2 $PostgreSQL: pgsql/doc/src/sgml/release.sgml,v 1.298 2004/10/01 02:00:43 neilc Exp $
5 <appendix id="release">
6 <title>Release Notes</title>
8 <sect1 id="release-8-0">
9 <title>Release 8.0</title>
12 <title>Release date</title>
13 <simpara>2004-??-??, current as of 2004-08-06</simpara>
17 <title>Overview</title>
20 Major changes in this release:
30 This is the first <productname>PostgreSQL</productname>
31 release to natively run on Microsoft Windows as a server. It
32 can run as a Windows service. This release supports NT-based
33 Windows releases like Win2000, XP, Win2003. Older releases
34 like Windows 95, 98, and ME are not supported because these
35 operating systems do not have the infrastructure to support
36 <productname>PostgreSQL</productname>. A separate installer
37 project has been created to ease installation on Windows:
38 <ulink url="http://pgfoundry.org/projects/pginstaller">
39 http://pgfoundry.org/projects/pginstaller</ulink>.
42 Although tested throughout our release cycle, the Windows port
43 does not have the benefit of years of use in production
44 environments that PostgreSQL has on Unix platforms and
45 therefore should be treated with the same level of caution as
46 you would a new product.
49 Previous releases required the Unix emulation toolkit Cygwin for
50 Win32 server support. <productname>PostgreSQL</productname>
51 has always supported clients on Win32.
63 Savepoints allow specific parts of a transaction to be aborted
64 without affecting the remainder of the transaction. Prior
65 releases had no such capability; there was no way to recover
66 from a statement failure within a transaction except by
67 aborting the whole transaction. This feature is valuable for
68 application writers who require error recovery within a
76 Point-In-Time Recovery
81 Though <productname>PostgreSQL</productname> is very reliable,
82 in previous releases there was no way to recover from disk
83 drive failure except to restore from a previous backup or use
84 a standby replication server. Point-in-time recovery allows
85 continuous backup of the server. You can recover either to
86 the point of failure or to some transaction in the past.
98 Tablespaces allow administrators to select the file systems
99 used for storage of databases, schemas, tables, or
100 indexes. This improves performance and control over disk space
101 usage. Prior releases used <application>initlocation</> and
102 manual symlink management for such tasks.
109 Improved Buffer Management, <command>CHECKPOINT</command>,
110 <command>VACUUM</command>
115 This release has a more intelligent buffer replacement strategy,
116 which will make better use of available shared buffers and
117 improve performance. The performance impact of vacuum and
118 checkpoints is also lessened.
130 A column's data type can now be changed with <command>ALTER
138 New Perl Server-Side Language
143 A new version of the <application>plperl</> server-side language now
144 supports a persistent shared storage area, triggers, returning records
145 and arrays of records, and SPI calls to access the database.
152 CSV support in <command>COPY</command>
157 <command>COPY</command> can now read and write
158 comma-separated-value (CSV) files. It has the flexibility to
159 interpret non-standard quoting and separation characters too.
169 <title>Migration to version 8.0</title>
172 A dump/restore using <application>pg_dump</application> is
173 required for those wishing to migrate data from any previous
178 Observe the following incompatibilities:
184 Server configuration parameters <varname>virtual_host</> and
185 <varname>tcpip_socket</> have been replaced with a more general
186 parameter <varname>listen_addresses</>. Also, the server now listens on
187 <literal>localhost</> by default, which eliminates the need for the
188 <literal>-i</> postmaster switch in many scenarios.
194 Server configuration parameters <varname>SortMem</> and
195 <varname>VacuumMem</> have been renamed to <varname>work_mem</>
196 and <varname>maintenance_work_mem</> to better reflect their
197 use. The original names are still supported in
198 <command>SET</command> and <command>SHOW</command>.
204 Server configuration parameters <varname>log_pid</>,
205 <varname>log_timestamp</>, and <varname>log_source_port</> have been
206 removed now that a more flexible <varname>log_line_prefix</> has been
213 Server configuration parameter <varname>syslog</> has been removed and
214 replaced with a more logical <varname>log_destination</> variable to
215 control the log output destination.
221 Server configuration parameter <varname>log_statement</> has been
222 changed so it can selectively log just database modification or
223 data definition statements.
229 Server configuration parameter <varname>max_expr_depth</> parameter has
230 been replaced with <varname>max_stack_depth</> which measures the
231 physical stack size rather than the expression nesting depth. This
232 helps prevent session termination due to stack overflow caused by
239 The length() function no longer counts trailing spaces in
246 Casting an integer to BIT(N) selects the rightmost N bits of the
247 integer, not the leftmost N bits as before.
253 UPDATE-ing an element or slice of a NULL array value now produces
254 a non-NULL array result, namely an array containing
255 just the assigned-to positions.
261 The server now warns of empty strings passed to
262 <type>oid</type>/<type>float4</type>/<type>float8</type> data
263 types. In the next major release, doing this will generate an
270 The extract() function (also called date_part) now returns the proper
271 year for BC dates. It previously returned one less than the current
272 year. The function now also returns the proper values for millennium
279 psql's \copy command now reads or writes the query stdin/stdout, rather
280 than psql's stdin/stdout. The previous behavior can be accessed via new
281 pstdin/pstdout parameters.
286 <para> The JDBC client interface has been removed from the core
287 distribution, and is now hosted at <ulink url=
288 "http://jdbc.postgresql.org">http://jdbc.postgresql.org</ulink>.
289 The TCL client interface has also been removed. There are several
290 TCL interfaces now hosted at <ulink url=
291 "http://gborg.postgresql.org">http://gborg.postgresql.org</ulink>.
297 The server now uses its own time zone database, rather than the
298 one supplied by the operating system. This will provide consistent
299 behavior across all platforms. In most cases, there should be
300 little noticeable difference in time zone behavior, except that
301 the time zone names used by SET/SHOW <varname>TimeZone</> may
302 be different from what your platform provides.
308 <command>EXECUTE</command> now returns a completion tag that
309 matches the executed statement.
315 configure's threading option no longer requires users to run
316 tests or edit configuration files; threading options are now detected
323 Now that tablespaces have been implemented,
324 <application>initlocation</> has been removed.
330 Syntax checking of array input values has been tightened up
331 considerably. Junk that was previously allowed in odd places with
332 odd results now causes an ERROR. Empty-string element values must
333 now be written as <literal>""</>, rather than writing nothing.
334 Also changed behavior with respect to whitespace surrounding array
335 elements: trailing whitespace is now ignored, for symmetry with leading
336 whitespace (which has always been ignored).
342 In <literal>READ COMMITTED</> serialization mode, volatile functions
343 now see the results of concurrent transactions committed up to the
344 beginning of each statement within the function, rather than up to the
345 beginning of the interactive command that called the function.
351 Functions declared <literal>STABLE</> or <literal>IMMUTABLE</> always
352 use the snapshot of the calling query, and therefore do not see the
353 effects of actions taken after the calling query starts, whether in
354 their own transaction or other transactions. Such a function must be
355 read-only, too, meaning that it cannot use any SQL commands other than
362 Non-deferred AFTER triggers are now fired immediately after completion
363 of the triggering query, rather than upon finishing the current
364 interactive command. This makes a difference when the triggering query
365 occurred within a function: the trigger is invoked before the function
366 proceeds to its next operation.
374 <title>Deprecated Features</title>
377 Some aspects of <productname>PostgreSQL</productname>'s behavior
378 have been determined to be suboptimal. For the sake of backward
379 compatibility these have not been removed in 8.0, but they are
380 considered deprecated and will be removed in the next major
387 The 8.1 release will remove the function
388 <literal>to_char(interval)</literal>.
394 By default, tables in <productname>PostgreSQL</productname> 8.0
395 and earlier are created with OIDs. In the next release, this
396 will <emphasis>not</emphasis> be the case: to create a table
397 that contains OIDs, the <literal>WITH OIDS</literal> clause must
398 be specified or the <varname>default_with_oids</varname>
399 configuration parameter must be enabled. Users are encouraged to
400 explicitly specify <literal>WITH OIDS</literal> if their tables
401 require OIDs for compatibility with future releases of
402 <productname>PostgreSQL</productname>.
409 <title>Changes</title>
412 Below you will find a detailed account of the changes between
413 release 8.0 and the previous major release.
417 <title>Performance Improvements</title>
422 Support cross-data-type index usage (Tom)
425 Before this change, many queries would not use an index if the data
426 types did not match exactly. This improvement makes index usage more
427 intuitive and consistent.
433 New buffer replacement strategy that improves caching (Jan)
436 Prior releases used a least-recently-used (LRU) cache to keep
437 recently referenced pages in memory. The LRU algorithm
438 did not consider the number of times a specific cache entry was
439 accessed, so large table scans could force out useful cache pages.
440 The new cache algorithm uses four separate lists to track most
441 recently used and most frequently used cache pages and dynamically
442 optimize their replacement based on the work load. This should
443 lead to much more efficient use of the shared buffer cache.
444 Administrators who have tested shared buffer sizes in the past
445 should retest with this new cache replacement policy.
451 Add subprocess to write dirty buffers periodically to reduce checkpoint writes (Jan)
454 In previous releases, the checkpoint process, which runs every few
455 minutes, would write all dirty buffers to the operating system's
456 buffer cache then flush all dirty operating system buffers to
457 disk. This often resulted in a periodic spike in disk usage that
458 hurt performance. The new code uses a background writer to trickle
459 disk writes at a steady pace so checkpoints have far fewer dirty
460 pages to write to disk. Also, the new code does not issue a global
461 sync() call, but instead fsync()s just the files written since the
462 last checkpoint. This should improve performance and
463 minimize degradation during checkpoints.
469 Add ability to prolong vacuum to reduce performance impact (Jan)
472 On busy systems, <command>VACUUM</command> performs many I/O
473 requests which can hurt performance for other users. This
474 release allows you to slow down <command>VACUUM</command> to
475 reduce its impact on other users, though this increases the
476 total duration of <command>VACUUM</command>.
482 Improve btree index performance for duplicate keys (Dmitry Tkach, Tom)
485 This improves the way indexes are scanned when many duplicate
486 values exist in the index.
492 Improved index usage with OR clauses (Tom)
495 This allows the optimizer to use indexes in statements with many OR
496 clauses that would not have been indexed in the past. It can also use
497 multi-column indexes where the first column is specified and the second
498 column is part of an OR clause.
504 Improve matching of partial index clauses (Tom)
507 The server is now smarter about using partial indexes in queries
508 involving complex WHERE clauses.
514 Improve performance of the GEQO optimizer (Tom)
517 The GEQO optimizer is used to plan queries involving many tables (by
518 default, twelve or more). This release speeds up the way queries are
519 analyzed to decrease time spent in optimization.
525 Miscellaneous optimizer improvements
528 There is not room here to list all the minor improvements made, but
529 numerous special cases work better than in prior releases.
535 Improve lookup speed for C functions (Tom)
538 This release uses a hash table to lookup information for dynamically
539 loaded C functions. This improves their speed so they perform nearly as
540 quickly as functions that are built into the server executable.
546 Add type-specific <command>ANALYZE</command> statistics
547 capability (Mark Cave-Ayland)
550 This feature allows more flexibility in generating statistics
551 for non-standard data types.
557 Allow collection of <command>ANALYZE</command> statistics for
558 expression indexes (Tom)
561 Expression indexes (also called functional indexes) allow users to
562 index not just columns but the results of expressions and function
563 calls. With this release, the optimizer can gather and use statistics
564 about the contents of expression indexes. This will greatly improve
565 the quality of planning for queries in which an expression index is
572 New two-stage sampling method for <command>ANALYZE</command>
576 This gives better statistics for asymmetric data distributions.
582 Speed up <command>TRUNCATE</command> (Tom)
585 This buys back some of the performance loss observed in 7.4, while still
586 keeping <command>TRUNCATE</command> transaction-safe.
595 <title>Server Changes</title>
600 Add WAL file archiving and point-in-time recovery (Simon Riggs)
606 Add tablespaces so admins can control disk layout (Gavin)
612 Add a built-in log rotation program (Andreas Pflug)
615 It is now possible to log server messages conveniently without
616 relying on either syslog or an external log rotation program.
622 Add new read-only server configuration parameters to query server
623 compile-time settings: <varname>block_size</>,
624 <varname>integer_datetimes</>, <varname>max_function_args</>,
625 <varname>max_identifier_length</>, <varname>max_index_keys</> (Joe)
631 Make quoting of "sameuser", "samegroup", and "all" remove
632 special meaning of these terms in pg_hba.conf (Andrew)
638 Use clearer IPv6 name <literal>::1/128</> for localhost in default
645 Rename server configuration parameters <varname>SortMem</> and
646 <varname>VacuumMem</> to <varname>work_mem</> and
647 <varname>maintenance_work_mem</> (Old names still supported) (Tom)
650 This change was made to clarify that bulk operations such as index and
651 foreign key creation use <varname>maintenance_work_mem</>, while
652 <varname>work_mem</> is for workspaces used during query execution.
658 Allow logging of session disconnections using server configuration
659 <varname>log_disconnections</> (Andrew)
665 Add new server configuration parameter <varname>log_line_prefix</> to
666 allow control of information emitted in each log line (Andrew)
669 Available information includes user name, database name, remote IP
670 address, and session start time.
676 Remove server configuration parameters <varname>log_pid</>,
677 <varname>log_timestamp</>, <varname>log_source_port</>; functionality
678 superseded by <varname>log_line_prefix</> (Andrew)
684 Replace the <varname>virtual_host</> and <varname>tcpip_socket</>
685 parameters with a unified <varname>listen_addresses</> parameter
692 Listen on localhost by default, which eliminates the need for the
693 <literal>-i</> postmaster switch in many scenarios (Andrew)
696 Listening on localhost (127.0.0.1) opens no new security holes but
697 allows configurations like Win32 and JDBC, which do not support
698 local sockets, to work without special adjustments.
704 Remove <varname>syslog</> server configuration parameter, and add more
705 logical <varname>log_destination</> variable to control log output
712 Change server configuration parameter <varname>log_statement</> to take
713 values <varname>all</>, <varname>mod</>, <varname>ddl</>, or
714 <varname>none</> to select which queries are logged (Bruce)
717 This allows administrators to log only data definition changes or
718 only data modification statements.
724 Allow configuration files to be placed outside the data directory (mlw)
727 By default, configuration files sit in the top server directory.
728 With this addition, configuration files can be placed outside the
729 data directory, easing administration.
735 Plan prepared queries only when first executed so constants can be
736 used for statistics (Oliver Jowett)
739 Prepared statements plan queries once and execute them many
740 times. While prepared queries avoid the overhead of re-planning
741 on each use, the quality of the plan suffers from not knowing the exact
742 parameters to be used in the query. In this release, planning of
743 unnamed prepared statements is delayed until the first execution,
744 and the actual parameter values of that execution are used as
751 Allow <command>DECLARE CURSOR</command> to take parameters
755 It is now useful to issue <command>DECLARE CURSOR</command> in a
756 Parse message with parameters. The parameter values sent at
757 Bind time will be substituted into the execution of the cursor's
764 Fix hash joins and aggregates of <type>inet</type> and
765 <type>cidr</type> data types (Tom)
768 Release 7.4 handled hashing of mixed <type>inet</type> and
769 <type>cidr</type> values incorrectly. (This bug did not exist
770 in prior releases because they wouldn't try to hash either
780 <title>Query Changes</title>
785 Add savepoints (nested transactions) (Alvaro)
791 Unsupported isolation levels are now accepted and promoted to the
792 nearest supported level (Peter)
795 The SQL specification states that if a database doesn't support a
796 specific isolation level, it should use the next more restrictive level.
797 This change complies with that recommendation.
803 Allow <command>BEGIN WORK</command> to specify transaction
804 isolation levels like <command>START TRANSACTION</command> does
811 Fix table permission checking for cases in which rules generate
812 a query type different from the originally submitted query (Tom)
818 Implement dollar quoting to simplify single-quote usage (Andrew, Tom,
822 In previous releases, because single quotes had to be used to
823 quote a function's body, the use of single quotes inside the
824 function text required use of two single quotes or other error-prone
825 notations. With this release we add the ability to use "dollar
826 quoting" to quote a block of text. The ability to use different
827 quoting delimiters at different nesting levels greatly simplifies
828 the task of quoting correctly, especially in complex functions.
829 Dollar quoting can be used anywhere quoted text is needed.
835 Make <literal>CASE val WHEN compval1 THEN ...</> evaluate <literal>val</> only once (Tom)
838 CASE no longer evaluates the test expression multiple
839 times. This has benefits when the expression is complex or is
846 Test HAVING before computing target list of an aggregate query (Tom)
849 Fixes improper failure of cases such as SELECT SUM(win)/SUM(lose)
850 ... GROUP BY ... HAVING SUM(lose) > 0. This should work but formerly
851 could fail with divide-by-zero.
857 Replace <varname>max_expr_depth</> parameter with
858 <varname>max_stack_depth</> parameter, measured in kilobytes of stack
862 This gives us a fairly bulletproof defense against crashing due to
863 runaway recursive functions. Instead of measuring the depth of expression
864 nesting, we now directly measure the size of the execution stack.
870 Allow arbitrary row expressions (Tom)
873 This release allows SQL expressions to contain arbitrary composite
874 types, that is, row values. It also allows functions to more easily
875 take rows as arguments and return row values.
881 Allow LIKE/ILIKE to be used as the operator in row and subselect
882 comparisons (Fabien Coelho)
888 Avoid locale-specific case conversion of basic ASCII letters in
889 identifiers and keywords (Tom)
892 This solves the <quote>Turkish problem</> with mangling of words
893 containing <literal>I</> and <literal>i</>. Folding of characters
894 outside the 7-bit-ASCII set is still locale-aware.
900 Improve syntax error reporting (Fabien, Tom)
903 Syntax error reports are more useful than before.
909 Change <command>EXECUTE</command> to return a completion tag
910 matching the executed statement (Kris Jurka)
913 Previous releases return an <command>EXECUTE</command> tag for
914 any <command>EXECUTE</command> call. In this release, the tag
915 returned will reflect the command executed.
921 Avoid emitting NATURAL CROSS JOIN in rule listings (Tom)
924 Such a clause makes no logical sense, but in some cases the rule
925 decompiler formerly produced this syntax.
934 <title>Object Manipulation Changes</title>
939 Add COMMENT ON casts, conversions, languages, operator classes, and
940 large objects (Christopher)
946 Add new server configuration parameter <varname>default_with_oids</> to
947 control whether tables are created with OIDs by default (Neil)
950 This allows administrators to default all <command>CREATE
951 TABLE</command> commands to create tables without OID columns.
957 Add WITH / WITHOUT OIDS clause to <command>CREATE TABLE
964 Allow ALTER TABLE DROP COLUMN to drop an OID column
965 (ALTER TABLE SET WITHOUT OIDS still works) (Tom)
971 Allow composite types as table columns (Tom)
977 Allow ALTER ... ADD COLUMN with defaults and NOT NULL constraints;
978 works per SQL spec (Rod)
981 It is now possible for ADD COLUMN to create a column that is not
982 initially filled with NULLs, but with a specified default value.
988 Add ALTER COLUMN TYPE to change column's type (Rod)
991 It is now possible to alter a column's datatype without dropping
992 and re-adding the column.
998 Allow multiple ALTER actions in a single <command>ALTER
999 TABLE</command> command (Rod)
1002 This is particularly useful for ALTER commands that rewrite the
1003 table (which include ALTER COLUMN TYPE and ADD COLUMN with a default).
1004 By grouping ALTER commands together, the table need be
1005 rewritten only once.
1011 Allow <command>ALTER TABLE</command> to add <type>serial</type>
1015 This is related to the new capability of adding defaults for new
1022 Allow changing the owners of aggregates, conversions, databases,
1023 functions, operators, operator classes, schemas, types, and tablespaces
1024 (Christopher, Euler Taveira de Oliveira)
1027 Previously this required modifying the system tables directly.
1033 Allow temporary object creation to be limited to SECURITY DEFINER
1034 functions (Sean Chittenden)
1040 Add ALTER TABLE ... SET WITHOUT CLUSTER (Christopher)
1043 Prior to this release, there was no way to clear an auto-cluster
1044 specification except to modify the system tables.
