3758 <listitem><para>Prevent crash in xml (John Gray, Michael Richards)</para></listitem>
3759 <listitem><para>Update oracle</para></listitem>
3760 <listitem><para>Update mysql</para></listitem>
3761 <listitem><para>Update cube (Bruno Wolff III)</para></listitem>
3762 <listitem><para>Update earthdistance to use cube (Bruno Wolff III)</para></listitem>
3763 <listitem><para>Update btree_gist (Oleg)</para></listitem>
3764 <listitem><para>New tsearch2 full-text search module (Oleg, Teodor)</para></listitem>
3765 <listitem><para>Add hash-based crosstab function to tablefuncs (Joe)</para></listitem>
3766 <listitem><para>Add serial column to order <function>connectby()</> siblings in tablefuncs (Nabil Sayegh,Joe)</para></listitem>
3767 <listitem><para>Add named persistent connections to dblink (Shridhar Daithanka)</para></listitem>
3768 <listitem><para>New pg_autovacuum allows automatic <command>VACUUM</command> (Matthew T. O'Connor)</para></listitem>
3769 <listitem><para>Make pgbench honor environment variables <envar>PGHOST</>, <envar>PGPORT</>, <envar>PGUSER</> (Tatsuo)</para></listitem>
3770 <listitem><para>Improve intarray (Teodor Sigaev)</para></listitem>
3771 <listitem><para>Improve pgstattuple (Rod)</para></listitem>
3772 <listitem><para>Fix bug in <function>metaphone()</function> in fuzzystrmatch</para></listitem>
3773 <listitem><para>Improve adddepend (Rod)</para></listitem>
3774 <listitem><para>Update spi/timetravel (Böjthe Zoltán)</para></listitem>
3775 <listitem><para>Fix dbase <option>-s</> option and improve non-ASCII handling (Thomas Behr, Márcio Smiderle)</para></listitem>
3776 <listitem><para>Remove array module because features now included by default (Joe)</para></listitem>
3782 <sect1 id="release-7-3-6">
3783 <title>Release 7.3.6</title>
3786 <title>Release date</title>
3787 <simpara>2004-03-02</simpara>
3791 This release contains a variety of fixes from 7.3.5.
3796 <title>Migration to version 7.3.6</title>
3799 A dump/restore is <emphasis>not</emphasis> required for those
3806 <title>Changes</title>
3809 <listitem><para>Revert erroneous changes in rule permissions checking</para>
3810 <para>A patch applied in 7.3.3 to fix a corner case in rule permissions checks
3811 turns out to have disabled rule-related permissions checks in many
3812 not-so-corner cases. This would for example allow users to insert into views
3813 they weren't supposed to have permission to insert into. We have therefore
3814 reverted the 7.3.3 patch. The original bug will be fixed in 8.0.
3816 <listitem><para>Repair incorrect order of operations in
3817 GetNewTransactionId()</para>
3819 This bug could result in failure under out-of-disk-space conditions, including
3820 inability to restart even after disk space is freed.
3822 <listitem><para>Ensure configure selects -fno-strict-aliasing even when
3823 an external value for CFLAGS is supplied</para>
3825 On some platforms, building with -fstrict-aliasing causes bugs.
3827 <listitem><para>Make pg_restore handle 64-bit off_t correctly</para>
3829 This bug prevented proper restoration from archive files exceeding 4Gb.
3831 <listitem><para>Make contrib/dblink not assume that local and remote type OIDs
3832 match (Joe)</para></listitem>
3833 <listitem><para>Quote connectby()'s start_with argument properly (Joe)</para></listitem>
3834 <listitem><para>Don't crash when a rowtype argument to a plpgsql function is
3835 NULL</para></listitem>
3836 <listitem><para>Avoid generating invalid character encoding sequences in
3837 corner cases when planning LIKE operations</para></listitem>
3838 <listitem><para>Ensure text_position() cannot scan past end of source string
3839 in multibyte cases (Korea PostgreSQL Users' Group)</para></listitem>
3840 <listitem><para>Fix index optimization and selectivity estimates for LIKE
3841 operations on bytea columns (Joe)</para></listitem>
3847 <sect1 id="release-7-3-5">
3848 <title>Release 7.3.5</title>
3851 <title>Release date</title>
3852 <simpara>2003-12-03</simpara>
3856 This has a variety of fixes from 7.3.4.
3861 <title>Migration to version 7.3.5</title>
3864 A dump/restore is <emphasis>not</emphasis> required for those
3870 <title>Changes</title>
3873 <listitem><para>Force zero_damaged_pages to be on during recovery from WAL</para></listitem>
3874 <listitem><para>Prevent some obscure cases of <quote>variable not in subplan target lists</quote></para></listitem>
3875 <listitem><para>Force stats processes to detach from shared memory, ensuring cleaner shutdown</para></listitem>
3876 <listitem><para>Make PQescapeBytea and byteaout consistent with each other (Joe)</para></listitem>
3877 <listitem><para>Added missing SPI_finish() calls to dblink's get_tuple_of_interest() (Joe)</para></listitem>
3878 <listitem><para>Fix for possible foreign key violation when rule rewrites INSERT (Jan)</para></listitem>
3879 <listitem><para>Support qualified type names in PL/Tcl's spi_prepare command (Jan)</para></listitem>
3880 <listitem><para>Make pg_dump handle a procedural language handler located in pg_catalog</para></listitem>
3881 <listitem><para>Make pg_dump handle cases where a custom opclass is in another schema</para></listitem>
3882 <listitem><para>Make pg_dump dump binary-compatible casts correctly (Jan)</para></listitem>
3883 <listitem><para>Fix insertion of expressions containing subqueries into rule bodies</para></listitem>
3884 <listitem><para>Fix incorrect argument processing in clusterdb script (Anand Ranganathan)</para></listitem>
3885 <listitem><para>Fix problems with dropped columns in plpython triggers</para></listitem>
3886 <listitem><para>Repair problems with to_char() reading past end of its input string (Karel)</para></listitem>
3887 <listitem><para>Fix GB18030 mapping errors (Tatsuo)</para></listitem>
3888 <listitem><para>Fix several problems with SSL error handling and asynchronous SSL I/O</para></listitem>
3889 <listitem><para>Remove ability to bind a list of values to a single parameter in JDBC
3890 (prevents possible SQL-injection attacks)</para></listitem>
3891 <listitem><para>Fix some errors in HAVE_INT64_TIMESTAMP code paths</para></listitem>
3892 <listitem><para>Fix corner case for btree search in parallel with first root page split</para></listitem>
3898 <sect1 id="release-7-3-4">
3899 <title>Release 7.3.4</title>
3902 <title>Release date</title>
3903 <simpara>2003-07-24</simpara>
3907 This has a variety of fixes from 7.3.3.
3912 <title>Migration to version 7.3.4</title>
3915 A dump/restore is <emphasis>not</emphasis> required for those
3921 <title>Changes</title>
3924 <listitem><para>Repair breakage in timestamp-to-date conversion for dates before 2000</para></listitem>
3925 <listitem><para>Prevent rare possibility of server startup failure (Tom)</para></listitem>
3926 <listitem><para>Fix bugs in interval-to-time conversion (Tom)</para></listitem>
3927 <listitem><para>Add constraint names in a few places in pg_dump (Rod)</para></listitem>
3928 <listitem><para>Improve performance of functions with many parameters (Tom)</para></listitem>
3929 <listitem><para>Fix to_ascii() buffer overruns (Tom)</para></listitem>
3930 <listitem><para>Prevent restore of database comments from throwing an error (Tom)</para></listitem>
3931 <listitem><para>Work around buggy strxfrm() present in some Solaris releases (Tom)</para></listitem>
3932 <listitem><para>Properly escape jdbc setObject() strings to improve security (Barry)</para></listitem>
3938 <sect1 id="release-7-3-3">
3939 <title>Release 7.3.3</title>
3942 <title>Release date</title>
3943 <simpara>2003-05-22</simpara>
3947 This release contains a variety of fixes for version 7.3.2.
3951 <title>Migration to version 7.3.3</title>
3954 A dump/restore is <emphasis>not</emphasis> required for those
3955 running version 7.3.*.
3960 <title>Changes</title>
3963 <listitem><para>Repair sometimes-incorrect computation of StartUpID after a crash</para></listitem>
3964 <listitem><para>Avoid slowness with lots of deferred triggers in one transaction (Stephan)</para></listitem>
3965 <listitem><para>Don't lock referenced row when <command>UPDATE</command> doesn't change foreign key's value (Jan)</para></listitem>
3966 <listitem><para>Use <command>-fPIC</command> not <command>-fpic</command> on Sparc (Tom Callaway)</para></listitem>
3967 <listitem><para>Repair lack of schema-awareness in contrib/reindexdb</para></listitem>
3968 <listitem><para>Fix contrib/intarray error for zero-element result array (Teodor)</para></listitem>
3969 <listitem><para>Ensure createuser script will exit on control-C (Oliver)</para></listitem>
3970 <listitem><para>Fix errors when the type of a dropped column has itself been dropped</para></listitem>
3971 <listitem><para><command>CHECKPOINT</command> does not cause database panic on failure in noncritical steps</para></listitem>
3972 <listitem><para>Accept 60 in seconds fields of timestamp, time, interval input values</para></listitem>
3973 <listitem><para>Issue notice, not error, if <type>TIMESTAMP</type>,
3974 <type> TIME</type>, or <type>INTERVAL</type> precision too large</para></listitem>
3975 <listitem><para>Fix <function>abstime-to-time</function> cast function (fix is
3976 not applied unless you <application>initdb</application>)</para></listitem>
3977 <listitem><para>Fix <application>pg_proc</application> entry for
3978 <type>timestampt_izone</type> (fix is not applied unless you
3979 <application>initdb</application>)</para></listitem>
3980 <listitem><para>Make <function>EXTRACT(EPOCH FROM timestamp without time zone)</function> treat input as local time</para></listitem>
3981 <listitem><para><command>'now'::timestamptz</command> gave wrong answer if timezone changed earlier in transaction</para></listitem>
3982 <listitem><para><envar>HAVE_INT64_TIMESTAMP</envar> code for time with timezone overwrote its input</para></listitem>
3983 <listitem><para>Accept <command>GLOBAL TEMP/TEMPORARY</command> as a
3984 synonym for <command>TEMPORARY</command></para></listitem>
3985 <listitem><para>Avoid improper schema-privilege-check failure in foreign-key triggers</para></listitem>
3986 <listitem><para>Fix bugs in foreign-key triggers for <command>SET DEFAULT</command> action</para></listitem>
3987 <listitem><para>Fix incorrect time-qual check in row fetch for
3988 <command>UPDATE</command> and <command>DELETE</command> triggers</para></listitem>
3989 <listitem><para>Foreign-key clauses were parsed but ignored in
3990 <command>ALTER TABLE ADD COLUMN</command></para></listitem>
3991 <listitem><para>Fix createlang script breakage for case where handler function already exists</para></listitem>
3992 <listitem><para>Fix misbehavior on zero-column tables in <application>pg_dump</application>, COPY, ANALYZE, other places</para></listitem>
3993 <listitem><para>Fix misbehavior of <function>func_error()</function> on type names containing '%'</para></listitem>
3994 <listitem><para>Fix misbehavior of <function>replace()</function> on strings containing '%'</para></listitem>
3995 <listitem><para>Regular-expression patterns containing certain multibyte characters failed</para></listitem>
3996 <listitem><para>Account correctly for <command>NULL</command>s in more cases in join size estimation</para></listitem>
3997 <listitem><para>Avoid conflict with system definition of <function>isblank()</function> function or macro</para></listitem>
3998 <listitem><para>Fix failure to convert large code point values in EUC_TW conversions (Tatsuo)</para></listitem>
3999 <listitem><para>Fix error recovery for <function>SSL_read</function>/<function>SSL_write</function> calls</para></listitem>
4000 <listitem><para>Don't do early constant-folding of type coercion expressions</para></listitem>
4001 <listitem><para>Validate page header fields immediately after reading in any page</para></listitem>
4002 <listitem><para>Repair incorrect check for ungrouped variables in unnamed joins</para></listitem>
4003 <listitem><para>Fix buffer overrun in <function>to_ascii</function> (Guido Notari)</para></listitem>
4004 <listitem><para>contrib/ltree fixes (Teodor)</para></listitem>
4005 <listitem><para>Fix core dump in deadlock detection on machines where char is unsigned</para></listitem>
4006 <listitem><para>Avoid running out of buffers in many-way indexscan (bug introduced in 7.3)</para></listitem>
4007 <listitem><para>Fix planner's selectivity estimation functions to handle domains properly</para></listitem>
4008 <listitem><para>Fix <application>dbmirror</application> memory-allocation bug (Steven Singer)</para></listitem>
4009 <listitem><para>Prevent infinite loop in <function>ln(numeric)</function> due to roundoff error</para></listitem>
4010 <listitem><para><command>GROUP BY</command> got confused if there were multiple equal GROUP BY items</para></listitem>
4011 <listitem><para>Fix bad plan when inherited <command>UPDATE</command>/<command>DELETE</command> references another inherited table</para></listitem>
4012 <listitem><para>Prevent clustering on incomplete (partial or non-NULL-storing) indexes</para></listitem>
4013 <listitem><para>Service shutdown request at proper time if it arrives while still starting up</para></listitem>
4014 <listitem><para>Fix left-links in temporary indexes (could make backwards scans miss entries)</para></listitem>
4015 <listitem><para>Fix incorrect handling of client_encoding setting in postgresql.conf (Tatsuo)</para></listitem>
4016 <listitem><para>Fix failure to respond to <command>pg_ctl stop -m fast</command> after Async_NotifyHandler runs</para></listitem>
4017 <listitem><para>Fix SPI for case where rule contains multiple statements of the same type</para></listitem>
4018 <listitem><para>Fix problem with checking for wrong type of access privilege in rule query</para></listitem>
4019 <listitem><para>Fix problem with <command>EXCEPT</command> in <command>CREATE RULE</command></para></listitem>
4020 <listitem><para>Prevent problem with dropping temp tables having serial columns</para></listitem>
4021 <listitem><para>Fix replace_vars_with_subplan_refs failure in complex views</para></listitem>
4022 <listitem><para>Fix regexp slowness in single-byte encodings (Tatsuo)</para></listitem>
4023 <listitem><para>Allow qualified type names in <command>CREATE CAST</command>
4024 and <command> DROP CAST</command></para></listitem>
4025 <listitem><para>Accept <function>SETOF type[]</function>, which formerly had to
4026 be written <function>SETOF _type</function></para></listitem>
4027 <listitem><para>Fix <application>pg_dump</application> core dump in some cases with procedural languages</para></listitem>
4028 <listitem><para>Force ISO datestyle in <application>pg_dump</application> output, for portability (Oliver)</para></listitem>
4029 <listitem><para><application>pg_dump</application> failed to handle error return
4030 from <function>lo_read</function> (Oleg Drokin)</para></listitem>
4031 <listitem><para><application>pg_dumpall</application> failed with groups having no members (Nick Eskelinen)</para></listitem>
4032 <listitem><para><application>pg_dumpall</application> failed to recognize --globals-only switch</para></listitem>
4033 <listitem><para>pg_restore failed to restore blobs if -X disable-triggers is specified</para></listitem>
4034 <listitem><para>Repair intrafunction memory leak in plpgsql</para></listitem>
4035 <listitem><para>pltcl's <command>elog</command> command dumped core if given wrong parameters (Ian Harding)</para></listitem>
4036 <listitem><para>plpython used wrong value of <envar>atttypmod</envar> (Brad McLean)</para></listitem>
4037 <listitem><para>Fix improper quoting of boolean values in Python interface (D'Arcy)</para></listitem>
4038 <listitem><para>Added <function>addDataType()</function> method to PGConnection interface for JDBC</para></listitem>
4039 <listitem><para>Fixed various problems with updateable ResultSets for JDBC (Shawn Green)</para></listitem>
4040 <listitem><para>Fixed various problems with DatabaseMetaData for JDBC (Kris Jurka, Peter Royal)</para></listitem>
4041 <listitem><para>Fixed problem with parsing table ACLs in JDBC</para></listitem>
4042 <listitem><para>Better error message for character set conversion problems in JDBC</para></listitem>
4048 <sect1 id="release-7-3-2">
4049 <title>Release 7.3.2</title>
4052 <title>Release date</title>
4053 <simpara>2003-02-04</simpara>
4057 This release contains a variety of fixes for version 7.3.1.
4062 <title>Migration to version 7.3.2</title>
4065 A dump/restore is <emphasis>not</emphasis> required for those
4066 running version 7.3.*.
4071 <title>Changes</title>
4074 <listitem><para>Restore creation of OID column in CREATE TABLE AS / SELECT INTO</para></listitem>
4075 <listitem><para>Fix <application>pg_dump</> core dump when dumping views having comments</para></listitem>
4076 <listitem><para>Dump DEFERRABLE/INITIALLY DEFERRED constraints properly</para></listitem>
4077 <listitem><para>Fix UPDATE when child table's column numbering differs from parent</para></listitem>
4078 <listitem><para>Increase default value of max_fsm_relations</para></listitem>
4079 <listitem><para>Fix problem when fetching backwards in a cursor for a single-row query</para></listitem>
4080 <listitem><para>Make backward fetch work properly with cursor on SELECT DISTINCT query</para></listitem>
4081 <listitem><para>Fix problems with loading <application>pg_dump</> files containing contrib/lo usage</para></listitem>
4082 <listitem><para>Fix problem with all-numeric user names</para></listitem>
4083 <listitem><para>Fix possible memory leak and core dump during disconnect in libpgtcl</para></listitem>
4084 <listitem><para>Make plpython's spi_execute command handle nulls properly (Andrew Bosma)</para></listitem>
4085 <listitem><para>Adjust plpython error reporting so that its regression test passes again</para></listitem>
4086 <listitem><para>Work with bison 1.875</para></listitem>
4087 <listitem><para>Handle mixed-case names properly in plpgsql's %type (Neil)</para></listitem>
4088 <listitem><para>Fix core dump in pltcl when executing a query rewritten by a rule</para></listitem>
4089 <listitem><para>Repair array subscript overruns (per report from Yichen Xie)</para></listitem>
4090 <listitem><para>Reduce MAX_TIME_PRECISION from 13 to 10 in floating-point case</para></listitem>
4091 <listitem><para>Correctly case-fold variable names in per-database and per-user settings</para></listitem>
4092 <listitem><para>Fix coredump in plpgsql's RETURN NEXT when SELECT into record returns no rows</para></listitem>
4093 <listitem><para>Fix outdated use of pg_type.typprtlen in python client interface</para></listitem>
4094 <listitem><para>Correctly handle fractional seconds in timestamps in JDBC driver</para></listitem>
4095 <listitem><para>Improve performance of getImportedKeys() in JDBC</para></listitem>
4096 <listitem><para>Make shared-library symlinks work standardly on HPUX (Giles)</para></listitem>
4097 <listitem><para>Repair inconsistent rounding behavior for timestamp, time, interval</para></listitem>
4098 <listitem><para>SSL negotiation fixes (Nathan Mueller)</para></listitem>
4099 <listitem><para>Make libpq's ~/.pgpass feature work when connecting with PQconnectDB</para></listitem>
4100 <listitem><para>Update my2pg, ora2pg</para></listitem>
4101 <listitem><para>Translation updates</para></listitem>
4102 <listitem><para>Add casts between types lo and oid in contrib/lo</para></listitem>
4103 <listitem><para>fastpath code now checks for privilege to call function</para></listitem>
4109 <sect1 id="release-7-3-1">
4110 <title>Release 7.3.1</title>
4113 <title>Release date</title>
4114 <simpara>2002-12-18</simpara>
4118 This release contains a variety of fixes for version 7.3.
4123 <title>Migration to version 7.3.1</title>
4126 A dump/restore is <emphasis>not</emphasis> required for those
4127 running version 7.3. However, it should be noted that the main
4128 <productname>PostgreSQL</productname> interface library, libpq,
4129 has a new major version number for this release, which may require
4130 recompilation of client code in certain cases.
4135 <title>Changes</title>
4138 <listitem><para>Fix a core dump of COPY TO when client/server encodings don't match (Tom)</para></listitem>
4139 <listitem><para>Allow <application>pg_dump</> to work with pre-7.2 servers (Philip)</para></listitem>
4140 <listitem><para>contrib/adddepend fixes (Tom)</para></listitem>
4141 <listitem><para>Fix problem with deletion of per-user/per-database config settings (Tom)</para></listitem>
4142 <listitem><para>contrib/vacuumlo fix (Tom)</para></listitem>
4143 <listitem><para>Allow 'password' encryption even when pg_shadow contains MD5 passwords (Bruce)</para></listitem>
4144 <listitem><para>contrib/dbmirror fix (Steven Singer)</para></listitem>
4145 <listitem><para>Optimizer fixes (Tom)</para></listitem>
4146 <listitem><para>contrib/tsearch fixes (Teodor Sigaev, Magnus)</para></listitem>
4147 <listitem><para>Allow locale names to be mixed case (Nicolai Tufar)</para></listitem>
4148 <listitem><para>Increment libpq library's major version number (Bruce)</para></listitem>
4149 <listitem><para>pg_hba.conf error reporting fixes (Bruce, Neil)</para></listitem>
4150 <listitem><para>Add SCO Openserver 5.0.4 as a supported platform (Bruce)</para></listitem>
4151 <listitem><para>Prevent EXPLAIN from crashing server (Tom)</para></listitem>
4152 <listitem><para>SSL fixes (Nathan Mueller)</para></listitem>
4153 <listitem><para>Prevent composite column creation via ALTER TABLE (Tom)</para></listitem>
4159 <sect1 id="release-7-3">
4160 <title>Release 7.3</title>
4163 <title>Release date</title>
4164 <simpara>2002-11-27</simpara>
4168 <title>Overview</title>
4171 Major changes in this release:
4175 <term>Schemas</term>
4178 Schemas allow users to create objects in separate namespaces,
4179 so two people or applications can have tables with the same
4180 name. There is also a public schema for shared tables.
4181 Table/index creation can be restricted by removing privileges
4182 on the public schema.
4188 <term>Drop Column</term>
4191 <productname>PostgreSQL</productname> now supports the
4192 <literal>ALTER TABLE ... DROP COLUMN</literal> functionality.
4198 <term>Table Functions</term>
4201 Functions returning multiple rows and/or multiple columns are
4202 now much easier to use than before. You can call such a
4203 <quote>table function</quote> in the <literal>SELECT</literal>
4204 <literal>FROM</literal> clause, treating its output like a
4205 table. Also, <application>PL/pgSQL</application> functions can
4212 <term>Prepared Queries</term>
4215 <productname>PostgreSQL</productname> now supports prepared
4216 queries, for improved performance.
4222 <term>Dependency Tracking</term>
4225 <productname>PostgreSQL</productname> now records object
4226 dependencies, which allows improvements in many areas.
4227 <command>DROP</command> statements now take either
4228 <literal>CASCADE</> or <literal>RESTRICT</> to control whether
4229 dependent objects are also dropped.
4235 <term>Privileges</term>
4238 Functions and procedural languages now have privileges, and
4239 functions can be defined to run with the privileges of their
4246 <term>Internationalization</term>
4249 Both multibyte and locale support are now always enabled.
4255 <term>Logging</term>
4258 A variety of logging options have been enhanced.
4264 <term>Interfaces</term>
4267 A large number of interfaces have been moved to <ulink
4268 url="http://gborg.postgresql.org">http://gborg.postgresql.org</>
4269 where they can be developed and released independently.
4275 <term>Functions/Identifiers</term>
4278 By default, functions can now take up to 32 parameters, and
4279 identifiers can be up to 63 bytes long. Also, <literal>OPAQUE</>
4280 is now deprecated: there are specific <quote>pseudo-datatypes</>
4281 to represent each of the former meanings of <literal>OPAQUE</>
4282 in function argument and result types.
4292 <title>Migration to version 7.3</title>
4295 A dump/restore using <application>pg_dump</> is required for those
4296 wishing to migrate data from any previous release. If your
4297 application examines the system catalogs, additional changes will
4298 be required due to the introduction of schemas in 7.3; for more
4299 information, see: <ulink
4300 url="http://developer.postgresql.org/~momjian/upgrade_tips_7.3">
4301 http://developer.postgresql.org/~momjian/upgrade_tips_7.3</>.
4305 Observe the following incompatibilities:
4310 Pre-6.3 clients are no longer supported.
4316 <filename>pg_hba.conf</filename> now has a column for the user
4317 name and additional features. Existing files need to be
4324 Several <filename>postgresql.conf</filename> logging parameters
4331 <literal>LIMIT #,#</literal> has been disabled; use
4332 <literal>LIMIT # OFFSET #</literal>.
4338 <command>INSERT</command> statements with column lists must
4339 specify a value for each specified column. For example,
4340 <literal>INSERT INTO tab (col1, col2) VALUES ('val1')</literal>
4341 is now invalid. It's still allowed to supply fewer columns than
4342 expected if the <command>INSERT</command> does not have a column list.
4348 <type>serial</type> columns are no longer automatically
4349 <literal>UNIQUE</>; thus, an index will not automatically be
4356 A <command>SET</command> command inside an aborted transaction
4363 <command>COPY</command> no longer considers missing trailing
4364 columns to be null. All columns need to be specified.
4365 (However, one may achieve a similar effect by specifying a
4366 column list in the <command>COPY</command> command.)
4372 The data type <type>timestamp</type> is now equivalent to
4373 <type>timestamp without time zone</type>, instead of
4374 <type>timestamp with time zone</type>.
4380 Pre-7.3 databases loaded into 7.3 will not have the new object
4381 dependencies for <type>serial</type> columns, unique
4382 constraints, and foreign keys. See the directory
4383 <filename>contrib/adddepend/</filename> for a detailed
4384 description and a script that will add such dependencies.