1050 Constraint/Index/SERIAL names are now table_column_type with numbers
1051 appended to guarantee uniqueness within the schema (Tom)
1054 The SQL specification states that such names should be unique
1061 Add pg_get_serial_sequence() to return a serial column's
1062 sequence name (Christopher)
1065 This allows automated scripts to reliabily find the serial
1072 Warn when primary/foreign key datatype mismatch requires costly lookup
1080 <title>Utility Command Changes</title>
1085 Allow <command>CREATE SCHEMA</command> to create triggers,
1086 indexes, and sequences (Neil)
1092 Add ALSO keyword to <command>CREATE RULE</command> (Fabien
1096 This allows ALSO to be added to rule creation to contrast it with
1103 Add NOWAIT option to <command>LOCK</command> (Tatsuo)
1106 This allows the <command>LOCK</command> command to fail if it
1107 would have to wait for the requested lock.
1113 Allow <command>COPY</command> to read and write
1114 comma-separated-value (CSV) files (Andrew, Bruce)
1120 Generate error if the <command>COPY</command> delimiter and NULL
1121 string conflict (Bruce)
1127 <command>GRANT</command>/<command>REVOKE</command> behavior
1128 follows the SQL spec more closely
1134 Avoid locking conflict between <command>CREATE INDEX</command>
1135 and <command>CHECKPOINT</command> (Tom)
1138 In 7.3 and 7.4, a long-running btree index build could block concurrent
1139 CHECKPOINTs from completing, thereby causing WAL bloat because the
1140 WAL log could not be recycled.
1146 Database-wide <command>ANALYZE</command> does not hold locks
1150 This reduces the potential for deadlocks against other backends
1151 that want exclusive locks on tables. To get the benefit of this
1152 change, do not execute database-wide <command>ANALYZE</command>
1153 inside a transaction block (<command>BEGIN</command> block); it
1154 must be able to commit and start a new transaction for each
1161 Erase MD5 user passwords when a user is renamed (Bruce)
1164 <productname>PostgreSQL</productname> uses the user name as salt
1165 when encrypting passwords via MD5. When a user name is changed,
1166 their salt no longer matches the stored MD5 password, so the
1167 stored password becomes useless. In this release a notice is
1168 generated and the password is cleared. A new password must then
1175 New pg_ctl <literal>kill</> option for Win32 (Andrew)
1178 Win32 does not have a <literal>kill</> command to send signals to
1179 backends so this capability was added to pg_ctl.
1185 Information schema improvements
1191 Add --pwfile option to <application>initdb</application> so the
1192 initial password can be set by GUI tools (Magnus)
1198 Detect locale/encoding mismatch in
1199 <application>initdb</application> (Peter)
1208 <title>Data Type and Function Changes</title>
1213 More complete support for composite types (row types) (Tom)
1216 Composite values can be used in many places where only scalar values
1223 Reject non-rectangular array values as erroneous (Joe)
1226 Formerly, <literal>array_in</literal> would silently build a
1233 Syntax checking of array input values considerably tightened up (Joe)
1236 Junk that was previously allowed in odd places with odd results now
1237 causes an ERROR: for example, non-whitespace after the closing
1244 Empty-string array element values must now be written as
1245 <literal>""</>, rather than writing nothing (Joe)
1248 Formerly, both ways of writing an empty-string element value were
1249 allowed, but now a quoted empty string is required. The case where
1250 nothing at all appears will probably be taken over to mean a NULL
1251 element value in some future release.
1257 Array element trailing whitespace is now ignored (Joe)
1260 Formerly leading whitespace was ignored, but trailing whitespace
1261 between an element value and the delimiter or right brace was
1262 significant. Now trailing whitespace is also ignored.
1268 Emit array values with explicit array bounds when lower bound is not one
1275 Accept YYYY-monthname-DD as a date string (Tom)
1281 Make netmask() and hostmask() functions return maximum-length
1288 Change factorial function to return <type>numeric</type> (Gavin)
1291 Returning <type>numeric</type> allows the factorial function to
1292 work for a wider range of input values.
1298 to_char/to_date date conversion improvements (Kurt Roeckx)
1304 Make length() disregard trailing spaces in CHAR(n) (Gavin)
1307 This change was made to improve consistency: trailing spaces are
1308 semantically insignificant in CHAR(n) data, so they should not be
1309 counted by length().
1315 Warn of empty string being passed to oid/float4/float8 data types (Neil)
1318 8.1 will throw an error instead.
1324 Allow int2/int4/int8/float4/float8 input routines to have leading
1325 or trailing whitespace (Neil)
1331 Better support for IEEE Infinity and NaN values in
1332 <type>float4</type>, <type>float8</type> (Neil)
1335 These should now work on all platforms that support IEEE-compliant
1336 floating point arithmetic.
1342 Add "week" to date_trunc options (Robert Creager)
1348 Fix to_char for 1 BC (previously it returned 1 AD) (Bruce)
1354 Fix to_char(year) for BC dates (previously it returned one less than
1355 the correct year) (Bruce)
1361 Fix date_part() to return the proper millennium and century (Fabien
1365 In previous versions, the century and millennium results had a wrong
1366 number and started in the wrong year, as compared to standard
1367 reckoning of such things.
1373 Add ceiling() as an alias for ceil(), and power() as an alias for pow()
1374 for standards compliance (Neil)
1380 Change ln(), log(), power(), and sqrt() to emit the correct SQLSTATE
1381 error codes for certain error conditions, as specified by SQL2003
1388 Add width_bucket() function as defined by SQL2003 (Neil)
1394 Add generate_series() functions to simplify working with numeric sets (Joe)
1400 Fix upper/lower/initcap functions to work with multibyte encodings (Tom)
1406 Add boolean and bitwise integer AND/OR aggregates (Fabien Coelho)
1412 New session information functions to return network addresses for client
1413 and server (Sean Chittenden)
1419 Add function to determine the area of a closed path (Sean Chittenden)
1425 Add function to send cancel request to other backends (Magnus)
1431 Add interval plus datetime operators (Tom)
1434 The reverse ordering, datetime plus interval, was already supported,
1435 but both are required by the SQL standard.
1441 Casting an integer to BIT(N) selects the rightmost N bits of the integer
1445 In prior releases, the leftmost N bits were selected, but this was
1446 deemed unhelpful, not to mention inconsistent with casting from bit
1456 <title>Server-Side Language Changes</title>
1461 In <literal>READ COMMITTED</> serialization mode, volatile functions
1462 now see the results of concurrent transactions committed up to the
1463 beginning of each statement within the function, rather than up to the
1464 beginning of the interactive command that called the function.
1470 Functions declared <literal>STABLE</> or <literal>IMMUTABLE</> always
1471 use the snapshot of the calling query, and therefore do not see the
1472 effects of actions taken after the calling query starts, whether in
1473 their own transaction or other transactions. Such a function must be
1474 read-only, too, meaning that it cannot use any SQL commands other than
1481 Non-deferred AFTER triggers are now fired immediately after completion
1482 of the triggering query, rather than upon finishing the current
1483 interactive command. This makes a difference when the triggering query
1484 occurred within a function: the trigger is invoked before the function
1485 proceeds to its next operation. For example, if a function inserts
1486 a new row into a table, any non-deferred foreign key checks occur
1487 before proceeding with the function.
1493 Allow function parameters to be declared with names (Dennis Bjorklund)
1496 This allows better documentation of functions. Whether the names
1497 actually do anything depends on the specific function language
1504 Allow PL/pgSQL parameter names to be referenced in the function (Dennis Bjorklund)
1507 This basically creates an automatic alias for each named parameter.
1513 Do minimal syntax checking of PL/pgSQL functions at creation time (Tom)
1516 This allows us to catch simple syntax errors sooner.
1522 More support for composite types (row and record variables) in PL/pgSQL
1525 For example, it now works to pass a rowtype variable to another function
1526 as a single variable.
1532 Default values for PL/pgSQL variables can now reference previously
1539 Improve parsing of PL/pgSQL FOR loops (Tom)
1542 Parsing is now driven by presence of ".." rather than datatype of FOR
1543 variable. This makes no difference for correct functions, but should
1544 result in more understandable error messages when a mistake is made.
1550 Major overhaul of PL/Perl server-side language (Command Prompt, Andrew Dunstan)
1559 <title>psql Changes</title>
1564 Improve psql information display about database objects (Christopher)
1570 Allow psql to display group membership in \du and \dg (Markus Bertheau)
1576 Prevent psql \dn from showing temporary schemas (Bruce)
1582 Allow psql to handle tilde user expansion for file names (Zach Irmen)
1588 Allow psql to display fancy prompts, including color, via readline
1589 (Reece Hart, Chet Ramey)
1595 Make psql \copy match <command>COPY</command> command syntax
1602 Show the location of syntax errors (Fabien Coelho, Tom)
1608 Add <command>CLUSTER</command> information to psql \d display
1615 Change psql \copy stdin/stdout to read from command input/output (Bruce)
1621 Add pstdin/pstdout to read from psql's stdin/stdout (Mark Feit)
1627 Add global psql config file, <filename>psqlrc.sample</filename>
1631 This allows a central file where global psql startup commands can
1638 Have psql \d+ indicate if the table has an OID column (Neil)
1644 On Windows, use binary mode in psql when reading files so control-Z
1645 is not seen as end-of-file
1651 Have \dn+ show permissions and description for schemas (Dennis
1661 <title>pg_dump Changes</title>
1666 Use dependency information to improve the reliability of pg_dump (Tom)
1672 Have pg_dump output objects in alphabetical order if possible (Tom)
1675 This should make it easier to identify changes between
1682 Allow pg_restore to ignore some SQL errors (Fabien Coelho)
1685 This makes pg_restore's behavior similar to the results of feeding
1686 a pg_dump output script to psql. In most cases, ignoring errors and
1687 plowing ahead is the most useful thing to do.
1693 New start/stop of dump markers in pg_dump (Bruce)
1699 Add start/stop times for pg_dump/pg_dumpall in verbose mode (Bruce)
1705 Allow most pg_dump options in pg_dumpall (Christopher)
1711 Have pg_dump use ALTER OWNER rather than SET SESSION
1712 AUTHORIZATION by default (Christopher)
1721 <title>libpq Changes</title>
1726 Make libpq SIGPIPE thread-safe (Bruce)
1732 Add PQmbdsplen() which returns the "display length" of a character (Tatsuo)
1738 Add thread locking to SSL and Kerberos connections (Manfred Spraul)
1744 Allow PQoidValue(), PQcmdTuples(), and PQoidStatus() to work on
1745 <command>EXECUTE</command> commands (Neil)
1754 <title>Source Code Changes</title>
1759 Allow the database server to run natively on Win32 (Claudio, Magnus, Andrew)
1765 Shell script commands converted to C versions for Win32 support (Andrew)
1771 Create an extension makefile framework (Fabien Coelho, Peter)
1774 This simplifies the task of building extensions outside the original
1781 Support relocatable installations (Bruce)
1784 Directory paths for installed files (such as the /share directory)
1785 are now computed relative to the actual location of the executables,
1786 so that an installation tree can be moved to another place without
1787 reconfiguring and rebuilding.
1793 Use --with-docdir to choose installation location of documentation; also
1794 allow --infodir (Peter)
1800 Add --without-docdir to prevent installation of documentation (Peter)
1806 Upgrade to DocBook V4.2 SGML (Peter)
1812 New "PostgreSQL" CVS tag (Marc)
1815 This was done to make it easier for organizations to manage
1816 their own copies of the <productname>PostgreSQL</productname>
1817 CVS repository. File version stamps from the master repository
1818 will not get munged by checking into or out of a copied
1825 Clarify locking code (Manfred Koizar)
1831 Buffer manager cleanup (Neil)
1837 Decouple platform tests from cpu spinlock code (Bruce, Tom)
1843 Add inlined test-and-set code on PA-RISC for gcc (ViSolve, Tom)
1849 Improve i386 spinlock code (Manfred Spraul)
1855 Clean up spinlock assembly code to avoid warnings from newer gcc
1862 Remove JDBC from source tree; now a separate project
1868 Remove the libpgtcl client interface; now a separate project
1875 More accurately estimate memory and file descriptor usage (Tom)
1881 Improvements to the Mac OS X startup scripts (Ray A.)
1887 New fsync test program (Bruce)
1893 Major documentation improvements (Neil, Peter)
1899 Remove contrib/rserv: obsoleted by various separate projects
1905 Remove <application>pg_encoding</application>; not needed
1912 Remove <application>pg_id</application>; not needed anymore
1918 Remove <application>initlocation</application>; not needed
1925 Auto detect thread flags (no more manual testing) (Bruce)
1931 Use Olson's public domain timezone library (Magnus)
1937 With threading enabled, use thread flags on Unixware for
1938 backend executables too (Bruce)
1941 Unixware can not mix threaded and non-threaded object files in the
1942 same executable, so everything must be compiled as threaded.
1948 psql now uses a flex-generated lexical analyzer to process command strings
1954 Reimplement the linked list data structure used throughout the
1958 This improves performance by allowing list append and length
1959 operations to be more efficient.
1965 Allow dynamically loaded modules to create their own server configuration
1966 parameters (Thomas Hallgren)
1972 New Brazilian version of FAQ (Euler Taveira de Oliveira)
1978 Add French FAQ (Guillaume Lelarge)
1984 New pgevent for Win32 logging
1990 Make libpq and ecpg build as proper shared libraries on OS X (Tom)
1999 <title>Contrib Changes</title>
2004 Many ecpg fixes, including SET DESCRIPTOR (Michael)
2010 Overhaul of /contrib/dblink (Joe)
2016 /contrib/dbmirror improvements
2022 New /contrib/xml2 (John Gray, Torchbox)
2028 Update /contrib/mysql
2034 New version of /contrib/btree_gist (Teodor)
2040 New /contrib/trgm, trigram matching for
2041 <productname>PostgreSQL</productname> (Teodor)
2047 Many /contrib/tsearch2 improvements (Teodor)
2053 Add double metaphone to /contrib/fuzzystrmatch (Andrew)
2063 <sect1 id="release-7-4-5">
2064 <title>Release 7.4.5</title>
2067 <title>Release date</title>
2068 <simpara>2004-08-18</simpara>
2072 This release contains one serious bug fix over 7.4.4.
2077 <title>Migration to version 7.4.5</title>
2080 A dump/restore is not required for those running 7.4.X.
2085 <title>Changes</title>
2088 <listitem><para>Repair possible crash during concurrent btree index insertions</para>
2090 This patch fixes a rare case in which concurrent insertions into a btree index
2091 could result in a server panic. No permanent damage would result, but it's
2092 still worth a re-release. The bug does not exist in pre-7.4 releases.
2099 <sect1 id="release-7-4-4">
2100 <title>Release 7.4.4</title>
2103 <title>Release date</title>
2104 <simpara>2004-08-16</simpara>
2108 This release contains a variety of fixes from 7.4.3.
2113 <title>Migration to version 7.4.4</title>
2116 A dump/restore is not required for those running 7.4.X.
2121 <title>Changes</title>
2124 <listitem><para>Prevent possible loss of committed transactions during crash</para>
2126 Due to insufficient interlocking between transaction commit and checkpointing,
2127 it was possible for transactions committed just before the most recent
2128 checkpoint to be lost, in whole or in part, following a database crash and
2129 restart. This is a serious bug that has existed
2130 since <productname>PostgreSQL</productname> 7.1.
2132 <listitem><para>Check HAVING restriction before evaluating result list of an
2133 aggregate plan</para></listitem>
2134 <listitem><para>Avoid crash when session's current userID is deleted</para></listitem>
2135 <listitem><para>Fix hashed crosstab for zero-rows case (Joe)</para></listitem>
2136 <listitem><para>Force cache update after renaming a column in a foreign key</para></listitem>
2137 <listitem><para>Pretty-print UNION queries correctly</para></listitem>
2138 <listitem><para>Make psql handle <literal>\r\n</> newlines properly in COPY IN</para></listitem>
2139 <listitem><para>pg_dump handled ACLs with grant options incorrectly</para></listitem>
2140 <listitem><para>Fix thread support for OS X and Solaris</para></listitem>
2141 <listitem><para>Updated JDBC driver (build 215) with various fixes</para></listitem>
2142 <listitem><para>ECPG fixes</para></listitem>
2143 <listitem><para>Translation updates (various contributors)</para></listitem>
2149 <sect1 id="release-7-4-3">
2150 <title>Release 7.4.3</title>
2153 <title>Release date</title>
2154 <simpara>2004-06-14</simpara>
2158 This release contains a variety of fixes from 7.4.2.
2163 <title>Migration to version 7.4.3</title>
2166 A dump/restore is not required for those running 7.4.X.
2171 <title>Changes</title>
2174 <listitem><para>Fix temporary memory leak when using non-hashed aggregates (Tom)</para></listitem>
2175 <listitem><para>ECPG fixes, including some for Informix compatibility (Michael)</para></listitem>
2176 <listitem><para>Fixes for compiling with thread-safety, particularly Solaris (Bruce)</para></listitem>
2177 <listitem><para>Fix error in COPY IN termination when using the old network protocol (ljb)</para></listitem>
2178 <listitem><para>Several important fixes in pg_autovacuum, including fixes for
2179 large tables, unsigned oids, stability, temp tables, and debug mode
2180 (Matthew T. O'Connor)</para></listitem>
2181 <listitem><para>Fix problem with reading tar-format dumps on NetBSD and BSD/OS (Bruce)</para></listitem>
2182 <listitem><para>Several JDBC fixes</para></listitem>
2183 <listitem><para>Fix ALTER SEQUENCE RESTART where last_value equals the restart value (Tom)</para></listitem>
2184 <listitem><para>Repair failure to recalculate nested sub-selects (Tom)</para></listitem>
2185 <listitem><para>Fix problems with non-constant expressions in LIMIT/OFFSET</para></listitem>
2186 <listitem><para>Support FULL JOIN with no join clause, such as X FULL JOIN Y ON TRUE (Tom)</para></listitem>
2187 <listitem><para>Fix another zero-column table bug (Tom)</para></listitem>
2188 <listitem><para>Improve handling of non-qualified identifiers in GROUP BY clauses in sub-selects (Tom)</para>
2190 Select-list aliases within the sub-select will now take precedence over
2191 names from outer query levels.
2193 <listitem><para>Do not generate <quote>NATURAL CROSS JOIN</> when decompiling rules (Tom)</para></listitem>
2194 <listitem><para>Add checks for invalid field length in binary COPY (Tom)</para>
2196 This fixes a difficult-to-exploit security hole.
2198 <listitem><para>Avoid locking conflict between <command>ANALYZE</command> and <command>LISTEN</command>/<command>NOTIFY</command></para></listitem>
2199 <listitem><para>Numerous translation updates (various contributors)</para></listitem>
2205 <sect1 id="release-7-4-2">
2206 <title>Release 7.4.2</title>
2209 <title>Release date</title>
2210 <simpara>2004-03-08</simpara>
2214 This release contains a variety of fixes from 7.4.1.
2219 <title>Migration to version 7.4.2</title>
2222 A dump/restore is not required for those running 7.4.X. However,
2223 it may be advisable as the easiest method of incorporating fixes for
2224 two errors that have been found in the initial contents of 7.4.X system
2225 catalogs. A dump/initdb/reload sequence using 7.4.2's initdb will
2226 automatically correct these problems.
2230 The more severe of the two errors is that data type <type>anyarray</>
2231 has the wrong alignment label; this is a problem because the
2232 <structname>pg_statistic</> system catalog uses <type>anyarray</>
2233 columns. The mislabeling can cause planner misestimations and even
2234 crashes when planning queries that involve <literal>WHERE</> clauses on
2235 double-aligned columns (such as <type>float8</> and <type>timestamp</>).
2236 It is strongly recommended that all installations repair this error,
2237 either by initdb or by following the manual repair procedure given
2242 The lesser error is that the system view <structname>pg_settings</>
2243 ought to be marked as having public update access, to allow
2244 <literal>UPDATE pg_settings</> to be used as a substitute for
2245 <command>SET</>. This can also be fixed either by initdb or manually,
2246 but it is not necessary to fix unless you want to use <literal>UPDATE
2251 If you wish not to do an initdb, the following procedure will work
2252 for fixing <structname>pg_statistic</>. As the database superuser,
2256 -- clear out old data in pg_statistic:
2257 DELETE FROM pg_statistic;
2258 VACUUM pg_statistic;
2259 -- this should update 1 row:
2260 UPDATE pg_type SET typalign = 'd' WHERE oid = 2277;
2261 -- this should update 6 rows:
2262 UPDATE pg_attribute SET attalign = 'd' WHERE atttypid = 2277;
2264 -- At this point you MUST start a fresh backend to avoid a crash!
2266 -- repopulate pg_statistic:
2270 This can be done in a live database, but beware that all backends
2271 running in the altered database must be restarted before it is safe to
2272 repopulate <structname>pg_statistic</>.
2276 To repair the <structname>pg_settings</> error, simply do:
2278 GRANT SELECT, UPDATE ON pg_settings TO PUBLIC;
2283 The above procedures must be carried out in <emphasis>each</> database
2284 of an installation, including <literal>template1</>, and ideally
2285 including <literal>template0</> as well. If you do not fix the
2286 template databases then any subsequently created databases will contain
2287 the same errors. <literal>template1</> can be fixed in the same way
2288 as any other database, but fixing <literal>template0</> requires
2289 additional steps. First, from any database issue
2291 UPDATE pg_database SET datallowconn = true WHERE datname = 'template0';
2293 Next connect to <literal>template0</> and perform the above repair
2294 procedures. Finally, do
2296 -- re-freeze template0:
2298 -- and protect it against future alterations:
2299 UPDATE pg_database SET datallowconn = false WHERE datname = 'template0';
2305 <title>Changes</title>
2308 Release 7.4.2 incorporates all the fixes included in release 7.3.6,
2309 plus the following fixes:
2313 <listitem><para>Fix <structname>pg_statistics</> alignment bug that could crash optimizer</para>
2314 <para>See above for details about this problem.</para></listitem>
2315 <listitem><para>Allow non-super users to update <structname>pg_settings</></para></listitem>
2316 <listitem><para>Fix several optimizer bugs, most of which led to
2317 <quote>variable not found in subplan target lists</> errors</para></listitem>
2318 <listitem><para>Avoid out-of-memory failure during startup of large multiple
2319 index scan</para></listitem>
2320 <listitem><para>Fix multibyte problem that could lead to <quote>out of
2321 memory</> error during <command>COPY IN</></para></listitem>
2322 <listitem><para>Fix problems with <command>SELECT INTO</> / <command>CREATE
2323 TABLE AS</> from tables without OIDs</para></listitem>
2324 <listitem><para>Fix problems with <filename>alter_table</> regression test
2325 during parallel testing</para></listitem>
2326 <listitem><para>Fix problems with hitting open file limit, especially on OS X (Tom)</para></listitem>
2327 <listitem><para>Partial fix for Turkish-locale issues</para>
2328 <para>initdb will succeed now in Turkish locale, but there are still some
2329 inconveniences associated with the <literal>i/I</> problem.</para></listitem>
2330 <listitem><para>Make pg_dump set client encoding on restore</para></listitem>
2331 <listitem><para>Other minor pg_dump fixes</para></listitem>
2332 <listitem><para>Allow ecpg to again use C keywords as column names (Michael)</para></listitem>
2333 <listitem><para>Added ecpg <literal>WHENEVER NOT_FOUND</> to
2334 <literal>SELECT/INSERT/UPDATE/DELETE</> (Michael)</para></listitem>
2335 <listitem><para>Fix ecpg crash for queries calling set-returning functions (Michael)</para></listitem>
2336 <listitem><para>Various other ecpg fixes (Michael)</para></listitem>
2337 <listitem><para>Fixes for Borland compiler</para></listitem>
2338 <listitem><para>Thread build improvements (Bruce)</para></listitem>
2339 <listitem><para>Various other build fixes</para></listitem>
2340 <listitem><para>Various JDBC fixes</para></listitem>
2346 <sect1 id="release-7-4-1">
2347 <title>Release 7.4.1</title>
2350 <title>Release date</title>
2351 <simpara>2003-12-22</simpara>
2355 This release contains a variety of fixes from 7.4.
2360 <title>Migration to version 7.4.1</title>
2363 A dump/restore is <emphasis>not</emphasis> required for those
2368 If you want to install the fixes in the information schema
2369 you need to reload it into the database.
2370 This is either accomplished by initializing a new cluster
2371 by running <command>initdb</command>, or by running the following
2372 sequence of SQL commands in each database (ideally including
2373 <literal>template1</literal>) as a superuser in
2374 <application>psql</application>, after installing the new release:
2376 DROP SCHEMA information_schema CASCADE;
2377 \i /usr/local/pgsql/share/information_schema.sql
2379 Substitute your installation path in the second command.
2385 <title>Changes</title>
2388 <listitem><para>Fixed bug in <command>CREATE SCHEMA</command> parsing in ECPG (Michael)</para></listitem>
2389 <listitem><para>Fix compile error when <option>--enable-thread-safety</option> and <option>--with-perl</option> are used together (Peter)</para></listitem>
2390 <listitem><para>Fix for subqueries that used hash joins (Tom)</para>
2392 Certain subqueries that used hash joins would crash because of
2393 improperly shared structures.
2395 <listitem><para>Fix free space map compaction bug (Tom)</para>
2397 This fixes a bug where compaction of the free space map could lead
2398 to a database server shutdown.
2401 <listitem><para>Fix for Borland compiler build of libpq (Bruce)</para></listitem>
2402 <listitem><para>Fix <function>netmask()</function> and <function>hostmask()</function> to return the maximum-length masklen (Tom)</para>
2404 Fix these functions to return values consistent with pre-7.4
2408 <listitem><para>Several <filename>contrib/pg_autovacuum</filename> fixes</para>
2410 Fixes include improper variable initialization, missing vacuum after
2411 <command>TRUNCATE</command>, and duration computation overflow for long vacuums.
2414 <listitem><para>Allow compile of <filename>contrib/cube</filename> under Cygwin (Jason Tishler)</para></listitem>
2415 <listitem><para>Fix Solaris use of password file when no passwords are defined (Tom)</para>
2417 Fix crash on Solaris caused by use of any type of password
2418 authentication when no passwords were defined.
2421 <listitem><para>JDBC fix for thread problems, other fixes</para></listitem>
2422 <listitem><para>Fix for <type>bytea</type> index lookups (Joe)</para></listitem>
2423 <listitem><para>Fix information schema for bit data types (Peter)</para></listitem>
2424 <listitem><para>Force zero_damaged_pages to be on during recovery from WAL</para></listitem>
2425 <listitem><para>Prevent some obscure cases of <quote>variable not in subplan target lists</quote></para></listitem>
2426 <listitem><para>Make <function>PQescapeBytea</function> and <function>byteaout</function> consistent with each other (Joe)</para></listitem>
2427 <listitem><para>Escape <type>bytea</type> output for bytes > 0x7e(Joe)</para>
2429 If different client encodings are used for <type>bytea</type> output and input, it
2430 is possible for <type>bytea</type> values to be corrupted by the differing
2431 encodings. This fix escapes all bytes that might be affected.
2434 <listitem><para>Added missing <function>SPI_finish()</function> calls to dblink's <function>get_tuple_of_interest()</function> (Joe)</para></listitem>
2435 <listitem><para>New Czech FAQ</para></listitem>
2436 <listitem><para>Fix information schema view <literal>constraint_column_usage</literal> for foreign keys (Peter)</para></listitem>
2437 <listitem><para>ECPG fixes (Michael)</para></listitem>
2438 <listitem><para>Fix bug with multiple <literal>IN</literal> subqueries and joins in the subqueries (Tom)</para></listitem>
2439 <listitem><para>Alllow <literal>COUNT('x')</literal> to work (Tom)</para></listitem>
2440 <listitem><para>Install ECPG include files for Informix compatibility into separate directory (Peter)</para>
2442 Some names of ECPG include files for Informix compatibility conflicted with operating system include files.
2443 By installing them in their own directory, name conflicts have been reduced.
2446 <listitem><para>Fix SSL memory leak (Neil)</para>
2448 This release fixes a bug in 7.4 where SSL didn't free all memory it allocated.
2451 <listitem><para>Prevent <filename>pg_service.conf</filename> from using service name as default dbname (Bruce)</para></listitem>
2452 <listitem><para>Fix local ident authentication on FreeBSD (Tom)</para></listitem>
2458 <sect1 id="release-7-4">
2459 <title>Release 7.4</title>
2462 <title>Release date</title>
2463 <simpara>2003-11-17</simpara>
2467 <title>Overview</title>
2470 Major changes in this release:
2475 <literal>IN</literal> / <literal>NOT IN</literal> subqueries are
2476 now much more efficient
2481 In previous releases, <literal>IN</literal>/<literal>NOT
2482 IN</literal> subqueries were joined to the upper query by
2483 sequentially scanning the subquery looking for a match. The
2484 7.4 code uses the same sophisticated techniques used by
2485 ordinary joins and so is much faster. An
2486 <literal>IN</literal> will now usually be as fast as or faster
2487 than an equivalent <literal>EXISTS</literal> subquery; this
2488 reverses the conventional wisdom that applied to previous
2496 Improved <literal>GROUP BY</literal> processing by using hash buckets
2501 In previous releases, rows to be grouped had to be sorted
2502 first. The 7.4 code can do <literal>GROUP BY</literal>
2503 without sorting, by accumulating results into a hash table
2504 with one entry per group. It will still use the sort
2505 technique, however, if the hash table is estimated to be too
2506 large to fit in <varname>sort_mem</>.
2513 New multikey hash join capability
2518 In previous releases, hash joins could only occur on single
2519 keys. This release allows multicolumn hash joins.
2526 Queries using the explicit <literal>JOIN</literal> syntax are
2527 now better optimized
2532 Prior releases evaluated queries using the explicit
2533 <literal>JOIN</literal> syntax only in the order implied by
2534 the syntax. 7.4 allows full optimization of these queries,
2535 meaning the optimizer considers all possible join orderings
2536 and chooses the most efficient. Outer joins, however, must
2537 still follow the declared ordering.
2544 Faster and more powerful regular expression code
2549 The entire regular expression module has been replaced with a
2550 new version by Henry Spencer, originally written for Tcl. The
2551 code greatly improves performance and supports several flavors
2552 of regular expressions.
2559 Function-inlining for simple SQL functions
2564 Simple SQL functions can now be inlined by including their SQL
2565 in the main query. This improves performance by eliminating
2566 per-call overhead. That means simple SQL functions now
2574 Full support for IPv6 connections and IPv6 address data types
2579 Previous releases allowed only IPv4 connections, and the IP
2580 data types only supported IPv4 addresses. This release adds
2581 full IPv6 support in both of these areas.
2588 Major improvements in SSL performance and reliability
2593 Several people very familiar with the SSL API have overhauled
2594 our SSL code to improve SSL key negotiation and error
2602 Make free space map efficiently reuse empty index pages,
2603 and other free space management improvements
2608 In previous releases, B-tree index pages that were left empty
2609 because of deleted rows could only be reused by rows with
2610 index values similar to the rows originally indexed on that
2611 page. In 7.4, <command>VACUUM</command> records empty index
2612 pages and allows them to be reused for any future index rows.
2619 SQL-standard information schema
2624 The information schema provides a standardized and stable way
2625 to access information about the schema objects defined in a
2633 Cursors conform more closely to the SQL standard
2638 The commands <command>FETCH</command> and
2639 <command>MOVE</command> have been overhauled to conform more
2640 closely to the SQL standard.
2647 Cursors can exist outside transactions
2652 These cursors are also called holdable cursors.
2659 New client-to-server protocol
2664 The new protocol adds error codes, more status information,
2665 faster startup, better support for binary data transmission,
2666 parameter values separated from SQL commands, prepared
2667 statements available at the protocol level, and cleaner
2668 recovery from <command>COPY</command> failures. The older
2669 protocol is still supported by both server and clients.
2676 <application>libpq</application> and
2677 <application>ECPG</application> applications are now fully
2683 While previous <application>libpq</application> releases
2684 already supported threads, this release improves thread safety
2685 by fixing some non-thread-safe code that was used during
2686 database connection startup. The <command>configure</command>
2687 option <option>--enable-thread-safety</option> must be used to
2688 enable this feature.
2695 New version of full-text indexing
2700 A new full-text indexing suite is available in
2701 <filename>contrib/tsearch2</filename>.
2713 The new autovacuum tool in
2714 <filename>contrib/autovacuum</filename> monitors the database
2715 statistics tables for
2716 <command>INSERT</command>/<command>UPDATE</command>/<command>DELETE</command>
2717 activity and automatically vacuums tables when needed.
2724 Array handling has been improved and moved into the server core
2729 Many array limitations have been removed, and arrays behave
2730 more like fully-supported data types.
2739 <title>Migration to version 7.4</title>
2742 A dump/restore using <application>pg_dump</application> is
2743 required for those wishing to migrate data from any previous
2748 Observe the following incompatibilities:
2753 The server-side autocommit setting was removed and
2754 reimplemented in client applications and languages.
2755 Server-side autocommit was causing too many problems with
2756 languages and applications that wanted to control their own
2757 autocommit behavior, so autocommit was removed from the server
2758 and added to individual client APIs as appropriate.
2764 Error message wording has changed substantially in this
2765 release. Significant effort was invested to make the messages
2766 more consistent and user-oriented. If your applications try to
2767 detect different error conditions by parsing the error message,
2768 you are strongly encouraged to use the new error code facility instead.
2774 Inner joins using the explicit <literal>JOIN</literal> syntax
2775 may behave differently because they are now better
2782 A number of server configuration parameters have been renamed
2783 for clarity, primarily those related to
2790 <literal>FETCH 0</literal> or <literal>MOVE 0</literal> now
2791 does nothing. In prior releases, <literal>FETCH 0</literal>
2792 would fetch all remaining rows, and <literal>MOVE 0</literal>
2793 would move to the end of the cursor.
2799 <command>FETCH</command> and <command>MOVE</command> now return
2800 the actual number of rows fetched/moved, or zero if at the
2801 beginning/end of the cursor. Prior releases would return the
2802 row count passed to the command, not the number of rows
2803 actually fetched or moved.
2809 <command>COPY</command> now can process files that use
2810 carriage-return or carriage-return/line-feed end-of-line
2811 sequences. Literal carriage-returns and line-feeds are no
2812 longer accepted in data values; use <literal>\r</literal> and
2813 <literal>\n</literal> instead.
2819 Trailing spaces are now trimmed when converting from type
2820 <type>char(<replaceable>n</>)</type> to
2821 <type>varchar(<replaceable>n</>)</type> or <type>text</type>.
2822 This is what most people always expected to happen anyway.
2828 The data type <type>float(<replaceable>p</>)</type> now
2829 measures <replaceable>p</> in binary digits, not decimal
2830 digits. The new behavior follows the SQL standard.
2836 Ambiguous date values now must match the ordering specified by
2837 the <varname>datestyle</varname> setting. In prior releases, a
2838 date specification of <literal>10/20/03</> was interpreted as a
2839 date in October even if <varname>datestyle</> specified that
2840 the day should be first. 7.4 will throw an error if a date
2841 specification is invalid for the current setting of
2842 <varname>datestyle</>.
2848 The functions <function>oidrand</function>,
2849 <function>oidsrand</function>, and
2850 <function>userfntest</function> have been removed. These
2851 functions were determined to be no longer useful.
2857 String literals specifying time-varying date/time values, such
2858 as <literal>'now'</literal> or <literal>'today'</literal> will
2859 no longer work as expected in column default expressions; they
2860 now cause the time of the table creation to be the default, not
2861 the time of the insertion. Functions such as
2862 <function>now()</>, <function>current_timestamp</>, or
2863 <function>current_date</function> should be used instead.
2867 In previous releases, there was special code so that strings
2868 such as <literal>'now'</literal> were interpreted at
2869 <command>INSERT</> time and not at table creation time, but
2870 this work around didn't cover all cases. Release 7.4 now
2871 requires that defaults be defined properly using functions such
2872 as <function>now()</> or <function>current_timestamp</>. These
2873 will work in all situations.
2879 The dollar sign (<literal>$</>) is no longer allowed in
2880 operator names. It can instead be a non-first character in
2881 identifiers. This was done to improve compatibility with other
2882 database systems, and to avoid syntax problems when parameter
2883 placeholders (<literal>$<replaceable>n</></>) are written
2884 adjacent to operators.
2892 <title>Changes</title>
2895 Below you will find a detailed account of the changes between
2896 release 7.4 and the previous major release.
2900 <title>Server Operation Changes</title>
2905 Allow IPv6 server connections (Nigel Kukard, Johan Jordaan,
2906 Bruce, Tom, Kurt Roeckx, Andrew Dunstan)
2912 Fix SSL to handle errors cleanly (Nathan Mueller)
2915 In prior releases, certain SSL API error reports were not
2916 handled correctly. This release fixes those problems.
2922 SSL protocol security and performance improvements (Sean Chittenden)
2925 SSL key renegotiation was happening too frequently, causing poor
2926 SSL performance. Also, initial key handling was improved.
2932 Print lock information when a deadlock is detected (Tom)
2935 This allows easier debugging of deadlock situations.
2941 Update <filename>/tmp</filename> socket modification times
2942 regularly to avoid their removal (Tom)
2945 This should help prevent <filename>/tmp</filename> directory
2946 cleaner administration scripts from removing server socket
2951 <listitem><para>Enable PAM for Mac OS X (Aaron Hillegass)</para></listitem>
2954 <para>Make B-tree indexes fully WAL-safe (Tom)</para>
2956 In prior releases, under certain rare cases, a server crash
2957 could cause B-tree indexes to become corrupt. This release
2958 removes those last few rare cases.
2962 <listitem><para>Allow B-tree index compaction and empty page reuse (Tom)</para></listitem>
2966 Fix inconsistent index lookups during split of first root page (Tom)
2969 In prior releases, when a single-page index split into two
2970 pages, there was a brief period when another database session
2971 could miss seeing an index entry. This release fixes that rare
2976 <listitem><para>Improve free space map allocation logic (Tom)</para></listitem>
2979 <para>Preserve free space information between server restarts (Tom)</para>
2981 In prior releases, the free space map was not saved when the
2982 postmaster was stopped, so newly started servers had no free
2983 space information. This release saves the free space map, and
2984 reloads it when the server is restarted.
2988 <listitem><para>Add start time to <literal>pg_stat_activity</literal> (Neil)</para></listitem>
2989 <listitem><para>New code to detect corrupt disk pages; erase with <varname>zero_damaged_pages</varname> (Tom)</para></listitem>
2990 <listitem><para>New client/server protocol: faster, no username length limit, allow clean exit from <command>COPY</command> (Tom)</para></listitem>
2991 <listitem><para>Add transaction status, table ID, column ID to client/server protocol (Tom)</para></listitem>
2992 <listitem><para>Add binary I/O to client/server protocol (Tom)</para></listitem>
2993 <listitem><para>Remove autocommit server setting; move to client applications (Tom)</para></listitem>
2994 <listitem><para>New error message wording, error codes, and three levels of error detail (Tom, Joe, Peter)</para></listitem>
2999 <title>Performance Improvements</title>
3002 <listitem><para>Add hashing for <literal>GROUP BY</literal> aggregates (Tom)</para></listitem>
3003 <listitem><para>Make nested-loop joins be smarter about multicolumn indexes (Tom)</para></listitem>
3004 <listitem><para>Allow multikey hash joins (Tom)</para></listitem>
3005 <listitem><para>Improve constant folding (Tom)</para></listitem>
3006 <listitem><para>Add ability to inline simple SQL functions (Tom)</para></listitem>
3009 <para>Reduce memory usage for queries using complex functions (Tom)</para>
3011 In prior releases, functions returning allocated memory would
3012 not free it until the query completed. This release allows the
3013 freeing of function-allocated memory when the function call
3014 completes, reducing the total memory used by functions.
3019 <para>Improve GEQO optimizer performance (Tom)</para>
3021 This release fixes several inefficiencies in the way the GEQO optimizer
3022 manages potential query paths.
3028 Allow <literal>IN</>/<literal>NOT IN</> to be handled via hash
3035 Improve <literal>NOT IN (<replaceable>subquery</>)</literal>
3042 Allow most <literal>IN</literal> subqueries to be processed as
3049 Pattern matching operations can use indexes regardless of
3053 There is no way for non-ASCII locales to use the standard
3054 indexes for <literal>LIKE</literal> comparisons. This release
3055 adds a way to create a special index for
3056 <literal>LIKE</literal>.
3061 <para>Allow the postmaster to preload libraries using <varname>preload_libraries</varname> (Joe)</para>
3063 For shared libraries that require a long time to load, this
3064 option is available so the library can be preloaded in the
3065 postmaster and inherited by all database sessions.