4390 An empty string (<literal>''</literal>) is no longer allowed as
4391 the input into an integer field. Formerly, it was silently
4402 <title>Changes</title>
4405 <title>Server Operation</title>
4407 <listitem><para>Add pg_locks view to show locks (Neil)</para></listitem>
4408 <listitem><para>Security fixes for password negotiation memory allocation (Neil)</para></listitem>
4409 <listitem><para>Remove support for version 0 FE/BE protocol (<productname>PostgreSQL</productname> 6.2 and earlier) (Tom)</para></listitem>
4410 <listitem><para>Reserve the last few backend slots for superusers, add parameter superuser_reserved_connections to control this (Nigel J. Andrews)</para></listitem>
4415 <title>Performance</title>
4417 <listitem><para>Improve startup by calling localtime() only once (Tom)</para></listitem>
4418 <listitem><para>Cache system catalog information in flat files for faster startup (Tom)</para></listitem>
4419 <listitem><para>Improve caching of index information (Tom)</para></listitem>
4420 <listitem><para>Optimizer improvements (Tom, Fernando Nasser)</para></listitem>
4421 <listitem><para>Catalog caches now store failed lookups (Tom)</para></listitem>
4422 <listitem><para>Hash function improvements (Neil)</para></listitem>
4423 <listitem><para>Improve performance of query tokenization and network handling (Peter)</para></listitem>
4424 <listitem><para>Speed improvement for large object restore (Mario Weilguni)</para></listitem>
4425 <listitem><para>Mark expired index entries on first lookup, saving later heap fetches (Tom)</para></listitem>
4426 <listitem><para>Avoid excessive NULL bitmap padding (Manfred Koizar)</para></listitem>
4427 <listitem><para>Add BSD-licensed qsort() for Solaris, for performance (Bruce)</para></listitem>
4428 <listitem><para>Reduce per-row overhead by four bytes (Manfred Koizar)</para></listitem>
4429 <listitem><para>Fix GEQO optimizer bug (Neil Conway)</para></listitem>
4430 <listitem><para>Make WITHOUT OID actually save four bytes per row (Manfred Koizar)</para></listitem>
4431 <listitem><para>Add default_statistics_target variable to specify ANALYZE buckets (Neil)</para></listitem>
4432 <listitem><para>Use local buffer cache for temporary tables so no WAL overhead (Tom)</para></listitem>
4433 <listitem><para>Improve free space map performance on large tables (Stephen Marshall, Tom)</para></listitem>
4434 <listitem><para>Improved WAL write concurrency (Tom)</para></listitem>
4439 <title>Privileges</title>
4441 <listitem><para>Add privileges on functions and procedural languages (Peter)</para></listitem>
4442 <listitem><para>Add OWNER to CREATE DATABASE so superusers can create databases on behalf of unprivileged users (Gavin Sherry, Tom)</para></listitem>
4443 <listitem><para>Add new object privilege bits EXECUTE and USAGE (Tom)</para></listitem>
4444 <listitem><para>Add SET SESSION AUTHORIZATION DEFAULT and RESET SESSION AUTHORIZATION (Tom)</para></listitem>
4445 <listitem><para>Allow functions to be executed with the privilege of the function owner (Peter)</para></listitem>
4450 <title>Server Configuration</title>
4452 <listitem><para>Server log messages now tagged with LOG, not DEBUG (Bruce)</para></listitem>
4453 <listitem><para>Add user column to pg_hba.conf (Bruce)</para></listitem>
4454 <listitem><para>Have log_connections output two lines in log file (Tom)</para></listitem>
4455 <listitem><para>Remove debug_level from postgresql.conf, now server_min_messages (Bruce)</para></listitem>
4456 <listitem><para>New ALTER DATABASE/USER ... SET command for per-user/database initialization (Peter)</para></listitem>
4457 <listitem><para>New parameters server_min_messages and client_min_messages to control which messages are sent to the server logs or client applications (Bruce)</para></listitem>
4458 <listitem><para>Allow pg_hba.conf to specify lists of users/databases separated by commas, group names prepended with +, and file names prepended with @ (Bruce)</para></listitem>
4459 <listitem><para>Remove secondary password file capability and pg_password utility (Bruce)</para></listitem>
4460 <listitem><para>Add variable db_user_namespace for database-local user names (Bruce)</para></listitem>
4461 <listitem><para>SSL improvements (Bear Giles)</para></listitem>
4462 <listitem><para>Make encryption of stored passwords the default (Bruce)</para></listitem>
4463 <listitem><para>Allow pg_statistics to be reset by calling pg_stat_reset() (Christopher)</para></listitem>
4464 <listitem><para>Add log_duration parameter (Bruce)</para></listitem>
4465 <listitem><para>Rename debug_print_query to log_statement (Bruce)</para></listitem>
4466 <listitem><para>Rename show_query_stats to show_statement_stats (Bruce)</para></listitem>
4467 <listitem><para>Add param log_min_error_statement to print commands to logs on error (Gavin)</para></listitem>
4472 <title>Queries</title>
4474 <listitem><para>Make cursors insensitive, meaning their contents do not change (Tom)</para></listitem>
4475 <listitem><para>Disable LIMIT #,# syntax; now only LIMIT # OFFSET # supported (Bruce)</para></listitem>
4476 <listitem><para>Increase identifier length to 63 (Neil, Bruce)</para></listitem>
4477 <listitem><para>UNION fixes for merging >= 3 columns of different lengths (Tom)</para></listitem>
4478 <listitem><para>Add DEFAULT key word to INSERT, e.g., INSERT ... (..., DEFAULT, ...) (Rod)</para></listitem>
4479 <listitem><para>Allow views to have default values using ALTER COLUMN ... SET DEFAULT (Neil)</para></listitem>
4480 <listitem><para>Fail on INSERTs with column lists that don't supply all column values, e.g., INSERT INTO tab (col1, col2) VALUES ('val1'); (Rod)</para></listitem>
4481 <listitem><para>Fix for join aliases (Tom)</para></listitem>
4482 <listitem><para>Fix for FULL OUTER JOINs (Tom)</para></listitem>
4483 <listitem><para>Improve reporting of invalid identifier and location (Tom, Gavin)</para></listitem>
4484 <listitem><para>Fix OPEN cursor(args) (Tom)</para></listitem>
4485 <listitem><para>Allow 'ctid' to be used in a view and currtid(viewname) (Hiroshi)</para></listitem>
4486 <listitem><para>Fix for CREATE TABLE AS with UNION (Tom)</para></listitem>
4487 <listitem><para>SQL99 syntax improvements (Thomas)</para></listitem>
4488 <listitem><para>Add statement_timeout variable to cancel queries (Bruce)</para></listitem>
4489 <listitem><para>Allow prepared queries with PREPARE/EXECUTE (Neil)</para></listitem>
4490 <listitem><para>Allow FOR UPDATE to appear after LIMIT/OFFSET (Bruce)</para></listitem>
4491 <listitem><para>Add variable autocommit (Tom, David Van Wie)</para></listitem>
4496 <title>Object Manipulation</title>
4498 <listitem><para>Make equals signs optional in CREATE DATABASE (Gavin Sherry)</para></listitem>
4499 <listitem><para>Make ALTER TABLE OWNER change index ownership too (Neil)</para></listitem>
4500 <listitem><para>New ALTER TABLE tabname ALTER COLUMN colname SET STORAGE controls TOAST storage, compression (John Gray)</para></listitem>
4501 <listitem><para>Add schema support, CREATE/DROP SCHEMA (Tom)</para></listitem>
4502 <listitem><para>Create schema for temporary tables (Tom)</para></listitem>
4503 <listitem><para>Add variable search_path for schema search (Tom)</para></listitem>
4504 <listitem><para>Add ALTER TABLE SET/DROP NOT NULL (Christopher)</para></listitem>
4505 <listitem><para>New CREATE FUNCTION volatility levels (Tom)</para></listitem>
4506 <listitem><para>Make rule names unique only per table (Tom)</para></listitem>
4507 <listitem><para>Add 'ON tablename' clause to DROP RULE and COMMENT ON RULE (Tom)</para></listitem>
4508 <listitem><para>Add ALTER TRIGGER RENAME (Joe)</para></listitem>
4509 <listitem><para>New current_schema() and current_schemas() inquiry functions (Tom)</para></listitem>
4510 <listitem><para>Allow functions to return multiple rows (table functions) (Joe)</para></listitem>
4511 <listitem><para>Make WITH optional in CREATE DATABASE, for consistency (Bruce)</para></listitem>
4512 <listitem><para>Add object dependency tracking (Rod, Tom)</para></listitem>
4513 <listitem><para>Add RESTRICT/CASCADE to DROP commands (Rod)</para></listitem>
4514 <listitem><para>Add ALTER TABLE DROP for non-CHECK CONSTRAINT (Rod)</para></listitem>
4515 <listitem><para>Autodestroy sequence on DROP of table with SERIAL (Rod)</para></listitem>
4516 <listitem><para>Prevent column dropping if column is used by foreign key (Rod)</para></listitem>
4517 <listitem><para>Automatically drop constraints/functions when object is dropped (Rod)</para></listitem>
4518 <listitem><para>Add CREATE/DROP OPERATOR CLASS (Bill Studenmund, Tom)</para></listitem>
4519 <listitem><para>Add ALTER TABLE DROP COLUMN (Christopher, Tom, Hiroshi)</para></listitem>
4520 <listitem><para>Prevent inherited columns from being removed or renamed (Alvaro Herrera)</para></listitem>
4521 <listitem><para>Fix foreign key constraints to not error on intermediate database states (Stephan)</para></listitem>
4522 <listitem><para>Propagate column or table renaming to foreign key constraints</para></listitem>
4523 <listitem><para>Add CREATE OR REPLACE VIEW (Gavin, Neil, Tom)</para></listitem>
4524 <listitem><para>Add CREATE OR REPLACE RULE (Gavin, Neil, Tom)</para></listitem>
4525 <listitem><para>Have rules execute alphabetically, returning more predictable values (Tom)</para></listitem>
4526 <listitem><para>Triggers are now fired in alphabetical order (Tom)</para></listitem>
4527 <listitem><para>Add /contrib/adddepend to handle pre-7.3 object dependencies (Rod)</para></listitem>
4528 <listitem><para>Allow better casting when inserting/updating values (Tom)</para></listitem>
4533 <title>Utility Commands</title>
4535 <listitem><para>Have COPY TO output embedded carriage returns and newlines as \r and \n (Tom)</para></listitem>
4536 <listitem><para>Allow DELIMITER in COPY FROM to be 8-bit clean (Tatsuo)</para></listitem>
4537 <listitem><para>Make <application>pg_dump</> use ALTER TABLE ADD PRIMARY KEY, for performance (Neil)</para></listitem>
4538 <listitem><para>Disable brackets in multistatement rules (Bruce)</para></listitem>
4539 <listitem><para>Disable VACUUM from being called inside a function (Bruce)</para></listitem>
4540 <listitem><para>Allow dropdb and other scripts to use identifiers with spaces (Bruce)</para></listitem>
4541 <listitem><para>Restrict database comment changes to the current database</para></listitem>
4542 <listitem><para>Allow comments on operators, independent of the underlying function (Rod)</para></listitem>
4543 <listitem><para>Rollback SET commands in aborted transactions (Tom)</para></listitem>
4544 <listitem><para>EXPLAIN now outputs as a query (Tom)</para></listitem>
4545 <listitem><para>Display condition expressions and sort keys in EXPLAIN (Tom)</para></listitem>
4546 <listitem><para>Add 'SET LOCAL var = value' to set configuration variables for a single transaction (Tom)</para></listitem>
4547 <listitem><para>Allow ANALYZE to run in a transaction (Bruce)</para></listitem>
4548 <listitem><para>Improve COPY syntax using new WITH clauses, keep backward compatibility (Bruce)</para></listitem>
4549 <listitem><para>Fix <application>pg_dump</> to consistently output tags in non-ASCII dumps (Bruce)</para></listitem>
4550 <listitem><para>Make foreign key constraints clearer in dump file (Rod)</para></listitem>
4551 <listitem><para>Add COMMENT ON CONSTRAINT (Rod)</para></listitem>
4552 <listitem><para>Allow COPY TO/FROM to specify column names (Brent Verner)</para></listitem>
4553 <listitem><para>Dump UNIQUE and PRIMARY KEY constraints as ALTER TABLE (Rod)</para></listitem>
4554 <listitem><para>Have SHOW output a query result (Joe)</para></listitem>
4555 <listitem><para>Generate failure on short COPY lines rather than pad NULLs (Neil)</para></listitem>
4556 <listitem><para>Fix CLUSTER to preserve all table attributes (Alvaro Herrera)</para></listitem>
4557 <listitem><para>New pg_settings table to view/modify GUC settings (Joe)</para></listitem>
4558 <listitem><para>Add smart quoting, portability improvements to <application>pg_dump</> output (Peter)</para></listitem>
4559 <listitem><para>Dump serial columns out as SERIAL (Tom)</para></listitem>
4560 <listitem><para>Enable large file support, >2G for <application>pg_dump</> (Peter, Philip Warner, Bruce)</para></listitem>
4561 <listitem><para>Disallow TRUNCATE on tables that are involved in referential constraints (Rod)</para></listitem>
4562 <listitem><para>Have TRUNCATE also auto-truncate the toast table of the relation (Tom)</para></listitem>
4563 <listitem><para>Add clusterdb utility that will auto-cluster an entire database based on previous CLUSTER operations (Alvaro Herrera)</para></listitem>
4564 <listitem><para>Overhaul pg_dumpall (Peter)</para></listitem>
4565 <listitem><para>Allow REINDEX of TOAST tables (Tom)</para></listitem>
4566 <listitem><para>Implemented START TRANSACTION, per SQL99 (Neil)</para></listitem>
4567 <listitem><para>Fix rare index corruption when a page split affects bulk delete (Tom)</para></listitem>
4568 <listitem><para>Fix ALTER TABLE ... ADD COLUMN for inheritance (Alvaro Herrera)</para></listitem>
4573 <title>Data Types and Functions</title>
4575 <listitem><para>Fix factorial(0) to return 1 (Bruce)</para></listitem>
4576 <listitem><para>Date/time/timezone improvements (Thomas)</para></listitem>
4577 <listitem><para>Fix for array slice extraction (Tom)</para></listitem>
4578 <listitem><para>Fix extract/date_part to report proper microseconds for timestamp (Tatsuo)</para></listitem>
4579 <listitem><para>Allow text_substr() and bytea_substr() to read TOAST values more efficiently (John Gray)</para></listitem>
4580 <listitem><para>Add domain support (Rod)</para></listitem>
4581 <listitem><para>Make WITHOUT TIME ZONE the default for TIMESTAMP and TIME data types (Thomas)</para></listitem>
4582 <listitem><para>Allow alternate storage scheme of 64-bit integers for date/time types using --enable-integer-datetimes in configure (Thomas)</para></listitem>
4583 <listitem><para>Make timezone(timestamptz) return timestamp rather than a string (Thomas)</para></listitem>
4584 <listitem><para>Allow fractional seconds in date/time types for dates prior to 1BC (Thomas)</para></listitem>
4585 <listitem><para>Limit timestamp data types to 6 decimal places of precision (Thomas)</para></listitem>
4586 <listitem><para>Change timezone conversion functions from timetz() to timezone() (Thomas)</para></listitem>
4587 <listitem><para>Add configuration variables datestyle and timezone (Tom)</para></listitem>
4588 <listitem><para>Add OVERLAY(), which allows substitution of a substring in a string (Thomas)</para></listitem>
4589 <listitem><para>Add SIMILAR TO (Thomas, Tom)</para></listitem>
4590 <listitem><para>Add regular expression SUBSTRING(string FROM pat FOR escape) (Thomas)</para></listitem>
4591 <listitem><para>Add LOCALTIME and LOCALTIMESTAMP functions (Thomas)</para></listitem>
4592 <listitem><para>Add named composite types using CREATE TYPE typename AS (column) (Joe)</para></listitem>
4593 <listitem><para>Allow composite type definition in the table alias clause (Joe)</para></listitem>
4594 <listitem><para>Add new API to simplify creation of C language table functions (Joe)</para></listitem>
4595 <listitem><para>Remove ODBC-compatible empty parentheses from calls to SQL99 functions for which these parentheses do not match the standard (Thomas)</para></listitem>
4596 <listitem><para>Allow macaddr data type to accept 12 hex digits with no separators (Mike Wyer)</para></listitem>
4597 <listitem><para>Add CREATE/DROP CAST (Peter)</para></listitem>
4598 <listitem><para>Add IS DISTINCT FROM operator (Thomas)</para></listitem>
4599 <listitem><para>Add SQL99 TREAT() function, synonym for CAST() (Thomas)</para></listitem>
4600 <listitem><para>Add pg_backend_pid() to output backend pid (Bruce)</para></listitem>
4601 <listitem><para>Add IS OF / IS NOT OF type predicate (Thomas)</para></listitem>
4602 <listitem><para>Allow bit string constants without fully-specified length (Thomas)</para></listitem>
4603 <listitem><para>Allow conversion between 8-byte integers and bit strings (Thomas)</para></listitem>
4604 <listitem><para>Implement hex literal conversion to bit string literal (Thomas)</para></listitem>
4605 <listitem><para>Allow table functions to appear in the FROM clause (Joe)</para></listitem>
4606 <listitem><para>Increase maximum number of function parameters to 32 (Bruce)</para></listitem>
4607 <listitem><para>No longer automatically create index for SERIAL column (Tom)</para></listitem>
4608 <listitem><para>Add current_database() (Rod)</para></listitem>
4609 <listitem><para>Fix cash_words() to not overflow buffer (Tom)</para></listitem>
4610 <listitem><para>Add functions replace(), split_part(), to_hex() (Joe)</para></listitem>
4611 <listitem><para>Fix LIKE for bytea as a right-hand argument (Joe)</para></listitem>
4612 <listitem><para>Prevent crashes caused by SELECT cash_out(2) (Tom)</para></listitem>
4613 <listitem><para>Fix to_char(1,'FM999.99') to return a period (Karel)</para></listitem>
4614 <listitem><para>Fix trigger/type/language functions returning OPAQUE to return proper type (Tom)</para></listitem>
4619 <title>Internationalization</title>
4621 <listitem><para>Add additional encodings: Korean (JOHAB), Thai (WIN874), Vietnamese (TCVN), Arabic (WIN1256), Simplified Chinese (GBK), Korean (UHC) (Eiji Tokuya)</para></listitem>
4622 <listitem><para>Enable locale support by default (Peter)</para></listitem>
4623 <listitem><para>Add locale variables (Peter)</para></listitem>
4624 <listitem><para>Escape byes >= 0x7f for multibyte in PQescapeBytea/PQunescapeBytea (Tatsuo)</para></listitem>
4625 <listitem><para>Add locale awareness to regular expression character classes</para></listitem>
4626 <listitem><para>Enable multibyte support by default (Tatsuo)</para></listitem>
4627 <listitem><para>Add GB18030 multibyte support (Bill Huang)</para></listitem>
4628 <listitem><para>Add CREATE/DROP CONVERSION, allowing loadable encodings (Tatsuo, Kaori)</para></listitem>
4629 <listitem><para>Add pg_conversion table (Tatsuo)</para></listitem>
4630 <listitem><para>Add SQL99 CONVERT() function (Tatsuo)</para></listitem>
4631 <listitem><para>pg_dumpall, pg_controldata, and pg_resetxlog now national-language aware (Peter)</para></listitem>
4632 <listitem><para>New and updated translations</para></listitem>
4637 <title>Server-side Languages</title>
4639 <listitem><para>Allow recursive SQL function (Peter)</para></listitem>
4640 <listitem><para>Change PL/Tcl build to use configured compiler and Makefile.shlib (Peter)</para></listitem>
4641 <listitem><para>Overhaul the PL/pgSQL FOUND variable to be more Oracle-compatible (Neil, Tom)</para></listitem>
4642 <listitem><para>Allow PL/pgSQL to handle quoted identifiers (Tom)</para></listitem>
4643 <listitem><para>Allow set-returning PL/pgSQL functions (Neil)</para></listitem>
4644 <listitem><para>Make PL/pgSQL schema-aware (Joe)</para></listitem>
4645 <listitem><para>Remove some memory leaks (Nigel J. Andrews, Tom)</para></listitem>
4652 <listitem><para>Don't lowercase psql \connect database name for 7.2.0 compatibility (Tom)</para></listitem>
4653 <listitem><para>Add psql \timing to time user queries (Greg Sabino Mullane)</para></listitem>
4654 <listitem><para>Have psql \d show index information (Greg Sabino Mullane)</para></listitem>
4655 <listitem><para>New psql \dD shows domains (Jonathan Eisler)</para></listitem>
4656 <listitem><para>Allow psql to show rules on views (Paul ?)</para></listitem>
4657 <listitem><para>Fix for psql variable substitution (Tom)</para></listitem>
4658 <listitem><para>Allow psql \d to show temporary table structure (Tom)</para></listitem>
4659 <listitem><para>Allow psql \d to show foreign keys (Rod)</para></listitem>
4660 <listitem><para>Fix \? to honor \pset pager (Bruce)</para></listitem>
4661 <listitem><para>Have psql reports its version number on startup (Tom)</para></listitem>
4662 <listitem><para>Allow \copy to specify column names (Tom)</para></listitem>
4667 <title>libpq</title>
4669 <listitem><para>Add $HOME/.pgpass to store host/user password combinations (Alvaro Herrera)</para></listitem>
4670 <listitem><para>Add PQunescapeBytea() function to libpq (Patrick Welche)</para></listitem>
4671 <listitem><para>Fix for sending large queries over non-blocking connections (Bernhard Herzog)</para></listitem>
4672 <listitem><para>Fix for libpq using timers on Win9X (David Ford)</para></listitem>
4673 <listitem><para>Allow libpq notify to handle servers with different-length identifiers (Tom)</para></listitem>
4674 <listitem><para>Add libpq PQescapeString() and PQescapeBytea() to Windows (Bruce)</para></listitem>
4675 <listitem><para>Fix for SSL with non-blocking connections (Jack Bates)</para></listitem>
4676 <listitem><para>Add libpq connection timeout parameter (Denis A Ustimenko)</para></listitem>
4683 <listitem><para>Allow JDBC to compile with JDK 1.4 (Dave)</para></listitem>
4684 <listitem><para>Add JDBC 3 support (Barry)</para></listitem>
4685 <listitem><para>Allows JDBC to set loglevel by adding ?loglevel=X to the connection URL (Barry)</para></listitem>
4686 <listitem><para>Add Driver.info() message that prints out the version number (Barry)</para></listitem>
4687 <listitem><para>Add updateable result sets (Raghu Nidagal, Dave)</para></listitem>
4688 <listitem><para>Add support for callable statements (Paul Bethe)</para></listitem>
4689 <listitem><para>Add query cancel capability</para></listitem>
4690 <listitem><para>Add refresh row (Dave)</para></listitem>
4691 <listitem><para>Fix MD5 encryption handling for multibyte servers (Jun Kawai)</para></listitem>
4692 <listitem><para>Add support for prepared statements (Barry)</para></listitem>
4697 <title>Miscellaneous Interfaces</title>
4699 <listitem><para>Fixed ECPG bug concerning octal numbers in single quotes (Michael)</para></listitem>
4700 <listitem><para>Move src/interfaces/libpgeasy to http://gborg.postgresql.org (Marc, Bruce)</para></listitem>
4701 <listitem><para>Improve Python interface (Elliot Lee, Andrew Johnson, Greg Copeland)</para></listitem>
4702 <listitem><para>Add libpgtcl connection close event (Gerhard Hintermayer)</para></listitem>
4703 <listitem><para>Move src/interfaces/libpq++ to http://gborg.postgresql.org (Marc, Bruce)</para></listitem>
4704 <listitem><para>Move src/interfaces/odbc to http://gborg.postgresql.org (Marc)</para></listitem>
4705 <listitem><para>Move src/interfaces/libpgeasy to http://gborg.postgresql.org (Marc, Bruce)</para></listitem>
4706 <listitem><para>Move src/interfaces/perl5 to http://gborg.postgresql.org (Marc, Bruce)</para></listitem>
4707 <listitem><para>Remove src/bin/pgaccess from main tree, now at http://www.pgaccess.org (Bruce)</para></listitem>
4708 <listitem><para>Add pg_on_connection_loss command to libpgtcl (Gerhard Hintermayer, Tom)</para></listitem>
4713 <title>Source Code</title>
4715 <listitem><para>Fix for parallel make (Peter)</para></listitem>
4716 <listitem><para>AIX fixes for linking Tcl (Andreas Zeugswetter)</para></listitem>
4717 <listitem><para>Allow PL/Perl to build under Cygwin (Jason Tishler)</para></listitem>
4718 <listitem><para>Improve MIPS compiles (Peter, Oliver Elphick)</para></listitem>
4719 <listitem><para>Require Autoconf version 2.53 (Peter)</para></listitem>
4720 <listitem><para>Require readline and zlib by default in configure (Peter)</para></listitem>
4721 <listitem><para>Allow Solaris to use Intimate Shared Memory (ISM), for performance (Scott Brunza, P.J. Josh Rovero)</para></listitem>
4722 <listitem><para>Always enable syslog in compile, remove --enable-syslog option (Tatsuo)</para></listitem>
4723 <listitem><para>Always enable multibyte in compile, remove --enable-multibyte option (Tatsuo)</para></listitem>
4724 <listitem><para>Always enable locale in compile, remove --enable-locale option (Peter)</para></listitem>
4725 <listitem><para>Fix for Win9x DLL creation (Magnus Naeslund)</para></listitem>
4726 <listitem><para>Fix for link() usage by WAL code on Windows, BeOS (Jason Tishler)</para></listitem>
4727 <listitem><para>Add sys/types.h to c.h, remove from main files (Peter, Bruce)</para></listitem>
4728 <listitem><para>Fix AIX hang on SMP machines (Tomoyuki Niijima)</para></listitem>
4729 <listitem><para>AIX SMP hang fix (Tomoyuki Niijima)</para></listitem>
4730 <listitem><para>Fix pre-1970 date handling on newer glibc libraries (Tom)</para></listitem>
4731 <listitem><para>Fix PowerPC SMP locking (Tom)</para></listitem>
4732 <listitem><para>Prevent gcc -ffast-math from being used (Peter, Tom)</para></listitem>
4733 <listitem><para>Bison >= 1.50 now required for developer builds</para></listitem>
4734 <listitem><para>Kerberos 5 support now builds with Heimdal (Peter)</para></listitem>
4735 <listitem><para>Add appendix in the User's Guide which lists SQL features (Thomas)</para></listitem>
4736 <listitem><para>Improve loadable module linking to use RTLD_NOW (Tom)</para></listitem>
4737 <listitem><para>New error levels WARNING, INFO, LOG, DEBUG[1-5] (Bruce)</para></listitem>
4738 <listitem><para>New src/port directory holds replaced libc functions (Peter, Bruce)</para></listitem>
4739 <listitem><para>New pg_namespace system catalog for schemas (Tom)</para></listitem>
4740 <listitem><para>Add pg_class.relnamespace for schemas (Tom)</para></listitem>
4741 <listitem><para>Add pg_type.typnamespace for schemas (Tom)</para></listitem>
4742 <listitem><para>Add pg_proc.pronamespace for schemas (Tom)</para></listitem>
4743 <listitem><para>Restructure aggregates to have pg_proc entries (Tom)</para></listitem>
4744 <listitem><para>System relations now have their own namespace, pg_* test not required (Fernando Nasser)</para></listitem>
4745 <listitem><para>Rename TOAST index names to be *_index rather than *_idx (Neil)</para></listitem>
4746 <listitem><para>Add namespaces for operators, opclasses (Tom)</para></listitem>
4747 <listitem><para>Add additional checks to server control file (Thomas)</para></listitem>
4748 <listitem><para>New Polish FAQ (Marcin Mazurek)</para></listitem>
4749 <listitem><para>Add Posix semaphore support (Tom)</para></listitem>
4750 <listitem><para>Document need for reindex (Bruce)</para></listitem>
4751 <listitem><para>Rename some internal identifiers to simplify Windows compile (Jan, Katherine Ward)</para></listitem>
4752 <listitem><para>Add documentation on computing disk space (Bruce)</para></listitem>
4753 <listitem><para>Remove KSQO from GUC (Bruce)</para></listitem>
4754 <listitem><para>Fix memory leak in rtree (Kenneth Been)</para></listitem>
4755 <listitem><para>Modify a few error messages for consistency (Bruce)</para></listitem>
4756 <listitem><para>Remove unused system table columns (Peter)</para></listitem>
4757 <listitem><para>Make system columns NOT NULL where appropriate (Tom)</para></listitem>
4758 <listitem><para>Clean up use of sprintf in favor of snprintf() (Neil, Jukka Holappa)</para></listitem>
4759 <listitem><para>Remove OPAQUE and create specific subtypes (Tom)</para></listitem>
4760 <listitem><para>Cleanups in array internal handling (Joe, Tom)</para></listitem>
4761 <listitem><para>Disallow pg_atoi('') (Bruce)</para></listitem>
4762 <listitem><para>Remove parameter wal_files because WAL files are now recycled (Bruce)</para></listitem>
4763 <listitem><para>Add version numbers to heap pages (Tom)</para></listitem>
4768 <title>Contrib</title>
4770 <listitem><para>Allow inet arrays in /contrib/array (Neil)</para></listitem>
4771 <listitem><para>GiST fixes (Teodor Sigaev, Neil)</para></listitem>
4772 <listitem><para>Upgrade /contrib/mysql</para></listitem>
4773 <listitem><para>Add /contrib/dbsize which shows table sizes without vacuum (Peter)</para></listitem>
4774 <listitem><para>Add /contrib/intagg, integer aggregator routines (mlw)</para></listitem>
4775 <listitem><para>Improve /contrib/oid2name (Neil, Bruce)</para></listitem>
4776 <listitem><para>Improve /contrib/tsearch (Oleg, Teodor Sigaev)</para></listitem>
4777 <listitem><para>Cleanups of /contrib/rserver (Alexey V. Borzov)</para></listitem>
4778 <listitem><para>Update /contrib/oracle conversion utility (Gilles Darold)</para></listitem>
4779 <listitem><para>Update /contrib/dblink (Joe)</para></listitem>
4780 <listitem><para>Improve options supported by /contrib/vacuumlo (Mario Weilguni)</para></listitem>
4781 <listitem><para>Improvements to /contrib/intarray (Oleg, Teodor Sigaev, Andrey Oktyabrski)</para></listitem>
4782 <listitem><para>Add /contrib/reindexdb utility (Shaun Thomas)</para></listitem>
4783 <listitem><para>Add indexing to /contrib/isbn_issn (Dan Weston)</para></listitem>
4784 <listitem><para>Add /contrib/dbmirror (Steven Singer)</para></listitem>
4785 <listitem><para>Improve /contrib/pgbench (Neil)</para></listitem>
4786 <listitem><para>Add /contrib/tablefunc table function examples (Joe)</para></listitem>
4787 <listitem><para>Add /contrib/ltree data type for tree structures (Teodor Sigaev, Oleg Bartunov)</para></listitem>
4788 <listitem><para>Move /contrib/pg_controldata, pg_resetxlog into main tree (Bruce)</para></listitem>
4789 <listitem><para>Fixes to /contrib/cube (Bruno Wolff)</para></listitem>
4790 <listitem><para>Improve /contrib/fulltextindex (Christopher)</para></listitem>
4798 <sect1 id="release-7-2-4">
4799 <title>Release 7.2.4</title>
4802 <title>Release date</title>
4803 <simpara>2003-01-30</simpara>
4807 This release contains a variety of fixes for version 7.2.3,
4808 including fixes to prevent possible data loss.
4812 <title>Migration to version 7.2.4</title>
4815 A dump/restore is <emphasis>not</emphasis> required for those
4816 running version 7.2.*.
4821 <title>Changes</title>
4824 <listitem><para>Fix some additional cases of VACUUM "No one parent tuple was found" error</para></listitem>
4825 <listitem><para>Prevent VACUUM from being called inside a function (Bruce)</para></listitem>
4826 <listitem><para>Ensure pg_clog updates are sync'd to disk before marking checkpoint complete</para></listitem>
4827 <listitem><para>Avoid integer overflow during large hash joins</para></listitem>
4828 <listitem><para>Make GROUP commands work when pg_group.grolist is large enough to be toasted</para></listitem>
4829 <listitem><para>Fix errors in datetime tables; some timezone names weren't being recognized</para></listitem>
4830 <listitem><para>Fix integer overflows in circle_poly(), path_encode(), path_add() (Neil)</para></listitem>
4831 <listitem><para>Repair long-standing logic errors in lseg_eq(), lseg_ne(), lseg_center()</para></listitem>
4837 <sect1 id="release-7-2-3">
4838 <title>Release 7.2.3</title>
4841 <title>Release date</title>
4842 <simpara>2002-10-01</simpara>
4846 This release contains a variety of fixes for version 7.2.2,
4847 including fixes to prevent possible data loss.
4851 <title>Migration to version 7.2.3</title>
4854 A dump/restore is <emphasis>not</emphasis> required for those
4855 running version 7.2.*.
4860 <title>Changes</title>
4863 <listitem><para>Prevent possible compressed transaction log loss (Tom)</para></listitem>
4864 <listitem><para>Prevent non-superuser from increasing most recent vacuum info (Tom)</para></listitem>
4865 <listitem><para>Handle pre-1970 date values in newer versions of glibc (Tom)</para></listitem>
4866 <listitem><para>Fix possible hang during server shutdown</para></listitem>
4867 <listitem><para>Prevent spinlock hangs on SMP PPC machines (Tomoyuki Niijima)</para></listitem>
4868 <listitem><para>Fix <application>pg_dump</> to properly dump FULL JOIN USING (Tom)</para></listitem>
4874 <sect1 id="release-7-2-2">
4875 <title>Release 7.2.2</title>
4878 <title>Release date</title>
4879 <simpara>2002-08-23</simpara>
4883 This release contains a variety of fixes for version 7.2.1.
4887 <title>Migration to version 7.2.2</title>
4890 A dump/restore is <emphasis>not</emphasis> required for those
4891 running version 7.2.*.
4896 <title>Changes</title>
4899 <listitem><para>Allow EXECUTE of "CREATE TABLE AS ... SELECT" in PL/pgSQL (Tom)</para></listitem>
4900 <listitem><para>Fix for compressed transaction log id wraparound (Tom)</para></listitem>
4901 <listitem><para>Fix PQescapeBytea/PQunescapeBytea so that they handle bytes > 0x7f (Tatsuo)</para></listitem>
4902 <listitem><para>Fix for psql and <application>pg_dump</> crashing when invoked with non-existent long options (Tatsuo)</para></listitem>
4903 <listitem><para>Fix crash when invoking geometric operators (Tom)</para></listitem>
4904 <listitem><para>Allow OPEN cursor(args) (Tom)</para></listitem>
4905 <listitem><para>Fix for rtree_gist index build (Teodor)</para></listitem>
4906 <listitem><para>Fix for dumping user-defined aggregates (Tom)</para></listitem>
4907 <listitem><para>contrib/intarray fixes (Oleg)</para></listitem>
4908 <listitem><para>Fix for complex UNION/EXCEPT/INTERSECT queries using parens (Tom)</para></listitem>
4909 <listitem><para>Fix to pg_convert (Tatsuo)</para></listitem>
4910 <listitem><para>Fix for crash with long DATA strings (Thomas, Neil)</para></listitem>
4911 <listitem><para>Fix for repeat(), lpad(), rpad() and long strings (Neil)</para></listitem>
4917 <sect1 id="release-7-2-1">
4918 <title>Release 7.2.1</title>
4921 <title>Release date</title>
4922 <simpara>2002-03-21</simpara>
4926 This release contains a variety of fixes for version 7.2.
4930 <title>Migration to version 7.2.1</title>
4933 A dump/restore is <emphasis>not</emphasis> required for those
4934 running version 7.2.
4939 <title>Changes</title>
4942 <listitem><para>Ensure that sequence counters do not go backwards after a crash (Tom)</para></listitem>
4943 <listitem><para>Fix pgaccess kanji-conversion key binding (Tatsuo)</para></listitem>
4944 <listitem><para>Optimizer improvements (Tom)</para></listitem>
4945 <listitem><para>Cash I/O improvements (Tom)</para></listitem>
4946 <listitem><para>New Russian FAQ</para></listitem>
4947 <listitem><para>Compile fix for missing AuthBlockSig (Heiko)</para></listitem>
4948 <listitem><para>Additional time zones and time zone fixes (Thomas)</para></listitem>
4949 <listitem><para>Allow psql \connect to handle mixed case database and user names (Tom)</para></listitem>
4950 <listitem><para>Return proper OID on command completion even with ON INSERT rules (Tom)</para></listitem>
4951 <listitem><para>Allow COPY FROM to use 8-bit DELIMITERS (Tatsuo)</para></listitem>
4952 <listitem><para>Fix bug in extract/date_part for milliseconds/microseconds (Tatsuo)</para></listitem>
4953 <listitem><para>Improve handling of multiple UNIONs with different lengths (Tom)</para></listitem>
4954 <listitem><para>contrib/btree_gist improvements (Teodor Sigaev)</para></listitem>
4955 <listitem><para>contrib/tsearch dictionary improvements, see README.tsearch for an additional installation step (Thomas T. Thai, Teodor Sigaev)</para></listitem>
4956 <listitem><para>Fix for array subscripts handling (Tom)</para></listitem>
4957 <listitem><para>Allow EXECUTE of "CREATE TABLE AS ... SELECT" in PL/pgSQL (Tom)</para></listitem>
4963 <sect1 id="release-7-2">
4964 <title>Release 7.2</title>
4967 <title>Release date</title>
4968 <simpara>2002-02-04</simpara>
4972 <title>Overview</title>
4975 This release improves <productname>PostgreSQL</> for use in
4976 high-volume applications.
4980 Major changes in this release:
4987 Vacuuming no longer locks tables, thus allowing normal user
4988 access during the vacuum. A new <command>VACUUM FULL</>
4989 command does old-style vacuum by locking the table and
4990 shrinking the on-disk copy of the table.
4996 <term>Transactions</term>
4999 There is no longer a problem with installations that exceed
5000 four billion transactions.
5009 OIDs are now optional. Users can now create tables without
5010 OIDs for cases where OID usage is excessive.
5016 <term>Optimizer</term>
5019 The system now computes histogram column statistics during
5020 <command>ANALYZE</>, allowing much better optimizer choices.
5026 <term>Security</term>
5029 A new MD5 encryption option allows more secure storage and
5030 transfer of passwords. A new Unix-domain socket
5031 authentication option is available on Linux and BSD systems.
5037 <term>Statistics</term>
5040 Administrators can use the new table access statistics module
5041 to get fine-grained information about table and index usage.
5047 <term>Internationalization</term>
5050 Program and library messages can now be displayed in several
5060 <title>Migration to version 7.2</title>
5063 A dump/restore using <command>pg_dump</command> is required for
5064 those wishing to migrate data from any previous release.
5068 Observe the following incompatibilities:
5073 The semantics of the <command>VACUUM</command> command have
5074 changed in this release. You may wish to update your
5075 maintenance procedures accordingly.
5081 In this release, comparisons using <literal>= NULL</literal>
5082 will always return false (or NULL, more precisely). Previous
5083 releases automatically transformed this syntax to <literal>IS
5084 NULL</literal>. The old behavior can be re-enabled using a
5085 <filename>postgresql.conf</filename> parameter.
5091 The <filename>pg_hba.conf</> and <filename>pg_ident.conf</>
5092 configuration is now only reloaded after receiving a
5093 <systemitem>SIGHUP</> signal, not with each connection.
5099 The function <filename>octet_length()</> now returns the uncompressed data length.
5105 The date/time value <literal>'current'</literal> is no longer
5106 available. You will need to rewrite your applications.
5112 The <literal>timestamp()</literal>, <literal>time()</literal>,
5113 and <literal>interval()</literal> functions are no longer
5114 available. Instead of <literal>timestamp()</literal>, use
5115 <literal>timestamp 'string'</literal> or <literal>CAST</literal>.
5122 The <literal>SELECT ... LIMIT #,#</literal> syntax will be removed
5123 in the next release. You should change your queries to use
5124 separate LIMIT and OFFSET clauses, e.g. <literal>LIMIT 10 OFFSET
5130 <title>Changes</title>
5133 <title>Server Operation</title>
5135 <listitem><para>Create temporary files in a separate directory (Bruce)</para></listitem>
5136 <listitem><para>Delete orphaned temporary files on postmaster startup (Bruce)</para></listitem>
5137 <listitem><para>Added unique indexes to some system tables (Tom)</para></listitem>
5138 <listitem><para>System table operator reorganization (Oleg Bartunov, Teodor Sigaev, Tom)</para></listitem>
5139 <listitem><para>Renamed pg_log to pg_clog (Tom)</para></listitem>
5140 <listitem><para>Enable SIGTERM, SIGQUIT to kill backends (Jan)</para></listitem>
5141 <listitem><para>Removed compile-time limit on number of backends (Tom)</para></listitem>
5142 <listitem><para>Better cleanup for semaphore resource failure (Tatsuo, Tom)</para></listitem>
5143 <listitem><para>Allow safe transaction ID wraparound (Tom)</para></listitem>
5144 <listitem><para>Removed OIDs from some system tables (Tom)</para></listitem>
5145 <listitem><para>Removed "triggered data change violation" error check (Tom)</para></listitem>
5146 <listitem><para>SPI portal creation of prepared/saved plans (Jan)</para></listitem>
5147 <listitem><para>Allow SPI column functions to work for system columns (Tom)</para></listitem>
5148 <listitem><para>Long value compression improvement (Tom)</para></listitem>
5149 <listitem><para>Statistics collector for table, index access (Jan)</para></listitem>
5150 <listitem><para>Truncate extra-long sequence names to a reasonable value (Tom)</para></listitem>
5151 <listitem><para>Measure transaction times in milliseconds (Thomas)</para></listitem>
5152 <listitem><para>Fix TID sequential scans (Hiroshi)</para></listitem>
5153 <listitem><para>Superuser ID now fixed at 1 (Peter E)</para></listitem>
5154 <listitem><para>New pg_ctl "reload" option (Tom)</para></listitem>
5159 <title>Performance</title>
5161 <listitem><para>Optimizer improvements (Tom)</para></listitem>
5162 <listitem><para>New histogram column statistics for optimizer (Tom)</para></listitem>
5163 <listitem><para>Reuse write-ahead log files rather than discarding them (Tom)</para></listitem>
5164 <listitem><para>Cache improvements (Tom)</para></listitem>
5165 <listitem><para>IS NULL, IS NOT NULL optimizer improvement (Tom)</para></listitem>
5166 <listitem><para>Improve lock manager to reduce lock contention (Tom)</para></listitem>
5167 <listitem><para>Keep relcache entries for index access support functions (Tom)</para></listitem>
5168 <listitem><para>Allow better selectivity with NaN and infinities in NUMERIC (Tom)</para></listitem>
5169 <listitem><para>R-tree performance improvements (Kenneth Been)</para></listitem>
5170 <listitem><para>B-tree splits more efficient (Tom)</para></listitem>
5175 <title>Privileges</title>
5177 <listitem><para>Change UPDATE, DELETE privileges to be distinct (Peter E)</para></listitem>
5178 <listitem><para>New REFERENCES, TRIGGER privileges (Peter E)</para></listitem>
5179 <listitem><para>Allow GRANT/REVOKE to/from more than one user at a time (Peter E)</para></listitem>
5180 <listitem><para>New has_table_privilege() function (Joe Conway)</para></listitem>
5181 <listitem><para>Allow non-superuser to vacuum database (Tom)</para></listitem>
5182 <listitem><para>New SET SESSION AUTHORIZATION command (Peter E)</para></listitem>
5183 <listitem><para>Fix bug in privilege modifications on newly created tables (Tom)</para></listitem>
5184 <listitem><para>Disallow access to pg_statistic for non-superuser, add user-accessible views (Tom)</para></listitem>
5189 <title>Client Authentication</title>
5191 <listitem><para>Fork postmaster before doing authentication to prevent hangs (Peter E)</para></listitem>
5192 <listitem><para>Add ident authentication over Unix domain sockets on Linux, *BSD (Helge Bahmann, Oliver Elphick, Teodor Sigaev, Bruce)</para></listitem>
5193 <listitem><para>Add a password authentication method that uses MD5 encryption (Bruce)</para></listitem>
5194 <listitem><para>Allow encryption of stored passwords using MD5 (Bruce)</para></listitem>
5195 <listitem><para>PAM authentication (Dominic J. Eidson)</para></listitem>
5196 <listitem><para>Load pg_hba.conf and pg_ident.conf only on startup and SIGHUP (Bruce)</para></listitem>
5201 <title>Server Configuration</title>
5203 <listitem><para>Interpretation of some time zone abbreviations as Australian rather than North American now settable at run time (Bruce)</para></listitem>
5204 <listitem><para>New parameter to set default transaction isolation level (Peter E)</para></listitem>
5205 <listitem><para>New parameter to enable conversion of "expr = NULL" into "expr IS NULL", off by default (Peter E)</para></listitem>
5206 <listitem><para>New parameter to control memory usage by VACUUM (Tom)</para></listitem>
5207 <listitem><para>New parameter to set client authentication timeout (Tom)</para></listitem>
5208 <listitem><para>New parameter to set maximum number of open files (Tom)</para></listitem>
5213 <title>Queries</title>
5215 <listitem><para>Statements added by INSERT rules now execute after the INSERT (Jan)</para></listitem>
5216 <listitem><para>Prevent unadorned relation names in target list (Bruce)</para></listitem>
5217 <listitem><para>NULLs now sort after all normal values in ORDER BY (Tom)</para></listitem>
5218 <listitem><para>New IS UNKNOWN, IS NOT UNKNOWN Boolean tests (Tom)</para></listitem>
5219 <listitem><para>New SHARE UPDATE EXCLUSIVE lock mode (Tom)</para></listitem>
5220 <listitem><para>New EXPLAIN ANALYZE command that shows run times and row counts (Martijn van Oosterhout)</para></listitem>
5221 <listitem><para>Fix problem with LIMIT and subqueries (Tom)</para></listitem>
5222 <listitem><para>Fix for LIMIT, DISTINCT ON pushed into subqueries (Tom)</para></listitem>
5223 <listitem><para>Fix nested EXCEPT/INTERSECT (Tom)</para></listitem>
5228 <title>Schema Manipulation</title>
5230 <listitem><para>Fix SERIAL in temporary tables (Bruce)</para></listitem>
5231 <listitem><para>Allow temporary sequences (Bruce)</para></listitem>
5232 <listitem><para>Sequences now use int8 internally (Tom)</para></listitem>
5233 <listitem><para>New SERIAL8 creates int8 columns with sequences, default still SERIAL4 (Tom)</para></listitem>
5234 <listitem><para>Make OIDs optional using WITHOUT OIDS (Tom)</para></listitem>
5235 <listitem><para>Add %TYPE syntax to CREATE TYPE (Ian Lance Taylor)</para></listitem>
5236 <listitem><para>Add ALTER TABLE / DROP CONSTRAINT for CHECK constraints (Christopher Kings-Lynne)</para></listitem>
5237 <listitem><para>New CREATE OR REPLACE FUNCTION to alter existing function (preserving the function OID) (Gavin Sherry)</para></listitem>
5238 <listitem><para>Add ALTER TABLE / ADD [ UNIQUE | PRIMARY ] (Christopher Kings-Lynne)</para></listitem>
5239 <listitem><para>Allow column renaming in views</para></listitem>
5240 <listitem><para>Make ALTER TABLE / RENAME COLUMN update column names of indexes (Brent Verner)</para></listitem>
5241 <listitem><para>Fix for ALTER TABLE / ADD CONSTRAINT ... CHECK with inherited tables (Stephan Szabo)</para></listitem>
5242 <listitem><para>ALTER TABLE RENAME update foreign-key trigger arguments correctly (Brent Verner)</para></listitem>
5243 <listitem><para>DROP AGGREGATE and COMMENT ON AGGREGATE now accept an aggtype (Tom)</para></listitem>
5244 <listitem><para>Add automatic return type data casting for SQL functions (Tom)</para></listitem>
5245 <listitem><para>Allow GiST indexes to handle NULLs and multikey indexes (Oleg Bartunov, Teodor Sigaev, Tom)</para></listitem>
5246 <listitem><para>Enable partial indexes (Martijn van Oosterhout)</para></listitem>
5251 <title>Utility Commands</title>
5253 <listitem><para>Add RESET ALL, SHOW ALL (Marko Kreen)</para></listitem>
5254 <listitem><para>CREATE/ALTER USER/GROUP now allow options in any order (Vince)</para></listitem>
5255 <listitem><para>Add LOCK A, B, C functionality (Neil Padgett)</para></listitem>
5256 <listitem><para>New ENCRYPTED/UNENCRYPTED option to CREATE/ALTER USER (Bruce)</para></listitem>
5257 <listitem><para>New light-weight VACUUM does not lock table; old semantics are available as VACUUM FULL (Tom)</para></listitem>
5258 <listitem><para>Disable COPY TO/FROM on views (Bruce)</para></listitem>
5259 <listitem><para>COPY DELIMITERS string must be exactly one character (Tom)</para></listitem>
5260 <listitem><para>VACUUM warning about index tuples fewer than heap now only appears when appropriate (Martijn van Oosterhout)</para></listitem>
5261 <listitem><para>Fix privilege checks for CREATE INDEX (Tom)</para></listitem>
5262 <listitem><para>Disallow inappropriate use of CREATE/DROP INDEX/TRIGGER/VIEW (Tom)</para></listitem>
5267 <title>Data Types and Functions</title>
5269 <listitem><para>SUM(), AVG(), COUNT() now uses int8 internally for speed (Tom)</para></listitem>
5270 <listitem><para>Add convert(), convert2() (Tatsuo)</para></listitem>
5271 <listitem><para>New function bit_length() (Peter E)</para></listitem>
5272 <listitem><para>Make the "n" in CHAR(n)/VARCHAR(n) represents letters, not bytes (Tatsuo)</para></listitem>
5273 <listitem><para>CHAR(), VARCHAR() now reject strings that are too long (Peter E)</para></listitem>
5274 <listitem><para>BIT VARYING now rejects bit strings that are too long (Peter E)</para></listitem>
5275 <listitem><para>BIT now rejects bit strings that do not match declared size (Peter E)</para></listitem>
5276 <listitem><para>INET, CIDR text conversion functions (Alex Pilosov)</para></listitem>
5277 <listitem><para>INET, CIDR operators << and <<= indexable (Alex Pilosov)</para></listitem>
5278 <listitem><para>Bytea \### now requires valid three digit octal number</para></listitem>
5279 <listitem><para>Bytea comparison improvements, now supports =, <>, >, >=, <, and <=</para></listitem>
5280 <listitem><para>Bytea now supports B-tree indexes</para></listitem>
5281 <listitem><para>Bytea now supports LIKE, LIKE...ESCAPE, NOT LIKE, NOT LIKE...ESCAPE</para></listitem>
5282 <listitem><para>Bytea now supports concatenation</para></listitem>
5283 <listitem><para>New bytea functions: position, substring, trim, btrim, and length</para></listitem>
5284 <listitem><para>New encode() function mode, "escaped", converts minimally escaped bytea to/from text</para></listitem>
5285 <listitem><para>Add pg_database_encoding_max_length() (Tatsuo)</para></listitem>
5286 <listitem><para>Add pg_client_encoding() function (Tatsuo)</para></listitem>
5287 <listitem><para>now() returns time with millisecond precision (Thomas)</para></listitem>
5288 <listitem><para>New TIMESTAMP WITHOUT TIMEZONE data type (Thomas)</para></listitem>
5289 <listitem><para>Add ISO date/time specification with "T", yyyy-mm-ddThh:mm:ss (Thomas)</para></listitem>
5290 <listitem><para>New xid/int comparison functions (Hiroshi)</para></listitem>
5291 <listitem><para>Add precision to TIME, TIMESTAMP, and INTERVAL data types (Thomas)</para></listitem>
5292 <listitem><para>Modify type coercion logic to attempt binary-compatible functions first (Tom)</para></listitem>
5293 <listitem><para>New encode() function installed by default (Marko Kreen)</para></listitem>
5294 <listitem><para>Improved to_*() conversion functions (Karel Zak)</para></listitem>
5295 <listitem><para>Optimize LIKE/ILIKE when using single-byte encodings (Tatsuo)</para></listitem>
5296 <listitem><para>New functions in contrib/pgcrypto: crypt(), hmac(), encrypt(), gen_salt() (Marko Kreen)</para></listitem>
5297 <listitem><para>Correct description of translate() function (Bruce)</para></listitem>
5298 <listitem><para>Add INTERVAL argument for SET TIME ZONE (Thomas)</para></listitem>
5299 <listitem><para>Add INTERVAL YEAR TO MONTH (etc.) syntax (Thomas)</para></listitem>
5300 <listitem><para>Optimize length functions when using single-byte encodings (Tatsuo)</para></listitem>
5301 <listitem><para>Fix path_inter, path_distance, path_length, dist_ppath to handle closed paths (Curtis Barrett, Tom)</para></listitem>
5302 <listitem><para>octet_length(text) now returns non-compressed length (Tatsuo, Bruce)</para></listitem>
5303 <listitem><para>Handle "July" full name in date/time literals (Greg Sabino Mullane)</para></listitem>
5304 <listitem><para>Some datatype() function calls now evaluated differently</para></listitem>
5305 <listitem><para>Add support for Julian and ISO time specifications (Thomas)</para></listitem>
5310 <title>Internationalization</title>
5312 <listitem><para>National language support in psql, <application>pg_dump</>, libpq, and server (Peter E)</para></listitem>
5313 <listitem><para>Message translations in Chinese (simplified, traditional), Czech, French, German, Hungarian, Russian, Swedish (Peter E, Serguei A. Mokhov, Karel Zak, Weiping He, Zhenbang Wei, Kovacs Zoltan)</para></listitem>
5314 <listitem><para>Make trim, ltrim, rtrim, btrim, lpad, rpad, translate multibyte aware (Tatsuo)</para></listitem>
5315 <listitem><para>Add LATIN5,6,7,8,9,10 support (Tatsuo)</para></listitem>
5316 <listitem><para>Add ISO 8859-5,6,7,8 support (Tatsuo)</para></listitem>
5317 <listitem><para>Correct LATIN5 to mean ISO-8859-9, not ISO-8859-5 (Tatsuo)</para></listitem>
5318 <listitem><para>Make mic2ascii() non-ASCII aware (Tatsuo)</para></listitem>
5319 <listitem><para>Reject invalid multibyte character sequences (Tatsuo)</para></listitem>
5324 <title><application>PL/pgSQL</></title>
5326 <listitem><para>Now uses portals for SELECT loops, allowing huge result sets (Jan)</para></listitem>
5327 <listitem><para>CURSOR and REFCURSOR support (Jan)</para></listitem>
5328 <listitem><para>Can now return open cursors (Jan)</para></listitem>
5329 <listitem><para>Add ELSEIF (Klaus Reger)</para></listitem>
5330 <listitem><para>Improve PL/pgSQL error reporting, including location of error (Tom)</para></listitem>
5331 <listitem><para>Allow IS or FOR key words in cursor declaration, for compatibility (Bruce)</para></listitem>
5332 <listitem><para>Fix for SELECT ... FOR UPDATE (Tom)</para></listitem>
5333 <listitem><para>Fix for PERFORM returning multiple rows (Tom)</para></listitem>
5334 <listitem><para>Make PL/pgSQL use the server's type coercion code (Tom)</para></listitem>
5335 <listitem><para>Memory leak fix (Jan, Tom)</para></listitem>
5336 <listitem><para>Make trailing semicolon optional (Tom)</para></listitem>
5341 <title>PL/Perl</title>
5343 <listitem><para>New untrusted PL/Perl (Alex Pilosov)</para></listitem>
5344 <listitem><para>PL/Perl is now built on some platforms even if libperl is not shared (Peter E)</para></listitem>
5349 <title>PL/Tcl</title>
5351 <listitem><para>Now reports errorInfo (Vsevolod Lobko)</para></listitem>
5352 <listitem><para>Add spi_lastoid function (bob@redivi.com)</para></listitem>
5357 <title>PL/Python</title>
5359 <listitem><para>...is new (Andrew Bosma)</para></listitem>
5364 <title><application>psql</></title>
5366 <listitem><para>\d displays indexes in unique, primary groupings (Christopher Kings-Lynne)</para></listitem>
5367 <listitem><para>Allow trailing semicolons in backslash commands (Greg Sabino Mullane)</para></listitem>
5368 <listitem><para>Read password from /dev/tty if possible</para></listitem>
5369 <listitem><para>Force new password prompt when changing user and database (Tatsuo, Tom)</para></listitem>
5370 <listitem><para>Format the correct number of columns for Unicode (Patrice)</para></listitem>
5375 <title><application>libpq</></title>
5377 <listitem><para>New function PQescapeString() to escape quotes in command strings (Florian Weimer)</para></listitem>
5378 <listitem><para>New function PQescapeBytea() escapes binary strings for use as SQL string literals</para></listitem>
5385 <listitem><para>Return OID of INSERT (Ken K)</para></listitem>
5386 <listitem><para>Handle more data types (Ken K)</para></listitem>
5387 <listitem><para>Handle single quotes and newlines in strings (Ken K)</para></listitem>
5388 <listitem><para>Handle NULL variables (Ken K)</para></listitem>
5389 <listitem><para>Fix for time zone handling (Barry Lind)</para></listitem>
5390 <listitem><para>Improved Druid support</para></listitem>
5391 <listitem><para>Allow eight-bit characters with non-multibyte server (Barry Lind)</para></listitem>
5392 <listitem><para>Support BIT, BINARY types (Ned Wolpert)</para></listitem>
5393 <listitem><para>Reduce memory usage (Michael Stephens, Dave Cramer)</para></listitem>
5394 <listitem><para>Update DatabaseMetaData (Peter E)</para></listitem>
5395 <listitem><para>Add DatabaseMetaData.getCatalogs() (Peter E)</para></listitem>
5396 <listitem><para>Encoding fixes (Anders Bengtsson)</para></listitem>
5397 <listitem><para>Get/setCatalog methods (Jason Davies)</para></listitem>
5398 <listitem><para>DatabaseMetaData.getColumns() now returns column defaults (Jason Davies)</para></listitem>
5399 <listitem><para>DatabaseMetaData.getColumns() performance improvement (Jeroen van Vianen)</para></listitem>
5400 <listitem><para>Some JDBC1 and JDBC2 merging (Anders Bengtsson)</para></listitem>
5401 <listitem><para>Transaction performance improvements (Barry Lind)</para></listitem>
5402 <listitem><para>Array fixes (Greg Zoller)</para></listitem>
5403 <listitem><para>Serialize addition </para></listitem>
5404 <listitem><para>Fix batch processing (Rene Pijlman)</para></listitem>
5405 <listitem><para>ExecSQL method reorganization (Anders Bengtsson)</para></listitem>
5406 <listitem><para>GetColumn() fixes (Jeroen van Vianen)</para></listitem>
5407 <listitem><para>Fix isWriteable() function (Rene Pijlman)</para></listitem>
5408 <listitem><para>Improved passage of JDBC2 conformance tests (Rene Pijlman)</para></listitem>
5409 <listitem><para>Add bytea type capability (Barry Lind)</para></listitem>
5410 <listitem><para>Add isNullable() (Rene Pijlman)</para></listitem>
5411 <listitem><para>JDBC date/time test suite fixes (Liam Stewart)</para></listitem>
5412 <listitem><para>Fix for SELECT 'id' AS xxx FROM table (Dave Cramer)</para></listitem>
5413 <listitem><para>Fix DatabaseMetaData to show precision properly (Mark Lillywhite)</para></listitem>
5414 <listitem><para>New getImported/getExported keys (Jason Davies)</para></listitem>
5415 <listitem><para>MD5 password encryption support (Jeremy Wohl)</para></listitem>
5416 <listitem><para>Fix to actually use type cache (Ned Wolpert)</para></listitem>
5423 <listitem><para>Remove query size limit (Hiroshi)</para></listitem>
5424 <listitem><para>Remove text field size limit (Hiroshi)</para></listitem>
5425 <listitem><para>Fix for SQLPrimaryKeys in multibyte mode (Hiroshi)</para></listitem>
5426 <listitem><para>Allow ODBC procedure calls (Hiroshi)</para></listitem>
5427 <listitem><para>Improve boolean handing (Aidan Mountford)</para></listitem>
5428 <listitem><para>Most configuration options now settable via DSN (Hiroshi)</para></listitem>
5429 <listitem><para>Multibyte, performance fixes (Hiroshi)</para></listitem>
5430 <listitem><para>Allow driver to be used with iODBC or unixODBC (Peter E)</para></listitem>
5431 <listitem><para>MD5 password encryption support (Bruce)</para></listitem>
5432 <listitem><para>Add more compatibility functions to odbc.sql (Peter E)</para></listitem>
5437 <title><application>ECPG</></title>
5439 <listitem><para>EXECUTE ... INTO implemented (Christof Petig)</para></listitem>
5440 <listitem><para>Multiple row descriptor support (e.g. CARDINALITY) (Christof Petig)</para></listitem>
5441 <listitem><para>Fix for GRANT parameters (Lee Kindness)</para></listitem>
5442 <listitem><para>Fix INITIALLY DEFERRED bug</para></listitem>
5443 <listitem><para>Various bug fixes (Michael, Christof Petig)</para></listitem>
5444 <listitem><para>Auto allocation for indicator variable arrays (int *ind_p=NULL)</para></listitem>
5445 <listitem><para>Auto allocation for string arrays (char **foo_pp=NULL)</para></listitem>
5446 <listitem><para>ECPGfree_auto_mem fixed</para></listitem>
5447 <listitem><para>All function names with external linkage are now prefixed by ECPG</para></listitem>
5448 <listitem><para>Fixes for arrays of structures (Michael)</para></listitem>
5453 <title>Misc. Interfaces</title>
5455 <listitem><para>Python fix fetchone() (Gerhard Haring)</para></listitem>
5456 <listitem><para>Use UTF, Unicode in Tcl where appropriate (Vsevolod Lobko, Reinhard Max)</para></listitem>
5457 <listitem><para>Add Tcl COPY TO/FROM (ljb)</para></listitem>
5458 <listitem><para>Prevent output of default index op class in <application>pg_dump</> (Tom)</para></listitem>
5459 <listitem><para>Fix libpgeasy memory leak (Bruce)</para></listitem>
5464 <title>Build and Install</title>
5466 <listitem><para>Configure, dynamic loader, and shared library fixes (Peter E)</para></listitem>
5467 <listitem><para>Fixes in QNX 4 port (Bernd Tegge)</para></listitem>
5468 <listitem><para>Fixes in Cygwin and Windows ports (Jason Tishler, Gerhard Haring, Dmitry Yurtaev, Darko Prenosil, Mikhail Terekhov)</para></listitem>
5469 <listitem><para>Fix for Windows socket communication failures (Magnus, Mikhail Terekhov)</para></listitem>
5470 <listitem><para>Hurd compile fix (Oliver Elphick)</para></listitem>
5471 <listitem><para>BeOS fixes (Cyril Velter)</para></listitem>
5472 <listitem><para>Remove configure --enable-unicode-conversion, now enabled by multibyte (Tatsuo)</para></listitem>
5473 <listitem><para>AIX fixes (Tatsuo, Andreas)</para></listitem>
5474 <listitem><para>Fix parallel make (Peter E)</para></listitem>
5475 <listitem><para>Install SQL language manual pages into OS-specific directories (Peter E)</para></listitem>
5476 <listitem><para>Rename config.h to pg_config.h (Peter E)</para></listitem>
5477 <listitem><para>Reorganize installation layout of header files (Peter E)</para></listitem>
5482 <title>Source Code</title>
5484 <listitem><para>Remove SEP_CHAR (Bruce)</para></listitem>
5485 <listitem><para>New GUC hooks (Tom)</para></listitem>
5486 <listitem><para>Merge GUC and command line handling (Marko Kreen)</para></listitem>
5487 <listitem><para>Remove EXTEND INDEX (Martijn van Oosterhout, Tom)</para></listitem>
5488 <listitem><para>New pgjindent utility to indent java code (Bruce)</para></listitem>
5489 <listitem><para>Remove define of true/false when compiling under C++ (Leandro Fanzone, Tom)</para></listitem>
5490 <listitem><para>pgindent fixes (Bruce, Tom)</para></listitem>
5491 <listitem><para>Replace strcasecmp() with strcmp() where appropriate (Peter E)</para></listitem>
5492 <listitem><para>Dynahash portability improvements (Tom)</para></listitem>
5493 <listitem><para>Add 'volatile' usage in spinlock structures</para></listitem>
5494 <listitem><para>Improve signal handling logic (Tom)</para></listitem>
5499 <title>Contrib</title>
5501 <listitem><para>New contrib/rtree_gist (Oleg Bartunov, Teodor Sigaev)</para></listitem>
5502 <listitem><para>New contrib/tsearch full-text indexing (Oleg, Teodor Sigaev)</para></listitem>
5503 <listitem><para>Add contrib/dblink for remote database access (Joe Conway)</para></listitem>
5504 <listitem><para>contrib/ora2pg Oracle conversion utility (Gilles Darold)</para></listitem>
5505 <listitem><para>contrib/xml XML conversion utility (John Gray)</para></listitem>
5506 <listitem><para>contrib/fulltextindex fixes (Christopher Kings-Lynne)</para></listitem>
5507 <listitem><para>New contrib/fuzzystrmatch with levenshtein and metaphone, soundex merged (Joe Conway)</para></listitem>
5508 <listitem><para>Add contrib/intarray boolean queries, binary search, fixes (Oleg Bartunov)</para></listitem>
5509 <listitem><para>New pg_upgrade utility (Bruce)</para></listitem>
5510 <listitem><para>Add new pg_resetxlog options (Bruce, Tom)</para></listitem>
5517 <sect1 id="release-7-1-3">
5518 <title>Release 7.1.3</title>
5521 <title>Release date</title>
5522 <simpara>2001-08-15</simpara>
5526 <title>Migration to version 7.1.3</title>
5529 A dump/restore is <emphasis>not</emphasis> required for those running
5535 <title>Changes</title>
5539 Remove unused WAL segements of large transactions (Tom)
5540 Multiaction rule fix (Tom)
5541 PL/pgSQL memory allocation fix (Jan)
5542 VACUUM buffer fix (Tom)
5543 Regression test fixes (Tom)
5544 pg_dump fixes for GRANT/REVOKE/comments on views, user-defined types (Tom)