3071 Improve optimizer cost computations, particularly for subqueries (Tom)
3077 Avoid sort when subquery <literal>ORDER BY</literal> matches upper query (Tom)
3083 Deduce that <literal>WHERE a.x = b.y AND b.y = 42</literal> also
3084 means <literal>a.x = 42</literal> (Tom)
3090 Allow hash/merge joins on complex joins (Tom)
3096 Allow hash joins for more data types (Tom)
3102 Allow join optimization of explicit inner joins, disable with
3103 <varname>join_collapse_limit</varname> (Tom)
3109 Add parameter <varname>from_collapse_limit</varname> to control
3110 conversion of subqueries to joins (Tom)
3116 Use faster and more powerful regular expression code from Tcl
3117 (Henry Spencer, Tom)
3123 Use bit-mapped relation sets in the optimizer (Tom)
3128 <para>Improve connection startup time (Tom)</para>
3130 The new client/server protocol requires fewer network packets to
3131 start a database session.
3137 Improve trigger/constraint performance (Stephan)
3143 Improve speed of <literal>col IN (const, const, const, ...)</literal> (Tom)
3149 Fix hash indexes which were broken in rare cases (Tom)
3153 <listitem><para>Improve hash index concurrency and speed (Tom)</para>
3155 Prior releases suffered from poor hash index performance,
3156 particularly for high concurrency situations. This release fixes
3157 that, and the development group is interested in reports
3158 comparing B-tree and hash index performance.
3163 <para>Align shared buffers on 32-byte boundary for copy speed improvement (Manfred Spraul)</para>
3165 Certain CPU's perform faster data copies when addresses are
3171 <para>Data type <type>numeric</type> reimplemented for better performance (Tom)</para>
3173 <type>numeric</type> used to be stored in base 100. The new code
3174 uses base 10000, for significantly better performance.
3181 <title>Server Configuration Changes</title>
3185 <para>Rename server parameter <varname>server_min_messages</> to <varname>log_min_messages</> (Bruce)</para>
3187 This was done so most parameters that control the server logs
3188 begin with <literal>log_</>.
3192 <listitem><para>Rename <varname>show_*_stats</> to <varname>log_*_stats</> (Bruce)</para></listitem>
3193 <listitem><para>Rename <varname>show_source_port</> to <varname>log_source_port</> (Bruce)</para></listitem>
3194 <listitem><para>Rename <varname>hostname_lookup</> to <varname>log_hostname</> (Bruce)</para></listitem>
3197 <para>Add <varname>checkpoint_warning</> to warn of excessive checkpointing (Bruce)</para>
3199 In prior releases, it was difficult to determine if checkpoint
3200 was happening too frequently. This feature adds a warning to the
3201 server logs when excessive checkpointing happens.
3205 <listitem><para>New read-only server parameters for localization (Tom)</para></listitem>
3209 Change debug server log messages to output as <literal>DEBUG</>
3210 rather than <literal>LOG</> (Bruce)
3215 <para>Prevent server log variables from being turned off by non-superusers (Bruce)</para>
3217 This is a security feature so non-superusers cannot disable
3218 logging that was enabled by the administrator.
3224 <varname>log_min_messages</>/<varname>client_min_messages</> now
3225 controls <varname>debug_*</> output (Bruce)
3228 This centralizes client debug information so all debug output
3229 can be sent to either the client or server logs.
3234 <para>Add Mac OS X Rendezvous server support (Chris Campbell)</para>
3236 This allows Mac OS X hosts to query the network for available
3237 <productname>PostgreSQL</productname> servers.
3243 Add ability to print only slow statements using
3244 <varname>log_min_duration_statement</varname>
3248 This is an often requested debugging feature that allows
3249 administrators to see only slow queries in their server logs.
3254 <para>Allow <filename>pg_hba.conf</filename> to accept netmasks in CIDR format (Andrew Dunstan)</para>
3256 This allows administrators to merge the host IP address and
3257 netmask fields into a single CIDR field in <filename>pg_hba.conf</filename>.
3261 <listitem><para>New read-only parameter <varname>is_superuser</varname> (Tom)</para></listitem>
3264 <para>New parameter <varname>log_error_verbosity</varname> to control error detail (Tom)</para>
3266 This works with the new error reporting feature to supply
3267 additional error information like hints, file names and line
3273 <para><literal>postgres --describe-config</literal> now dumps server config variables (Aizaz Ahmed, Peter)</para>
3275 This option is useful for administration tools that need to know
3276 the configuration variable names and their minimums, maximums,
3277 defaults, and descriptions.
3283 Add new columns in <literal>pg_settings</literal>:
3284 <literal>context</>, <literal>type</>, <literal>source</>,
3285 <literal>min_val</>, <literal>max_val</> (Joe)
3291 Make default <varname>shared_buffers</> 1000 and
3292 <varname>max_connections</> 100, if possible (Tom)
3295 Prior versions defaulted to 64 shared buffers so <productname>PostgreSQL</productname>
3296 would start on even very old systems. This release tests the
3297 amount of shared memory allowed by the platform and selects more
3298 reasonable default values if possible. Of course, users are
3299 still encouraged to evaluate their resource load and size
3300 <varname>shared_buffers</varname> accordingly.
3306 New <filename>pg_hba.conf</filename> record type
3307 <literal>hostnossl</> to prevent SSL connections (Jon
3311 In prior releases, there was no way to prevent SSL connections
3312 if both the client and server supported SSL. This option allows
3319 Remove parameter <varname>geqo_random_seed</varname>
3326 Add server parameter <varname>regex_flavor</varname> to control regular expression processing (Tom)
3332 Make <command>pg_ctl</command> better handle nonstandard ports (Greg)
3339 <title>Query Changes</title>
3342 <listitem><para>New SQL-standard information schema (Peter)</para></listitem>
3343 <listitem><para>Add read-only transactions (Peter)</para></listitem>
3344 <listitem><para>Print key name and value in foreign-key violation messages (Dmitry Tkach)</para></listitem>
3347 <para>Allow users to see their own queries in <literal>pg_stat_activity</literal> (Kevin Brown)</para>
3349 In prior releases, only the superuser could see query strings
3350 using <literal>pg_stat_activity</literal>. Now ordinary users
3351 can see their own query strings.
3356 <para>Fix aggregates in subqueries to match SQL standard (Tom)</para>
3358 The SQL standard says that an aggregate function appearing
3359 within a nested subquery belongs to the outer query if its
3360 argument contains only outer-query variables. Prior
3361 <productname>PostgreSQL</productname> releases did not handle
3362 this fine point correctly.
3367 <para>Add option to prevent auto-addition of tables referenced in query (Nigel J. Andrews)</para>
3369 By default, tables mentioned in the query are automatically
3370 added to the <literal>FROM</> clause if they are not already
3371 there. This is compatible with historic
3372 <productname>POSTGRES</productname> behavior but is contrary to
3373 the SQL standard. This option allows selecting
3374 standard-compatible behavior.
3379 <para>Allow <literal>UPDATE ... SET col = DEFAULT</literal> (Rod)</para>
3381 This allows <command>UPDATE</command> to set a column to its
3382 declared default value.
3387 <para>Allow expressions to be used in <literal>LIMIT</>/<literal>OFFSET</> (Tom)</para>
3389 In prior releases, <literal>LIMIT</>/<literal>OFFSET</> could
3390 only use constants, not expressions.
3395 <para>Implement <literal>CREATE TABLE AS EXECUTE</literal> (Neil, Peter)</para>
3401 <title>Object Manipulation Changes</title>
3405 <para>Make <command>CREATE SEQUENCE</command> grammar more conforming to SQL 2003 (Neil)</para>
3409 <para>Add statement-level triggers (Neil)</para>
3411 While this allows a trigger to fire at the end of a statement,
3412 it does not allow the trigger to access all rows modified by the
3413 statement. This capability is planned for a future release.
3418 <para>Add check constraints for domains (Rod)</para>
3420 This greatly increases the usefulness of domains by allowing
3421 them to use check constraints.
3426 <para>Add <command>ALTER DOMAIN</command> (Rod)</para>
3428 This allows manipulation of existing domains.
3433 <para>Fix several zero-column table bugs (Tom)</para>
3435 <productname>PostgreSQL</productname> supports zero-column tables. This fixes various bugs
3436 that occur when using such tables.
3441 <para>Have <literal>ALTER TABLE ... ADD PRIMARY KEY</literal> add not-null constraint (Rod)</para>
3443 In prior releases, <literal>ALTER TABLE ... ADD
3444 PRIMARY</literal> would add a unique index, but not a not-null
3445 constraint. That is fixed in this release.
3449 <listitem><para>Add <literal>ALTER TABLE ... WITHOUT OIDS</literal> (Rod)</para>
3451 This allows control over whether new and updated rows will have
3452 an OID column. This is most useful for saving storage space.
3458 Add <literal>ALTER SEQUENCE</literal> to modify minimum, maximum,
3459 increment, cache, cycle values (Rod)
3464 <para>Add <literal>ALTER TABLE ... CLUSTER ON</literal> (Alvaro Herrera)</para>
3466 This command is used by <command>pg_dump</command> to record the
3467 cluster column for each table previously clustered. This
3468 information is used by database-wide cluster to cluster all
3469 previously clustered tables.
3473 <listitem><para>Improve automatic type casting for domains (Rod, Tom)</para></listitem>
3474 <listitem><para>Allow dollar signs in identifiers, except as first character (Tom)</para></listitem>
3475 <listitem><para>Disallow dollar signs in operator names, so <literal>x=$1</> works (Tom)</para></listitem>
3479 Allow copying table schema using <literal>LIKE
3480 <replaceable>subtable</replaceable></literal>, also SQL 2003
3481 feature <literal>INCLUDING DEFAULTS</literal> (Rod)
3487 Add <literal>WITH GRANT OPTION</literal> clause to
3488 <command>GRANT</command> (Peter)
3491 This enabled <command>GRANT</command> to give other users the
3492 ability to grant privileges on a object.
3499 <title>Utility Command Changes</title>
3503 <para>Add <literal>ON COMMIT</literal> clause to <command>CREATE TABLE</command> for temporary tables (Gavin)</para>
3505 This adds the ability for a table to be dropped or all rows
3506 deleted on transaction commit.
3511 <para>Allow cursors outside transactions using <literal>WITH HOLD</literal> (Neil)</para>
3513 In previous releases, cursors were removed at the end of the
3514 transaction that created them. Cursors can now be created with
3515 the <literal>WITH HOLD</literal> option, which allows them to
3516 continue to be accessed after the creating transaction has
3522 <para><literal>FETCH 0</literal> and <literal>MOVE 0 </literal> now do nothing (Bruce)</para>
3524 In previous releases, <literal>FETCH 0</literal> fetched all
3525 remaining rows, and <literal>MOVE 0</literal> moved to the end
3532 Cause <command>FETCH</command> and <command>MOVE</command> to
3533 return the number of rows fetched/moved, or zero if at the
3534 beginning/end of cursor, per SQL standard (Bruce)
3537 In prior releases, the row count returned by
3538 <command>FETCH</command> and <command>MOVE</command> did not
3539 accurately reflect the number of rows processed.
3544 <para>Properly handle <literal>SCROLL</literal> with cursors, or
3545 report an error (Neil)</para>
3547 Allowing random access (both forward and backward scrolling) to
3548 some kinds of queries cannot be done without some additional
3549 work. If <literal>SCROLL</literal> is specified when the cursor
3550 is created, this additional work will be performed. Furthermore,
3551 if the cursor has been created with <literal>NO SCROLL</literal>,
3552 no random access is allowed.
3558 Implement SQL-compatible options <literal>FIRST</>,
3559 <literal>LAST</>, <literal>ABSOLUTE <replaceable>n</></>,
3560 <literal>RELATIVE <replaceable>n</></> for
3561 <command>FETCH</command> and <command>MOVE</command> (Tom)
3566 <para>Allow <command>EXPLAIN</command> on <command>DECLARE CURSOR</command> (Tom)</para>
3570 <para>Allow <command>CLUSTER</command> to use index marked as pre-clustered by default (Alvaro Herrera)</para>
3574 <para>Allow <command>CLUSTER</command> to cluster all tables (Alvaro Herrera)</para>
3576 This allows all previously clustered tables in a database to be
3577 reclustered with a single command.
3581 <listitem><para>Prevent <command>CLUSTER</command> on partial indexes (Tom)</para></listitem>
3583 <listitem><para>Allow DOS and Mac line-endings in <command>COPY</> files (Bruce)</para></listitem>
3587 Disallow literal carriage return as a data value,
3588 backslash-carriage-return and <literal>\r</> are still allowed
3594 <para><command>COPY</> changes (binary, <literal>\.</>) (Tom)</para>
3598 <para>Recover from <command>COPY</command> failure cleanly (Tom)</para>
3602 <para>Prevent possible memory leaks in <command>COPY</command> (Tom)</para>
3606 <para>Make <command>TRUNCATE</command> transaction-safe (Rod)</para>
3608 <command>TRUNCATE</command> can now be used inside a
3609 transaction. If the transaction aborts, the changes made by the
3610 <command>TRUNCATE</command> are automatically rolled back.
3616 Allow prepare/bind of utility commands like
3617 <command>FETCH</command> and <command>EXPLAIN</command> (Tom)
3622 <para>Add <command>EXPLAIN EXECUTE</command> (Neil)</para>
3626 <para>Improve <command>VACUUM</command> performance on indexes by reducing WAL traffic (Tom)</para>
3630 <para>Functional indexes have been generalized into indexes on expressions (Tom)</para>
3632 In prior releases, functional indexes only supported a simple
3633 function applied to one or more column names. This release
3634 allows any type of scalar expression.
3640 Have <command>SHOW TRANSACTION ISOLATION</command> match input
3641 to <command>SET TRANSACTION ISOLATION</command>
3648 Have <command>COMMENT ON DATABASE</command> on nonlocal
3649 database generate a warning, rather than an error (Rod)
3653 Database comments are stored in database-local tables so
3654 comments on a database have to be stored in each database.
3660 Improve reliability of <command>LISTEN</>/<command>NOTIFY</> (Tom)
3665 <para>Allow <command>REINDEX</command> to reliably reindex nonshared system catalog indexes (Tom)</para>
3667 This allows system tables to be reindexed without the
3668 requirement of a standalone session, which was necessary in
3669 previous releases. The only tables that now require a standalone
3670 session for reindexing are the global system tables
3671 <literal>pg_database</>, <literal>pg_shadow</>, and
3672 <literal>pg_group</>.
3679 <title>Data Type and Function Changes</title>
3684 New server parameter <varname>extra_float_digits</varname> to
3685 control precision display of floating-point numbers (Pedro
3689 This controls output precision which was causing regression
3694 <listitem><para>Allow <literal>+1300</literal> as a numeric time-zone specifier, for FJST (Tom)</para></listitem>
3698 Remove rarely used functions <function>oidrand</>,
3699 <function>oidsrand</>, and <function>userfntest</> functions
3705 <para>Add <function>md5()</> function to main server, already in <filename>contrib/pgcrypto</filename> (Joe)</para>
3707 An MD5 function was frequently requested. For more complex
3708 encryption capabilities, use
3709 <filename>contrib/pgcrypto</filename>.
3713 <listitem><para>Increase date range of <type>timestamp</type> (John Cochran)</para></listitem>
3717 Change <literal>EXTRACT(EPOCH FROM timestamp)</literal> so
3718 <type>timestamp without time zone</type> is assumed to be in
3719 local time, not GMT (Tom)
3723 <listitem><para>Trap division by zero in case the operating system doesn't prevent it (Tom)</para></listitem>
3724 <listitem><para>Change the <type>numeric</type> data type internally to base 10000 (Tom)</para></listitem>
3725 <listitem><para>New <function>hostmask()</function> function (Greg Wickham)</para></listitem>
3726 <listitem><para>Fixes for <function>to_char()</function> and <function>to_timestamp()</function> (Karel)</para></listitem>
3730 Allow functions that can take any argument data type and return
3731 any data type, using <type>anyelement</type> and
3732 <type>anyarray</type> (Joe)
3735 This allows the creation of functions that can work with any
3742 Arrays may now be specified as <literal>ARRAY[1,2,3]</literal>,
3743 <literal>ARRAY[['a','b'],['c','d']]</literal>, or
3744 <literal>ARRAY[ARRAY[ARRAY[2]]]</literal> (Joe)
3750 Allow proper comparisons for arrays, including <literal>ORDER
3751 BY</literal> and <literal>DISTINCT</literal> support
3756 <listitem><para>Allow indexes on array columns (Joe)</para></listitem>
3757 <listitem><para>Allow array concatenation with <literal>||</literal> (Joe)</para></listitem>
3761 Allow <literal>WHERE</literal> qualification
3762 <literal><replaceable>expr</> <replaceable>op</> ANY/SOME/ALL
3763 (<replaceable>array_expr</>)</literal> (Joe)
3766 This allows arrays to behave like a list of values, for purposes
3767 like <literal>SELECT * FROM tab WHERE col IN
3768 (array_val)</literal>.
3774 New array functions <function>array_append</>,
3775 <function>array_cat</>, <function>array_lower</>,
3776 <function>array_prepend</>, <function>array_to_string</>,
3777 <function>array_upper</>, <function>string_to_array</> (Joe)
3781 <listitem><para>Allow user defined aggregates to use polymorphic functions (Joe)</para></listitem>
3782 <listitem><para>Allow assignments to empty arrays (Joe)</para></listitem>
3786 Allow 60 in seconds fields of <type>time</type>,
3787 <type>timestamp</type>, and <type>interval</type> input values
3791 Sixty-second values are needed for leap seconds.
3795 <listitem><para>Allow <type>cidr</type> data type to be cast to <type>text</type> (Tom)</para></listitem>
3797 <listitem><para>Disallow invalid time zone names in SET TIMEZONE</para></listitem>
3801 Trim trailing spaces when <type>char</type> is cast to
3802 <type>varchar</> or <type>text</> (Tom)
3808 Make <type>float(<replaceable>p</>)</> measure the precision
3809 <replaceable>p</> in binary digits, not decimal digits
3815 <para>Add IPv6 support to the <type>inet</type> and <type>cidr</type> data types (Michael Graff)</para>
3819 <para>Add <function>family()</function> function to report whether address is IPv4 or IPv6 (Michael Graff)</para>
3824 Have <literal>SHOW datestyle</literal> generate output similar
3825 to that used by <literal>SET datestyle</literal> (Tom)
3831 Make <literal>EXTRACT(TIMEZONE)</literal> and <literal>SET/SHOW
3832 TIME ZONE</literal> follow the SQL convention for the sign of
3833 time zone offsets, i.e., positive is east from UTC (Tom)
3838 <para>Fix <literal>date_trunc('quarter', ...)</literal> (Böjthe Zoltán)</para>
3840 Prior releases returned an incorrect value for this function call.
3845 <para>Make <function>initcap()</function> more compatible with Oracle (Mike Nolan)</para>
3847 <function>initcap()</function> now uppercases a letter appearing
3848 after any non-alphanumeric character, rather than only after
3854 <para>Allow only <varname>datestyle</varname> field order for date values not in ISO-8601 format (Greg)</para>
3859 Add new <varname>datestyle</varname> values <literal>MDY</>,
3860 <literal>DMY</>, and <literal>YMD</> to set input field order;
3861 honor <literal>US</> and <literal>European</> for backward
3868 String literals like <literal>'now'</literal> or
3869 <literal>'today'</literal> will no longer work as a column
3870 default. Use functions such as <function>now()</function>,
3871 <function>current_timestamp</function> instead. (change
3872 required for prepared statements) (Tom)
3877 <para>Treat NaN as larger than any other value in <function>min()</>/<function>max()</> (Tom)</para>
3879 NaN was already sorted after ordinary numeric values for most
3880 purposes, but <function>min()</> and <function>max()</> didn't
3886 <para>Prevent interval from suppressing <literal>:00</literal>
3887 seconds display</para>
3892 New functions <function>pg_get_triggerdef(prettyprint)</function>
3893 and <function>pg_conversion_is_visible()</function> (Christopher)
3898 <para>Allow time to be specified as <literal>040506</> or <literal>0405</> (Tom)</para>
3903 Input date order must now be <literal>YYYY-MM-DD</literal> (with 4-digit year) or
3904 match <varname>datestyle</varname>
3910 Make <function>pg_get_constraintdef</function> support
3911 unique, primary-key, and check constraints (Christopher)
3918 <title>Server-Side Language Changes</title>
3923 Prevent PL/pgSQL crash when <literal>RETURN NEXT</literal> is
3924 used on a zero-row record variable (Tom)
3930 Make PL/Python's <function>spi_execute</function> interface
3931 handle null values properly (Andrew Bosma)
3936 <para>Allow PL/pgSQL to declare variables of composite types without <literal>%ROWTYPE</literal> (Tom)</para>
3940 <para>Fix PL/Python's <function>_quote()</function> function to handle big integers</para>
3944 <para>Make PL/Python an untrusted language, now called <literal>plpythonu</literal> (Kevin Jacobs, Tom)</para>
3946 The Python language no longer supports a restricted execution
3947 environment, so the trusted version of PL/Python was removed. If
3948 this situation changes, a version of PL/Python that can be used
3949 by non-superusers will be readded.