5545 Fix subselects with DISTINCT ON or LIMIT (Tom)
5547 Disable COPY TO/FROM a view (Tom)
5548 Cygwin build (Jason Tishler)
5555 <sect1 id="release-7-1-2">
5556 <title>Release 7.1.2</title>
5559 <title>Release date</title>
5560 <simpara>2001-05-11</simpara>
5564 This has one fix from 7.1.1.
5569 <title>Migration to version 7.1.2</title>
5572 A dump/restore is <emphasis>not</emphasis> required for those running
5578 <title>Changes</title>
5582 Fix PL/pgSQL SELECTs when returning no rows
5583 Fix for psql backslash core dump
5584 Referential integrity privilege fix
5593 <sect1 id="release-7-1-1">
5594 <title>Release 7.1.1</title>
5597 <title>Release date</title>
5598 <simpara>2001-05-05</simpara>
5602 This has a variety of fixes from 7.1.
5607 <title>Migration to version 7.1.1</title>
5610 A dump/restore is <emphasis>not</emphasis> required for those running
5616 <title>Changes</title>
5620 Fix for numeric MODULO operator (Tom)
5621 pg_dump fixes (Philip)
5622 pg_dump can dump 7.0 databases (Philip)
5623 readline 4.2 fixes (Peter E)
5625 AIX, MSWIN, VAX, N32K fixes (Tom)
5626 Multibytes fixes (Tom)
5627 Unicode fixes (Tatsuo)
5628 Optimizer improvements (Tom)
5629 Fix for whole rows in functions (Tom)
5630 Fix for pg_ctl and option strings with spaces (Peter E)
5631 ODBC fixes (Hiroshi)
5632 EXTRACT can now take string argument (Thomas)
5633 Python fixes (Darcy)
5640 <sect1 id="release-7-1">
5641 <title>Release 7.1</title>
5644 <title>Release date</title>
5645 <simpara>2001-04-13</simpara>
5649 This release focuses on removing limitations that have existed in the
5650 <productname>PostgreSQL</productname> code for many years.
5654 Major changes in this release:
5659 Write-ahead Log (WAL)
5663 To maintain database consistency in case of an operating system crash,
5664 previous releases of <productname>PostgreSQL</productname> have forced
5665 all data modifications to disk before each transaction commit. With
5666 WAL, only one log file must be flushed to disk, greatly improving
5667 performance. If you have been using -F in previous releases to
5668 disable disk flushes, you may want to consider discontinuing its use.
5679 TOAST - Previous releases had a compiled-in row length limit,
5680 typically 8k - 32k. This limit made storage of long text fields
5681 difficult. With TOAST, long rows of any length can be stored with good
5693 We now support outer joins. The UNION/NOT IN
5694 workaround for outer joins is no longer required. We use the SQL92
5706 The previous C function manager did not
5707 handle null values properly, nor did it support 64-bit <acronym>CPU</acronym>'s (Alpha). The new
5708 function manager does. You can continue using your old custom
5709 functions, but you may want to rewrite them in the future to use the new
5710 function manager call interface.
5721 A large number of complex queries that were
5722 unsupported in previous releases now work. Many combinations of views,
5723 aggregates, UNION, LIMIT, cursors, subqueries, and inherited tables
5724 now work properly. Inherited tables are now accessed by default.
5725 Subqueries in FROM are now supported.
5734 <title>Migration to version 7.1</title>
5737 A dump/restore using pg_dump is required for those wishing to migrate
5738 data from any previous release.
5743 <title>Changes</title>
5749 Many multibyte/Unicode/locale fixes (Tatsuo and others)
5750 More reliable ALTER TABLE RENAME (Tom)
5751 Kerberos V fixes (David Wragg)
5752 Fix for INSERT INTO...SELECT where targetlist has subqueries (Tom)
5753 Prompt username/password on standard error (Bruce)
5754 Large objects inv_read/inv_write fixes (Tom)
5755 Fixes for to_char(), to_date(), to_ascii(), and to_timestamp() (Karel,
5757 Prevent query expressions from leaking memory (Tom)
5758 Allow UPDATE of arrays elements (Tom)
5759 Wake up lock waiters during cancel (Hiroshi)
5760 Fix rare cursor crash when using hash join (Tom)
5761 Fix for DROP TABLE/INDEX in rolled-back transaction (Hiroshi)
5762 Fix psql crash from \l+ if MULTIBYTE enabled (Peter E)
5763 Fix truncation of rule names during CREATE VIEW (Ross Reedstrom)
5764 Fix PL/perl (Alex Kapranoff)
5765 Disallow LOCK on views (Mark Hollomon)
5766 Disallow INSERT/UPDATE/DELETE on views (Mark Hollomon)
5767 Disallow DROP RULE, CREATE INDEX, TRUNCATE on views (Mark Hollomon)
5768 Allow PL/pgSQL accept non-ASCII identifiers (Tatsuo)
5769 Allow views to proper handle GROUP BY, aggregates, DISTINCT (Tom)
5770 Fix rare failure with TRUNCATE command (Tom)
5771 Allow UNION/INTERSECT/EXCEPT to be used with ALL, subqueries, views,
5772 DISTINCT, ORDER BY, SELECT...INTO (Tom)
5773 Fix parser failures during aborted transactions (Tom)
5774 Allow temporary relations to properly clean up indexes (Bruce)
5775 Fix VACUUM problem with moving rows in same page (Tom)
5776 Modify pg_dump to better handle user-defined items in template1 (Philip)
5777 Allow LIMIT in VIEW (Tom)
5778 Require cursor FETCH to honor LIMIT (Tom)
5779 Allow PRIMARY/FOREIGN Key definitions on inherited columns (Stephan)
5780 Allow ORDER BY, LIMIT in subqueries (Tom)
5781 Allow UNION in CREATE RULE (Tom)
5782 Make ALTER/DROP TABLE rollback-able (Vadim, Tom)
5783 Store initdb collation in pg_control so collation cannot be changed (Tom)
5784 Fix INSERT...SELECT with rules (Tom)
5785 Fix FOR UPDATE inside views and subselects (Tom)
5786 Fix OVERLAPS operators conform to SQL92 spec regarding NULLs (Tom)
5787 Fix lpad() and rpad() to handle length less than input string (Tom)
5788 Fix use of NOTIFY in some rules (Tom)
5789 Overhaul btree code (Tom)
5790 Fix NOT NULL use in Pl/pgSQL variables (Tom)
5791 Overhaul GIST code (Oleg)
5792 Fix CLUSTER to preserve constraints and column default (Tom)
5793 Improved deadlock detection handling (Tom)
5794 Allow multiple SERIAL columns in a table (Tom)
5795 Prevent occasional index corruption (Vadim)
5799 Add OUTER JOINs (Tom)
5800 Function manager overhaul (Tom)
5801 Allow ALTER TABLE RENAME on indexes (Tom)
5802 Improve CLUSTER (Tom)
5803 Improve ps status display for more platforms (Peter E, Marc)
5804 Improve CREATE FUNCTION failure message (Ross)
5805 JDBC improvements (Peter, Travis Bauer, Christopher Cain, William Webber,
5807 Grand Unified Configuration scheme/GUC. Many options can now be set in
5808 data/postgresql.conf, postmaster/postgres flags, or SET commands (Peter E)
5809 Improved handling of file descriptor cache (Tom)
5810 New warning code about auto-created table alias entries (Bruce)
5811 Overhaul initdb process (Tom, Peter E)
5812 Overhaul of inherited tables; inherited tables now accessed by default;
5813 new ONLY key word prevents it (Chris Bitmead, Tom)
5814 ODBC cleanups/improvements (Nick Gorham, Stephan Szabo, Zoltan Kovacs,
5816 Allow renaming of temp tables (Tom)
5817 Overhaul memory manager contexts (Tom)
5818 pg_dumpall uses CREATE USER or CREATE GROUP rather using COPY (Peter E)
5819 Overhaul pg_dump (Philip Warner)
5820 Allow pg_hba.conf secondary password file to specify only username (Peter E)
5821 Allow TEMPORARY or TEMP key word when creating temporary tables (Bruce)
5822 New memory leak checker (Karel)
5823 New SET SESSION CHARACTERISTICS (Thomas)
5824 Allow nested block comments (Thomas)
5825 Add WITHOUT TIME ZONE type qualifier (Thomas)
5826 New ALTER TABLE ADD CONSTRAINT (Stephan)
5827 Use NUMERIC accumulators for INTEGER aggregates (Tom)
5828 Overhaul aggregate code (Tom)
5829 New VARIANCE and STDDEV() aggregates
5830 Improve dependency ordering of pg_dump (Philip)
5831 New pg_restore command (Philip)
5832 New pg_dump tar output option (Philip)
5833 New pg_dump of large objects (Philip)
5834 New ESCAPE option to LIKE (Thomas)
5835 New case-insensitive LIKE - ILIKE (Thomas)
5836 Allow functional indexes to use binary-compatible type (Tom)
5837 Allow SQL functions to be used in more contexts (Tom)
5838 New pg_config utility (Peter E)
5839 New PL/pgSQL EXECUTE command which allows dynamic SQL and utility statements
5841 New PL/pgSQL GET DIAGNOSTICS statement for SPI value access (Jan)
5842 New quote_identifiers() and quote_literal() functions (Jan)
5843 New ALTER TABLE table OWNER TO user command (Mark Hollomon)
5844 Allow subselects in FROM, i.e. FROM (SELECT ...) [AS] alias (Tom)
5845 Update PyGreSQL to version 3.1 (D'Arcy)
5846 Store tables as files named by OID (Vadim)
5847 New SQL function setval(seq,val,bool) for use in pg_dump (Philip)
5848 Require DROP VIEW to remove views, no DROP TABLE (Mark)
5849 Allow DROP VIEW view1, view2 (Mark)
5850 Allow multiple objects in DROP INDEX, DROP RULE, and DROP TYPE (Tom)
5851 Allow automatic conversion to/from Unicode (Tatsuo, Eiji)
5852 New /contrib/pgcrypto hashing functions (Marko Kreen)
5853 New pg_dumpall --globals-only option (Peter E)
5854 New CHECKPOINT command for WAL which creates new WAL log file (Vadim)
5855 New AT TIME ZONE syntax (Thomas)
5856 Allow location of Unix domain socket to be configurable (David J. MacKenzie)
5857 Allow postmaster to listen on a specific IP address (David J. MacKenzie)
5858 Allow socket path name to be specified in hostname by using leading slash
5859 (David J. MacKenzie)
5860 Allow CREATE DATABASE to specify template database (Tom)
5861 New utility to convert MySQL schema dumps to SQL92 and PostgreSQL (Thomas)
5862 New /contrib/rserv replication toolkit (Vadim)
5863 New file format for COPY BINARY (Tom)
5864 New /contrib/oid2name to map numeric files to table names (B Palmer)
5865 New "idle in transaction" ps status message (Marc)
5866 Update to pgaccess 0.98.7 (Constantin Teodorescu)
5867 pg_ctl now defaults to -w (wait) on shutdown, new -l (log) option
5868 Add rudimentary dependency checking to pg_dump (Philip)
5872 Fix INET/CIDR type ordering and add new functions (Tom)
5873 Make OID behave as an unsigned type (Tom)
5874 Allow BIGINT as synonym for INT8 (Peter E)
5875 New int2 and int8 comparison operators (Tom)
5876 New BIT and BIT VARYING types (Adriaan Joubert, Tom, Peter E)
5877 CHAR() no longer faster than VARCHAR() because of TOAST (Tom)
5878 New GIST seg/cube examples (Gene Selkov)
5879 Improved round(numeric) handling (Tom)
5880 Fix CIDR output formatting (Tom)
5881 New CIDR abbrev() function (Tom)
5885 Write-Ahead Log (WAL) to provide crash recovery with less performance
5887 ANALYZE stage of VACUUM no longer exclusively locks table (Bruce)
5888 Reduced file seeks (Denis Perchine)
5889 Improve BTREE code for duplicate keys (Tom)
5890 Store all large objects in a single table (Denis Perchine, Tom)
5891 Improve memory allocation performance (Karel, Tom)
5895 New function manager call conventions (Tom)
5896 SGI portability fixes (David Kaelbling)
5897 New configure --enable-syslog option (Peter E)
5898 New BSDI README (Bruce)
5899 configure script moved to top level, not /src (Peter E)
5900 Makefile/configuration/compilation overhaul (Peter E)
5901 New configure --with-python option (Peter E)
5902 Solaris cleanups (Peter E)
5903 Overhaul /contrib Makefiles (Karel)
5904 New OpenSSL configuration option (Magnus, Peter E)
5906 QNX fixes (Maurizio)
5907 New heap_open(), heap_openr() API (Tom)
5908 Remove colon and semi-colon operators (Thomas)
5909 New pg_class.relkind value for views (Mark Hollomon)
5910 Rename ichar() to chr() (Karel)
5911 New documentation for btrim(), ascii(), chr(), repeat() (Karel)
5912 Fixes for NT/Cygwin (Pete Forman)
5913 AIX port fixes (Andreas)
5914 New BeOS port (David Reid, Cyril Velter)
5915 Add proofreader's changes to docs (Addison-Wesley, Bruce)
5916 New Alpha spinlock code (Adriaan Joubert, Compaq)
5917 UnixWare port overhaul (Peter E)
5918 New Darwin/MacOS X port (Peter Bierman, Bruce Hartzler)
5919 New FreeBSD Alpha port (Alfred)
5920 Overhaul shared memory segments (Tom)
5921 Add IBM S/390 support (Neale Ferguson)
5922 Moved macmanuf to /contrib (Larry Rosenman)
5923 Syslog improvements (Larry Rosenman)
5924 New template0 database that contains no user additions (Tom)
5925 New /contrib/cube and /contrib/seg GIST sample code (Gene Selkov)
5926 Allow NetBSD's libedit instead of readline (Peter)
5927 Improved assembly language source code format (Bruce)
5928 New contrib/pg_logger
5929 New --template option to createdb
5930 New contrib/pg_control utility (Oliver)
5931 New FreeBSD tools ipc_check, start-scripts/freebsd
5938 <sect1 id="release-7-0-3">
5939 <title>Release 7.0.3</title>
5942 <title>Release date</title>
5943 <simpara>2000-11-11</simpara>
5947 This has a variety of fixes from 7.0.2.
5952 <title>Migration to version 7.0.3</title>
5955 A dump/restore is <emphasis>not</emphasis> required for those running
5961 <title>Changes</title>
5966 Large object fix (Tom)
5967 Fix lean in COPY WITH OIDS leak (Tom)
5968 Fix backwards-index-scan (Tom)
5969 Fix SELECT ... FOR UPDATE so it checks for duplicate keys (Hiroshi)
5970 Add --enable-syslog to configure (Marc)
5971 Fix abort transaction at backend exit in rare cases (Tom)
5972 Fix for psql \l+ when multibyte enabled (Tatsuo)
5973 Allow PL/pgSQL to accept non ascii identifiers (Tatsuo)
5974 Make vacuum always flush buffers (Tom)
5975 Fix to allow cancel while waiting for a lock (Hiroshi)
5976 Fix for memory aloocation problem in user authentication code (Tom)
5977 Remove bogus use of int4out() (Tom)
5978 Fixes for multiple subqueries in COALESCE or BETWEEN (Tom)
5979 Fix for failure of triggers on heap open in certain cases (Jeroen van
5981 Fix for erroneous selectivity of not-equals (Tom)
5982 Fix for erroneous use of strcmp() (Tom)
5983 Fix for bug where storage manager accesses items beyond end of file
5985 Fix to include kernel errno message in all smgr elog messages (Tom)
5986 Fix for '.' not in PATH at build time (SL Baur)
5987 Fix for out-of-file-descriptors error (Tom)
5988 Fix to make pg_dump dump 'iscachable' flag for functions (Tom)
5989 Fix for subselect in targetlist of Append node (Tom)
5990 Fix for mergejoin plans (Tom)
5991 Fix TRUNCATE failure on relations with indexes (Tom)
5992 Avoid database-wide restart on write error (Hiroshi)
5993 Fix nodeMaterial to honor chgParam by recomputing its output (Tom)
5994 Fix VACUUM problem with moving chain of update row versions when source
5995 and destination of a row version lie on the same page (Tom)
5996 Fix user.c CommandCounterIncrement (Tom)
5997 Fix for AM/PM boundary problem in to_char() (Karel Zak)
5998 Fix TIME aggregate handling (Tom)
5999 Fix to_char() to avoid coredump on NULL input (Tom)
6001 Fix for inserting/copying longer multibyte strings into char() data
6003 Fix for crash of backend, on abort (Tom)
6010 <sect1 id="release-7-0-2">
6011 <title>Release 7.0.2</title>
6014 <title>Release date</title>
6015 <simpara>2000-06-05</simpara>
6019 This is a repackaging of 7.0.1 with added documentation.
6024 <title>Migration to version 7.0.2</title>
6027 A dump/restore is <emphasis>not</emphasis> required for those running
6033 <title>Changes</title>
6037 Added documentation to tarball.
6044 <sect1 id="release-7-0-1">
6045 <title>Release 7.0.1</title>
6048 <title>Release date</title>
6049 <simpara>2000-06-01</simpara>
6053 This is a cleanup release for 7.0.
6057 <title>Migration to version 7.0.1</title>
6060 A dump/restore is <emphasis>not</emphasis> required for those running
6066 <title>Changes</title>
6070 Fix many CLUSTER failures (Tom)
6071 Allow ALTER TABLE RENAME works on indexes (Tom)
6072 Fix plpgsql to handle datetime->timestamp and timespan->interval (Bruce)
6073 New configure --with-setproctitle switch to use setproctitle() (Marc, Bruce)
6074 Fix the off by one errors in ResultSet from 6.5.3, and more.
6075 jdbc ResultSet fixes (Joseph Shraibman)
6076 optimizer tunings (Tom)
6077 Fix create user for pgaccess
6078 Fix for UNLISTEN failure
6079 IRIX fixes (David Kaelbling)
6080 QNX fixes (Andreas Kardos)
6081 Reduce COPY IN lock level (Tom)
6082 Change libpqeasy to use PQconnectdb() style parameters (Bruce)
6083 Fix pg_dump to handle OID indexes (Tom)
6084 Fix small memory leak (Tom)
6085 Solaris fix for createdb/dropdb (Tatsuo)
6086 Fix for non-blocking connections (Alfred Perlstein)
6087 Fix improper recovery after RENAME TABLE failures (Tom)
6088 Copy pg_ident.conf.sample into /lib directory in install (Bruce)
6089 Add SJIS UDC (NEC selection IBM kanji) support (Eiji Tokuya)
6090 Fix too long syslog message (Tatsuo)
6091 Fix problem with quoted indexes that are too long (Tom)
6092 JDBC ResultSet.getTimestamp() fix (Gregory Krasnow & Floyd Marinescu)
6093 ecpg changes (Michael)
6099 <sect1 id="release-7-0">
6100 <title>Release 7.0</title>
6103 <title>Release date</title>
6104 <simpara>2000-05-08</simpara>
6108 This release contains improvements in many areas, demonstrating
6109 the continued growth of <productname>PostgreSQL</productname>.
6110 There are more improvements and fixes in 7.0 than in any previous
6111 release. The developers have confidence that this is the best
6112 release yet; we do our best to put out only solid releases, and
6113 this one is no exception.
6117 Major changes in this release:
6126 Foreign keys are now implemented, with the exception of PARTIAL MATCH
6127 foreign keys. Many users have been asking for this feature, and we are
6128 pleased to offer it.
6139 Continuing on work started a year ago, the optimizer has been
6140 improved, allowing better query plan selection and faster performance
6141 with less memory usage.
6148 Updated <application>psql</application>
6152 <application>psql</application>, our interactive terminal monitor, has been
6153 updated with a variety of new features. See the <application>psql</application> manual page for details.
6164 SQL92 join syntax is now supported, though only as
6165 <literal>INNER JOIN</> for this release. <literal>JOIN</>,
6166 <literal>NATURAL JOIN</>, <literal>JOIN</>/<literal>USING</>,
6167 and <literal>JOIN</>/<literal>ON</> are available, as are
6168 column correlation names.
6176 <title>Migration to version 7.0</title>
6179 A dump/restore using <application>pg_dump</application>
6180 is required for those wishing to migrate data from any
6181 previous release of <productname>PostgreSQL</productname>.
6182 For those upgrading from 6.5.*, you may instead use
6183 <application>pg_upgrade</application> to upgrade to this
6184 release; however, a full dump/reload installation is always the
6185 most robust method for upgrades.
6189 Interface and compatibility issues to consider for the new
6195 The date/time types <type>datetime</type> and
6196 <type>timespan</type> have been superseded by the
6197 SQL92-defined types <type>timestamp</type> and
6198 <type>interval</type>. Although there has been some effort to
6199 ease the transition by allowing
6200 <productname>PostgreSQL</productname> to recognize
6201 the deprecated type names and translate them to the new type
6202 names, this mechanism may not be completely transparent to
6203 your existing application.
6209 The optimizer has been substantially improved in the area of
6210 query cost estimation. In some cases, this will result in
6211 decreased query times as the optimizer makes a better choice
6212 for the preferred plan. However, in a small number of cases,
6213 usually involving pathological distributions of data, your
6214 query times may go up. If you are dealing with large amounts
6215 of data, you may want to check your queries to verify
6222 The <acronym>JDBC</acronym> and <acronym>ODBC</acronym>
6223 interfaces have been upgraded and extended.
6229 The string function <function>CHAR_LENGTH</function> is now a
6230 native function. Previous versions translated this into a call
6231 to <function>LENGTH</function>, which could result in
6232 ambiguity with other types implementing
6233 <function>LENGTH</function> such as the geometric types.
6242 <title>Changes</title>
6248 Prevent function calls exceeding maximum number of arguments (Tom)
6249 Improve CASE construct (Tom)
6250 Fix SELECT coalesce(f1,0) FROM int4_tbl GROUP BY f1 (Tom)
6251 Fix SELECT sentence.words[0] FROM sentence GROUP BY sentence.words[0] (Tom)
6252 Fix GROUP BY scan bug (Tom)
6253 Improvements in SQL grammar processing (Tom)
6254 Fix for views involved in INSERT ... SELECT ... (Tom)
6255 Fix for SELECT a/2, a/2 FROM test_missing_target GROUP BY a/2 (Tom)
6256 Fix for subselects in INSERT ... SELECT (Tom)
6257 Prevent INSERT ... SELECT ... ORDER BY (Tom)
6258 Fixes for relations greater than 2GB, including vacuum
6259 Improve propagating system table changes to other backends (Tom)
6260 Improve propagating user table changes to other backends (Tom)
6261 Fix handling of temp tables in complex situations (Bruce, Tom)
6262 Allow table locking at table open, improving concurrent reliability (Tom)
6263 Properly quote sequence names in pg_dump (Ross J. Reedstrom)
6264 Prevent DROP DATABASE while others accessing
6265 Prevent any rows from being returned by GROUP BY if no rows processed (Tom)
6266 Fix SELECT COUNT(1) FROM table WHERE ...' if no rows matching WHERE (Tom)
6267 Fix pg_upgrade so it works for MVCC (Tom)
6268 Fix for SELECT ... WHERE x IN (SELECT ... HAVING SUM(x) > 1) (Tom)
6269 Fix for "f1 datetime DEFAULT 'now'" (Tom)
6270 Fix problems with CURRENT_DATE used in DEFAULT (Tom)
6271 Allow comment-only lines, and ;;; lines too. (Tom)
6272 Improve recovery after failed disk writes, disk full (Hiroshi)
6273 Fix cases where table is mentioned in FROM but not joined (Tom)
6274 Allow HAVING clause without aggregate functions (Tom)
6275 Fix for "--" comment and no trailing newline, as seen in perl interface
6276 Improve pg_dump failure error reports (Bruce)
6277 Allow sorts and hashes to exceed 2GB file sizes (Tom)
6278 Fix for pg_dump dumping of inherited rules (Tom)
6279 Fix for NULL handling comparisons (Tom)
6280 Fix inconsistent state caused by failed CREATE/DROP commands (Hiroshi)
6281 Fix for dbname with dash
6282 Prevent DROP INDEX from interfering with other backends (Tom)
6283 Fix file descriptor leak in verify_password()
6284 Fix for "Unable to identify an operator =$" problem
6285 Fix ODBC so no segfault if CommLog and Debug enabled (Dirk Niggemann)
6286 Fix for recursive exit call (Massimo)
6287 Fix for extra-long timezones (Jeroen van Vianen)
6288 Make pg_dump preserve primary key information (Peter E)
6289 Prevent databases with single quotes (Peter E)
6290 Prevent DROP DATABASE inside transaction (Peter E)
6291 ecpg memory leak fixes (Stephen Birch)
6292 Fix for SELECT null::text, SELECT int4fac(null) and SELECT 2 + (null) (Tom)
6293 Y2K timestamp fix (Massimo)
6294 Fix for VACUUM 'HEAP_MOVED_IN was not expected' errors (Tom)
6295 Fix for views with tables/columns containing spaces (Tom)
6296 Prevent privileges on indexes (Peter E)
6297 Fix for spinlock stuck problem when error is generated (Hiroshi)
6298 Fix ipcclean on Linux
6299 Fix handling of NULL constraint conditions (Tom)
6300 Fix memory leak in odbc driver (Nick Gorham)
6301 Fix for privilege check on UNION tables (Tom)
6302 Fix to allow SELECT 'a' LIKE 'a' (Tom)
6303 Fix for SELECT 1 + NULL (Tom)
6305 Fix log() on numeric type (Tom)
6306 Deprecate ':' and ';' operators
6307 Allow vacuum of temporary tables
6308 Disallow inherited columns with the same name as new columns
6309 Recover or force failure when disk space is exhausted (Hiroshi)
6310 Fix INSERT INTO ... SELECT with AS columns matching result columns
6311 Fix INSERT ... SELECT ... GROUP BY groups by target columns not source columns (Tom)
6312 Fix CREATE TABLE test (a char(5) DEFAULT text '', b int4) with INSERT (Tom)
6313 Fix UNION with LIMIT
6314 Fix CREATE TABLE x AS SELECT 1 UNION SELECT 2
6315 Fix CREATE TABLE test(col char(2) DEFAULT user)
6316 Fix mismatched types in CREATE TABLE ... DEFAULT
6317 Fix SELECT * FROM pg_class where oid in (0,-1)
6318 Fix SELECT COUNT('asdf') FROM pg_class WHERE oid=12
6319 Prevent user who can create databases can modifying pg_database table (Peter E)
6320 Fix btree to give a useful elog when key > 1/2 (page - overhead) (Tom)
6321 Fix INSERT of 0.0 into DECIMAL(4,4) field (Tom)
6325 New CLI interface include file sqlcli.h, based on SQL3/SQL98
6326 Remove all limits on query length, row length limit still exists (Tom)
6327 Update jdbc protocol to 2.0 (Jens Glaser <email>jens@jens.de</email>)
6328 Add TRUNCATE command to quickly truncate relation (Mike Mascari)
6329 Fix to give super user and createdb user proper update catalog rights (Peter E)
6330 Allow ecpg bool variables to have NULL values (Christof)
6331 Issue ecpg error if NULL value for variable with no NULL indicator (Christof)
6332 Allow ^C to cancel COPY command (Massimo)
6333 Add SET FSYNC and SHOW PG_OPTIONS commands(Massimo)
6334 Function name overloading for dynamically-loaded C functions (Frankpitt)
6335 Add CmdTuples() to libpq++(Vince)
6336 New CREATE CONSTRAINT TRIGGER and SET CONSTRAINTS commands(Jan)
6337 Allow CREATE FUNCTION/WITH clause to be used for all language types
6338 configure --enable-debug adds -g (Peter E)
6339 configure --disable-debug removes -g (Peter E)
6340 Allow more complex default expressions (Tom)
6341 First real FOREIGN KEY constraint trigger functionality (Jan)
6342 Add FOREIGN KEY ... MATCH FULL ... ON DELETE CASCADE (Jan)
6343 Add FOREIGN KEY ... MATCH <unspecified> referential actions (Don Baccus)
6344 Allow WHERE restriction on ctid (physical heap location) (Hiroshi)
6345 Move pginterface from contrib to interface directory, rename to pgeasy (Bruce)
6346 Change pgeasy connectdb() parameter ordering (Bruce)
6347 Require SELECT DISTINCT target list to have all ORDER BY columns (Tom)
6348 Add Oracle's COMMENT ON command (Mike Mascari <email>mascarim@yahoo.com</email>)
6349 libpq's PQsetNoticeProcessor function now returns previous hook(Peter E)
6350 Prevent PQsetNoticeProcessor from being set to NULL (Peter E)
6351 Make USING in COPY optional (Bruce)
6352 Allow subselects in the target list (Tom)
6353 Allow subselects on the left side of comparison operators (Tom)
6354 New parallel regression test (Jan)
6355 Change backend-side COPY to write files with permissions 644 not 666 (Tom)
6356 Force permissions on PGDATA directory to be secure, even if it exists (Tom)
6357 Added psql LASTOID variable to return last inserted oid (Peter E)
6358 Allow concurrent vacuum and remove pg_vlock vacuum lock file (Tom)
6359 Add privilege check for vacuum (Peter E)
6360 New libpq functions to allow asynchronous connections: PQconnectStart(),
6361 PQconnectPoll(), PQresetStart(), PQresetPoll(), PQsetenvStart(),
6362 PQsetenvPoll(), PQsetenvAbort (Ewan Mellor)
6363 New libpq PQsetenv() function (Ewan Mellor)
6364 create/alter user extension (Peter E)
6365 New postmaster.pid and postmaster.opts under $PGDATA (Tatsuo)
6366 New scripts for create/drop user/db (Peter E)
6367 Major psql overhaul (Peter E)
6368 Add const to libpq interface (Peter E)
6369 New libpq function PQoidValue (Peter E)
6370 Show specific non-aggregate causing problem with GROUP BY (Tom)
6371 Make changes to pg_shadow recreate pg_pwd file (Peter E)
6372 Add aggregate(DISTINCT ...) (Tom)
6373 Allow flag to control COPY input/output of NULLs (Peter E)
6374 Make postgres user have a password by default (Peter E)
6375 Add CREATE/ALTER/DROP GROUP (Peter E)
6376 All administration scripts now support --long options (Peter E, Karel)
6377 Vacuumdb script now supports --all option (Peter E)
6378 ecpg new portable FETCH syntax
6379 Add ecpg EXEC SQL IFDEF, EXEC SQL IFNDEF, EXEC SQL ELSE, EXEC SQL ELIF
6380 and EXEC SQL ENDIF directives
6381 Add pg_ctl script to control backend start-up (Tatsuo)
6382 Add postmaster.opts.default file to store start-up flags (Tatsuo)
6383 Allow --with-mb=SQL_ASCII
6384 Increase maximum number of index keys to 16 (Bruce)
6385 Increase maximum number of function arguments to 16 (Bruce)
6386 Allow configuration of maximum number of index keys and arguments (Bruce)
6387 Allow unprivileged users to change their passwords (Peter E)
6388 Password authentication enabled; required for new users (Peter E)
6389 Disallow dropping a user who owns a database (Peter E)
6390 Change initdb option --with-mb to --enable-multibyte
6391 Add option for initdb to prompts for superuser password (Peter E)
6392 Allow complex type casts like col::numeric(9,2) and col::int2::float8 (Tom)
6393 Updated user interfaces on initdb, initlocation, pg_dump, ipcclean (Peter E)
6394 New pg_char_to_encoding() and pg_encoding_to_char() functions (Tatsuo)
6395 libpq non-blocking mode (Alfred Perlstein)
6396 Improve conversion of types in casts that don't specify a length
6397 New plperl internal programming language (Mark Hollomon)
6398 Allow COPY IN to read file that do not end with a newline (Tom)
6399 Indicate when long identifiers are truncated (Tom)
6400 Allow aggregates to use type equivalency (Peter E)
6401 Add Oracle's to_char(), to_date(), to_datetime(), to_timestamp(), to_number()
6402 conversion functions (Karel Zak <zakkr@zf.jcu.cz>)
6403 Add SELECT DISTINCT ON (expr [, expr ...]) targetlist ... (Tom)
6404 Check to be sure ORDER BY is compatible with the DISTINCT operation (Tom)
6405 Add NUMERIC and int8 types to ODBC
6406 Improve EXPLAIN results for Append, Group, Agg, Unique (Tom)
6407 Add ALTER TABLE ... ADD FOREIGN KEY (Stephan Szabo)
6408 Allow SELECT .. FOR UPDATE in PL/pgSQL (Hiroshi)
6409 Enable backward sequential scan even after reaching EOF (Hiroshi)
6410 Add btree indexing of boolean values, >= and <= (Don Baccus)
6411 Print current line number when COPY FROM fails (Massimo)
6412 Recognize POSIX time zone e.g. "PST+8" and "GMT-8" (Thomas)
6413 Add DEC as synonym for DECIMAL (Thomas)
6414 Add SESSION_USER as SQL92 key word, same as CURRENT_USER (Thomas)
6415 Implement SQL92 column aliases (aka correlation names) (Thomas)
6416 Implement SQL92 join syntax (Thomas)
6417 Make INTERVAL reserved word allowed as a column identifier (Thomas)
6418 Implement REINDEX command (Hiroshi)
6419 Accept ALL in aggregate function SUM(ALL col) (Tom)
6420 Prevent GROUP BY from using column aliases (Tom)
6421 New psql \encoding option (Tatsuo)
6422 Allow PQrequestCancel() to terminate when in waiting-for-lock state (Hiroshi)
6423 Allow negation of a negative number in all cases
6424 Add ecpg descriptors (Christof, Michael)
6425 Allow CREATE VIEW v AS SELECT f1::char(8) FROM tbl
6426 Allow casts with length, like foo::char(8)
6427 New libpq functions PQsetClientEncoding(), PQclientEncoding() (Tatsuo)
6428 Add support for SJIS user defined characters (Tatsuo)
6429 Larger views/rules supported
6430 Make libpq's PQconndefaults() thread-safe (Tom)
6431 Disable // as comment to be ANSI conforming, should use -- (Tom)
6432 Allow column aliases on views CREATE VIEW name (collist)
6433 Fixes for views with subqueries (Tom)
6434 Allow UPDATE table SET fld = (SELECT ...) (Tom)
6435 SET command options no longer require quotes
6436 Update pgaccess to 0.98.6
6437 New SET SEED command
6438 New pg_options.sample file
6439 New SET FSYNC command (Massimo)
6440 Allow pg_descriptions when creating tables
6441 Allow pg_descriptions when creating types, columns, and functions
6442 Allow psql \copy to allow delimiters (Peter E)
6443 Allow psql to print nulls as distinct from "" [null] (Peter E)
6447 Many array fixes (Tom)
6448 Allow bare column names to be subscripted as arrays (Tom)
6449 Improve type casting of int and float constants (Tom)
6450 Cleanups for int8 inputs, range checking, and type conversion (Tom)
6451 Fix for SELECT timespan('21:11:26'::time) (Tom)
6452 netmask('x.x.x.x/0') is 255.255.255.255 instead of 0.0.0.0 (Oleg Sharoiko)
6453 Add btree index on NUMERIC (Jan)
6454 Perl fix for large objects containing NUL characters (Douglas Thomson)
6455 ODBC fix for for large objects (free)
6456 Fix indexing of cidr data type
6457 Fix for Ethernet MAC addresses (macaddr type) comparisons
6458 Fix for date/time types when overflows happened in computations (Tom)
6459 Allow array on int8 (Peter E)
6460 Fix for rounding/overflow of NUMERIC type, like NUMERIC(4,4) (Tom)
6461 Allow NUMERIC arrays
6462 Fix bugs in NUMERIC ceil() and floor() functions (Tom)
6463 Make char_length()/octet_length including trailing blanks (Tom)
6464 Made abstime/reltime use int4 instead of time_t (Peter E)
6465 New lztext data type for compressed text fields
6466 Revise code to handle coercion of int and float constants (Tom)
6467 Start at new code to implement a BIT and BIT VARYING type (Adriaan Joubert)
6468 NUMERIC now accepts scientific notation (Tom)
6469 NUMERIC to int4 rounds (Tom)
6470 Convert float4/8 to NUMERIC properly (Tom)
6471 Allow type conversion with NUMERIC (Thomas)
6472 Make ISO date style (2000-02-16 09:33) the default (Thomas)
6473 Add NATIONAL CHAR [ VARYING ] (Thomas)
6474 Allow NUMERIC round and trunc to accept negative scales (Tom)
6475 New TIME WITH TIME ZONE type (Thomas)
6476 Add MAX()/MIN() on time type (Thomas)
6477 Add abs(), mod(), fac() for int8 (Thomas)
6478 Rename functions to round(), sqrt(), cbrt(), pow() for float8 (Thomas)
6479 Add transcendental math functions (e.g. sin(), acos()) for float8 (Thomas)
6480 Add exp() and ln() for NUMERIC type
6481 Rename NUMERIC power() to pow() (Thomas)
6482 Improved TRANSLATE() function (Edwin Ramirez, Tom)
6483 Allow X=-Y operators (Tom)
6484 Allow SELECT float8(COUNT(*))/(SELECT COUNT(*) FROM t) FROM t GROUP BY f1; (Tom)
6485 Allow LOCALE to use indexes in regular expression searches (Tom)
6486 Allow creation of functional indexes to use default types
6490 Prevent exponential space consumption with many AND's and OR's (Tom)
6491 Collect attribute selectivity values for system columns (Tom)
6492 Reduce memory usage of aggregates (Tom)
6493 Fix for LIKE optimization to use indexes with multibyte encodings (Tom)
6494 Fix r-tree index optimizer selectivity (Thomas)
6495 Improve optimizer selectivity computations and functions (Tom)
6496 Optimize btree searching for cases where many equal keys exist (Tom)
6497 Enable fast LIKE index processing only if index present (Tom)
6498 Re-use free space on index pages with duplicates (Tom)
6499 Improve hash join processing (Tom)
6500 Prevent descending sort if result is already sorted(Hiroshi)
6501 Allow commuting of index scan query qualifications (Tom)
6502 Prefer index scans in cases where ORDER BY/GROUP BY is required (Tom)
6503 Allocate large memory requests in fix-sized chunks for performance (Tom)
6504 Fix vacuum's performance by reducing memory allocation requests (Tom)
6505 Implement constant-expression simplification (Bernard Frankpitt, Tom)
6506 Use secondary columns to be used to determine start of index scan (Hiroshi)
6507 Prevent quadruple use of disk space when doing internal sorting (Tom)
6508 Faster sorting by calling fewer functions (Tom)
6509 Create system indexes to match all system caches (Bruce, Hiroshi)
6510 Make system caches use system indexes (Bruce)
6511 Make all system indexes unique (Bruce)
6512 Improve pg_statistics management for VACUUM speed improvement (Tom)
6513 Flush backend cache less frequently (Tom, Hiroshi)
6514 COPY now reuses previous memory allocation, improving performance (Tom)
6515 Improve optimization cost estimation (Tom)
6516 Improve optimizer estimate of range queries x > lowbound AND x < highbound (Tom)
6517 Use DNF instead of CNF where appropriate (Tom, Taral)
6518 Further cleanup for OR-of-AND WHERE-clauses (Tom)
6519 Make use of index in OR clauses (x = 1 AND y = 2) OR (x = 2 AND y = 4) (Tom)
6520 Smarter optimizer computations for random index page access (Tom)
6521 New SET variable to control optimizer costs (Tom)
6522 Optimizer queries based on LIMIT, OFFSET, and EXISTS qualifications (Tom)
6523 Reduce optimizer internal housekeeping of join paths for speedup (Tom)
6524 Major subquery speedup (Tom)
6525 Fewer fsync writes when fsync is not disabled (Tom)
6526 Improved LIKE optimizer estimates (Tom)
6527 Prevent fsync in SELECT-only queries (Vadim)
6528 Make index creation use psort code, because it is now faster (Tom)
6529 Allow creation of sort temp tables > 1 Gig
6533 Fix for linux PPC compile
6534 New generic expression-tree-walker subroutine (Tom)
6535 Change form() to varargform() to prevent portability problems
6536 Improved range checking for large integers on Alphas
6537 Clean up #include in /include directory (Bruce)
6538 Add scripts for checking includes (Bruce)
6539 Remove un-needed #include's from *.c files (Bruce)
6540 Change #include's to use <> and "" as appropriate (Bruce)
6541 Enable Windows compilation of libpq
6542 Alpha spinlock fix from Uncle George <email>gatgul@voicenet.com</email>
6543 Overhaul of optimizer data structures (Tom)
6544 Fix to cygipc library (Yutaka Tanida)
6545 Allow pgsql to work on newer Cygwin snapshots (Dan)
6546 New catalog version number (Tom)
6548 Rename heap_replace to heap_update
6549 Update for QNX (Dr. Andreas Kardos)
6550 New platform-specific regression handling (Tom)
6551 Rename oid8 -> oidvector and int28 -> int2vector (Bruce)
6552 Included all yacc and lex files into the distribution (Peter E.)
6553 Remove lextest, no longer needed (Peter E)
6554 Fix for libpq and psql on Windows (Magnus)
6555 Internally change datetime and timespan into timestamp and interval (Thomas)
6556 Fix for plpgsql on BSD/OS
6557 Add SQL_ASCII test case to the regression test (Tatsuo)
6558 configure --with-mb now deprecated (Tatsuo)
6560 NetBSD fixes (Johnny C. Lam <email>lamj@stat.cmu.edu</email>)
6561 Fixes for Alpha compiles
6562 New multibyte encodings
6568 <sect1 id="release-6-5-3">
6569 <title>Release 6.5.3</title>
6572 <title>Release date</title>
6573 <simpara>1999-10-13</simpara>
6577 This is basically a cleanup release for 6.5.2. We have added a new
6578 <application>PgAccess</> that was missing in 6.5.2, and installed an NT-specific fix.
6583 <title>Migration to version 6.5.3</title>
6586 A dump/restore is <emphasis>not</emphasis> required for those running
6591 <title>Changes</title>
6595 Updated version of pgaccess 0.98
6597 Fix dumping rules on inherited tables
6604 <sect1 id="release-6-5-2">
6605 <title>Release 6.5.2</title>
6608 <title>Release date</title>
6609 <simpara>1999-09-15</simpara>
6613 This is basically a cleanup release for 6.5.1. We have fixed a variety of
6614 problems reported by 6.5.1 users.
6619 <title>Migration to version 6.5.2</title>
6622 A dump/restore is <emphasis>not</emphasis> required for those running
6628 <title>Changes</title>
6632 subselect+CASE fixes(Tom)
6633 Add SHLIB_LINK setting for solaris_i386 and solaris_sparc ports(Daren Sefcik)
6634 Fixes for CASE in WHERE join clauses(Tom)
6635 Fix BTScan abort(Tom)
6636 Repair the check for redundant UNIQUE and PRIMARY KEY indexes(Thomas)
6637 Improve it so that it checks for multicolumn constraints(Thomas)
6638 Fix for Windows making problem with MB enabled(Hiroki Kataoka)
6639 Allow BSD yacc and bison to compile pl code(Bruce)
6640 Fix SET NAMES working
6642 Fix vacuum's memory consumption(Hiroshi,Tatsuo)
6643 Reduce the total memory consumption of vacuum(Tom)
6644 Fix for timestamp(datetime)
6645 Rule deparsing bugfixes(Tom)
6646 Fix quoting problems in mkMakefile.tcldefs.sh.in and mkMakefile.tkdefs.sh.in(Tom)
6647 This is to re-use space on index pages freed by vacuum(Vadim)
6648 document -x for pg_dump(Bruce)
6649 Fix for unary operators in rule deparser(Tom)
6650 Comment out FileUnlink of excess segments during mdtruncate()(Tom)
6651 IRIX linking fix from Yu Cao >yucao@falcon.kla-tencor.com<
6652 Repair logic error in LIKE: should not return LIKE_ABORT
6653 when reach end of pattern before end of text(Tom)
6654 Repair incorrect cleanup of heap memory allocation during transaction abort(Tom)
6655 Updated version of pgaccess 0.98
6661 <sect1 id="release-6-5-1">
6662 <title>Release 6.5.1</title>
6665 <title>Release date</title>
6666 <simpara>1999-07-15</simpara>
6670 This is basically a cleanup release for 6.5. We have fixed a variety of
6671 problems reported by 6.5 users.
6675 <title>Migration to version 6.5.1</title>
6678 A dump/restore is <emphasis>not</emphasis> required for those running
6684 <title>Changes</title>
6689 Portability fixes for linux_ppc, IRIX, linux_alpha, OpenBSD, alpha
6690 Remove QUERY_LIMIT, use SELECT...LIMIT
6691 Fix for EXPLAIN on inheritance(Tom)
6692 Patch to allow vacuum on multisegment tables(Hiroshi)
6693 R-Tree optimizer selectivity fix(Tom)
6694 ACL file descriptor leak fix(Atsushi Ogawa)
6695 New expresssion subtree code(Tom)
6696 Avoid disk writes for read-only transactions(Vadim)
6697 Fix for removal of temp tables if last transaction was aborted(Bruce)
6698 Fix to prevent too large row from being created(Bruce)
6700 Allow port numbers 32k - 64k(Bruce)
6701 Add ^ precidence(Bruce)
6702 Rename sort files called pg_temp to pg_sorttemp(Bruce)
6703 Fix for microseconds in time values(Tom)
6704 Tutorial source cleanup
6706 Fix for sorting of NULL's in some cases(Tom)
6707 Shared library dependencies fixed (Tom)
6708 Fixed glitches affecting GROUP BY in subselects(Tom)
6709 Fix some compiler warnings (Tomoaki Nishiyama)
6710 Add Win1250 (Czech) support (Pavel Behal)
6716 <sect1 id="release-6-5">
6717 <title>Release 6.5</title>
6720 <title>Release date</title>
6721 <simpara>1999-06-09</simpara>
6725 This release marks a major step in the development team's mastery of the source
6726 code we inherited from Berkeley. You will see we are now easily adding
6727 major features, thanks to the increasing size and experience of our
6728 world-wide development team.
6732 Here is a brief summary of the more notable changes:
6737 Multiversion concurrency control(MVCC)
6741 This removes our old table-level locking, and replaces it with
6742 a locking system that is superior to most commercial database
6743 systems. In a traditional system, each row that is modified
6744 is locked until committed, preventing reads by other users.
6745 MVCC uses the natural multiversion nature of
6746 <productname>PostgreSQL</productname> to allow readers to
6747 continue reading consistent data during writer activity.
6748 Writers continue to use the compact pg_log transaction system.
6749 This is all performed without having to allocate a lock for
6750 every row like traditional database systems. So, basically,
6751 we no longer are restricted by simple table-level locking; we
6752 have something better than row-level locking.
6759 Hot backups from <application>pg_dump</application>
6763 <application>pg_dump</application> takes advantage of the new
6764 MVCC features to give a consistent database dump/backup while
6765 the database stays online and available for queries.
6776 We now have a true numeric data type, with
6777 user-specified precision.
6788 Temporary tables are guaranteed to have unique names
6789 within a database session, and are destroyed on session exit.
6800 We now have CASE, INTERSECT, and EXCEPT statement
6801 support. We have new LIMIT/OFFSET, SET TRANSACTION ISOLATION LEVEL,
6802 SELECT ... FOR UPDATE, and an improved LOCK TABLE command.
6813 We continue to speed up <productname>PostgreSQL</productname>,
6814 thanks to the variety of talents within our team. We have
6815 sped up memory allocation, optimization, table joins, and row
6827 We continue to expand our port list, this time including
6828 <systemitem class="osname">Windows NT</>/<systemitem>ix86</> and <systemitem class="osname">NetBSD</>/<systemitem>arm32</>.
6839 Most interfaces have new versions, and existing functionality
6851 New and updated material is present throughout the
6852 documentation. New <acronym>FAQ</acronym>s have been
6853 contributed for <systemitem class="osname">SGI</> and <systemitem class="osname">AIX</> platforms.
6854 The <citetitle>Tutorial</citetitle> has introductory information
6855 on <acronym>SQL</acronym> from Stefan Simkovics.
6856 For the <citetitle>User's Guide</citetitle>, there are
6857 reference pages covering the postmaster and more utility
6858 programs, and a new appendix
6859 contains details on date/time behavior.
6860 The <citetitle>Administrator's Guide</citetitle> has a new
6861 chapter on troubleshooting from Tom Lane.
6862 And the <citetitle>Programmer's Guide</citetitle> has a
6863 description of query processing, also from Stefan, and details
6864 on obtaining the <productname>PostgreSQL</productname> source
6865 tree via anonymous <productname>CVS</productname> and
6866 <productname>CVSup</productname>.
6874 <title>Migration to version 6.5</title>
6877 A dump/restore using <application>pg_dump</application>
6878 is required for those wishing to migrate data from any
6879 previous release of <productname>PostgreSQL</productname>.
6880 <application>pg_upgrade</application> can <emphasis>not</emphasis>
6881 be used to upgrade to this release because the on-disk structure
6882 of the tables has changed compared to previous releases.
6886 The new Multiversion Concurrency Control (MVCC) features can
6887 give somewhat different behaviors in multiuser
6888 environments. <emphasis>Read and understand the following section
6889 to ensure that your existing applications will give you the
6890 behavior you need.</emphasis>
6894 <title>Multiversion Concurrency Control</title>
6897 Because readers in 6.5 don't lock data, regardless of transaction
6898 isolation level, data read by one transaction can be overwritten by
6899 another. In other words, if a row is returned by
6900 <command>SELECT</command> it doesn't mean that this row really exists
6901 at the time it is returned (i.e. sometime after the statement or
6902 transaction began) nor that the row is protected from being deleted or
6903 updated by concurrent transactions before the current transaction does
6904 a commit or rollback.
6908 To ensure the actual existence of a row and protect it against
6909 concurrent updates one must use <command>SELECT FOR UPDATE</command> or
6910 an appropriate <command>LOCK TABLE</command> statement. This should be
6911 taken into account when porting applications from previous releases of
6912 <productname>PostgreSQL</productname> and other environments.
6916 Keep the above in mind if you are using
6917 <filename>contrib/refint.*</filename> triggers for
6918 referential integrity. Additional techniques are required now. One way is
6919 to use <command>LOCK parent_table IN SHARE ROW EXCLUSIVE MODE</command>
6920 command if a transaction is going to update/delete a primary key and
6921 use <command>LOCK parent_table IN SHARE MODE</command> command if a
6922 transaction is going to update/insert a foreign key.
6926 Note that if you run a transaction in SERIALIZABLE mode then you must
6927 execute the <command>LOCK</command> commands above before execution of any
6928 <acronym>DML</acronym> statement
6929 (<command>SELECT/INSERT/DELETE/UPDATE/FETCH/COPY_TO</command>) in the
6936 These inconveniences will disappear in the future
6937 when the ability to read dirty
6938 (uncommitted) data (regardless of isolation level) and true referential
6939 integrity will be implemented.