3954 <para>Allow polymorphic PL/pgSQL functions (Joe, Tom)</para>
3958 <para>Allow polymorphic SQL functions (Joe)</para>
3963 Improved compiled function caching mechanism in PL/pgSQL with
3964 full support for polymorphism (Joe)
3970 Add new parameter <literal>$0</> in PL/pgSQL representing the
3971 function's actual return type (Joe)
3977 Allow PL/Tcl and PL/Python to use the same trigger on multiple tables (Tom)
3983 Fixed PL/Tcl's <function>spi_prepare</function> to accept fully
3984 qualified type names in the parameter type list
3992 <title>psql Changes</title>
3996 <para>Add <literal>\pset pager always</literal> to always use pager (Greg)</para>
3998 This forces the pager to be used even if the number of rows is
3999 less than the screen height. This is valuable for rows that
4000 wrap across several screen rows.
4004 <listitem><para>Improve tab completion (Rod, Ross Reedstrom, Ian Barwick)</para></listitem>
4005 <listitem><para>Reorder <literal>\?</> help into groupings (Harald Armin Massa, Bruce)</para></listitem>
4006 <listitem><para>Add backslash commands for listing schemas, casts, and conversions (Christopher)</para></listitem>
4010 <command>\encoding</> now changes based on the server parameter
4011 <varname>client_encoding</varname> (Tom)
4014 In previous versions, <command>\encoding</command> was not aware
4015 of encoding changes made using <literal>SET
4016 client_encoding</literal>.
4021 <para>Save editor buffer into readline history (Ross)</para>
4023 When <command>\e</> is used to edit a query, the result is saved
4024 in the readline history for retrieval using the up arrow.
4028 <listitem><para>Improve <command>\d</command> display (Christopher)</para></listitem>
4029 <listitem><para>Enhance HTML mode to be more standards-conforming (Greg)</para></listitem>
4032 <para>New <command>\set AUTOCOMMIT off</command> capability (Tom)</para>
4034 This takes the place of the removed server parameter <varname>autocommit</varname>.
4039 <para>New <command>\set VERBOSITY</command> to control error detail (Tom)</para>
4041 This controls the new error reporting details.
4045 <listitem><para>New prompt escape sequence <literal>%x</literal> to show transaction status (Tom)</para></listitem>
4046 <listitem><para>Long options for <application>psql</application> are now available on all platforms</para></listitem>
4051 <title>pg_dump Changes</title>
4054 <listitem><para>Multiple pg_dump fixes, including tar format and large objects</para></listitem>
4055 <listitem><para>Allow pg_dump to dump specific schemas (Neil)</para></listitem>
4058 <para>Make pg_dump preserve column storage characteristics (Christopher)</para>
4060 This preserves <literal>ALTER TABLE ... SET STORAGE</literal> information.
4064 <listitem><para>Make pg_dump preserve <command>CLUSTER</command> characteristics (Christopher)</para></listitem>
4068 Have pg_dumpall use <command>GRANT</>/<command>REVOKE</> to dump database-level privleges (Tom)
4074 Allow pg_dumpall to support the options <option>-a</>,
4075 <option>-s</>, <option>-x</> of pg_dump (Tom)
4079 <listitem><para>Prevent pg_dump from lowercasing identifiers specified on the command line (Tom)</para></listitem>
4083 pg_dump options <option>--use-set-session-authorization</option>
4084 and <option>--no-reconnect</option> now do nothing, all dumps
4085 use <command>SET SESSION AUTHORIZATION</command>
4088 pg_dump no longer reconnects to switch users, but instead always
4089 uses <command>SET SESSION AUTHORIZATION</command>. This will
4090 reduce password prompting during restores.
4095 <para>Long options for <application>pg_dump</application> are now available on all platforms</para>
4097 <productname>PostgreSQL</productname> now includes its own
4098 long-option processing routines.
4105 <title>libpq Changes</title>
4110 Add function <function>PQfreemem</function> for freeing memory on
4111 Windows, suggested for <command>NOTIFY</command> (Bruce)
4114 Windows requires that memory allocated in a library be freed by
4115 a function in the same library, hence
4116 <function>free()</function> doesn't work for freeing memory
4117 allocated by libpq. <function>PQfreemem</function> is the proper
4118 way to free libpq memory, especially on Windows, and is
4119 recommended for other platforms as well.
4124 <para>Document service capability, and add sample file (Bruce)</para>
4126 This allows clients to look up connection information in a
4127 central file on the client machine.
4133 Make <function>PQsetdbLogin</function> have the same defaults as
4134 <function>PQconnectdb</function> (Tom)
4138 <listitem><para>Allow libpq to cleanly fail when result sets are too large (Tom)</para></listitem>
4142 Improve performance of function <function>PQunescapeBytea</function> (Ben Lamb)
4148 Allow thread-safe libpq with <filename>configure</filename>
4149 option <option>--enable-thread-safety</option> (Lee Kindness,
4156 Allow function <function>pqInternalNotice</function> to accept a
4157 format string and arguments instead of just a preformatted
4158 message (Tom, Sean Chittenden)
4164 Control SSL negotiation with <literal>sslmode</literal> values
4165 <literal>disable</literal>, <literal>allow</literal>,
4166 <literal>prefer</literal>, and <literal>require</literal> (Jon
4172 <para>Allow new error codes and levels of text (Tom)</para>
4176 <para>Allow access to the underlying table and column of a query result (Tom)</para>
4178 This is helpful for query-builder applications that want to know
4179 the underlying table and column names associated with a specific
4184 <listitem><para>Allow access to the current transaction status (Tom)</para></listitem>
4185 <listitem><para>Add ability to pass binary data directly to the server (Tom)</para></listitem>
4189 Add function <function>PQexecPrepared</function> and
4190 <function>PQsendQueryPrepared</function> functions which perform
4191 bind/execute of previously prepared statements (Tom)
4198 <title>JDBC Changes</title>
4201 <listitem><para>Allow <function>setNull</function> on updateable result sets</para></listitem>
4202 <listitem><para>Allow <function>executeBatch</function> on a prepared statement (Barry)</para></listitem>
4203 <listitem><para>Support SSL connections (Barry)</para></listitem>
4204 <listitem><para>Handle schema names in result sets (Paul Sorenson)</para></listitem>
4205 <listitem><para>Add refcursor support (Nic Ferrier)</para></listitem>
4210 <title>Miscellaneous Interface Changes</title>
4214 <para>Prevent possible memory leak or core dump during libpgtcl shutdown (Tom)</para>
4217 <para>Add Informix compatibility to ECPG (Michael)</para>
4219 This allows ECPG to process embedded C programs that were
4220 written using certain Informix extensions.
4225 <para>Add type <type>decimal</type> to ECPG that is fixed length, for Informix (Michael)</para>
4230 Allow thread-safe embedded SQL programs with
4231 <filename>configure</filename> option
4232 <option>--enable-thread-safety</option> (Lee Kindness, Bruce)
4235 This allows multiple threads to access the database at the same
4241 <para>Moved Python client PyGreSQL to <ulink url="http://www.pygresql.org"></ulink> (Marc)</para>
4247 <title>Source Code Changes</title>
4250 <listitem><para>Prevent need for separate platform geometry regression result files (Tom)</para></listitem>
4251 <listitem><para>Improved PPC locking primitive (Reinhard Max)</para></listitem>
4252 <listitem><para>New function <function>palloc0</function> to allocate and clear memory (Bruce)</para></listitem>
4253 <listitem><para>Fix locking code for s390x CPU (64-bit) (Tom)</para></listitem>
4254 <listitem><para>Allow OpenBSD to use local ident credentials (William Ahern)</para></listitem>
4255 <listitem><para>Make query plan trees read-only to executor (Tom)</para></listitem>
4256 <listitem><para>Add Darwin startup scripts (David Wheeler)</para></listitem>
4257 <listitem><para>Allow libpq to compile with Borland C++ compiler (Lester Godwin, Karl Waclawek)</para></listitem>
4258 <listitem><para>Use our own version of <function>getopt_long()</function> if needed (Peter)</para></listitem>
4259 <listitem><para>Convert administration scripts to C (Peter)</para></listitem>
4260 <listitem><para> Bison >= 1.85 is now required to build the <productname>PostgreSQL</> grammar, if building from CVS</para></listitem>
4261 <listitem><para>Merge documentation into one book (Peter)</para></listitem>
4262 <listitem><para>Add Windows compatibility functions (Bruce)</para></listitem>
4263 <listitem><para>Allow client interfaces to compile under MinGW (Bruce)</para></listitem>
4264 <listitem><para>New <function>ereport()</function> function for error reporting (Tom)</para></listitem>
4265 <listitem><para>Support Intel compiler on Linux (Peter)</para></listitem>
4266 <listitem><para>Improve Linux startup scripts (Slawomir Sudnik, Darko Prenosil)</para></listitem>
4267 <listitem><para>Add support for AMD Opteron and Itanium (Jeffrey W. Baker, Bruce)</para></listitem>
4269 <para>Remove <option>--enable-recode</option> option from <command>configure</command></para>
4271 This was no longer needed now that we have <command>CREATE CONVERSION</command>.
4275 <para>Generate a compile error if spinlock code is not found (Bruce)</para>
4277 Platforms without spinlock code will now fail to compile, rather
4278 than silently using semaphores. This failure can be disabled
4279 with a new <command>configure</command> option.
4286 <title>Contrib Changes</title>
4289 <listitem><para>Change dbmirror license to BSD</para></listitem>
4290 <listitem><para>Improve earthdistance (Bruno Wolff III)</para></listitem>
4291 <listitem><para>Portability improvements to pgcrypto (Marko Kreen)</para></listitem>
4292 <listitem><para>Prevent crash in xml (John Gray, Michael Richards)</para></listitem>
4293 <listitem><para>Update oracle</para></listitem>
4294 <listitem><para>Update mysql</para></listitem>
4295 <listitem><para>Update cube (Bruno Wolff III)</para></listitem>
4296 <listitem><para>Update earthdistance to use cube (Bruno Wolff III)</para></listitem>
4297 <listitem><para>Update btree_gist (Oleg)</para></listitem>
4298 <listitem><para>New tsearch2 full-text search module (Oleg, Teodor)</para></listitem>
4299 <listitem><para>Add hash-based crosstab function to tablefuncs (Joe)</para></listitem>
4300 <listitem><para>Add serial column to order <function>connectby()</> siblings in tablefuncs (Nabil Sayegh,Joe)</para></listitem>
4301 <listitem><para>Add named persistent connections to dblink (Shridhar Daithanka)</para></listitem>
4302 <listitem><para>New pg_autovacuum allows automatic <command>VACUUM</command> (Matthew T. O'Connor)</para></listitem>
4303 <listitem><para>Make pgbench honor environment variables <envar>PGHOST</>, <envar>PGPORT</>, <envar>PGUSER</> (Tatsuo)</para></listitem>
4304 <listitem><para>Improve intarray (Teodor Sigaev)</para></listitem>
4305 <listitem><para>Improve pgstattuple (Rod)</para></listitem>
4306 <listitem><para>Fix bug in <function>metaphone()</function> in fuzzystrmatch</para></listitem>
4307 <listitem><para>Improve adddepend (Rod)</para></listitem>
4308 <listitem><para>Update spi/timetravel (Böjthe Zoltán)</para></listitem>
4309 <listitem><para>Fix dbase <option>-s</> option and improve non-ASCII handling (Thomas Behr, Márcio Smiderle)</para></listitem>
4310 <listitem><para>Remove array module because features now included by default (Joe)</para></listitem>
4316 <sect1 id="release-7-3-7">
4317 <title>Release 7.3.7</title>
4320 <title>Release date</title>
4321 <simpara>2004-08-16</simpara>
4325 This release contains one critical fix over 7.3.6, and some minor items.
4330 <title>Migration to version 7.3.7</title>
4333 A dump/restore is not required for those running 7.3.X.
4338 <title>Changes</title>
4341 <listitem><para>Prevent possible loss of committed transactions during crash</para>
4343 Due to insufficient interlocking between transaction commit and checkpointing,
4344 it was possible for transactions committed just before the most recent
4345 checkpoint to be lost, in whole or in part, following a database crash and
4346 restart. This is a serious bug that has existed
4347 since <productname>PostgreSQL</productname> 7.1.
4349 <listitem><para>Remove asymmetrical word processing in tsearch (Teodor)</para></listitem>
4350 <listitem><para>Properly schema-qualify function names when pg_dump'ing a CAST</para></listitem>
4356 <sect1 id="release-7-3-6">
4357 <title>Release 7.3.6</title>
4360 <title>Release date</title>
4361 <simpara>2004-03-02</simpara>
4365 This release contains a variety of fixes from 7.3.5.
4370 <title>Migration to version 7.3.6</title>
4373 A dump/restore is <emphasis>not</emphasis> required for those
4380 <title>Changes</title>
4383 <listitem><para>Revert erroneous changes in rule permissions checking</para>
4384 <para>A patch applied in 7.3.3 to fix a corner case in rule permissions checks
4385 turns out to have disabled rule-related permissions checks in many
4386 not-so-corner cases. This would for example allow users to insert into views
4387 they weren't supposed to have permission to insert into. We have therefore
4388 reverted the 7.3.3 patch. The original bug will be fixed in 8.0.
4390 <listitem><para>Repair incorrect order of operations in
4391 GetNewTransactionId()</para>
4393 This bug could result in failure under out-of-disk-space conditions, including
4394 inability to restart even after disk space is freed.
4396 <listitem><para>Ensure configure selects -fno-strict-aliasing even when
4397 an external value for CFLAGS is supplied</para>
4399 On some platforms, building with -fstrict-aliasing causes bugs.
4401 <listitem><para>Make pg_restore handle 64-bit off_t correctly</para>
4403 This bug prevented proper restoration from archive files exceeding 4Gb.
4405 <listitem><para>Make contrib/dblink not assume that local and remote type OIDs
4406 match (Joe)</para></listitem>
4407 <listitem><para>Quote connectby()'s start_with argument properly (Joe)</para></listitem>
4408 <listitem><para>Don't crash when a rowtype argument to a plpgsql function is
4409 NULL</para></listitem>
4410 <listitem><para>Avoid generating invalid character encoding sequences in
4411 corner cases when planning LIKE operations</para></listitem>
4412 <listitem><para>Ensure text_position() cannot scan past end of source string
4413 in multibyte cases (Korea PostgreSQL Users' Group)</para></listitem>
4414 <listitem><para>Fix index optimization and selectivity estimates for LIKE
4415 operations on bytea columns (Joe)</para></listitem>
4421 <sect1 id="release-7-3-5">
4422 <title>Release 7.3.5</title>
4425 <title>Release date</title>
4426 <simpara>2003-12-03</simpara>
4430 This has a variety of fixes from 7.3.4.
4435 <title>Migration to version 7.3.5</title>
4438 A dump/restore is <emphasis>not</emphasis> required for those
4444 <title>Changes</title>
4447 <listitem><para>Force zero_damaged_pages to be on during recovery from WAL</para></listitem>
4448 <listitem><para>Prevent some obscure cases of <quote>variable not in subplan target lists</quote></para></listitem>
4449 <listitem><para>Force stats processes to detach from shared memory, ensuring cleaner shutdown</para></listitem>
4450 <listitem><para>Make PQescapeBytea and byteaout consistent with each other (Joe)</para></listitem>
4451 <listitem><para>Added missing SPI_finish() calls to dblink's get_tuple_of_interest() (Joe)</para></listitem>
4452 <listitem><para>Fix for possible foreign key violation when rule rewrites INSERT (Jan)</para></listitem>
4453 <listitem><para>Support qualified type names in PL/Tcl's spi_prepare command (Jan)</para></listitem>
4454 <listitem><para>Make pg_dump handle a procedural language handler located in pg_catalog</para></listitem>
4455 <listitem><para>Make pg_dump handle cases where a custom opclass is in another schema</para></listitem>
4456 <listitem><para>Make pg_dump dump binary-compatible casts correctly (Jan)</para></listitem>
4457 <listitem><para>Fix insertion of expressions containing subqueries into rule bodies</para></listitem>
4458 <listitem><para>Fix incorrect argument processing in clusterdb script (Anand Ranganathan)</para></listitem>
4459 <listitem><para>Fix problems with dropped columns in plpython triggers</para></listitem>
4460 <listitem><para>Repair problems with to_char() reading past end of its input string (Karel)</para></listitem>
4461 <listitem><para>Fix GB18030 mapping errors (Tatsuo)</para></listitem>
4462 <listitem><para>Fix several problems with SSL error handling and asynchronous SSL I/O</para></listitem>
4463 <listitem><para>Remove ability to bind a list of values to a single parameter in JDBC
4464 (prevents possible SQL-injection attacks)</para></listitem>
4465 <listitem><para>Fix some errors in HAVE_INT64_TIMESTAMP code paths</para></listitem>
4466 <listitem><para>Fix corner case for btree search in parallel with first root page split</para></listitem>
4472 <sect1 id="release-7-3-4">
4473 <title>Release 7.3.4</title>
4476 <title>Release date</title>
4477 <simpara>2003-07-24</simpara>
4481 This has a variety of fixes from 7.3.3.
4486 <title>Migration to version 7.3.4</title>
4489 A dump/restore is <emphasis>not</emphasis> required for those
4495 <title>Changes</title>
4498 <listitem><para>Repair breakage in timestamp-to-date conversion for dates before 2000</para></listitem>
4499 <listitem><para>Prevent rare possibility of server startup failure (Tom)</para></listitem>
4500 <listitem><para>Fix bugs in interval-to-time conversion (Tom)</para></listitem>
4501 <listitem><para>Add constraint names in a few places in pg_dump (Rod)</para></listitem>
4502 <listitem><para>Improve performance of functions with many parameters (Tom)</para></listitem>
4503 <listitem><para>Fix to_ascii() buffer overruns (Tom)</para></listitem>
4504 <listitem><para>Prevent restore of database comments from throwing an error (Tom)</para></listitem>
4505 <listitem><para>Work around buggy strxfrm() present in some Solaris releases (Tom)</para></listitem>
4506 <listitem><para>Properly escape jdbc setObject() strings to improve security (Barry)</para></listitem>
4512 <sect1 id="release-7-3-3">
4513 <title>Release 7.3.3</title>
4516 <title>Release date</title>
4517 <simpara>2003-05-22</simpara>
4521 This release contains a variety of fixes for version 7.3.2.
4525 <title>Migration to version 7.3.3</title>
4528 A dump/restore is <emphasis>not</emphasis> required for those
4529 running version 7.3.*.