6945 <title>Changes</title>
6951 Fix text<->float8 and text<->float4 conversion functions(Thomas)
6952 Fix for creating tables with mixed-case constraints(Billy)
6953 Change exp()/pow() behavior to generate error on underflow/overflow(Jan)
6954 Fix bug in pg_dump -z
6955 Memory overrun cleanups(Tatsuo)
6956 Fix for lo_import crash(Tatsuo)
6957 Adjust handling of data type names to suppress double quotes(Thomas)
6958 Use type coercion for matching columns and DEFAULT(Thomas)
6959 Fix deadlock so it only checks once after one second of sleep(Bruce)
6960 Fixes for aggregates and PL/pgsql(Hiroshi)
6961 Fix for subquery crash(Vadim)
6962 Fix for libpq function PQfnumber and case-insensitive names(Bahman Rafatjoo)
6963 Fix for large object write-in-middle, no extra block, memory consumption(Tatsuo)
6964 Fix for pg_dump -d or -D and quote special characters in INSERT
6965 Repair serious problems with dynahash(Tom)
6966 Fix INET/CIDR portability problems
6967 Fix problem with selectivity error in ALTER TABLE ADD COLUMN(Bruce)
6968 Fix executor so mergejoin of different column types works(Tom)
6969 Fix for Alpha OR selectivity bug
6970 Fix OR index selectivity problem(Bruce)
6971 Fix so \d shows proper length for char()/varchar()(Ryan)
6972 Fix tutorial code(Clark)
6973 Improve destroyuser checking(Oliver)
6974 Fix for Kerberos(Rodney McDuff)
6975 Fix for dropping database while dirty buffers(Bruce)
6976 Fix so sequence nextval() can be case-sensitive(Bruce)
6978 Drop buffers before destroying database files(Bruce)
6979 Fix case where executor evaluates functions twice(Tatsuo)
6980 Allow sequence nextval actions to be case-sensitive(Bruce)
6981 Fix optimizer indexing not working for negative numbers(Bruce)
6982 Fix for memory leak in executor with fjIsNull
6983 Fix for aggregate memory leaks(Erik Riedel)
6984 Allow user name containing a dash to grant privileges
6985 Cleanup of NULL in inet types
6986 Clean up system table bugs(Tom)
6987 Fix problems of PAGER and \? command(Masaaki Sakaida)
6988 Reduce default multisegment file size limit to 1GB(Peter)
6989 Fix for dumping of CREATE OPERATOR(Tom)
6990 Fix for backward scanning of cursors(Hiroshi Inoue)
6991 Fix for COPY FROM STDIN when using \i(Tom)
6992 Fix for subselect is compared inside an expression(Jan)
6993 Fix handling of error reporting while returning rows(Tom)
6994 Fix problems with reference to array types(Tom,Jan)
6995 Prevent UPDATE SET oid(Jan)
6996 Fix pg_dump so -t option can handle case-sensitive tablenames
6997 Fixes for GROUP BY in special cases(Tom, Jan)
6998 Fix for memory leak in failed queries(Tom)
6999 DEFAULT now supports mixed-case identifiers(Tom)
7000 Fix for multisegment uses of DROP/RENAME table, indexes(Ole Gjerde)
7001 Disable use of pg_dump with both -o and -d options(Bruce)
7002 Allow pg_dump to properly dump group privileges(Bruce)
7003 Fix GROUP BY in INSERT INTO table SELECT * FROM table2(Jan)
7004 Fix for computations in views(Jan)
7005 Fix for aggregates on array indexes(Tom)
7006 Fix for DEFAULT handles single quotes in value requiring too many quotes
7007 Fix security problem with non-super users importing/exporting large objects(Tom)
7008 Rollback of transaction that creates table cleaned up properly(Tom)
7009 Fix to allow long table and column names to generate proper serial names(Tom)
7013 Add "vacuumdb" utility
7014 Speed up libpq by allocating memory better(Tom)
7015 EXPLAIN all indexes used(Tom)
7016 Implement CASE, COALESCE, NULLIF expression(Thomas)
7017 New pg_dump table output format(Constantin)
7018 Add string min()/max() functions(Thomas)
7019 Extend new type coercion techniques to aggregates(Thomas)
7020 New moddatetime contrib(Terry)
7021 Update to pgaccess 0.96(Constantin)
7022 Add routines for single-byte "char" type(Thomas)
7023 Improved substr() function(Thomas)
7024 Improved multibyte handling(Tatsuo)
7025 Multiversion concurrency control/MVCC(Vadim)
7026 New Serialized mode(Vadim)
7027 Fix for tables over 2gigs(Peter)
7028 New SET TRANSACTION ISOLATION LEVEL(Vadim)
7029 New LOCK TABLE IN ... MODE(Vadim)
7030 Update ODBC driver(Byron)
7031 New NUMERIC data type(Jan)
7032 New SELECT FOR UPDATE(Vadim)
7033 Handle "NaN" and "Infinity" for input values(Jan)
7034 Improved date/year handling(Thomas)
7035 Improved handling of backend connections(Magnus)
7036 New options ELOG_TIMESTAMPS and USE_SYSLOG options for log files(Massimo)
7037 New TCL_ARRAYS option(Massimo)
7038 New INTERSECT and EXCEPT(Stefan)
7039 New pg_index.indisprimary for primary key tracking(D'Arcy)
7040 New pg_dump option to allow dropping of tables before creation(Brook)
7041 Speedup of row output routines(Tom)
7042 New READ COMMITTED isolation level(Vadim)
7043 New TEMP tables/indexes(Bruce)
7044 Prevent sorting if result is already sorted(Jan)
7045 New memory allocation optimization(Jan)
7046 Allow psql to do \p\g(Bruce)
7047 Allow multiple rule actions(Jan)
7048 Added LIMIT/OFFSET functionality(Jan)
7049 Improve optimizer when joining a large number of tables(Bruce)
7050 New intro to SQL from S. Simkovics' Master's Thesis (Stefan, Thomas)
7051 New intro to backend processing from S. Simkovics' Master's Thesis (Stefan)
7052 Improved int8 support(Ryan Bradetich, Thomas, Tom)
7053 New routines to convert between int8 and text/varchar types(Thomas)
7054 New bushy plans, where meta-tables are joined(Bruce)
7055 Enable right-hand queries by default(Bruce)
7056 Allow reliable maximum number of backends to be set at configure time
7057 (--with-maxbackends and postmaster switch (-N backends))(Tom)
7058 GEQO default now 10 tables because of optimizer speedups(Tom)
7059 Allow NULL=Var for MS-SQL portability(Michael, Bruce)
7060 Modify contrib check_primary_key() so either "automatic" or "dependent"(Anand)
7061 Allow psql \d on a view show query(Ryan)
7062 Speedup for LIKE(Bruce)
7063 Ecpg fixes/features, see src/interfaces/ecpg/ChangeLog file(Michael)
7064 JDBC fixes/features, see src/interfaces/jdbc/CHANGELOG(Peter)
7065 Make % operator have precedence like /(Bruce)
7066 Add new postgres -O option to allow system table structure changes(Bruce)
7067 Update contrib/pginterface/findoidjoins script(Tom)
7068 Major speedup in vacuum of deleted rows with indexes(Vadim)
7069 Allow non-SQL functions to run different versions based on arguments(Tom)
7070 Add -E option that shows actual queries sent by \dt and friends(Masaaki Sakaida)
7071 Add version number in start-up banners for psql(Masaaki Sakaida)
7072 New contrib/vacuumlo removes large objects not referenced(Peter)
7073 New initialization for table sizes so non-vacuumed tables perform better(Tom)
7074 Improve error messages when a connection is rejected(Tom)
7075 Support for arrays of char() and varchar() fields(Massimo)
7076 Overhaul of hash code to increase reliability and performance(Tom)
7077 Update to PyGreSQL 2.4(D'Arcy)
7078 Changed debug options so -d4 and -d5 produce different node displays(Jan)
7079 New pg_options: pretty_plan, pretty_parse, pretty_rewritten(Jan)
7080 Better optimization statistics for system table access(Tom)
7081 Better handling of non-default block sizes(Massimo)
7082 Improve GEQO optimizer memory consumption(Tom)
7083 UNION now suppports ORDER BY of columns not in target list(Jan)
7084 Major libpq++ improvements(Vince Vielhaber)
7085 pg_dump now uses -z(ACL's) as default(Bruce)
7086 backend cache, memory speedups(Tom)
7087 have pg_dump do everything in one snapshot transaction(Vadim)
7088 fix for large object memory leakage, fix for pg_dumping(Tom)
7089 INET type now respects netmask for comparisons
7090 Make VACUUM ANALYZE only use a readlock(Vadim)
7091 Allow VIEWs on UNIONS(Jan)
7092 pg_dump now can generate consistent snapshots on active databases(Vadim)
7096 Improve port matching(Tom)
7097 Portability fixes for SunOS
7098 Add Windows NT backend port and enable dynamic loading(Magnus and Daniel Horak)
7099 New port to Cobalt Qube(Mips) running Linux(Tatsuo)
7100 Port to NetBSD/m68k(Mr. Mutsuki Nakajima)
7101 Port to NetBSD/sun3(Mr. Mutsuki Nakajima)
7102 Port to NetBSD/macppc(Toshimi Aoki)
7103 Fix for tcl/tk configuration(Vince)
7104 Removed CURRENT key word for rule queries(Jan)
7105 NT dynamic loading now works(Daniel Horak)
7106 Add ARM32 support(Andrew McMurry)
7107 Better support for HP-UX 11 and UnixWare
7108 Improve file handling to be more uniform, prevent file descriptor leak(Tom)
7109 New install commands for plpgsql(Jan)
7116 <sect1 id="release-6-4-2">
7117 <title>Release 6.4.2</title>
7120 <title>Release date</title>
7121 <simpara>1998-12-20</simpara>
7125 The 6.4.1 release was improperly packaged. This also has one additional
7131 <title>Migration to version 6.4.2</title>
7134 A dump/restore is <emphasis>not</emphasis> required for those running
7139 <title>Changes</title>
7143 Fix for datetime constant problem on some platforms(Thomas)
7151 <sect1 id="release-6-4-1">
7152 <title>Release 6.4.1</title>
7155 <title>Release date</title>
7156 <simpara>1998-12-18</simpara>
7160 This is basically a cleanup release for 6.4. We have fixed a variety of
7161 problems reported by 6.4 users.
7166 <title>Migration to version 6.4.1</title>
7169 A dump/restore is <emphasis>not</emphasis> required for those running
7174 <title>Changes</title>
7178 Add pg_dump -N flag to force double quotes around identifiers. This is
7180 Fix for NOT in where clause causing crash(Bruce)
7181 EXPLAIN VERBOSE coredump fix(Vadim)
7182 Fix shared-library problems on Linux
7183 Fix test for table existence to allow mixed-case and whitespace in
7184 the table name(Thomas)
7185 Fix a couple of pg_dump bugs
7186 Configure matches template/.similar entries better(Tom)
7187 Change builtin function names from SPI_* to spi_*
7188 OR WHERE clause fix(Vadim)
7189 Fixes for mixed-case table names(Billy)
7190 contrib/linux/postgres.init.csh/sh fix(Thomas)
7191 libpq memory overrun fix
7193 Change exp() behavior to generate error on underflow(Thomas)
7194 pg_dump fixes for memory leak, inheritance constraints, layout change
7195 update pgaccess to 0.93
7196 Fix prototype for 64-bit platforms
7197 Multibyte fixes(Tatsuo)
7199 Fix memory overruns(Tatsuo)
7200 Fix for lo_import() crash(Bruce)
7201 Better search for install program(Tom)
7204 Use implicit type coercion for matching DEFAULT values(Thomas)
7205 Add routines to help with single-byte (internal) character type(Thomas)
7206 Compilation of libpq for Windows fixes(Magnus)
7207 Upgrade to PyGreSQL 2.2(D'Arcy)
7215 <sect1 id="release-6-4">
7216 <title>Release 6.4</title>
7219 <title>Release date</title>
7220 <simpara>1998-10-30</simpara>
7224 There are <emphasis>many</emphasis> new features and improvements in this release.
7225 Thanks to our developers and maintainers, nearly every aspect of the system
7226 has received some attention since the previous release.
7227 Here is a brief, incomplete summary:
7232 Views and rules are now functional thanks to extensive new code in the
7233 rewrite rules system from Jan Wieck. He also wrote a chapter on it
7234 for the <citetitle>Programmer's Guide</citetitle>.
7239 Jan also contributed a second procedural language, <application>PL/pgSQL</application>, to go with the
7240 original <application>PL/pgTCL</application> procedural language he contributed last release.
7246 We have optional multiple-byte character set support from Tatsuo Ishii
7247 to complement our existing locale support.
7253 Client/server communications has been cleaned up, with better support for
7254 asynchronous messages and interrupts thanks to Tom Lane.
7260 The parser will now perform automatic type coercion to match arguments
7261 to available operators and functions, and to match columns and expressions
7262 with target columns. This uses a generic mechanism which supports
7263 the type extensibility features of <productname>PostgreSQL</productname>.
7264 There is a new chapter in the <citetitle>User's Guide</citetitle>
7265 which covers this topic.
7271 Three new data types have been added.
7272 Two types, <type>inet</type> and <type>cidr</type>, support various forms
7273 of IP network, subnet, and machine addressing. There is now an 8-byte integer
7274 type available on some platforms. See the chapter on data types
7275 in the <citetitle>User's Guide</citetitle> for details.
7276 A fourth type, <type>serial</type>, is now supported by the parser as an
7277 amalgam of the <type>int4</type> type, a sequence, and a unique index.
7283 Several more <acronym>SQL92</acronym>-compatible syntax features have been
7284 added, including <command>INSERT DEFAULT VALUES</command>
7290 The automatic configuration and installation system has received some
7291 attention, and should be more robust for more platforms than it has ever
7300 <title>Migration to version 6.4</title>
7303 A dump/restore using <application>pg_dump</application>
7304 or <application>pg_dumpall</application>
7305 is required for those wishing to migrate data from any
7306 previous release of <productname>PostgreSQL</productname>.
7311 <title>Changes</title>
7317 Fix for a tiny memory leak in PQsetdb/PQfinish(Bryan)
7318 Remove char2-16 data types, use char/varchar(Darren)
7319 Pqfn not handles a NOTICE message(Anders)
7320 Reduced busywaiting overhead for spinlocks with many backends (dg)
7321 Stuck spinlock detection (dg)
7322 Fix up "ISO-style" timespan decoding and encoding(Thomas)
7323 Fix problem with table drop after rollback of transaction(Vadim)
7324 Change error message and remove non-functional update message(Vadim)
7325 Fix for COPY array checking
7326 Fix for SELECT 1 UNION SELECT NULL
7327 Fix for buffer leaks in large object calls(Pascal)
7328 Change owner from oid to int4 type(Bruce)
7329 Fix a bug in the oracle compatibility functions btrim() ltrim() and rtrim()
7330 Fix for shared invalidation cache overflow(Massimo)
7331 Prevent file descriptor leaks in failed COPY's(Bruce)
7332 Fix memory leak in libpgtcl's pg_select(Constantin)
7333 Fix problems with username/passwords over 8 characters(Tom)
7334 Fix problems with handling of asynchronous NOTIFY in backend(Tom)
7335 Fix of many bad system table entries(Tom)
7339 Upgrade ecpg and ecpglib,see src/interfaces/ecpc/ChangeLog(Michael)
7340 Show the index used in an EXPLAIN(Zeugswetter)
7341 EXPLAIN invokes rule system and shows plan(s) for rewritten queries(Jan)
7342 Multibyte awareness of many data types and functions, via configure(Tatsuo)
7343 New configure --with-mb option(Tatsuo)
7344 New initdb --pgencoding option(Tatsuo)
7345 New createdb -E multibyte option(Tatsuo)
7346 Select version(); now returns PostgreSQL version(Jeroen)
7347 libpq now allows asynchronous clients(Tom)
7348 Allow cancel from client of backend query(Tom)
7349 psql now cancels query with Control-C(Tom)
7350 libpq users need not issue dummy queries to get NOTIFY messages(Tom)
7351 NOTIFY now sends sender's PID, so you can tell whether it was your own(Tom)
7352 PGresult struct now includes associated error message, if any(Tom)
7353 Define "tz_hour" and "tz_minute" arguments to date_part()(Thomas)
7354 Add routines to convert between varchar and bpchar(Thomas)
7355 Add routines to allow sizing of varchar and bpchar into target columns(Thomas)
7356 Add bit flags to support timezonehour and minute in data retrieval(Thomas)
7357 Allow more variations on valid floating point numbers (e.g. ".1", "1e6")(Thomas)
7358 Fixes for unary minus parsing with leading spaces(Thomas)
7359 Implement TIMEZONE_HOUR, TIMEZONE_MINUTE per SQL92 specs(Thomas)
7360 Check for and properly ignore FOREIGN KEY column constraints(Thomas)
7361 Define USER as synonym for CURRENT_USER per SQL92 specs(Thomas)
7362 Enable HAVING clause but no fixes elsewhere yet.
7363 Make "char" type a synonym for "char(1)" (actually implemented as bpchar)(Thomas)
7364 Save string type if specified for DEFAULT clause handling(Thomas)
7365 Coerce operations involving different data types(Thomas)
7366 Allow some index use for columns of different types(Thomas)
7367 Add capabilities for automatic type conversion(Thomas)
7368 Cleanups for large objects, so file is truncated on open(Peter)
7369 Readline cleanups(Tom)
7370 Allow psql \f \ to make spaces as delimiter(Bruce)
7371 Pass pg_attribute.atttypmod to the frontend for column field lengths(Tom,Bruce)
7372 Msql compatibility library in /contrib(Aldrin)
7373 Remove the requirement that ORDER/GROUP BY clause identifiers be
7374 included in the target list(David)
7375 Convert columns to match columns in UNION clauses(Thomas)
7376 Remove fork()/exec() and only do fork()(Bruce)
7377 Jdbc cleanups(Peter)
7378 Show backend status on ps command line(only works on some platforms)(Bruce)
7379 Pg_hba.conf now has a sameuser option in the database field
7380 Make lo_unlink take oid param, not int4
7381 New DISABLE_COMPLEX_MACRO for compilers that can't handle our macros(Bruce)
7382 Libpgtcl now handles NOTIFY as a Tcl event, need not send dummy queries(Tom)
7383 libpgtcl cleanups(Tom)
7384 Add -error option to libpgtcl's pg_result command(Tom)
7385 New locale patch, see docs/README/locale(Oleg)
7386 Fix for pg_dump so CONSTRAINT and CHECK syntax is correct(ccb)
7387 New contrib/lo code for large object orphan removal(Peter)
7388 New psql command "SET CLIENT_ENCODING TO 'encoding'" for multibytes
7389 feature, see /doc/README.mb(Tatsuo)
7390 contrib/noupdate code to revoke update permission on a column
7391 libpq can now be compiled on Windows(Magnus)
7392 Add PQsetdbLogin() in libpq
7393 New 8-byte integer type, checked by configure for OS support(Thomas)
7394 Better support for quoted table/column names(Thomas)
7395 Surround table and column names with double-quotes in pg_dump(Thomas)
7396 PQreset() now works with passwords(Tom)
7397 Handle case of GROUP BY target list column number out of range(David)
7398 Allow UNION in subselects
7399 Add auto-size to screen to \d? commands(Bruce)
7400 Use UNION to show all \d? results in one query(Bruce)
7401 Add \d? field search feature(Bruce)
7402 Pg_dump issues fewer \connect requests(Tom)
7403 Make pg_dump -z flag work better, document it in manual page(Tom)
7404 Add HAVING clause with full support for subselects and unions(Stephan)
7405 Full text indexing routines in contrib/fulltextindex(Maarten)
7406 Transaction ids now stored in shared memory(Vadim)
7407 New PGCLIENTENCODING when issuing COPY command(Tatsuo)
7408 Support for SQL92 syntax "SET NAMES"(Tatsuo)
7409 Support for LATIN2-5(Tatsuo)
7410 Add UNICODE regression test case(Tatsuo)
7411 Lock manager cleanup, new locking modes for LLL(Vadim)
7412 Allow index use with OR clauses(Bruce)
7413 Allows "SELECT NULL ORDER BY 1;"
7414 Explain VERBOSE prints the plan, and now pretty-prints the plan to
7415 the postmaster log file(Bruce)
7416 Add indexes display to \d command(Bruce)
7417 Allow GROUP BY on functions(David)
7418 New pg_class.relkind for large objects(Bruce)
7419 New way to send libpq NOTICE messages to a different location(Tom)
7420 New \w write command to psql(Bruce)
7421 New /contrib/findoidjoins scans oid columns to find join relationships(Bruce)
7422 Allow binary-compatible indexes to be considered when checking for valid
7423 Indexes for restriction clauses containing a constant(Thomas)
7424 New ISBN/ISSN code in /contrib/isbn_issn
7425 Allow NOT LIKE, IN, NOT IN, BETWEEN, and NOT BETWEEN constraint(Thomas)
7426 New rewrite system fixes many problems with rules and views(Jan)
7427 * Rules on relations work
7428 * Event qualifications on insert/update/delete work
7429 * New OLD variable to reference CURRENT, CURRENT will be remove in future
7430 * Update rules can reference NEW and OLD in rule qualifications/actions
7431 * Insert/update/delete rules on views work
7432 * Multiple rule actions are now supported, surrounded by parentheses
7433 * Regular users can create views/rules on tables they have RULE permits
7434 * Rules and views inherit the privileges of the creator
7435 * No rules at the column level
7436 * No UPDATE NEW/OLD rules
7437 * New pg_tables, pg_indexes, pg_rules and pg_views system views
7438 * Only a single action on SELECT rules
7439 * Total rewrite overhaul, perhaps for 6.5
7441 * handle aggregates on views
7442 * handle insert into select from view works
7443 System indexes are now multikey(Bruce)
7444 Oidint2, oidint4, and oidname types are removed(Bruce)
7445 Use system cache for more system table lookups(Bruce)
7446 New backend programming language PL/pgSQL in backend/pl(Jan)
7447 New SERIAL data type, auto-creates sequence/index(Thomas)
7448 Enable assert checking without a recompile(Massimo)
7449 User lock enhancements(Massimo)
7450 New setval() command to set sequence value(Massimo)
7451 Auto-remove unix socket file on start-up if no postmaster running(Massimo)
7452 Conditional trace package(Massimo)
7453 New UNLISTEN command(Massimo)
7454 psql and libpq now compile under Windows using win32.mak(Magnus)
7455 Lo_read no longer stores trailing NULL(Bruce)
7456 Identifiers are now truncated to 31 characters internally(Bruce)
7457 Createuser options now availble on the command line
7458 Code for 64-bit integer supported added, configure tested, int8 type(Thomas)
7459 Prevent file descriptor leaf from failed COPY(Bruce)
7460 New pg_upgrade command(Bruce)
7461 Updated /contrib directories(Massimo)
7462 New CREATE TABLE DEFAULT VALUES statement available(Thomas)
7463 New INSERT INTO TABLE DEFAULT VALUES statement available(Thomas)
7464 New DECLARE and FETCH feature(Thomas)
7465 libpq's internal structures now not exported(Tom)
7466 Allow up to 8 key indexes(Bruce)
7467 Remove ARCHIVE key word, that is no longer used(Thomas)
7468 pg_dump -n flag to supress quotes around indentifiers
7469 disable system columns for views(Jan)
7470 new INET and CIDR types for network addresses(TomH, Paul)
7471 no more double quotes in psql output
7472 pg_dump now dumps views(Terry)
7473 new SET QUERY_LIMIT(Tatsuo,Jan)
7477 /contrib cleanup(Jun)
7478 Inline some small functions called for every row(Bruce)
7481 Multibyte regression tests(Soonmyung.)
7482 Remove --disabled options from configure
7483 Define PGDOC to use POSTGRESDIR by default
7484 Make regression optional
7485 Remove extra braces code to pgindent(Bruce)
7486 Add bsdi shared library support(Bruce)
7487 New --without-CXX support configure option(Brook)
7489 Update backend flowchart in tools/backend(Bruce)
7490 Change atttypmod from int16 to int32(Bruce, Tom)
7491 Getrusage() fix for platforms that do not have it(Tom)
7492 Add PQconnectdb, PGUSER, PGPASSWORD to libpq man page
7493 NS32K platform fixes(Phil Nelson, John Buller)
7494 SCO 7/UnixWare 2.x fixes(Billy,others)
7495 Sparc/Solaris 2.5 fixes(Ryan)
7496 Pgbuiltin.3 is obsolete, move to doc files(Thomas)
7497 Even more documention(Thomas)
7498 Nextstep support(Jacek)
7500 pginterface manual page(Bruce)
7501 shared libraries all have version numbers
7502 merged all OS-specific shared library defines into one file
7503 smarter TCL/TK configuration checking(Billy)
7504 smarter perl configuration(Brook)
7505 configure uses supplied install-sh if no install script found(Tom)
7506 new Makefile.shlib for shared library configuration(Tom)
7512 <sect1 id="release-6-3-2">
7513 <title>Release 6.3.2</title>
7516 <title>Release date</title>
7517 <simpara>1998-04-07</simpara>
7521 This is a bug-fix release for 6.3.x.
7522 Refer to the release notes for version 6.3 for a more complete summary of new features.
7530 Repairs automatic configuration support for some platforms, including Linux,
7531 from breakage inadvertently introduced in version 6.3.1.
7537 Correctly handles function calls on the left side of BETWEEN and LIKE clauses.
7544 A dump/restore is NOT required for those running 6.3 or 6.3.1. A
7545 <literal>make distclean</>, <literal>make</>, and <literal>make install</> is all that is required.
7546 This last step should be performed while the postmaster is not running.
7547 You should re-link any custom applications that use <productname>PostgreSQL</productname> libraries.
7550 For upgrades from pre-6.3 installations,
7551 refer to the installation and migration instructions for version 6.3.
7555 <title>Changes</title>
7559 Configure detection improvements for tcl/tk(Brook Milligan, Alvin)
7560 Manual page improvements(Bruce)
7561 BETWEEN and LIKE fix(Thomas)
7562 fix for psql \connect used by pg_dump(Oliver Elphick)
7564 pgaccess, version 0.86
7565 qsort removed, now uses libc version, cleanups(Jeroen)
7566 fix for buffer over-runs detected(Maurice Gittens)
7567 fix for buffer overrun in libpgtcl(Randy Kunkee)
7568 fix for UNION with DISTINCT or ORDER BY(Bruce)
7569 gettimeofday configure check(Doug Winterburn)
7570 Fix "indexes not used" bug(Vadim)
7571 docs additions(Thomas)
7572 Fix for backend memory leak(Bruce)
7573 libreadline cleanup(Erwan MAS)
7574 Remove DISTDIR(Bruce)
7575 Makefile dependency cleanup(Jeroen van Vianen)
7582 <sect1 id="release-6-3-1">
7583 <title>Release 6.3.1</title>
7586 <title>Release date</title>
7587 <simpara>1998-03-23</simpara>
7596 Additional support for multibyte character sets.
7602 Repair byte ordering for mixed-endian clients and servers.
7608 Minor updates to allowed SQL syntax.
7614 Improvements to the configuration autodetection for installation.
7621 A dump/restore is NOT required for those running 6.3. A
7622 <literal>make distclean</>, <literal>make</>, and <literal>make install</> is all that is required.
7623 This last step should be performed while the postmaster is not running.
7624 You should re-link any custom applications that use <productname>PostgreSQL</productname> libraries.
7627 For upgrades from pre-6.3 installations,
7628 refer to the installation and migration instructions for version 6.3.
7632 <title>Changes</title>
7636 ecpg cleanup/fixes, now version 1.1(Michael Meskes)
7637 pg_user cleanup(Bruce)
7638 large object fix for pg_dump and tclsh (alvin)
7639 LIKE fix for multiple adjacent underscores
7640 fix for redefining builtin functions(Thomas)
7642 upgrade to pg_access 0.83
7643 updated CLUSTER manual page
7644 multibyte character set support, see doc/README.mb(Tatsuo)
7645 configure --with-pgport fix
7647 big-endian fix for backend communications(Kataoka)
7648 SUBSTR() and substring() fix(Jan)
7649 several jdbc fixes(Peter)
7650 libpgtcl improvements, see libptcl/README(Randy Kunkee)
7651 Fix for "Datasize = 0" error(Vadim)
7652 Prevent \do from wrapping(Bruce)
7653 Remove duplicate Russian character set entries
7655 Allow optional TABLE key word in LOCK and SELECT INTO(Thomas)
7656 CREATE SEQUENCE options to allow a negative integer(Thomas)
7657 Add "PASSWORD" as an allowed column identifier(Thomas)
7658 Add checks for UNION target fields(Bruce)
7659 Fix Alpha port(Dwayne Bailey)
7660 Fix for text arrays containing quotes(Doug Gibson)
7661 Solaris compile fix(Albert Chin-A-Young)
7662 Better identify tcl and tk libs and includes(Bruce)
7668 <sect1 id="release-6-3">
7669 <title>Release 6.3</title>
7672 <title>Release date</title>
7673 <simpara>1998-03-01</simpara>
7677 There are <emphasis>many</emphasis> new features and improvements in this release.
7678 Here is a brief, incomplete summary:
7683 Many new SQL features, including
7684 full <acronym>SQL92</acronym> subselect capability
7685 (everything is here but target-list subselects).
7691 Support for client-side environment variables to specify time zone and date style.
7697 Socket interface for client/server connection. This is the default now
7698 so you may need to start <application>postmaster</application> with the
7699 <option>-i</option> flag.
7705 Better password authorization mechanisms. Default table privileges have changed.
7711 Old-style <firstterm>time travel</firstterm>
7712 has been removed. Performance has been improved.
7720 Bruce Momjian wrote the following notes to introduce the new release.
7725 There are some general 6.3 issues that I want to mention. These are
7726 only the big items that can not be described in one sentence. A review
7727 of the detailed changes list is still needed.
7730 First, we now have subselects. Now that we have them, I would like to
7731 mention that without subselects, SQL is a very limited language.
7732 Subselects are a major feature, and you should review your code for
7733 places where subselects provide a better solution for your queries. I
7734 think you will find that there are more uses for subselects than you may
7735 think. Vadim has put us on the big SQL map with subselects, and fully
7736 functional ones too. The only thing you can't do with subselects is to
7737 use them in the target list.
7740 Second, 6.3 uses Unix domain sockets rather than TCP/IP by default. To
7741 enable connections from other machines, you have to use the new
7742 postmaster -i option, and of course edit <filename>pg_hba.conf</filename>. Also, for this
7743 reason, the format of <filename>pg_hba.conf</filename> has changed.
7746 Third, <type>char()</type> fields will now allow faster access than <type>varchar()</type> or
7747 <type>text</type>. Specifically, the <type>text</> and <type>varchar()</type> have a penalty for access to
7748 any columns after the first column of this type. <type>char()</type> used to also
7749 have this access penalty, but it no longer does. This may suggest that
7750 you redesign some of your tables, especially if you have short character
7751 columns that you have defined as <type>varchar()</type> or <type>text</type>. This and other
7752 changes make 6.3 even faster than earlier releases.
7755 We now have passwords definable independent of any Unix file. There are
7756 new SQL USER commands.
7757 See the <citetitle>Administrator's Guide</citetitle> for more
7758 information. There is a new table, pg_shadow, which is used to store
7759 user information and user passwords, and it by default only SELECT-able
7760 by the <systemitem>postgres</systemitem> super-user. pg_user is now a view of pg_shadow, and is
7761 SELECT-able by PUBLIC. You should keep using pg_user in your
7762 application without changes.
7765 User-created tables now no longer have SELECT privilege to PUBLIC by
7766 default. This was done because the ANSI standard requires it. You can
7767 of course GRANT any privileges you want after the table is created.
7768 System tables continue to be SELECT-able by PUBLIC.
7771 We also have real deadlock detection code. No more sixty-second
7772 timeouts. And the new locking code implements a <acronym>FIFO</acronym> better, so there
7773 should be less resource starvation during heavy use.
7776 Many complaints have been made about inadequate documentation in previous
7777 releases. Thomas has put much effort into many new manuals for this
7778 release. Check out the doc/ directory.
7781 For performance reasons, time travel is gone, but can be implemented
7782 using triggers (see <filename>pgsql/contrib/spi/README</filename>). Please check out the new
7783 \d command for types, operators, etc. Also, views have their own
7784 privileges now, not based on the underlying tables, so privileges on
7785 them have to be set separately. Check <filename>/pgsql/interfaces</filename> for some new
7786 ways to talk to <productname>PostgreSQL</productname>.
7789 This is the first release that really required an explanation for
7790 existing users. In many ways, this was necessary because the new
7791 release removes many limitations, and the work-arounds people were using
7792 are no longer needed.
7796 <title>Migration to version 6.3</title>
7799 A dump/restore using <application>pg_dump</application>
7800 or <application>pg_dumpall</application>
7801 is required for those wishing to migrate data from any
7802 previous release of <productname>PostgreSQL</productname>.
7807 <title>Changes</title>
7813 Fix binary cursors broken by MOVE implementation(Vadim)
7814 Fix for tcl library crash(Jan)
7815 Fix for array handling, from Gerhard Hintermayer
7816 Fix acl error, and remove duplicate pqtrace(Bruce)
7817 Fix psql \e for empty file(Bruce)
7818 Fix for textcat on varchar() fields(Bruce)
7819 Fix for DBT Sendproc (Zeugswetter Andres)
7820 Fix vacuum analyze syntax problem(Bruce)
7821 Fix for international identifiers(Tatsuo)
7822 Fix aggregates on inherited tables(Bruce)
7823 Fix substr() for out-of-bounds data
7824 Fix for select 1=1 or 2=2, select 1=1 and 2=2, and select sum(2+2)(Bruce)
7825 Fix notty output to show status result. -q option still turns it off(Bruce)
7826 Fix for count(*), aggs with views and multiple tables and sum(3)(Bruce)
7828 Fix for PQtrace start/stop several times(Bruce)
7829 Fix a variety of locking problems like newer lock waiters getting
7830 lock before older waiters, and having readlock people not share
7831 locks if a writer is waiting for a lock, and waiting writers not
7832 getting priority over waiting readers(Bruce)
7833 Fix crashes in psql when executing queries from external files(James)
7834 Fix problem with multiple order by columns, with the first one having
7836 Use correct hash table support functions for float8 and int4(Thomas)
7837 Re-enable JOIN= option in CREATE OPERATOR statement (Thomas)
7838 Change precedence for boolean operators to match expected behavior(Thomas)
7839 Generate elog(ERROR) on over-large integer(Bruce)
7840 Allow multiple-argument functions in constraint clauses(Thomas)
7841 Check boolean input literals for 'true','false','yes','no','1','0'
7842 and throw elog(ERROR) if unrecognized(Thomas)
7843 Major large objects fix
7844 Fix for GROUP BY showing duplicates(Vadim)
7845 Fix for index scans in MergeJion(Vadim)
7849 Subselects with EXISTS, IN, ALL, ANY key words (Vadim, Bruce, Thomas)
7850 New User Manual(Thomas, others)
7851 Speedup by inlining some frequently-called functions
7852 Real deadlock detection, no more timeouts(Bruce)
7853 Add SQL92 "constants" CURRENT_DATE, CURRENT_TIME, CURRENT_TIMESTAMP,
7854 CURRENT_USER(Thomas)
7855 Modify constraint syntax to be SQL92-compliant(Thomas)
7856 Implement SQL92 PRIMARY KEY and UNIQUE clauses using indexes(Thomas)
7857 Recognize SQL92 syntax for FOREIGN KEY. Throw elog notice(Thomas)
7858 Allow NOT NULL UNIQUE constraint clause (each allowed separately before)(Thomas)
7859 Allow PostgreSQL-style casting ("::") of non-constants(Thomas)
7860 Add support for SQL3 TRUE and FALSE boolean constants(Thomas)
7861 Support SQL92 syntax for IS TRUE/IS FALSE/IS NOT TRUE/IS NOT FALSE(Thomas)
7862 Allow shorter strings for boolean literals (e.g. "t", "tr", "tru")(Thomas)
7863 Allow SQL92 delimited identifiers(Thomas)
7864 Implement SQL92 binary and hexadecimal string decoding (b'10' and x'1F')(Thomas)
7865 Support SQL92 syntax for type coercion of literal strings
7866 (e.g. "DATETIME 'now'")(Thomas)
7867 Add conversions for int2, int4, and OID types to and from text(Thomas)
7868 Use shared lock when building indexes(Vadim)
7869 Free memory allocated for an user query inside transaction block after
7870 this query is done, was turned off in <= 6.2.1(Vadim)
7871 New SQL statement CREATE PROCEDURAL LANGUAGE(Jan)
7872 New <productname>PostgreSQL</productname> Procedural Language (PL) backend interface(Jan)
7873 Rename pg_dump -H option to -h(Bruce)
7874 Add Java support for passwords, European dates(Peter)
7875 Use indexes for LIKE and ~, !~ operations(Bruce)
7876 Add hash functions for datetime and timespan(Thomas)
7877 Time Travel removed(Vadim, Bruce)
7878 Add paging for \d and \z, and fix \i(Bruce)
7879 Add Unix domain socket support to backend and to frontend library(Goran)
7880 Implement CREATE DATABASE/WITH LOCATION and initlocation utility(Thomas)
7881 Allow more SQL92 and/or <productname>PostgreSQL</productname> reserved words as column identifiers(Thomas)
7882 Augment support for SQL92 SET TIME ZONE...(Thomas)
7883 SET/SHOW/RESET TIME ZONE uses TZ backend environment variable(Thomas)
7884 Implement SET keyword = DEFAULT and SET TIME ZONE DEFAULT(Thomas)
7885 Enable SET TIME ZONE using TZ environment variable(Thomas)
7886 Add PGDATESTYLE environment variable to frontend and backend initialization(Thomas)
7887 Add PGTZ, PGCOSTHEAP, PGCOSTINDEX, PGRPLANS, PGGEQO
7888 frontend library initialization environment variables(Thomas)
7889 Regression tests time zone automatically set with "setenv PGTZ PST8PDT"(Thomas)
7890 Add pg_description table for info on tables, columns, operators, types, and
7892 Increase 16 char limit on system table/index names to 32 characters(Bruce)
7893 Rename system indexes(Bruce)
7894 Add 'GERMAN' option to SET DATESTYLE(Thomas)
7895 Define an "ISO-style" timespan output format with "hh:mm:ss" fields(Thomas)
7896 Allow fractional values for delta times (e.g. '2.5 days')(Thomas)
7897 Validate numeric input more carefully for delta times(Thomas)
7898 Implement day of year as possible input to date_part()(Thomas)
7899 Define timespan_finite() and text_timespan() functions(Thomas)
7900 Remove archive stuff(Bruce)
7901 Allow for a pg_password authentication database that is separate from
7902 the system password file(Todd)
7903 Dump ACLs, GRANT, REVOKE privileges(Matt)
7904 Define text, varchar, and bpchar string length functions(Thomas)
7905 Fix Query handling for inheritance, and cost computations(Bruce)
7906 Implement CREATE TABLE/AS SELECT (alternative to SELECT/INTO)(Thomas)
7907 Allow NOT, IS NULL, IS NOT NULL in constraints(Thomas)
7908 Implement UNIONs for SELECT(Bruce)
7909 Add UNION, GROUP, DISTINCT to INSERT(Bruce)
7910 varchar() stores only necessary bytes on disk(Bruce)
7911 Fix for BLOBs(Peter)
7912 Mega-Patch for JDBC...see README_6.3 for list of changes(Peter)
7913 Remove unused "option" from PQconnectdb()
7914 New LOCK command and lock manual page describing deadlocks(Bruce)
7915 Add new psql \da, \dd, \df, \do, \dS, and \dT commands(Bruce)
7916 Enhance psql \z to show sequences(Bruce)
7917 Show NOT NULL and DEFAULT in psql \d table(Bruce)
7918 New psql .psqlrc file start-up(Andrew)
7919 Modify sample start-up script in contrib/linux to show syslog(Thomas)
7920 New types for IP and MAC addresses in contrib/ip_and_mac(TomH)
7921 Unix system time conversions with date/time types in contrib/unixdate(Thomas)
7922 Update of contrib stuff(Massimo)
7923 Add Unix socket support to DBD::Pg(Goran)
7924 New python interface (PyGreSQL 2.0)(D'Arcy)
7925 New frontend/backend protocol has a version number, network byte order(Phil)
7926 Security features in pg_hba.conf enhanced and documented, many cleanups(Phil)
7927 CHAR() now faster access than VARCHAR() or TEXT
7928 ecpg embedded SQL preprocessor
7929 Reduce system column overhead(Vadmin)
7930 Remove pg_time table(Vadim)
7931 Add pg_type attribute to identify types that need length (bpchar, varchar)
7932 Add report of offending line when COPY command fails
7933 Allow VIEW privileges to be set separately from the underlying tables.
7934 For security, use GRANT/REVOKE on views as appropriate(Jan)
7935 Tables now have no default GRANT SELECT TO PUBLIC. You must
7936 explicitly grant such privileges.
7937 Clean up tutorial examples(Darren)
7941 Add new html development tools, and flow chart in /tools/backend
7942 Fix for SCO compiles
7943 Stratus computer port Robert Gillies
7944 Added support for shlib for BSD44_derived & i386_solaris
7945 Make configure more automated(Brook)
7946 Add script to check regression test results
7947 Break parser functions into smaller files, group together(Bruce)
7948 Rename heap_create to heap_create_and_catalog, rename heap_creatr
7949 to heap_create()(Bruce)
7950 Sparc/Linux patch for locking(TomS)
7951 Remove PORTNAME and reorganize port-specific stuff(Marc)
7952 Add optimizer README file(Bruce)
7953 Remove some recursion in optimizer and clean up some code there(Bruce)
7954 Fix for NetBSD locking(Henry)
7955 Fix for libptcl make(Tatsuo)
7957 Change IS TRUE, IS FALSE, ... to expressions using "=" rather than
7958 function calls to istrue() or isfalse() to allow optimization(Thomas)
7959 Various fixes NetBSD/Sparc related(TomH)
7960 Alpha linux locking(Travis,Ryan)
7961 Change elog(WARN) to elog(ERROR)(Bruce)
7962 FAQ for FreeBSD(Marc)
7963 Bring in the PostODBC source tree as part of our standard distribution(Marc)
7964 A minor patch for HP/UX 10 vs 9(Stan)
7965 New pg_attribute.atttypmod for type-specific info like varchar length(Bruce)
7966 UnixWare patches(Billy)
7967 New i386 'lock' for spinlock asm(Billy)
7968 Support for multiplexed backends is removed
7969 Start an OpenBSD port
7972 Add string functions to regression suite(Thomas)
7973 Expand a few function names formerly truncated to 16 characters(Thomas)
7974 Remove un-needed malloc() calls and replace with palloc()(Bruce)
7980 <sect1 id="release-6-2-1">
7981 <title>Release 6.2.1</title>
7984 <title>Release date</title>
7985 <simpara>1997-10-17</simpara>
7989 6.2.1 is a bug-fix and usability release on 6.2.
7997 Allow strings to span lines, per <acronym>SQL92</acronym>.
8003 Include example trigger function for inserting user names on table updates.
8010 This is a minor bug-fix release on 6.2.
8011 For upgrades from pre-6.2 systems, a full dump/reload is required.
8012 Refer to the 6.2 release notes for instructions.
8016 <title>Migration from version 6.2 to version 6.2.1</title>
8019 This is a minor bug-fix release. A dump/reload is not required from version 6.2,
8020 but is required from any release prior to 6.2.
8023 In upgrading from version 6.2, if you choose to dump/reload you will find that
8024 avg(money) is now calculated correctly. All other bug fixes take effect
8025 upon updating the executables.
8028 Another way to avoid dump/reload is to use the following SQL command
8029 from <command>psql</command> to update the existing system table:
8032 update pg_aggregate set aggfinalfn = 'cash_div_flt8'
8033 where aggname = 'avg' and aggbasetype = 790;
8037 This will need to be done to every existing database, including template1.
8042 <title>Changes</title>
8046 Allow TIME and TYPE column names(Thomas)
8047 Allow larger range of true/false as boolean values(Thomas)
8048 Support output of "now" and "current"(Thomas)
8049 Handle DEFAULT with INSERT of NULL properly(Vadim)
8050 Fix for relation reference counts problem in buffer manager(Vadim)
8051 Allow strings to span lines, like ANSI(Thomas)
8052 Fix for backward cursor with ORDER BY(Vadim)
8053 Fix avg(cash) computation(Thomas)
8054 Fix for specifying a column twice in ORDER/GROUP BY(Vadim)
8055 Documented new libpq function to return affected rows, PQcmdTuples(Bruce)
8056 Trigger function for inserting user names for INSERT/UPDATE(Brook Milligan)
8062 <sect1 id="release-6-2">
8063 <title>Release 6.2</title>
8066 <title>Release date</title>
8067 <simpara>1997-10-02</simpara>
8071 A dump/restore is required for those wishing to migrate data from
8072 previous releases of <productname>PostgreSQL</productname>.
8076 <title>Migration from version 6.1 to version 6.2</title>
8079 This migration requires a complete dump of the 6.1 database and a
8080 restore of the database in 6.2.
8083 Note that the <command>pg_dump</command> and <command>pg_dumpall</command> utility from 6.2 should be used
8084 to dump the 6.1 database.
8089 <title>Migration from version 1.<replaceable>x</> to version 6.2</title>
8092 Those migrating from earlier 1.* releases should first upgrade to 1.09
8093 because the COPY output format was improved from the 1.02 release.
8098 <title>Changes</title>
8104 Fix problems with pg_dump for inheritance, sequences, archive tables(Bruce)
8105 Fix compile errors on overflow due to shifts, unsigned, and bad prototypes
8106 from Solaris(Diab Jerius)
8107 Fix bugs in geometric line arithmetic (bad intersection calculations)(Thomas)
8108 Check for geometric intersections at endpoints to avoid rounding ugliness(Thomas)
8109 Catch non-functional delete attempts(Vadim)
8110 Change time function names to be more consistent(Michael Reifenberg)
8111 Check for zero divides(Michael Reifenberg)
8112 Fix very old bug which made rows changed/inserted by a command
8113 visible to the command itself (so we had multiple update of
8114 updated rows, etc.)(Vadim)
8115 Fix for SELECT null, 'fail' FROM pg_am (Patrick)
8116 SELECT NULL as EMPTY_FIELD now allowed(Patrick)
8117 Remove un-needed signal stuff from contrib/pginterface
8118 Fix OR (where x != 1 or x isnull didn't return rows with x NULL) (Vadim)
8119 Fix time_cmp function (Vadim)
8120 Fix handling of functions with non-attribute first argument in
8121 WHERE clauses (Vadim)
8122 Fix GROUP BY when order of entries is different from order
8123 in target list (Vadim)
8124 Fix pg_dump for aggregates without sfunc1 (Vadim)
8128 Default genetic optimizer GEQO parameter is now 8(Bruce)
8129 Allow use parameters in target list having aggregates in functions(Vadim)
8130 Added JDBC driver as an interface(Adrian & Peter)
8132 Return number of rows inserted/affected by INSERT/UPDATE/DELETE etc.(Vadim)
8133 Triggers implemented with CREATE TRIGGER (SQL3)(Vadim)
8134 SPI (Server Programming Interface) allows execution of queries inside
8136 NOT NULL implemented (SQL92)(Robson Paniago de Miranda)
8137 Include reserved words for string handling, outer joins, and unions(Thomas)
8138 Implement extended comments ("/* ... */") using exclusive states(Thomas)
8139 Add "//" single-line comments(Bruce)
8140 Remove some restrictions on characters in operator names(Thomas)
8141 DEFAULT and CONSTRAINT for tables implemented (SQL92)(Vadim & Thomas)
8142 Add text concatenation operator and function (SQL92)(Thomas)
8143 Support WITH TIME ZONE syntax (SQL92)(Thomas)
8144 Support INTERVAL unit TO unit syntax (SQL92)(Thomas)
8145 Define types DOUBLE PRECISION, INTERVAL, CHARACTER,
8146 and CHARACTER VARYING (SQL92)(Thomas)
8147 Define type FLOAT(p) and rudimentary DECIMAL(p,s), NUMERIC(p,s) (SQL92)(Thomas)
8148 Define EXTRACT(), POSITION(), SUBSTRING(), and TRIM() (SQL92)(Thomas)
8149 Define CURRENT_DATE, CURRENT_TIME, CURRENT_TIMESTAMP (SQL92)(Thomas)
8150 Add syntax and warnings for UNION, HAVING, INNER and OUTER JOIN (SQL92)(Thomas)
8151 Add more reserved words, mostly for SQL92 compliance(Thomas)
8152 Allow hh:mm:ss time entry for timespan/reltime types(Thomas)
8153 Add center() routines for lseg, path, polygon(Thomas)
8154 Add distance() routines for circle-polygon, polygon-polygon(Thomas)
8155 Check explicitly for points and polygons contained within polygons
8156 using an axis-crossing algorithm(Thomas)
8157 Add routine to convert circle-box(Thomas)
8158 Merge conflicting operators for different geometric data types(Thomas)
8159 Replace distance operator "<===>" with "<->"(Thomas)
8160 Replace "above" operator "!^" with ">^" and "below" operator "!|" with "<^"(Thomas)
8161 Add routines for text trimming on both ends, substring, and string position(Thomas)
8162 Added conversion routines circle(box) and poly(circle)(Thomas)
8163 Allow internal sorts to be stored in memory rather than in files(Bruce & Vadim)
8164 Allow functions and operators on internally-identical types to succeed(Bruce)
8165 Speed up backend start-up after profiling analysis(Bruce)
8166 Inline frequently called functions for performance(Bruce)
8167 Reduce open() calls(Bruce)
8168 psql: Add PAGER for \h and \?,\C fix
8169 Fix for psql pager when no tty(Bruce)
8170 New entab utility(Bruce)
8171 General trigger functions for referential integrity (Vadim)
8172 General trigger functions for time travel (Vadim)
8173 General trigger functions for AUTOINCREMENT/IDENTITY feature (Vadim)
8174 MOVE implementation (Vadim)
8178 HP-UX 10 patches (Vladimir Turin)
8179 Added SCO support, (Daniel Harris)
8180 MkLinux patches (Tatsuo Ishii)
8181 Change geometric box terminology from "length" to "width"(Thomas)
8182 Deprecate temporary unstored slope fields in geometric code(Thomas)
8183 Remove restart instructions from INSTALL(Bruce)
8184 Look in /usr/ucb first for install(Bruce)
8185 Fix c++ copy example code(Thomas)
8186 Add -o to psql manual page(Bruce)
8187 Prevent relname unallocated string length from being copied into database(Bruce)
8188 Cleanup for NAMEDATALEN use(Bruce)
8189 Fix pg_proc names over 15 chars in output(Bruce)
8190 Add strNcpy() function(Bruce)
8191 remove some (void) casts that are unnecessary(Bruce)
8192 new interfaces directory(Marc)
8193 Replace fopen() calls with calls to fd.c functions(Bruce)
8194 Make functions static where possible(Bruce)
8195 enclose unused functions in #ifdef NOT_USED(Bruce)
8196 Remove call to difftime() in timestamp support to fix SunOS(Bruce & Thomas)
8197 Changes for Digital Unix
8198 Portability fix for pg_dumpall(Bruce)
8199 Rename pg_attribute.attnvals to attdispersion(Bruce)
8200 "intro/unix" manual page now "pgintro"(Bruce)
8201 "built-in" manual page now "pgbuiltin"(Bruce)
8202 "drop" manual page now "drop_table"(Bruce)
8203 Add "create_trigger", "drop_trigger" manual pages(Thomas)
8204 Add constraints regression test(Vadim & Thomas)
8205 Add comments syntax regression test(Thomas)
8206 Add PGINDENT and support program(Bruce)
8207 Massive commit to run PGINDENT on all *.c and *.h files(Bruce)
8208 Files moved to /src/tools directory(Bruce)
8209 SPI and Trigger programming guides (Vadim & D'Arcy)
8215 <sect1 id="release-6-1-1">
8216 <title>Release 6.1.1</title>
8219 <title>Release date</title>
8220 <simpara>1997-07-22</simpara>
8224 <title>Migration from version 6.1 to version 6.1.1</title>
8227 This is a minor bug-fix release. A dump/reload is not required from version 6.1,
8228 but is required from any release prior to 6.1.
8229 Refer to the release notes for 6.1 for more details.
8234 <title>Changes</title>
8238 fix for SET with options (Thomas)
8239 allow pg_dump/pg_dumpall to preserve ownership of all tables/objects(Bruce)
8240 new psql \connect option allows changing usernames without changing databases
8241 fix for initdb --debug option(Yoshihiko Ichikawa))
8242 lextest cleanup(Bruce)
8244 fix date/time month boundary arithmetic(Thomas)
8245 fix timezone daylight handling for some ports(Thomas, Bruce, Tatsuo)
8246 timestamp overhauled to use standard functions(Thomas)
8247 other code cleanup in date/time routines(Thomas)
8248 psql's \d now case-insensitive(Bruce)
8249 psql's backslash commands can now have trailing semicolon(Bruce)
8250 fix memory leak in psql when using \g(Bruce)
8251 major fix for endian handling of communication to server(Thomas, Tatsuo)
8252 Fix for Solaris assembler and include files(Yoshihiko Ichikawa)
8253 allow underscores in usernames(Bruce)
8254 pg_dumpall now returns proper status, portability fix(Bruce)
8260 <sect1 id="release-6-1">
8261 <title>Release 6.1</title>
8264 <title>Release date</title>
8265 <simpara>1997-06-08</simpara>
8269 The regression tests have been adapted and extensively modified for the
8270 6.1 release of <productname>PostgreSQL</productname>.
8274 Three new data types (<type>datetime</type>, <type>timespan</type>, and <type>circle</type>) have been added to
8275 the native set of <productname>PostgreSQL</productname> types. Points, boxes, paths, and polygons
8276 have had their output formats made consistent across the data types.
8277 The polygon output in misc.out has only been spot-checked for correctness
8278 relative to the original regression output.
8282 <productname>PostgreSQL</productname> 6.1 introduces a new, alternate
8283 optimizer which uses <firstterm>genetic</firstterm>
8284 algorithms. These algorithms introduce a random behavior in the ordering
8285 of query results when the query contains multiple qualifiers or multiple
8286 tables (giving the optimizer a choice on order of evaluation). Several
8287 regression tests have been modified to explicitly order the results, and
8288 hence are insensitive to optimizer choices. A few regression tests are
8289 for data types which are inherently unordered (e.g. points and time
8290 intervals) and tests involving those types are explicitly bracketed with
8291 <command>set geqo to 'off'</command> and <command>reset geqo</command>.
8295 The interpretation of array specifiers (the curly braces around atomic
8296 values) appears to have changed sometime after the original regression
8297 tests were generated. The current <filename>./expected/*.out</filename> files reflect this
8298 new interpretation, which may not be correct!
8302 The float8 regression test fails on at least some platforms. This is due
8303 to differences in implementations of <function>pow()</function> and <function>exp()</function> and the signaling
8304 mechanisms used for overflow and underflow conditions.
8308 The <quote>random</> results in the random test should cause the
8309 <quote>random</quote> test to be <quote>failed</quote>, since the
8310 regression tests are evaluated using a simple diff. However,
8311 <quote>random</> does not seem to produce random results on my test
8312 machine (Linux/<application>gcc</>/i686).
8316 <title>Migration to version 6.1</title>
8319 This migration requires a complete dump of the 6.0 database and a
8320 restore of the database in 6.1.
8323 Those migrating from earlier 1.* releases should first upgrade to 1.09
8324 because the COPY output format was improved from the 1.02 release.
8329 <title>Changes</title>
8335 packet length checking in library routines
8336 lock manager priority patch
8337 check for under/over flow of float8(Bruce)
8338 multitable join fix(Vadim)
8339 SIGPIPE crash fix(Darren)
8340 large object fixes(Sven)
8341 allow btree indexes to handle NULLs(Vadim)
8342 timezone fixes(D'Arcy)
8343 select SUM(x) can return NULL on no rows(Thomas)
8344 internal optimizer, executor bug fixes(Vadim)
8345 fix problem where inner loop in < or <= has no rows(Vadim)
8346 prevent re-commuting join index clauses(Vadim)
8347 fix join clauses for multiple tables(Vadim)
8348 fix hash, hashjoin for arrays(Vadim)
8349 fix btree for abstime type(Vadim)
8350 large object fixes(Raymond)
8351 fix buffer leak in hash indexes (Vadim)
8352 fix rtree for use in inner scan (Vadim)
8353 fix gist for use in inner scan, cleanups (Vadim, Andrea)
8354 avoid unnecessary local buffers allocation (Vadim, Massimo)
8355 fix local buffers leak in transaction aborts (Vadim)
8356 fix file manager memmory leaks, cleanups (Vadim, Massimo)
8357 fix storage manager memmory leaks (Vadim)
8358 fix btree duplicates handling (Vadim)
8359 fix deleted rows reincarnation caused by vacuum (Vadim)
8360 fix SELECT varchar()/char() INTO TABLE made zero-length fields(Bruce)
8361 many psql, pg_dump, and libpq memory leaks fixed using Purify (Igor)
8365 attribute optimization statistics(Bruce)
8366 much faster new btree bulk load code(Paul)
8367 BTREE UNIQUE added to bulk load code(Vadim)
8368 new lock debug code(Massimo)
8369 massive changes to libpg++(Leo)
8370 new GEQO optimizer speeds table multitable optimization(Martin)
8371 new WARN message for non-unique insert into unique key(Marc)
8372 update x=-3, no spaces, now valid(Bruce)
8373 remove case-sensitive identifier handling(Bruce,Thomas,Dan)
8374 debug backend now pretty-prints tree(Darren)
8375 new Oracle character functions(Edmund)
8376 new plaintext password functions(Dan)
8377 no such class or insufficient privilege changed to distinct messages(Dan)
8378 new ANSI timestamp function(Dan)
8379 new ANSI Time and Date types (Thomas)
8380 move large chunks of data in backend(Martin)
8381 multicolumn btree indexes(Vadim)
8382 new SET var TO value command(Martin)
8383 update transaction status on reads(Dan)
8384 new locale settings for character types(Oleg)
8385 new SEQUENCE serial number generator(Vadim)
8386 GROUP BY function now possible(Vadim)
8387 re-organize regression test(Thomas,Marc)
8388 new optimizer operation weights(Vadim)
8389 new psql \z grant/permit option(Marc)
8390 new MONEY data type(D'Arcy,Thomas)
8391 tcp socket communication speed improved(Vadim)
8392 new VACUUM option for attribute statistics, and for certain columns (Vadim)
8393 many geometric type improvements(Thomas,Keith)
8394 additional regression tests(Thomas)
8395 new datestyle variable(Thomas,Vadim,Martin)
8396 more comparison operators for sorting types(Thomas)
8397 new conversion functions(Thomas)
8398 new more compact btree format(Vadim)
8399 allow pg_dumpall to preserve database ownership(Bruce)
8400 new SET GEQO=# and R_PLANS variable(Vadim)
8401 old (!GEQO) optimizer can use right-sided plans (Vadim)
8402 typechecking improvement in SQL parser(Bruce)
8403 new SET, SHOW, RESET commands(Thomas,Vadim)
8404 new \connect database USER option
8405 new destroydb -i option (Igor)
8406 new \dt and \di psql commands (Darren)
8407 SELECT "\n" now escapes newline (A. Duursma)
8408 new geometry conversion functions from old format (Thomas)
8412 new configuration script(Marc)
8413 readline configuration option added(Marc)
8414 OS-specific configuration options removed(Marc)
8415 new OS-specific template files(Marc)
8416 no more need to edit Makefile.global(Marc)
8417 re-arrange include files(Marc)
8418 nextstep patches (Gregor Hoffleit)
8419 removed Windows-specific code(Bruce)
8420 removed postmaster -e option, now only postgres -e option (Bruce)
8421 merge duplicate library code in front/backends(Martin)
8422 now works with eBones, international Kerberos(Jun)
8423 more shared library support
8424 c++ include file cleanup(Bruce)
8425 warn about buggy flex(Bruce)
8426 DG/UX, Ultrix, IRIX, AIX portability fixes
8432 <sect1 id="release-6-0">
8433 <title>Release 6.0</title>
8436 <title>Release date</title>
8437 <simpara>1997-01-29</simpara>
8441 A dump/restore is required for those wishing to migrate data from
8442 previous releases of <productname>PostgreSQL</productname>.