4534 <title>Changes</title>
4537 <listitem><para>Repair sometimes-incorrect computation of StartUpID after a crash</para></listitem>
4538 <listitem><para>Avoid slowness with lots of deferred triggers in one transaction (Stephan)</para></listitem>
4539 <listitem><para>Don't lock referenced row when <command>UPDATE</command> doesn't change foreign key's value (Jan)</para></listitem>
4540 <listitem><para>Use <command>-fPIC</command> not <command>-fpic</command> on Sparc (Tom Callaway)</para></listitem>
4541 <listitem><para>Repair lack of schema-awareness in contrib/reindexdb</para></listitem>
4542 <listitem><para>Fix contrib/intarray error for zero-element result array (Teodor)</para></listitem>
4543 <listitem><para>Ensure createuser script will exit on control-C (Oliver)</para></listitem>
4544 <listitem><para>Fix errors when the type of a dropped column has itself been dropped</para></listitem>
4545 <listitem><para><command>CHECKPOINT</command> does not cause database panic on failure in noncritical steps</para></listitem>
4546 <listitem><para>Accept 60 in seconds fields of timestamp, time, interval input values</para></listitem>
4547 <listitem><para>Issue notice, not error, if <type>TIMESTAMP</type>,
4548 <type> TIME</type>, or <type>INTERVAL</type> precision too large</para></listitem>
4549 <listitem><para>Fix <function>abstime-to-time</function> cast function (fix is
4550 not applied unless you <application>initdb</application>)</para></listitem>
4551 <listitem><para>Fix <application>pg_proc</application> entry for
4552 <type>timestampt_izone</type> (fix is not applied unless you
4553 <application>initdb</application>)</para></listitem>
4554 <listitem><para>Make <function>EXTRACT(EPOCH FROM timestamp without time zone)</function> treat input as local time</para></listitem>
4555 <listitem><para><command>'now'::timestamptz</command> gave wrong answer if timezone changed earlier in transaction</para></listitem>
4556 <listitem><para><envar>HAVE_INT64_TIMESTAMP</envar> code for time with timezone overwrote its input</para></listitem>
4557 <listitem><para>Accept <command>GLOBAL TEMP/TEMPORARY</command> as a
4558 synonym for <command>TEMPORARY</command></para></listitem>
4559 <listitem><para>Avoid improper schema-privilege-check failure in foreign-key triggers</para></listitem>
4560 <listitem><para>Fix bugs in foreign-key triggers for <command>SET DEFAULT</command> action</para></listitem>
4561 <listitem><para>Fix incorrect time-qual check in row fetch for
4562 <command>UPDATE</command> and <command>DELETE</command> triggers</para></listitem>
4563 <listitem><para>Foreign-key clauses were parsed but ignored in
4564 <command>ALTER TABLE ADD COLUMN</command></para></listitem>
4565 <listitem><para>Fix createlang script breakage for case where handler function already exists</para></listitem>
4566 <listitem><para>Fix misbehavior on zero-column tables in <application>pg_dump</application>, COPY, ANALYZE, other places</para></listitem>
4567 <listitem><para>Fix misbehavior of <function>func_error()</function> on type names containing '%'</para></listitem>
4568 <listitem><para>Fix misbehavior of <function>replace()</function> on strings containing '%'</para></listitem>
4569 <listitem><para>Regular-expression patterns containing certain multibyte characters failed</para></listitem>
4570 <listitem><para>Account correctly for <command>NULL</command>s in more cases in join size estimation</para></listitem>
4571 <listitem><para>Avoid conflict with system definition of <function>isblank()</function> function or macro</para></listitem>
4572 <listitem><para>Fix failure to convert large code point values in EUC_TW conversions (Tatsuo)</para></listitem>
4573 <listitem><para>Fix error recovery for <function>SSL_read</function>/<function>SSL_write</function> calls</para></listitem>
4574 <listitem><para>Don't do early constant-folding of type coercion expressions</para></listitem>
4575 <listitem><para>Validate page header fields immediately after reading in any page</para></listitem>
4576 <listitem><para>Repair incorrect check for ungrouped variables in unnamed joins</para></listitem>
4577 <listitem><para>Fix buffer overrun in <function>to_ascii</function> (Guido Notari)</para></listitem>
4578 <listitem><para>contrib/ltree fixes (Teodor)</para></listitem>
4579 <listitem><para>Fix core dump in deadlock detection on machines where char is unsigned</para></listitem>
4580 <listitem><para>Avoid running out of buffers in many-way indexscan (bug introduced in 7.3)</para></listitem>
4581 <listitem><para>Fix planner's selectivity estimation functions to handle domains properly</para></listitem>
4582 <listitem><para>Fix <application>dbmirror</application> memory-allocation bug (Steven Singer)</para></listitem>
4583 <listitem><para>Prevent infinite loop in <function>ln(numeric)</function> due to roundoff error</para></listitem>
4584 <listitem><para><command>GROUP BY</command> got confused if there were multiple equal GROUP BY items</para></listitem>
4585 <listitem><para>Fix bad plan when inherited <command>UPDATE</command>/<command>DELETE</command> references another inherited table</para></listitem>
4586 <listitem><para>Prevent clustering on incomplete (partial or non-NULL-storing) indexes</para></listitem>
4587 <listitem><para>Service shutdown request at proper time if it arrives while still starting up</para></listitem>
4588 <listitem><para>Fix left-links in temporary indexes (could make backwards scans miss entries)</para></listitem>
4589 <listitem><para>Fix incorrect handling of client_encoding setting in postgresql.conf (Tatsuo)</para></listitem>
4590 <listitem><para>Fix failure to respond to <command>pg_ctl stop -m fast</command> after Async_NotifyHandler runs</para></listitem>
4591 <listitem><para>Fix SPI for case where rule contains multiple statements of the same type</para></listitem>
4592 <listitem><para>Fix problem with checking for wrong type of access privilege in rule query</para></listitem>
4593 <listitem><para>Fix problem with <command>EXCEPT</command> in <command>CREATE RULE</command></para></listitem>
4594 <listitem><para>Prevent problem with dropping temp tables having serial columns</para></listitem>
4595 <listitem><para>Fix replace_vars_with_subplan_refs failure in complex views</para></listitem>
4596 <listitem><para>Fix regexp slowness in single-byte encodings (Tatsuo)</para></listitem>
4597 <listitem><para>Allow qualified type names in <command>CREATE CAST</command>
4598 and <command> DROP CAST</command></para></listitem>
4599 <listitem><para>Accept <function>SETOF type[]</function>, which formerly had to
4600 be written <function>SETOF _type</function></para></listitem>
4601 <listitem><para>Fix <application>pg_dump</application> core dump in some cases with procedural languages</para></listitem>
4602 <listitem><para>Force ISO datestyle in <application>pg_dump</application> output, for portability (Oliver)</para></listitem>
4603 <listitem><para><application>pg_dump</application> failed to handle error return
4604 from <function>lo_read</function> (Oleg Drokin)</para></listitem>
4605 <listitem><para><application>pg_dumpall</application> failed with groups having no members (Nick Eskelinen)</para></listitem>
4606 <listitem><para><application>pg_dumpall</application> failed to recognize --globals-only switch</para></listitem>
4607 <listitem><para>pg_restore failed to restore blobs if -X disable-triggers is specified</para></listitem>
4608 <listitem><para>Repair intrafunction memory leak in plpgsql</para></listitem>
4609 <listitem><para>pltcl's <command>elog</command> command dumped core if given wrong parameters (Ian Harding)</para></listitem>
4610 <listitem><para>plpython used wrong value of <envar>atttypmod</envar> (Brad McLean)</para></listitem>
4611 <listitem><para>Fix improper quoting of boolean values in Python interface (D'Arcy)</para></listitem>
4612 <listitem><para>Added <function>addDataType()</function> method to PGConnection interface for JDBC</para></listitem>
4613 <listitem><para>Fixed various problems with updateable ResultSets for JDBC (Shawn Green)</para></listitem>
4614 <listitem><para>Fixed various problems with DatabaseMetaData for JDBC (Kris Jurka, Peter Royal)</para></listitem>
4615 <listitem><para>Fixed problem with parsing table ACLs in JDBC</para></listitem>
4616 <listitem><para>Better error message for character set conversion problems in JDBC</para></listitem>
4622 <sect1 id="release-7-3-2">
4623 <title>Release 7.3.2</title>
4626 <title>Release date</title>
4627 <simpara>2003-02-04</simpara>
4631 This release contains a variety of fixes for version 7.3.1.
4636 <title>Migration to version 7.3.2</title>
4639 A dump/restore is <emphasis>not</emphasis> required for those
4640 running version 7.3.*.
4645 <title>Changes</title>
4648 <listitem><para>Restore creation of OID column in CREATE TABLE AS / SELECT INTO</para></listitem>
4649 <listitem><para>Fix <application>pg_dump</> core dump when dumping views having comments</para></listitem>
4650 <listitem><para>Dump DEFERRABLE/INITIALLY DEFERRED constraints properly</para></listitem>
4651 <listitem><para>Fix UPDATE when child table's column numbering differs from parent</para></listitem>
4652 <listitem><para>Increase default value of max_fsm_relations</para></listitem>
4653 <listitem><para>Fix problem when fetching backwards in a cursor for a single-row query</para></listitem>
4654 <listitem><para>Make backward fetch work properly with cursor on SELECT DISTINCT query</para></listitem>
4655 <listitem><para>Fix problems with loading <application>pg_dump</> files containing contrib/lo usage</para></listitem>
4656 <listitem><para>Fix problem with all-numeric user names</para></listitem>
4657 <listitem><para>Fix possible memory leak and core dump during disconnect in libpgtcl</para></listitem>
4658 <listitem><para>Make plpython's spi_execute command handle nulls properly (Andrew Bosma)</para></listitem>
4659 <listitem><para>Adjust plpython error reporting so that its regression test passes again</para></listitem>
4660 <listitem><para>Work with bison 1.875</para></listitem>
4661 <listitem><para>Handle mixed-case names properly in plpgsql's %type (Neil)</para></listitem>
4662 <listitem><para>Fix core dump in pltcl when executing a query rewritten by a rule</para></listitem>
4663 <listitem><para>Repair array subscript overruns (per report from Yichen Xie)</para></listitem>
4664 <listitem><para>Reduce MAX_TIME_PRECISION from 13 to 10 in floating-point case</para></listitem>
4665 <listitem><para>Correctly case-fold variable names in per-database and per-user settings</para></listitem>
4666 <listitem><para>Fix coredump in plpgsql's RETURN NEXT when SELECT into record returns no rows</para></listitem>
4667 <listitem><para>Fix outdated use of pg_type.typprtlen in python client interface</para></listitem>
4668 <listitem><para>Correctly handle fractional seconds in timestamps in JDBC driver</para></listitem>
4669 <listitem><para>Improve performance of getImportedKeys() in JDBC</para></listitem>
4670 <listitem><para>Make shared-library symlinks work standardly on HPUX (Giles)</para></listitem>
4671 <listitem><para>Repair inconsistent rounding behavior for timestamp, time, interval</para></listitem>
4672 <listitem><para>SSL negotiation fixes (Nathan Mueller)</para></listitem>
4673 <listitem><para>Make libpq's ~/.pgpass feature work when connecting with PQconnectDB</para></listitem>
4674 <listitem><para>Update my2pg, ora2pg</para></listitem>
4675 <listitem><para>Translation updates</para></listitem>
4676 <listitem><para>Add casts between types lo and oid in contrib/lo</para></listitem>
4677 <listitem><para>fastpath code now checks for privilege to call function</para></listitem>
4683 <sect1 id="release-7-3-1">
4684 <title>Release 7.3.1</title>
4687 <title>Release date</title>
4688 <simpara>2002-12-18</simpara>
4692 This release contains a variety of fixes for version 7.3.
4697 <title>Migration to version 7.3.1</title>
4700 A dump/restore is <emphasis>not</emphasis> required for those
4701 running version 7.3. However, it should be noted that the main
4702 <productname>PostgreSQL</productname> interface library, libpq,
4703 has a new major version number for this release, which may require
4704 recompilation of client code in certain cases.
4709 <title>Changes</title>
4712 <listitem><para>Fix a core dump of COPY TO when client/server encodings don't match (Tom)</para></listitem>
4713 <listitem><para>Allow <application>pg_dump</> to work with pre-7.2 servers (Philip)</para></listitem>
4714 <listitem><para>contrib/adddepend fixes (Tom)</para></listitem>
4715 <listitem><para>Fix problem with deletion of per-user/per-database config settings (Tom)</para></listitem>
4716 <listitem><para>contrib/vacuumlo fix (Tom)</para></listitem>
4717 <listitem><para>Allow 'password' encryption even when pg_shadow contains MD5 passwords (Bruce)</para></listitem>
4718 <listitem><para>contrib/dbmirror fix (Steven Singer)</para></listitem>
4719 <listitem><para>Optimizer fixes (Tom)</para></listitem>
4720 <listitem><para>contrib/tsearch fixes (Teodor Sigaev, Magnus)</para></listitem>
4721 <listitem><para>Allow locale names to be mixed case (Nicolai Tufar)</para></listitem>
4722 <listitem><para>Increment libpq library's major version number (Bruce)</para></listitem>
4723 <listitem><para>pg_hba.conf error reporting fixes (Bruce, Neil)</para></listitem>
4724 <listitem><para>Add SCO Openserver 5.0.4 as a supported platform (Bruce)</para></listitem>
4725 <listitem><para>Prevent EXPLAIN from crashing server (Tom)</para></listitem>
4726 <listitem><para>SSL fixes (Nathan Mueller)</para></listitem>
4727 <listitem><para>Prevent composite column creation via ALTER TABLE (Tom)</para></listitem>
4733 <sect1 id="release-7-3">
4734 <title>Release 7.3</title>
4737 <title>Release date</title>
4738 <simpara>2002-11-27</simpara>
4742 <title>Overview</title>
4745 Major changes in this release:
4749 <term>Schemas</term>
4752 Schemas allow users to create objects in separate namespaces,
4753 so two people or applications can have tables with the same
4754 name. There is also a public schema for shared tables.
4755 Table/index creation can be restricted by removing privileges
4756 on the public schema.
4762 <term>Drop Column</term>
4765 <productname>PostgreSQL</productname> now supports the
4766 <literal>ALTER TABLE ... DROP COLUMN</literal> functionality.
4772 <term>Table Functions</term>
4775 Functions returning multiple rows and/or multiple columns are
4776 now much easier to use than before. You can call such a
4777 <quote>table function</quote> in the <literal>SELECT</literal>
4778 <literal>FROM</literal> clause, treating its output like a
4779 table. Also, <application>PL/pgSQL</application> functions can
4786 <term>Prepared Queries</term>
4789 <productname>PostgreSQL</productname> now supports prepared
4790 queries, for improved performance.
4796 <term>Dependency Tracking</term>
4799 <productname>PostgreSQL</productname> now records object
4800 dependencies, which allows improvements in many areas.
4801 <command>DROP</command> statements now take either
4802 <literal>CASCADE</> or <literal>RESTRICT</> to control whether
4803 dependent objects are also dropped.
4809 <term>Privileges</term>
4812 Functions and procedural languages now have privileges, and
4813 functions can be defined to run with the privileges of their
4820 <term>Internationalization</term>
4823 Both multibyte and locale support are now always enabled.
4829 <term>Logging</term>
4832 A variety of logging options have been enhanced.
4838 <term>Interfaces</term>
4841 A large number of interfaces have been moved to <ulink
4842 url="http://gborg.postgresql.org">http://gborg.postgresql.org</>
4843 where they can be developed and released independently.
4849 <term>Functions/Identifiers</term>
4852 By default, functions can now take up to 32 parameters, and
4853 identifiers can be up to 63 bytes long. Also, <literal>OPAQUE</>
4854 is now deprecated: there are specific <quote>pseudo-datatypes</>
4855 to represent each of the former meanings of <literal>OPAQUE</>
4856 in function argument and result types.
4866 <title>Migration to version 7.3</title>
4869 A dump/restore using <application>pg_dump</> is required for those
4870 wishing to migrate data from any previous release. If your
4871 application examines the system catalogs, additional changes will
4872 be required due to the introduction of schemas in 7.3; for more
4873 information, see: <ulink
4874 url="http://developer.postgresql.org/~momjian/upgrade_tips_7.3">
4875 http://developer.postgresql.org/~momjian/upgrade_tips_7.3</>.
4879 Observe the following incompatibilities:
4884 Pre-6.3 clients are no longer supported.
4890 <filename>pg_hba.conf</filename> now has a column for the user
4891 name and additional features. Existing files need to be
4898 Several <filename>postgresql.conf</filename> logging parameters
4905 <literal>LIMIT #,#</literal> has been disabled; use
4906 <literal>LIMIT # OFFSET #</literal>.
4912 <command>INSERT</command> statements with column lists must
4913 specify a value for each specified column. For example,
4914 <literal>INSERT INTO tab (col1, col2) VALUES ('val1')</literal>
4915 is now invalid. It's still allowed to supply fewer columns than
4916 expected if the <command>INSERT</command> does not have a column list.
4922 <type>serial</type> columns are no longer automatically
4923 <literal>UNIQUE</>; thus, an index will not automatically be
4930 A <command>SET</command> command inside an aborted transaction
4937 <command>COPY</command> no longer considers missing trailing
4938 columns to be null. All columns need to be specified.
4939 (However, one may achieve a similar effect by specifying a
4940 column list in the <command>COPY</command> command.)
4946 The data type <type>timestamp</type> is now equivalent to
4947 <type>timestamp without time zone</type>, instead of
4948 <type>timestamp with time zone</type>.
4954 Pre-7.3 databases loaded into 7.3 will not have the new object
4955 dependencies for <type>serial</type> columns, unique
4956 constraints, and foreign keys. See the directory
4957 <filename>contrib/adddepend/</filename> for a detailed
4958 description and a script that will add such dependencies.