8446 <title>Migration from version 1.09 to version 6.0</title>
8449 This migration requires a complete dump of the 1.09 database and a
8450 restore of the database in 6.0.
8455 <title>Migration from pre-1.09 to version 6.0</title>
8458 Those migrating from earlier 1.* releases should first upgrade to 1.09
8459 because the COPY output format was improved from the 1.02 release.
8464 <title>Changes</title>
8470 ALTER TABLE bug - running postgress process needs to re-read table definition
8471 Allow vacuum to be run on one table or entire database(Bruce)
8473 Fix array over-runs of memory writes(Kurt)
8474 Fix elusive btree range/non-range bug(Dan)
8475 Fix for hash indexes on some types like time and date
8476 Fix for pg_log size explosion
8477 Fix permissions on lo_export()(Bruce)
8478 Fix unitialized reads of memory(Kurt)
8479 Fixed ALTER TABLE ... char(3) bug(Bruce)
8480 Fixed a few small memory leaks
8481 Fixed EXPLAIN handling of options and changed full_path option name
8482 Fixed output of group acl privileges
8483 Memory leaks (hunt and destroy with tools like Purify(Kurt)
8484 Minor improvements to rules system
8486 New asserts for run-checking
8487 Overhauled parser/analyze code to properly report errors and increase speed
8488 Pg_dump -d now handles NULL's properly(Bruce)
8489 Prevent SELECT NULL from crashing server (Bruce)
8490 Properly report errors when INSERT ... SELECT columns did not match
8491 Properly report errors when insert column names were not correct
8492 psql \g filename now works(Bruce)
8493 psql fixed problem with multiple statements on one line with multiple outputs
8494 Removed duplicate system OIDs
8495 SELECT * INTO TABLE . GROUP/ORDER BY gives unlink error if table exists(Bruce)
8496 Several fixes for queries that crashed the backend
8497 Starting quote in insert string errors(Bruce)
8498 Submitting an empty query now returns empty status, not just " " query(Bruce)
8502 Add EXPLAIN manual page(Bruce)
8503 Add UNIQUE index capability(Dan)
8504 Add hostname/user level access control rather than just hostname and user
8505 Add synonym of != for <>(Bruce)
8506 Allow "select oid,* from table"
8507 Allow BY,ORDER BY to specify columns by number, or by non-alias table.column(Bruce)
8508 Allow COPY from the frontend(Bryan)
8509 Allow GROUP BY to use alias column name(Bruce)
8510 Allow actual compression, not just reuse on the same page(Vadim)
8511 Allow installation-configuration option to auto-add all local users(Bryan)
8512 Allow libpq to distinguish between text value '' and null(Bruce)
8513 Allow non-postgres users with createdb privs to destroydb's
8514 Allow restriction on who can create C functions(Bryan)
8515 Allow restriction on who can do backend COPY(Bryan)
8516 Can shrink tables, pg_time and pg_log(Vadim & Erich)
8517 Change debug level 2 to print queries only, changed debug heading layout(Bruce)
8518 Change default decimal constant representation from float4 to float8(Bruce)
8519 European date format now set when postmaster is started
8520 Execute lowercase function names if not found with exact case
8521 Fixes for aggregate/GROUP processing, allow 'select sum(func(x),sum(x+y) from z'
8522 Gist now included in the distrubution(Marc)
8523 Idend authentication of local users(Bryan)
8524 Implement BETWEEN qualifier(Bruce)
8525 Implement IN qualifier(Bruce)
8526 libpq has PQgetisnull()(Bruce)
8527 libpq++ improvements
8528 New options to initdb(Bryan)
8529 Pg_dump allow dump of OIDs(Bruce)
8530 Pg_dump create indexes after tables are loaded for speed(Bruce)
8531 Pg_dumpall dumps all databases, and the user table
8532 Pginterface additions for NULL values(Bruce)
8533 Prevent postmaster from being run as root
8534 psql \h and \? is now readable(Bruce)
8535 psql allow backslashed, semicolons anywhere on the line(Bruce)
8536 psql changed command prompt for lines in query or in quotes(Bruce)
8537 psql char(3) now displays as (bp)char in \d output(Bruce)
8538 psql return code now more accurate(Bryan?)
8539 psql updated help syntax(Bruce)
8540 Re-visit and fix vacuum(Vadim)
8541 Reduce size of regression diffs, remove timezone name difference(Bruce)
8542 Remove compile-time parameters to enable binary distributions(Bryan)
8543 Reverse meaning of HBA masks(Bryan)
8544 Secure Authentication of local users(Bryan)
8545 Speed up vacuum(Vadim)
8546 Vacuum now had VERBOSE option(Bruce)
8550 All functions now have prototypes that are compared against the calls
8551 Allow asserts to be disabled easly from Makefile.global(Bruce)
8552 Change oid constants used in code to #define names
8553 Decoupled sparc and solaris defines(Kurt)
8554 Gcc -Wall compiles cleanly with warnings only from unfixable constructs
8555 Major include file reorganization/reduction(Marc)
8556 Make now stops on compile failure(Bryan)
8557 Makefile restructuring(Bryan, Marc)
8558 Merge bsdi_2_1 to bsdi(Bruce)
8559 Monitor program removed
8560 Name change from Postgres95 to PostgreSQL
8561 New config.h file(Marc, Bryan)
8562 PG_VERSION now set to 6.0 and used by postmaster
8563 Portability additions, including Ultrix, DG/UX, AIX, and Solaris
8564 Reduced the number of #define's, centeralized #define's
8565 Remove duplicate OIDS in system tables(Dan)
8566 Remove duplicate system catalog info or report mismatches(Dan)
8567 Removed many os-specific #define's
8568 Restructured object file generation/location(Bryan, Marc)
8569 Restructured port-specific file locations(Bryan, Marc)
8570 Unused/uninialized variables corrected
8576 <sect1 id="release-1-09">
8577 <title>Release 1.09</title>
8580 <title>Release date</title>
8581 <simpara>1996-11-04</simpara>
8585 Sorry, we didn't keep track of changes from 1.02 to 1.09. Some of
8586 the changes listed in 6.0 were actually included in the 1.02.1 to 1.09
8591 <sect1 id="release-1-02">
8592 <title>Release 1.02</title>
8595 <title>Release date</title>
8596 <simpara>1996-08-01</simpara>
8600 <title>Migration from version 1.02 to version 1.02.1</title>
8603 Here is a new migration file for 1.02.1. It includes the 'copy' change
8604 and a script to convert old <acronym>ASCII</acronym> files.
8608 The following notes are for the benefit of users who want to migrate
8609 databases from <productname>Postgres95</> 1.01 and 1.02 to <productname>Postgres95</> 1.02.1.
8612 If you are starting afresh with <productname>Postgres95</> 1.02.1 and do not need
8613 to migrate old databases, you do not need to read any further.
8618 In order to upgrade older <productname>Postgres95</> version 1.01 or 1.02 databases to
8619 version 1.02.1, the following steps are required:
8624 Start up a new 1.02.1 postmaster
8629 Add the new built-in functions and operators of 1.02.1 to 1.01 or 1.02
8630 databases. This is done by running the new 1.02.1 server against
8631 your own 1.01 or 1.02 database and applying the queries attached at
8632 the end of the file. This can be done easily through <command>psql</>. If your
8633 1.01 or 1.02 database is named <literal>testdb</literal> and you have cut the commands
8634 from the end of this file and saved them in <filename>addfunc.sql</filename>:
8636 % psql testdb -f addfunc.sql
8639 Those upgrading 1.02 databases will get a warning when executing the
8640 last two statements in the file because they are already present in 1.02. This is
8641 not a cause for concern.
8648 <title>Dump/Reload Procedure</title>
8651 If you are trying to reload a pg_dump or text-mode, <literal>copy tablename to
8652 stdout</literal> generated with a previous version, you will need to run the
8653 attached <command>sed</command> script on the ASCII file before loading it into the
8654 database. The old format used '.' as end-of-data, while '\.' is now the
8655 end-of-data marker. Also, empty strings are now loaded in as '' rather
8656 than NULL. See the copy manual page for full details.
8659 sed 's/^\.$/\\./g' <in_file >out_file
8663 If you are loading an older binary copy or non-<systemitem>stdout</> copy, there is no
8664 end-of-data character, and hence no conversion necessary.
8667 -- following lines added by agc to reflect the case-insensitive
8668 -- regexp searching for varchar (in 1.02), and bpchar (in 1.02.1)
8669 create operator ~* (leftarg = bpchar, rightarg = text, procedure = texticregexeq);
8670 create operator !~* (leftarg = bpchar, rightarg = text, procedure = texticregexne);
8671 create operator ~* (leftarg = varchar, rightarg = text, procedure = texticregexeq);
8672 create operator !~* (leftarg = varchar, rightarg = text, procedure = texticregexne);
8678 <title>Changes</title>
8682 Source code maintenance and development
8683 * worldwide team of volunteers
8684 * the source tree now in CVS at ftp.ki.net
8687 * psql (and underlying libpq library) now has many more options for
8688 formatting output, including HTML
8689 * pg_dump now output the schema and/or the data, with many fixes to
8690 enhance completeness.
8691 * psql used in place of monitor in administration shell scripts.
8692 monitor to be deprecated in next release.
8693 * date/time functions enhanced
8694 * NULL insert/update/comparison fixed/enhanced
8695 * TCL/TK lib and shell fixed to work with both tck7.4/tk4.0 and tcl7.5/tk4.1
8697 Bug Fixes (almost too numerous to mention)
8699 * storage management
8700 * check for NULL pointer before dereferencing
8704 * added SolarisX86 port
8705 * added BSD/OS 2.1 port
8710 Contributors (appologies to any missed)
8711 * Kurt J. Lidl <lidl@va.pubnix.com>
8712 (missed in first run, but no less important)
8713 * Erich Stamberger <eberger@gewi.kfunigraz.ac.at>
8714 * Jason Wright <jason@shiloh.vnet.net>
8715 * Cees de Groot <C.deGroot@inter.NL.net>
8716 * ernst.molitor@uni-bonn.de
8717 * michael.siebenborn@ae3.Hypo.DE (Michael Siebenborn (6929))
8718 * Brian E. Gallew <geek+@cmu.edu>
8719 * Vadim B. Mikheev <vadim@sable.krasnoyarsk.su>
8720 * Adam Sussman <myddryn@vidya.com>
8721 * Chris Dunlop <chris@onthe.net.au>
8722 * Marc G. Fournier <scrappy@ki.net>
8723 * Dan McGuirk <mcguirk@indirect.com>
8724 * Dr_George_D_Detlefsen <drgeorge@ilt.com>
8725 * Erich Stamberger <eberger@gewi.kfunigraz.ac.at>
8726 * Massimo Dal Zotto <dz@cs.unitn.it>
8727 * Randy Kunkee <kunkee@Starbase.NeoSoft.COM>
8728 * Rick Weldon <rick@wisetech.com>
8729 * Thomas van Reimersdahl <reimersd@dali.techinfo.rwth-aachen.de>
8730 * david bennett <dave@bensoft.com>
8731 * ernst.molitor@uni-bonn.de
8732 * Julian Assange <proff@suburbia.net>
8733 * Bruce Momjian <pgman@candle.pha.pa.us>
8734 * Paul "Shag" Walmsley <ccshag@cclabs.missouri.edu>
8735 * "Alistair G. Crooks" <azcb0@sde.uts.amdahl.com>
8740 <sect1 id="release-1-01">
8741 <title>Release 1.01</title>
8744 <title>Release date</title>
8745 <simpara>1996-02-23</simpara>
8750 <title>Migration from version 1.0 to version 1.01</title>
8753 The following notes are for the benefit of users who want to migrate
8754 databases from <productname>Postgres95</> 1.0 to <productname>Postgres95</> 1.01.
8757 If you are starting afresh with <productname>Postgres95</> 1.01 and do not need
8758 to migrate old databases, you do not need to read any further.
8761 In order to <productname>Postgres95</> version 1.01 with databases created with
8762 <productname>Postgres95</> version 1.0, the following steps are required:
8767 Set the definition of <symbol>NAMEDATALEN</symbol> in <filename>src/Makefile.global</filename> to 16
8768 and <symbol>OIDNAMELEN</symbol> to 20.
8773 Decide whether you want to use Host based authentication.
8778 If you do, you must create a file name <literal>pg_hba</literal> in your top-level data
8779 directory (typically the value of your <envar>$PGDATA</envar>). <filename>src/libpq/pg_hba</filename>
8780 shows an example syntax.
8785 If you do not want host-based authentication, you can comment out
8790 in <filename>src/Makefile.global</filename>
8793 Note that host-based authentication is turned on by default, and if
8794 you do not take steps A or B above, the out-of-the-box 1.01 will
8795 not allow you to connect to 1.0 databases.
8803 Compile and install 1.01, but DO NOT do the <command>initdb</command> step.
8808 Before doing anything else, terminate your 1.0 postmaster, and
8809 backup your existing <envar>$PGDATA</envar> directory.
8814 Set your <envar>PGDATA</envar> environment variable to your 1.0 databases, but set up
8815 path up so that 1.01 binaries are being used.
8820 Modify the file <filename><envar>$PGDATA</envar>/PG_VERSION</filename> from 5.0 to 5.1
8825 Start up a new 1.01 postmaster
8830 Add the new built-in functions and operators of 1.01 to 1.0
8831 databases. This is done by running the new 1.01 server against
8832 your own 1.0 database and applying the queries attached and saving
8833 in the file 1.0_to_1.01.sql. This can be done easily through <command>psql</command>.
8834 If your 1.0 database is name <literal>testdb</literal>:
8837 % psql testdb -f 1.0_to_1.01.sql
8840 and then execute the following commands (cut and paste from here):
8843 -- add builtin functions that are new to 1.01
8845 create function int4eqoid (int4, oid) returns bool as 'foo'
8846 language 'internal';
8847 create function oideqint4 (oid, int4) returns bool as 'foo'
8848 language 'internal';
8849 create function char2icregexeq (char2, text) returns bool as 'foo'
8850 language 'internal';
8851 create function char2icregexne (char2, text) returns bool as 'foo'
8852 language 'internal';
8853 create function char4icregexeq (char4, text) returns bool as 'foo'
8854 language 'internal';
8855 create function char4icregexne (char4, text) returns bool as 'foo'
8856 language 'internal';
8857 create function char8icregexeq (char8, text) returns bool as 'foo'
8858 language 'internal';
8859 create function char8icregexne (char8, text) returns bool as 'foo'
8860 language 'internal';
8861 create function char16icregexeq (char16, text) returns bool as 'foo'
8862 language 'internal';
8863 create function char16icregexne (char16, text) returns bool as 'foo'
8864 language 'internal';
8865 create function texticregexeq (text, text) returns bool as 'foo'
8866 language 'internal';
8867 create function texticregexne (text, text) returns bool as 'foo'
8868 language 'internal';
8870 -- add builtin functions that are new to 1.01
8872 create operator = (leftarg = int4, rightarg = oid, procedure = int4eqoid);
8873 create operator = (leftarg = oid, rightarg = int4, procedure = oideqint4);
8874 create operator ~* (leftarg = char2, rightarg = text, procedure = char2icregexeq);
8875 create operator !~* (leftarg = char2, rightarg = text, procedure = char2icregexne);
8876 create operator ~* (leftarg = char4, rightarg = text, procedure = char4icregexeq);
8877 create operator !~* (leftarg = char4, rightarg = text, procedure = char4icregexne);
8878 create operator ~* (leftarg = char8, rightarg = text, procedure = char8icregexeq);
8879 create operator !~* (leftarg = char8, rightarg = text, procedure = char8icregexne);
8880 create operator ~* (leftarg = char16, rightarg = text, procedure = char16icregexeq);
8881 create operator !~* (leftarg = char16, rightarg = text, procedure = char16icregexne);
8882 create operator ~* (leftarg = text, rightarg = text, procedure = texticregexeq);
8883 create operator !~* (leftarg = text, rightarg = text, procedure = texticregexne);
8891 <title>Changes</title>
8896 * 1.01 is backwards compatible with 1.0 database provided the user
8897 follow the steps outlined in the MIGRATION_from_1.0_to_1.01 file.
8898 If those steps are not taken, 1.01 is not compatible with 1.0 database.
8901 * added PQdisplayTuples() to libpq and changed monitor and psql to use it
8902 * added NeXT port (requires SysVIPC implementation)
8903 * added CAST .. AS ... syntax
8904 * added ASC and DESC key words
8905 * added 'internal' as a possible language for CREATE FUNCTION
8906 internal functions are C functions which have been statically linked
8907 into the postgres backend.
8908 * a new type "name" has been added for system identifiers (table names,
8909 attribute names, etc.) This replaces the old char16 type. The
8910 of name is set by the NAMEDATALEN #define in src/Makefile.global
8911 * a readable reference manual that describes the query language.
8912 * added host-based access control. A configuration file ($PGDATA/pg_hba)
8913 is used to hold the configuration data. If host-based access control
8914 is not desired, comment out HBA=1 in src/Makefile.global.
8915 * changed regex handling to be uniform use of Henry Spencer's regex code
8916 regardless of platform. The regex code is included in the distribution
8917 * added functions and operators for case-insensitive regular expressions.
8918 The operators are ~* and !~*.
8919 * pg_dump uses COPY instead of SELECT loop for better performance
8922 * fixed an optimizer bug that was causing core dumps when
8923 functions calls were used in comparisons in the WHERE clause
8924 * changed all uses of getuid to geteuid so that effective uids are used
8925 * psql now returns non-zero status on errors when using -c
8926 * applied public patches 1-14
8932 <sect1 id="release-1-0">
8933 <title>Release 1.0</title>
8936 <title>Release date</title>
8937 <simpara>1995-09-05</simpara>
8941 <title>Changes</title>
8946 * The copyright of <productname>Postgres</productname> 1.0 has been loosened to be freely modifiable
8947 and modifiable for any purpose. Please read the COPYRIGHT file.
8948 Thanks to Professor Michael Stonebraker for making this possible.
8951 * date formats have to be MM-DD-YYYY (or DD-MM-YYYY if you're using
8952 EUROPEAN STYLE). This follows SQL-92 specs.
8953 * "delimiters" is now a key word
8956 * sql LIKE syntax has been added
8957 * copy command now takes an optional USING DELIMITER specification.
8958 delimiters can be any single-character string.
8959 * IRIX 5.3 port has been added.
8960 Thanks to Paul Walmsley and others.
8961 * updated pg_dump to work with new libpq
8962 * \d has been added psql
8963 Thanks to Keith Parks
8964 * regexp performance for architectures that use POSIX regex has been
8965 improved due to caching of precompiled patterns.
8966 Thanks to Alistair Crooks
8967 * a new version of libpq++
8968 Thanks to William Wanders
8971 * arbitrary userids can be specified in the createuser script
8972 * \c to connect to other databases in psql now works.
8973 * bad pg_proc entry for float4inc() is fixed
8974 * users with usecreatedb field set can now create databases without
8975 having to be usesuper
8976 * remove access control entries when the entry no longer has any
8978 * fixed non-portable datetimes implementation
8979 * added kerberos flags to the src/backend/Makefile
8980 * libpq now works with kerberos
8981 * typographic errors in the user manual have been corrected.
8982 * btrees with multiple index never worked, now we tell you they don't
8983 work when you try to use them
8989 <sect1 id="release-0-03">
8990 <title><productname>Postgres95</productname> Release 0.03</title>
8993 <title>Release date</title>
8994 <simpara>1995-07-21</simpara>
8998 <title>Changes</title>
9001 Incompatible changes:
9002 * BETA-0.3 IS INCOMPATIBLE WITH DATABASES CREATED WITH PREVIOUS VERSIONS
9003 (due to system catalog changes and indexing structure changes).
9004 * double-quote (") is deprecated as a quoting character for string literals;
9005 you need to convert them to single quotes ('). <!-- " -->
9006 * name of aggregates (eg. int4sum) are renamed in accordance with the
9007 SQL standard (eg. sum).
9008 * CHANGE ACL syntax is replaced by GRANT/REVOKE syntax.
9009 * float literals (eg. 3.14) are now of type float4 (instead of float8 in
9010 previous releases); you might have to do typecasting if you depend on it
9011 being of type float8. If you neglect to do the typecasting and you assign
9012 a float literal to a field of type float8, you may get incorrect values
9014 * LIBPQ has been totally revamped so that frontend applications
9015 can connect to multiple backends
9016 * the usesysid field in pg_user has been changed from int2 to int4 to
9017 allow wider range of Unix user ids.
9018 * the netbsd/freebsd/bsd o/s ports have been consolidated into a
9019 single BSD44_derived port. (thanks to Alistair Crooks)
9021 SQL standard-compliance (the following details changes that makes postgres95
9022 more compliant to the SQL-92 standard):
9023 * the following SQL types are now built-in: smallint, int(eger), float, real,
9024 char(N), varchar(N), date and time.
9026 The following are aliases to existing postgres types:
9028 integer, int -> int4
9029 float, real -> float4
9030 char(N) and varchar(N) are implemented as truncated text types. In
9031 addition, char(N) does blank-padding.
9032 * single-quote (') is used for quoting string literals; '' (in addition to
9033 \') is supported as means of inserting a single quote in a string
9034 * SQL standard aggregate names (MAX, MIN, AVG, SUM, COUNT) are used
9035 (Also, aggregates can now be overloaded, i.e. you can define your
9036 own MAX aggregate to take in a user-defined type.)
9037 * CHANGE ACL removed. GRANT/REVOKE syntax added.
9038 - Privileges can be given to a group using the "GROUP" key word.
9040 GRANT SELECT ON foobar TO GROUP my_group;
9041 The key word 'PUBLIC' is also supported to mean all users.
9043 Privileges can only be granted or revoked to one user or group
9046 "WITH GRANT OPTION" is not supported. Only class owners can change
9048 - The default access control is to to grant users readonly access.
9049 You must explicitly grant insert/update access to users. To change
9050 this, modify the line in
9051 src/backend/utils/acl.h
9052 that defines ACL_WORLD_DEFAULT
9055 * the bug where aggregates of empty tables were not run has been fixed. Now,
9056 aggregates run on empty tables will return the initial conditions of the
9057 aggregates. Thus, COUNT of an empty table will now properly return 0.
9058 MAX/MIN of an empty table will return a row of value NULL.
9059 * allow the use of \; inside the monitor
9060 * the LISTEN/NOTIFY asynchronous notification mechanism now work
9061 * NOTIFY in rule action bodies now work
9062 * hash indexes work, and access methods in general should perform better.
9063 creation of large btree indexes should be much faster. (thanks to Paul
9066 Other changes and enhancements:
9067 * addition of an EXPLAIN statement used for explaining the query execution
9068 plan (eg. "EXPLAIN SELECT * FROM EMP" prints out the execution plan for
9070 * WARN and NOTICE messages no longer have timestamps on them. To turn on
9071 timestamps of error messages, uncomment the line in
9072 src/backend/utils/elog.h:
9073 /* define ELOG_TIMESTAMPS */
9074 * On an access control violation, the message
9075 "Either no such class or insufficient privilege"
9076 will be given. This is the same message that is returned when
9077 a class is not found. This dissuades non-privileged users from
9078 guessing the existence of privileged classes.
9079 * some additional system catalog changes have been made that are not
9080 visible to the user.
9083 * The -oid option has been added to the "pg_result" tcl command.
9084 pg_result -oid returns oid of the last row inserted. If the
9085 last command was not an INSERT, then pg_result -oid returns "".
9086 * the large object interface is available as pg_lo* tcl commands:
9087 pg_lo_open, pg_lo_close, pg_lo_creat, etc.
9089 Portability enhancements and New Ports:
9090 * flex/lex problems have been cleared up. Now, you should be able to use
9091 flex instead of lex on any platforms. We no longer make assumptions of
9092 what lexer you use based on the platform you use.
9093 * The Linux-ELF port is now supported. Various configuration have been
9094 tested: The following configuration is known to work:
9095 kernel 1.2.10, gcc 2.6.3, libc 4.7.2, flex 2.5.2, bison 1.24
9096 with everything in ELF format,
9099 * ipcclean added to the distribution
9100 ipcclean usually does not need to be run, but if your backend crashes
9101 and leaves shared memory segments hanging around, ipcclean will
9102 clean them up for you.
9105 * the user manual has been revised and libpq documentation added.
9111 <sect1 id="release-0-02">
9112 <title><productname>Postgres95</productname> Release 0.02</title>
9115 <title>Release date</title>
9116 <simpara>1995-05-25</simpara>
9120 <title>Changes</title>
9124 Incompatible changes:
9125 * The SQL statement for creating a database is 'CREATE DATABASE' instead
9126 of 'CREATEDB'. Similarly, dropping a database is 'DROP DATABASE' instead
9127 of 'DESTROYDB'. However, the names of the executables 'createdb' and
9128 'destroydb' remain the same.
9131 * pgperl - a Perl (4.036) interface to Postgres95
9132 * pg_dump - a utility for dumping out a postgres database into a
9133 script file containing query commands. The script files are in a ASCII
9134 format and can be used to reconstruct the database, even on other
9135 machines and other architectures. (Also good for converting
9136 a Postgres 4.2 database to Postgres95 database.)
9138 The following ports have been incorporated into postgres95-beta-0.02:
9139 * the NetBSD port by Alistair Crooks
9140 * the AIX port by Mike Tung
9141 * the Windows NT port by Jon Forrest (more stuff but not done yet)
9142 * the Linux ELF port by Brian Gallew
9144 The following bugs have been fixed in postgres95-beta-0.02:
9145 * new lines not escaped in COPY OUT and problem with COPY OUT when first
9147 * cannot type return to use the default user id in createuser
9148 * SELECT DISTINCT on big tables crashes
9149 * Linux installation problems
9150 * monitor doesn't allow use of 'localhost' as PGHOST
9151 * psql core dumps when doing \c or \l
9152 * the "pgtclsh" target missing from src/bin/pgtclsh/Makefile
9153 * libpgtcl has a hard-wired default port number
9154 * SELECT DISTINCT INTO TABLE hangs
9155 * CREATE TYPE doesn't accept 'variable' as the internallength
9156 * wrong result using more than 1 aggregate in a SELECT
9162 <sect1 id="release-0-01">
9163 <title><productname>Postgres95</productname> Release 0.01</title>
9166 <title>Release date</title>
9167 <simpara>1995-05-01</simpara>
9176 <sect1 id="timing-results">
9177 <title>Timing Results</title>
9180 These timing results are from running the regression test with the commands
9183 % cd src/test/regress
9189 Timing under Linux 2.0.27 seems to have a roughly 5% variation from run
9190 to run, presumably due to the scheduling vagaries of multitasking systems.
9194 <title>Version 6.5</title>
9197 As has been the case for previous releases, timing between
9198 releases is not directly comparable since new regression tests
9199 have been added. In general, 6.5 is faster than previous
9204 Timing with <function>fsync()</function> disabled:
9208 02:00 Dual Pentium Pro 180, 224MB, UW-SCSI, Linux 2.0.36, gcc 2.7.2.3 -O2 -m486
9209 04:38 Sparc Ultra 1 143MHz, 64MB, Solaris 2.6
9214 Timing with <function>fsync()</function> enabled:
9218 04:21 Dual Pentium Pro 180, 224MB, UW-SCSI, Linux 2.0.36, gcc 2.7.2.3 -O2 -m486
9221 For the <systemitem class="osname">Linux</systemitem> system above, using <acronym>UW-SCSI</acronym> disks rather than (older) <acronym>IDE</acronym>
9222 disks leads to a 50% improvement in speed on the regression test.
9227 <title>Version 6.4beta</title>
9230 The times for this release are not directly comparable to those for previous releases
9231 since some additional regression tests have been included.
9232 In general, however, 6.4 should be slightly faster than the previous release (thanks, Bruce!).
9237 02:26 Dual Pentium Pro 180, 96MB, UW-SCSI, Linux 2.0.30, gcc 2.7.2.1 -O2 -m486
9243 <title>Version 6.3</title>
9246 The times for this release are not directly comparable to those for previous releases
9247 since some additional regression tests have been included and some obsolete tests involving
9248 time travel have been removed.
9249 In general, however, 6.3 is substantially faster than previous releases (thanks, Bruce!).
9254 02:30 Dual Pentium Pro 180, 96MB, UW-SCSI, Linux 2.0.30, gcc 2.7.2.1 -O2 -m486
9255 04:12 Dual Pentium Pro 180, 96MB, EIDE, Linux 2.0.30, gcc 2.7.2.1 -O2 -m486
9261 <title>Version 6.1</title>
9266 06:12 Pentium Pro 180, 32MB, EIDE, Linux 2.0.30, gcc 2.7.2 -O2 -m486
9267 12:06 P-100, 48MB, Linux 2.0.29, gcc
9268 39:58 Sparc IPC 32MB, Solaris 2.5, gcc 2.7.2.1 -O -g
9276 <!-- Keep this comment at the end of the file
9281 sgml-minimize-attributes:nil
9282 sgml-always-quote-attributes:t
9285 sgml-parent-document:nil
9286 sgml-default-dtd-file:"./reference.ced"
9287 sgml-exposed-tags:nil
9288 sgml-local-catalogs:("/usr/lib/sgml/catalog")
9289 sgml-local-ecat-files:nil
projects / postgresql / blob