4964 An empty string (<literal>''</literal>) is no longer allowed as
4965 the input into an integer field. Formerly, it was silently
4976 <title>Changes</title>
4979 <title>Server Operation</title>
4981 <listitem><para>Add pg_locks view to show locks (Neil)</para></listitem>
4982 <listitem><para>Security fixes for password negotiation memory allocation (Neil)</para></listitem>
4983 <listitem><para>Remove support for version 0 FE/BE protocol (<productname>PostgreSQL</productname> 6.2 and earlier) (Tom)</para></listitem>
4984 <listitem><para>Reserve the last few backend slots for superusers, add parameter superuser_reserved_connections to control this (Nigel J. Andrews)</para></listitem>
4989 <title>Performance</title>
4991 <listitem><para>Improve startup by calling localtime() only once (Tom)</para></listitem>
4992 <listitem><para>Cache system catalog information in flat files for faster startup (Tom)</para></listitem>
4993 <listitem><para>Improve caching of index information (Tom)</para></listitem>
4994 <listitem><para>Optimizer improvements (Tom, Fernando Nasser)</para></listitem>
4995 <listitem><para>Catalog caches now store failed lookups (Tom)</para></listitem>
4996 <listitem><para>Hash function improvements (Neil)</para></listitem>
4997 <listitem><para>Improve performance of query tokenization and network handling (Peter)</para></listitem>
4998 <listitem><para>Speed improvement for large object restore (Mario Weilguni)</para></listitem>
4999 <listitem><para>Mark expired index entries on first lookup, saving later heap fetches (Tom)</para></listitem>
5000 <listitem><para>Avoid excessive NULL bitmap padding (Manfred Koizar)</para></listitem>
5001 <listitem><para>Add BSD-licensed qsort() for Solaris, for performance (Bruce)</para></listitem>
5002 <listitem><para>Reduce per-row overhead by four bytes (Manfred Koizar)</para></listitem>
5003 <listitem><para>Fix GEQO optimizer bug (Neil Conway)</para></listitem>
5004 <listitem><para>Make WITHOUT OID actually save four bytes per row (Manfred Koizar)</para></listitem>
5005 <listitem><para>Add default_statistics_target variable to specify ANALYZE buckets (Neil)</para></listitem>
5006 <listitem><para>Use local buffer cache for temporary tables so no WAL overhead (Tom)</para></listitem>
5007 <listitem><para>Improve free space map performance on large tables (Stephen Marshall, Tom)</para></listitem>
5008 <listitem><para>Improved WAL write concurrency (Tom)</para></listitem>
5013 <title>Privileges</title>
5015 <listitem><para>Add privileges on functions and procedural languages (Peter)</para></listitem>
5016 <listitem><para>Add OWNER to CREATE DATABASE so superusers can create databases on behalf of unprivileged users (Gavin Sherry, Tom)</para></listitem>
5017 <listitem><para>Add new object privilege bits EXECUTE and USAGE (Tom)</para></listitem>
5018 <listitem><para>Add SET SESSION AUTHORIZATION DEFAULT and RESET SESSION AUTHORIZATION (Tom)</para></listitem>
5019 <listitem><para>Allow functions to be executed with the privilege of the function owner (Peter)</para></listitem>
5024 <title>Server Configuration</title>
5026 <listitem><para>Server log messages now tagged with LOG, not DEBUG (Bruce)</para></listitem>
5027 <listitem><para>Add user column to pg_hba.conf (Bruce)</para></listitem>
5028 <listitem><para>Have log_connections output two lines in log file (Tom)</para></listitem>
5029 <listitem><para>Remove debug_level from postgresql.conf, now server_min_messages (Bruce)</para></listitem>
5030 <listitem><para>New ALTER DATABASE/USER ... SET command for per-user/database initialization (Peter)</para></listitem>
5031 <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>
5032 <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>
5033 <listitem><para>Remove secondary password file capability and pg_password utility (Bruce)</para></listitem>
5034 <listitem><para>Add variable db_user_namespace for database-local user names (Bruce)</para></listitem>
5035 <listitem><para>SSL improvements (Bear Giles)</para></listitem>
5036 <listitem><para>Make encryption of stored passwords the default (Bruce)</para></listitem>
5037 <listitem><para>Allow pg_statistics to be reset by calling pg_stat_reset() (Christopher)</para></listitem>
5038 <listitem><para>Add log_duration parameter (Bruce)</para></listitem>
5039 <listitem><para>Rename debug_print_query to log_statement (Bruce)</para></listitem>
5040 <listitem><para>Rename show_query_stats to show_statement_stats (Bruce)</para></listitem>
5041 <listitem><para>Add param log_min_error_statement to print commands to logs on error (Gavin)</para></listitem>
5046 <title>Queries</title>
5048 <listitem><para>Make cursors insensitive, meaning their contents do not change (Tom)</para></listitem>
5049 <listitem><para>Disable LIMIT #,# syntax; now only LIMIT # OFFSET # supported (Bruce)</para></listitem>
5050 <listitem><para>Increase identifier length to 63 (Neil, Bruce)</para></listitem>
5051 <listitem><para>UNION fixes for merging >= 3 columns of different lengths (Tom)</para></listitem>
5052 <listitem><para>Add DEFAULT key word to INSERT, e.g., INSERT ... (..., DEFAULT, ...) (Rod)</para></listitem>
5053 <listitem><para>Allow views to have default values using ALTER COLUMN ... SET DEFAULT (Neil)</para></listitem>
5054 <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>
5055 <listitem><para>Fix for join aliases (Tom)</para></listitem>
5056 <listitem><para>Fix for FULL OUTER JOINs (Tom)</para></listitem>
5057 <listitem><para>Improve reporting of invalid identifier and location (Tom, Gavin)</para></listitem>
5058 <listitem><para>Fix OPEN cursor(args) (Tom)</para></listitem>
5059 <listitem><para>Allow 'ctid' to be used in a view and currtid(viewname) (Hiroshi)</para></listitem>
5060 <listitem><para>Fix for CREATE TABLE AS with UNION (Tom)</para></listitem>
5061 <listitem><para>SQL99 syntax improvements (Thomas)</para></listitem>
5062 <listitem><para>Add statement_timeout variable to cancel queries (Bruce)</para></listitem>
5063 <listitem><para>Allow prepared queries with PREPARE/EXECUTE (Neil)</para></listitem>
5064 <listitem><para>Allow FOR UPDATE to appear after LIMIT/OFFSET (Bruce)</para></listitem>
5065 <listitem><para>Add variable autocommit (Tom, David Van Wie)</para></listitem>
5070 <title>Object Manipulation</title>
5072 <listitem><para>Make equals signs optional in CREATE DATABASE (Gavin Sherry)</para></listitem>
5073 <listitem><para>Make ALTER TABLE OWNER change index ownership too (Neil)</para></listitem>
5074 <listitem><para>New ALTER TABLE tabname ALTER COLUMN colname SET STORAGE controls TOAST storage, compression (John Gray)</para></listitem>
5075 <listitem><para>Add schema support, CREATE/DROP SCHEMA (Tom)</para></listitem>
5076 <listitem><para>Create schema for temporary tables (Tom)</para></listitem>
5077 <listitem><para>Add variable search_path for schema search (Tom)</para></listitem>
5078 <listitem><para>Add ALTER TABLE SET/DROP NOT NULL (Christopher)</para></listitem>
5079 <listitem><para>New CREATE FUNCTION volatility levels (Tom)</para></listitem>
5080 <listitem><para>Make rule names unique only per table (Tom)</para></listitem>
5081 <listitem><para>Add 'ON tablename' clause to DROP RULE and COMMENT ON RULE (Tom)</para></listitem>
5082 <listitem><para>Add ALTER TRIGGER RENAME (Joe)</para></listitem>
5083 <listitem><para>New current_schema() and current_schemas() inquiry functions (Tom)</para></listitem>
5084 <listitem><para>Allow functions to return multiple rows (table functions) (Joe)</para></listitem>
5085 <listitem><para>Make WITH optional in CREATE DATABASE, for consistency (Bruce)</para></listitem>
5086 <listitem><para>Add object dependency tracking (Rod, Tom)</para></listitem>
5087 <listitem><para>Add RESTRICT/CASCADE to DROP commands (Rod)</para></listitem>
5088 <listitem><para>Add ALTER TABLE DROP for non-CHECK CONSTRAINT (Rod)</para></listitem>
5089 <listitem><para>Autodestroy sequence on DROP of table with SERIAL (Rod)</para></listitem>
5090 <listitem><para>Prevent column dropping if column is used by foreign key (Rod)</para></listitem>
5091 <listitem><para>Automatically drop constraints/functions when object is dropped (Rod)</para></listitem>
5092 <listitem><para>Add CREATE/DROP OPERATOR CLASS (Bill Studenmund, Tom)</para></listitem>
5093 <listitem><para>Add ALTER TABLE DROP COLUMN (Christopher, Tom, Hiroshi)</para></listitem>
5094 <listitem><para>Prevent inherited columns from being removed or renamed (Alvaro Herrera)</para></listitem>
5095 <listitem><para>Fix foreign key constraints to not error on intermediate database states (Stephan)</para></listitem>
5096 <listitem><para>Propagate column or table renaming to foreign key constraints</para></listitem>
5097 <listitem><para>Add CREATE OR REPLACE VIEW (Gavin, Neil, Tom)</para></listitem>
5098 <listitem><para>Add CREATE OR REPLACE RULE (Gavin, Neil, Tom)</para></listitem>
5099 <listitem><para>Have rules execute alphabetically, returning more predictable values (Tom)</para></listitem>
5100 <listitem><para>Triggers are now fired in alphabetical order (Tom)</para></listitem>
5101 <listitem><para>Add /contrib/adddepend to handle pre-7.3 object dependencies (Rod)</para></listitem>
5102 <listitem><para>Allow better casting when inserting/updating values (Tom)</para></listitem>
5107 <title>Utility Commands</title>
5109 <listitem><para>Have COPY TO output embedded carriage returns and newlines as \r and \n (Tom)</para></listitem>
5110 <listitem><para>Allow DELIMITER in COPY FROM to be 8-bit clean (Tatsuo)</para></listitem>
5111 <listitem><para>Make <application>pg_dump</> use ALTER TABLE ADD PRIMARY KEY, for performance (Neil)</para></listitem>
5112 <listitem><para>Disable brackets in multistatement rules (Bruce)</para></listitem>
5113 <listitem><para>Disable VACUUM from being called inside a function (Bruce)</para></listitem>
5114 <listitem><para>Allow dropdb and other scripts to use identifiers with spaces (Bruce)</para></listitem>
5115 <listitem><para>Restrict database comment changes to the current database</para></listitem>
5116 <listitem><para>Allow comments on operators, independent of the underlying function (Rod)</para></listitem>
5117 <listitem><para>Rollback SET commands in aborted transactions (Tom)</para></listitem>
5118 <listitem><para>EXPLAIN now outputs as a query (Tom)</para></listitem>
5119 <listitem><para>Display condition expressions and sort keys in EXPLAIN (Tom)</para></listitem>
5120 <listitem><para>Add 'SET LOCAL var = value' to set configuration variables for a single transaction (Tom)</para></listitem>
5121 <listitem><para>Allow ANALYZE to run in a transaction (Bruce)</para></listitem>
5122 <listitem><para>Improve COPY syntax using new WITH clauses, keep backward compatibility (Bruce)</para></listitem>
5123 <listitem><para>Fix <application>pg_dump</> to consistently output tags in non-ASCII dumps (Bruce)</para></listitem>
5124 <listitem><para>Make foreign key constraints clearer in dump file (Rod)</para></listitem>
5125 <listitem><para>Add COMMENT ON CONSTRAINT (Rod)</para></listitem>
5126 <listitem><para>Allow COPY TO/FROM to specify column names (Brent Verner)</para></listitem>
5127 <listitem><para>Dump UNIQUE and PRIMARY KEY constraints as ALTER TABLE (Rod)</para></listitem>
5128 <listitem><para>Have SHOW output a query result (Joe)</para></listitem>
5129 <listitem><para>Generate failure on short COPY lines rather than pad NULLs (Neil)</para></listitem>
5130 <listitem><para>Fix CLUSTER to preserve all table attributes (Alvaro Herrera)</para></listitem>
5131 <listitem><para>New pg_settings table to view/modify GUC settings (Joe)</para></listitem>
5132 <listitem><para>Add smart quoting, portability improvements to <application>pg_dump</> output (Peter)</para></listitem>
5133 <listitem><para>Dump serial columns out as SERIAL (Tom)</para></listitem>
5134 <listitem><para>Enable large file support, >2G for <application>pg_dump</> (Peter, Philip Warner, Bruce)</para></listitem>
5135 <listitem><para>Disallow TRUNCATE on tables that are involved in referential constraints (Rod)</para></listitem>
5136 <listitem><para>Have TRUNCATE also auto-truncate the toast table of the relation (Tom)</para></listitem>
5137 <listitem><para>Add clusterdb utility that will auto-cluster an entire database based on previous CLUSTER operations (Alvaro Herrera)</para></listitem>
5138 <listitem><para>Overhaul pg_dumpall (Peter)</para></listitem>
5139 <listitem><para>Allow REINDEX of TOAST tables (Tom)</para></listitem>
5140 <listitem><para>Implemented START TRANSACTION, per SQL99 (Neil)</para></listitem>
5141 <listitem><para>Fix rare index corruption when a page split affects bulk delete (Tom)</para></listitem>
5142 <listitem><para>Fix ALTER TABLE ... ADD COLUMN for inheritance (Alvaro Herrera)</para></listitem>
5147 <title>Data Types and Functions</title>
5149 <listitem><para>Fix factorial(0) to return 1 (Bruce)</para></listitem>
5150 <listitem><para>Date/time/timezone improvements (Thomas)</para></listitem>
5151 <listitem><para>Fix for array slice extraction (Tom)</para></listitem>
5152 <listitem><para>Fix extract/date_part to report proper microseconds for timestamp (Tatsuo)</para></listitem>
5153 <listitem><para>Allow text_substr() and bytea_substr() to read TOAST values more efficiently (John Gray)</para></listitem>
5154 <listitem><para>Add domain support (Rod)</para></listitem>
5155 <listitem><para>Make WITHOUT TIME ZONE the default for TIMESTAMP and TIME data types (Thomas)</para></listitem>
5156 <listitem><para>Allow alternate storage scheme of 64-bit integers for date/time types using --enable-integer-datetimes in configure (Thomas)</para></listitem>
5157 <listitem><para>Make timezone(timestamptz) return timestamp rather than a string (Thomas)</para></listitem>
5158 <listitem><para>Allow fractional seconds in date/time types for dates prior to 1BC (Thomas)</para></listitem>
5159 <listitem><para>Limit timestamp data types to 6 decimal places of precision (Thomas)</para></listitem>
5160 <listitem><para>Change timezone conversion functions from timetz() to timezone() (Thomas)</para></listitem>
5161 <listitem><para>Add configuration variables datestyle and timezone (Tom)</para></listitem>
5162 <listitem><para>Add OVERLAY(), which allows substitution of a substring in a string (Thomas)</para></listitem>
5163 <listitem><para>Add SIMILAR TO (Thomas, Tom)</para></listitem>
5164 <listitem><para>Add regular expression SUBSTRING(string FROM pat FOR escape) (Thomas)</para></listitem>
5165 <listitem><para>Add LOCALTIME and LOCALTIMESTAMP functions (Thomas)</para></listitem>
5166 <listitem><para>Add named composite types using CREATE TYPE typename AS (column) (Joe)</para></listitem>
5167 <listitem><para>Allow composite type definition in the table alias clause (Joe)</para></listitem>
5168 <listitem><para>Add new API to simplify creation of C language table functions (Joe)</para></listitem>
5169 <listitem><para>Remove ODBC-compatible empty parentheses from calls to SQL99 functions for which these parentheses do not match the standard (Thomas)</para></listitem>
5170 <listitem><para>Allow macaddr data type to accept 12 hex digits with no separators (Mike Wyer)</para></listitem>
5171 <listitem><para>Add CREATE/DROP CAST (Peter)</para></listitem>
5172 <listitem><para>Add IS DISTINCT FROM operator (Thomas)</para></listitem>
5173 <listitem><para>Add SQL99 TREAT() function, synonym for CAST() (Thomas)</para></listitem>
5174 <listitem><para>Add pg_backend_pid() to output backend pid (Bruce)</para></listitem>
5175 <listitem><para>Add IS OF / IS NOT OF type predicate (Thomas)</para></listitem>
5176 <listitem><para>Allow bit string constants without fully-specified length (Thomas)</para></listitem>
5177 <listitem><para>Allow conversion between 8-byte integers and bit strings (Thomas)</para></listitem>
5178 <listitem><para>Implement hex literal conversion to bit string literal (Thomas)</para></listitem>
5179 <listitem><para>Allow table functions to appear in the FROM clause (Joe)</para></listitem>
5180 <listitem><para>Increase maximum number of function parameters to 32 (Bruce)</para></listitem>
5181 <listitem><para>No longer automatically create index for SERIAL column (Tom)</para></listitem>
5182 <listitem><para>Add current_database() (Rod)</para></listitem>
5183 <listitem><para>Fix cash_words() to not overflow buffer (Tom)</para></listitem>
5184 <listitem><para>Add functions replace(), split_part(), to_hex() (Joe)</para></listitem>
5185 <listitem><para>Fix LIKE for bytea as a right-hand argument (Joe)</para></listitem>
5186 <listitem><para>Prevent crashes caused by SELECT cash_out(2) (Tom)</para></listitem>
5187 <listitem><para>Fix to_char(1,'FM999.99') to return a period (Karel)</para></listitem>
5188 <listitem><para>Fix trigger/type/language functions returning OPAQUE to return proper type (Tom)</para></listitem>
5193 <title>Internationalization</title>
5195 <listitem><para>Add additional encodings: Korean (JOHAB), Thai (WIN874), Vietnamese (TCVN), Arabic (WIN1256), Simplified Chinese (GBK), Korean (UHC) (Eiji Tokuya)</para></listitem>
5196 <listitem><para>Enable locale support by default (Peter)</para></listitem>
5197 <listitem><para>Add locale variables (Peter)</para></listitem>
5198 <listitem><para>Escape byes >= 0x7f for multibyte in PQescapeBytea/PQunescapeBytea (Tatsuo)</para></listitem>
5199 <listitem><para>Add locale awareness to regular expression character classes</para></listitem>
5200 <listitem><para>Enable multibyte support by default (Tatsuo)</para></listitem>
5201 <listitem><para>Add GB18030 multibyte support (Bill Huang)</para></listitem>
5202 <listitem><para>Add CREATE/DROP CONVERSION, allowing loadable encodings (Tatsuo, Kaori)</para></listitem>
5203 <listitem><para>Add pg_conversion table (Tatsuo)</para></listitem>
5204 <listitem><para>Add SQL99 CONVERT() function (Tatsuo)</para></listitem>
5205 <listitem><para>pg_dumpall, pg_controldata, and pg_resetxlog now national-language aware (Peter)</para></listitem>
5206 <listitem><para>New and updated translations</para></listitem>
5211 <title>Server-side Languages</title>
5213 <listitem><para>Allow recursive SQL function (Peter)</para></listitem>
5214 <listitem><para>Change PL/Tcl build to use configured compiler and Makefile.shlib (Peter)</para></listitem>
5215 <listitem><para>Overhaul the PL/pgSQL FOUND variable to be more Oracle-compatible (Neil, Tom)</para></listitem>
5216 <listitem><para>Allow PL/pgSQL to handle quoted identifiers (Tom)</para></listitem>
5217 <listitem><para>Allow set-returning PL/pgSQL functions (Neil)</para></listitem>
5218 <listitem><para>Make PL/pgSQL schema-aware (Joe)</para></listitem>
5219 <listitem><para>Remove some memory leaks (Nigel J. Andrews, Tom)</para></listitem>
5226 <listitem><para>Don't lowercase psql \connect database name for 7.2.0 compatibility (Tom)</para></listitem>
5227 <listitem><para>Add psql \timing to time user queries (Greg Sabino Mullane)</para></listitem>
5228 <listitem><para>Have psql \d show index information (Greg Sabino Mullane)</para></listitem>
5229 <listitem><para>New psql \dD shows domains (Jonathan Eisler)</para></listitem>
5230 <listitem><para>Allow psql to show rules on views (Paul ?)</para></listitem>
5231 <listitem><para>Fix for psql variable substitution (Tom)</para></listitem>
5232 <listitem><para>Allow psql \d to show temporary table structure (Tom)</para></listitem>
5233 <listitem><para>Allow psql \d to show foreign keys (Rod)</para></listitem>
5234 <listitem><para>Fix \? to honor \pset pager (Bruce)</para></listitem>
5235 <listitem><para>Have psql reports its version number on startup (Tom)</para></listitem>
5236 <listitem><para>Allow \copy to specify column names (Tom)</para></listitem>
5241 <title>libpq</title>
5243 <listitem><para>Add $HOME/.pgpass to store host/user password combinations (Alvaro Herrera)</para></listitem>
5244 <listitem><para>Add PQunescapeBytea() function to libpq (Patrick Welche)</para></listitem>
5245 <listitem><para>Fix for sending large queries over non-blocking connections (Bernhard Herzog)</para></listitem>
5246 <listitem><para>Fix for libpq using timers on Win9X (David Ford)</para></listitem>
5247 <listitem><para>Allow libpq notify to handle servers with different-length identifiers (Tom)</para></listitem>
5248 <listitem><para>Add libpq PQescapeString() and PQescapeBytea() to Windows (Bruce)</para></listitem>
5249 <listitem><para>Fix for SSL with non-blocking connections (Jack Bates)</para></listitem>
5250 <listitem><para>Add libpq connection timeout parameter (Denis A Ustimenko)</para></listitem>
5257 <listitem><para>Allow JDBC to compile with JDK 1.4 (Dave)</para></listitem>
5258 <listitem><para>Add JDBC 3 support (Barry)</para></listitem>
5259 <listitem><para>Allows JDBC to set loglevel by adding ?loglevel=X to the connection URL (Barry)</para></listitem>
5260 <listitem><para>Add Driver.info() message that prints out the version number (Barry)</para></listitem>
5261 <listitem><para>Add updateable result sets (Raghu Nidagal, Dave)</para></listitem>
5262 <listitem><para>Add support for callable statements (Paul Bethe)</para></listitem>
5263 <listitem><para>Add query cancel capability</para></listitem>
5264 <listitem><para>Add refresh row (Dave)</para></listitem>
5265 <listitem><para>Fix MD5 encryption handling for multibyte servers (Jun Kawai)</para></listitem>
5266 <listitem><para>Add support for prepared statements (Barry)</para></listitem>
5271 <title>Miscellaneous Interfaces</title>
5273 <listitem><para>Fixed ECPG bug concerning octal numbers in single quotes (Michael)</para></listitem>
5274 <listitem><para>Move src/interfaces/libpgeasy to http://gborg.postgresql.org (Marc, Bruce)</para></listitem>
5275 <listitem><para>Improve Python interface (Elliot Lee, Andrew Johnson, Greg Copeland)</para></listitem>
5276 <listitem><para>Add libpgtcl connection close event (Gerhard Hintermayer)</para></listitem>
5277 <listitem><para>Move src/interfaces/libpq++ to http://gborg.postgresql.org (Marc, Bruce)</para></listitem>
5278 <listitem><para>Move src/interfaces/odbc to http://gborg.postgresql.org (Marc)</para></listitem>
5279 <listitem><para>Move src/interfaces/libpgeasy to http://gborg.postgresql.org (Marc, Bruce)</para></listitem>
5280 <listitem><para>Move src/interfaces/perl5 to http://gborg.postgresql.org (Marc, Bruce)</para></listitem>
5281 <listitem><para>Remove src/bin/pgaccess from main tree, now at http://www.pgaccess.org (Bruce)</para></listitem>
5282 <listitem><para>Add pg_on_connection_loss command to libpgtcl (Gerhard Hintermayer, Tom)</para></listitem>
5287 <title>Source Code</title>
5289 <listitem><para>Fix for parallel make (Peter)</para></listitem>
5290 <listitem><para>AIX fixes for linking Tcl (Andreas Zeugswetter)</para></listitem>
5291 <listitem><para>Allow PL/Perl to build under Cygwin (Jason Tishler)</para></listitem>
5292 <listitem><para>Improve MIPS compiles (Peter, Oliver Elphick)</para></listitem>
5293 <listitem><para>Require Autoconf version 2.53 (Peter)</para></listitem>
5294 <listitem><para>Require readline and zlib by default in configure (Peter)</para></listitem>
5295 <listitem><para>Allow Solaris to use Intimate Shared Memory (ISM), for performance (Scott Brunza, P.J. Josh Rovero)</para></listitem>
5296 <listitem><para>Always enable syslog in compile, remove --enable-syslog option (Tatsuo)</para></listitem>
5297 <listitem><para>Always enable multibyte in compile, remove --enable-multibyte option (Tatsuo)</para></listitem>
5298 <listitem><para>Always enable locale in compile, remove --enable-locale option (Peter)</para></listitem>
5299 <listitem><para>Fix for Win9x DLL creation (Magnus Naeslund)</para></listitem>
5300 <listitem><para>Fix for link() usage by WAL code on Windows, BeOS (Jason Tishler)</para></listitem>
5301 <listitem><para>Add sys/types.h to c.h, remove from main files (Peter, Bruce)</para></listitem>
5302 <listitem><para>Fix AIX hang on SMP machines (Tomoyuki Niijima)</para></listitem>
5303 <listitem><para>AIX SMP hang fix (Tomoyuki Niijima)</para></listitem>
5304 <listitem><para>Fix pre-1970 date handling on newer glibc libraries (Tom)</para></listitem>
5305 <listitem><para>Fix PowerPC SMP locking (Tom)</para></listitem>
5306 <listitem><para>Prevent gcc -ffast-math from being used (Peter, Tom)</para></listitem>
5307 <listitem><para>Bison >= 1.50 now required for developer builds</para></listitem>
5308 <listitem><para>Kerberos 5 support now builds with Heimdal (Peter)</para></listitem>
5309 <listitem><para>Add appendix in the User's Guide which lists SQL features (Thomas)</para></listitem>
5310 <listitem><para>Improve loadable module linking to use RTLD_NOW (Tom)</para></listitem>
5311 <listitem><para>New error levels WARNING, INFO, LOG, DEBUG[1-5] (Bruce)</para></listitem>
5312 <listitem><para>New src/port directory holds replaced libc functions (Peter, Bruce)</para></listitem>
5313 <listitem><para>New pg_namespace system catalog for schemas (Tom)</para></listitem>
5314 <listitem><para>Add pg_class.relnamespace for schemas (Tom)</para></listitem>
5315 <listitem><para>Add pg_type.typnamespace for schemas (Tom)</para></listitem>
5316 <listitem><para>Add pg_proc.pronamespace for schemas (Tom)</para></listitem>
5317 <listitem><para>Restructure aggregates to have pg_proc entries (Tom)</para></listitem>
5318 <listitem><para>System relations now have their own namespace, pg_* test not required (Fernando Nasser)</para></listitem>
5319 <listitem><para>Rename TOAST index names to be *_index rather than *_idx (Neil)</para></listitem>
5320 <listitem><para>Add namespaces for operators, opclasses (Tom)</para></listitem>
5321 <listitem><para>Add additional checks to server control file (Thomas)</para></listitem>
5322 <listitem><para>New Polish FAQ (Marcin Mazurek)</para></listitem>
5323 <listitem><para>Add Posix semaphore support (Tom)</para></listitem>
5324 <listitem><para>Document need for reindex (Bruce)</para></listitem>
5325 <listitem><para>Rename some internal identifiers to simplify Windows compile (Jan, Katherine Ward)</para></listitem>
5326 <listitem><para>Add documentation on computing disk space (Bruce)</para></listitem>
5327 <listitem><para>Remove KSQO from GUC (Bruce)</para></listitem>
5328 <listitem><para>Fix memory leak in rtree (Kenneth Been)</para></listitem>
5329 <listitem><para>Modify a few error messages for consistency (Bruce)</para></listitem>
5330 <listitem><para>Remove unused system table columns (Peter)</para></listitem>
5331 <listitem><para>Make system columns NOT NULL where appropriate (Tom)</para></listitem>
5332 <listitem><para>Clean up use of sprintf in favor of snprintf() (Neil, Jukka Holappa)</para></listitem>
5333 <listitem><para>Remove OPAQUE and create specific subtypes (Tom)</para></listitem>
5334 <listitem><para>Cleanups in array internal handling (Joe, Tom)</para></listitem>
5335 <listitem><para>Disallow pg_atoi('') (Bruce)</para></listitem>
5336 <listitem><para>Remove parameter wal_files because WAL files are now recycled (Bruce)</para></listitem>
5337 <listitem><para>Add version numbers to heap pages (Tom)</para></listitem>
5342 <title>Contrib</title>
5344 <listitem><para>Allow inet arrays in /contrib/array (Neil)</para></listitem>
5345 <listitem><para>GiST fixes (Teodor Sigaev, Neil)</para></listitem>
5346 <listitem><para>Upgrade /contrib/mysql</para></listitem>
5347 <listitem><para>Add /contrib/dbsize which shows table sizes without vacuum (Peter)</para></listitem>
5348 <listitem><para>Add /contrib/intagg, integer aggregator routines (mlw)</para></listitem>
5349 <listitem><para>Improve /contrib/oid2name (Neil, Bruce)</para></listitem>
5350 <listitem><para>Improve /contrib/tsearch (Oleg, Teodor Sigaev)</para></listitem>
5351 <listitem><para>Cleanups of /contrib/rserver (Alexey V. Borzov)</para></listitem>
5352 <listitem><para>Update /contrib/oracle conversion utility (Gilles Darold)</para></listitem>
5353 <listitem><para>Update /contrib/dblink (Joe)</para></listitem>
5354 <listitem><para>Improve options supported by /contrib/vacuumlo (Mario Weilguni)</para></listitem>
5355 <listitem><para>Improvements to /contrib/intarray (Oleg, Teodor Sigaev, Andrey Oktyabrski)</para></listitem>
5356 <listitem><para>Add /contrib/reindexdb utility (Shaun Thomas)</para></listitem>
5357 <listitem><para>Add indexing to /contrib/isbn_issn (Dan Weston)</para></listitem>
5358 <listitem><para>Add /contrib/dbmirror (Steven Singer)</para></listitem>
5359 <listitem><para>Improve /contrib/pgbench (Neil)</para></listitem>
5360 <listitem><para>Add /contrib/tablefunc table function examples (Joe)</para></listitem>
5361 <listitem><para>Add /contrib/ltree data type for tree structures (Teodor Sigaev, Oleg Bartunov)</para></listitem>
5362 <listitem><para>Move /contrib/pg_controldata, pg_resetxlog into main tree (Bruce)</para></listitem>
5363 <listitem><para>Fixes to /contrib/cube (Bruno Wolff)</para></listitem>
5364 <listitem><para>Improve /contrib/fulltextindex (Christopher)</para></listitem>
5372 <sect1 id="release-7-2-5">
5373 <title>Release 7.2.5</title>
5376 <title>Release date</title>
5377 <simpara>2004-08-16</simpara>
5381 This release contains a variety of fixes from 7.2.4.
5386 <title>Migration to version 7.2.5</title>
5389 A dump/restore is not required for those running 7.2.X.
5394 <title>Changes</title>
5397 <listitem><para>Prevent possible loss of committed transactions during crash</para>
5399 Due to insufficient interlocking between transaction commit and checkpointing,
5400 it was possible for transactions committed just before the most recent
5401 checkpoint to be lost, in whole or in part, following a database crash and
5402 restart. This is a serious bug that has existed
5403 since <productname>PostgreSQL</productname> 7.1.
5405 <listitem><para>Fix corner case for btree search in parallel with first root page split</para></listitem>
5406 <listitem><para>Fix buffer overrun in <function>to_ascii</function> (Guido Notari)</para></listitem>
5407 <listitem><para>Fix core dump in deadlock detection on machines where char is unsigned</para></listitem>
5408 <listitem><para>Fix failure to respond to <command>pg_ctl stop -m fast</command> after Async_NotifyHandler runs</para></listitem>
5409 <listitem><para>Repair memory leaks in pg_dump</para></listitem>
5410 <listitem><para>Avoid conflict with system definition of <function>isblank()</function> function or macro</para></listitem>
5415 <sect1 id="release-7-2-4">
5416 <title>Release 7.2.4</title>
5419 <title>Release date</title>
5420 <simpara>2003-01-30</simpara>
5424 This release contains a variety of fixes for version 7.2.3,
5425 including fixes to prevent possible data loss.
5429 <title>Migration to version 7.2.4</title>
5432 A dump/restore is <emphasis>not</emphasis> required for those
5433 running version 7.2.*.
5438 <title>Changes</title>
5441 <listitem><para>Fix some additional cases of VACUUM "No one parent tuple was found" error</para></listitem>
5442 <listitem><para>Prevent VACUUM from being called inside a function (Bruce)</para></listitem>
5443 <listitem><para>Ensure pg_clog updates are sync'd to disk before marking checkpoint complete</para></listitem>
5444 <listitem><para>Avoid integer overflow during large hash joins</para></listitem>
5445 <listitem><para>Make GROUP commands work when pg_group.grolist is large enough to be toasted</para></listitem>
5446 <listitem><para>Fix errors in datetime tables; some timezone names weren't being recognized</para></listitem>
5447 <listitem><para>Fix integer overflows in circle_poly(), path_encode(), path_add() (Neil)</para></listitem>
5448 <listitem><para>Repair long-standing logic errors in lseg_eq(), lseg_ne(), lseg_center()</para></listitem>
5454 <sect1 id="release-7-2-3">
5455 <title>Release 7.2.3</title>
5458 <title>Release date</title>
5459 <simpara>2002-10-01</simpara>
5463 This release contains a variety of fixes for version 7.2.2,
5464 including fixes to prevent possible data loss.
5468 <title>Migration to version 7.2.3</title>
5471 A dump/restore is <emphasis>not</emphasis> required for those
5472 running version 7.2.*.
5477 <title>Changes</title>
5480 <listitem><para>Prevent possible compressed transaction log loss (Tom)</para></listitem>
5481 <listitem><para>Prevent non-superuser from increasing most recent vacuum info (Tom)</para></listitem>
5482 <listitem><para>Handle pre-1970 date values in newer versions of glibc (Tom)</para></listitem>
5483 <listitem><para>Fix possible hang during server shutdown</para></listitem>
5484 <listitem><para>Prevent spinlock hangs on SMP PPC machines (Tomoyuki Niijima)</para></listitem>
5485 <listitem><para>Fix <application>pg_dump</> to properly dump FULL JOIN USING (Tom)</para></listitem>
5491 <sect1 id="release-7-2-2">
5492 <title>Release 7.2.2</title>
5495 <title>Release date</title>
5496 <simpara>2002-08-23</simpara>
5500 This release contains a variety of fixes for version 7.2.1.
5504 <title>Migration to version 7.2.2</title>
5507 A dump/restore is <emphasis>not</emphasis> required for those
5508 running version 7.2.*.
5513 <title>Changes</title>
5516 <listitem><para>Allow EXECUTE of "CREATE TABLE AS ... SELECT" in PL/pgSQL (Tom)</para></listitem>
5517 <listitem><para>Fix for compressed transaction log id wraparound (Tom)</para></listitem>
5518 <listitem><para>Fix PQescapeBytea/PQunescapeBytea so that they handle bytes > 0x7f (Tatsuo)</para></listitem>
5519 <listitem><para>Fix for psql and <application>pg_dump</> crashing when invoked with non-existent long options (Tatsuo)</para></listitem>
5520 <listitem><para>Fix crash when invoking geometric operators (Tom)</para></listitem>
5521 <listitem><para>Allow OPEN cursor(args) (Tom)</para></listitem>
5522 <listitem><para>Fix for rtree_gist index build (Teodor)</para></listitem>
5523 <listitem><para>Fix for dumping user-defined aggregates (Tom)</para></listitem>
5524 <listitem><para>contrib/intarray fixes (Oleg)</para></listitem>
5525 <listitem><para>Fix for complex UNION/EXCEPT/INTERSECT queries using parens (Tom)</para></listitem>
5526 <listitem><para>Fix to pg_convert (Tatsuo)</para></listitem>
5527 <listitem><para>Fix for crash with long DATA strings (Thomas, Neil)</para></listitem>
5528 <listitem><para>Fix for repeat(), lpad(), rpad() and long strings (Neil)</para></listitem>
5534 <sect1 id="release-7-2-1">
5535 <title>Release 7.2.1</title>
5538 <title>Release date</title>
5539 <simpara>2002-03-21</simpara>
5543 This release contains a variety of fixes for version 7.2.
5547 <title>Migration to version 7.2.1</title>
5550 A dump/restore is <emphasis>not</emphasis> required for those
5551 running version 7.2.
5556 <title>Changes</title>
5559 <listitem><para>Ensure that sequence counters do not go backwards after a crash (Tom)</para></listitem>
5560 <listitem><para>Fix pgaccess kanji-conversion key binding (Tatsuo)</para></listitem>
5561 <listitem><para>Optimizer improvements (Tom)</para></listitem>
5562 <listitem><para>Cash I/O improvements (Tom)</para></listitem>
5563 <listitem><para>New Russian FAQ</para></listitem>
5564 <listitem><para>Compile fix for missing AuthBlockSig (Heiko)</para></listitem>
5565 <listitem><para>Additional time zones and time zone fixes (Thomas)</para></listitem>
5566 <listitem><para>Allow psql \connect to handle mixed case database and user names (Tom)</para></listitem>
5567 <listitem><para>Return proper OID on command completion even with ON INSERT rules (Tom)</para></listitem>
5568 <listitem><para>Allow COPY FROM to use 8-bit DELIMITERS (Tatsuo)</para></listitem>
5569 <listitem><para>Fix bug in extract/date_part for milliseconds/microseconds (Tatsuo)</para></listitem>
5570 <listitem><para>Improve handling of multiple UNIONs with different lengths (Tom)</para></listitem>
5571 <listitem><para>contrib/btree_gist improvements (Teodor Sigaev)</para></listitem>
5572 <listitem><para>contrib/tsearch dictionary improvements, see README.tsearch for an additional installation step (Thomas T. Thai, Teodor Sigaev)</para></listitem>
5573 <listitem><para>Fix for array subscripts handling (Tom)</para></listitem>
5574 <listitem><para>Allow EXECUTE of "CREATE TABLE AS ... SELECT" in PL/pgSQL (Tom)</para></listitem>
5580 <sect1 id="release-7-2">
5581 <title>Release 7.2</title>
5584 <title>Release date</title>
5585 <simpara>2002-02-04</simpara>
5589 <title>Overview</title>
5592 This release improves <productname>PostgreSQL</> for use in
5593 high-volume applications.
5597 Major changes in this release:
5604 Vacuuming no longer locks tables, thus allowing normal user
5605 access during the vacuum. A new <command>VACUUM FULL</>
5606 command does old-style vacuum by locking the table and
5607 shrinking the on-disk copy of the table.
5613 <term>Transactions</term>
5616 There is no longer a problem with installations that exceed
5617 four billion transactions.
5626 OIDs are now optional. Users can now create tables without
5627 OIDs for cases where OID usage is excessive.
5633 <term>Optimizer</term>
5636 The system now computes histogram column statistics during
5637 <command>ANALYZE</>, allowing much better optimizer choices.
5643 <term>Security</term>
5646 A new MD5 encryption option allows more secure storage and
5647 transfer of passwords. A new Unix-domain socket
5648 authentication option is available on Linux and BSD systems.
5654 <term>Statistics</term>
5657 Administrators can use the new table access statistics module
5658 to get fine-grained information about table and index usage.
5664 <term>Internationalization</term>
5667 Program and library messages can now be displayed in several
5677 <title>Migration to version 7.2</title>
5680 A dump/restore using <command>pg_dump</command> is required for
5681 those wishing to migrate data from any previous release.
5685 Observe the following incompatibilities:
5690 The semantics of the <command>VACUUM</command> command have
5691 changed in this release. You may wish to update your
5692 maintenance procedures accordingly.
5698 In this release, comparisons using <literal>= NULL</literal>
5699 will always return false (or NULL, more precisely). Previous
5700 releases automatically transformed this syntax to <literal>IS
5701 NULL</literal>. The old behavior can be re-enabled using a
5702 <filename>postgresql.conf</filename> parameter.
5708 The <filename>pg_hba.conf</> and <filename>pg_ident.conf</>
5709 configuration is now only reloaded after receiving a
5710 <systemitem>SIGHUP</> signal, not with each connection.
5716 The function <filename>octet_length()</> now returns the uncompressed data length.
5722 The date/time value <literal>'current'</literal> is no longer
5723 available. You will need to rewrite your applications.
5729 The <literal>timestamp()</literal>, <literal>time()</literal>,
5730 and <literal>interval()</literal> functions are no longer
5731 available. Instead of <literal>timestamp()</literal>, use
5732 <literal>timestamp 'string'</literal> or <literal>CAST</literal>.
5739 The <literal>SELECT ... LIMIT #,#</literal> syntax will be removed
5740 in the next release. You should change your queries to use
5741 separate LIMIT and OFFSET clauses, e.g. <literal>LIMIT 10 OFFSET
5747 <title>Changes</title>
5750 <title>Server Operation</title>
5752 <listitem><para>Create temporary files in a separate directory (Bruce)</para></listitem>
5753 <listitem><para>Delete orphaned temporary files on postmaster startup (Bruce)</para></listitem>
5754 <listitem><para>Added unique indexes to some system tables (Tom)</para></listitem>
5755 <listitem><para>System table operator reorganization (Oleg Bartunov, Teodor Sigaev, Tom)</para></listitem>
5756 <listitem><para>Renamed pg_log to pg_clog (Tom)</para></listitem>
5757 <listitem><para>Enable SIGTERM, SIGQUIT to kill backends (Jan)</para></listitem>
5758 <listitem><para>Removed compile-time limit on number of backends (Tom)</para></listitem>
5759 <listitem><para>Better cleanup for semaphore resource failure (Tatsuo, Tom)</para></listitem>
5760 <listitem><para>Allow safe transaction ID wraparound (Tom)</para></listitem>
5761 <listitem><para>Removed OIDs from some system tables (Tom)</para></listitem>
5762 <listitem><para>Removed "triggered data change violation" error check (Tom)</para></listitem>
5763 <listitem><para>SPI portal creation of prepared/saved plans (Jan)</para></listitem>
5764 <listitem><para>Allow SPI column functions to work for system columns (Tom)</para></listitem>
5765 <listitem><para>Long value compression improvement (Tom)</para></listitem>
5766 <listitem><para>Statistics collector for table, index access (Jan)</para></listitem>
5767 <listitem><para>Truncate extra-long sequence names to a reasonable value (Tom)</para></listitem>
5768 <listitem><para>Measure transaction times in milliseconds (Thomas)</para></listitem>
5769 <listitem><para>Fix TID sequential scans (Hiroshi)</para></listitem>
5770 <listitem><para>Superuser ID now fixed at 1 (Peter E)</para></listitem>
5771 <listitem><para>New pg_ctl "reload" option (Tom)</para></listitem>
5776 <title>Performance</title>
5778 <listitem><para>Optimizer improvements (Tom)</para></listitem>
5779 <listitem><para>New histogram column statistics for optimizer (Tom)</para></listitem>
5780 <listitem><para>Reuse write-ahead log files rather than discarding them (Tom)</para></listitem>
5781 <listitem><para>Cache improvements (Tom)</para></listitem>
5782 <listitem><para>IS NULL, IS NOT NULL optimizer improvement (Tom)</para></listitem>
5783 <listitem><para>Improve lock manager to reduce lock contention (Tom)</para></listitem>
5784 <listitem><para>Keep relcache entries for index access support functions (Tom)</para></listitem>
5785 <listitem><para>Allow better selectivity with NaN and infinities in NUMERIC (Tom)</para></listitem>
5786 <listitem><para>R-tree performance improvements (Kenneth Been)</para></listitem>
5787 <listitem><para>B-tree splits more efficient (Tom)</para></listitem>
5792 <title>Privileges</title>
5794 <listitem><para>Change UPDATE, DELETE privileges to be distinct (Peter E)</para></listitem>
5795 <listitem><para>New REFERENCES, TRIGGER privileges (Peter E)</para></listitem>
5796 <listitem><para>Allow GRANT/REVOKE to/from more than one user at a time (Peter E)</para></listitem>
5797 <listitem><para>New has_table_privilege() function (Joe Conway)</para></listitem>
5798 <listitem><para>Allow non-superuser to vacuum database (Tom)</para></listitem>
5799 <listitem><para>New SET SESSION AUTHORIZATION command (Peter E)</para></listitem>
5800 <listitem><para>Fix bug in privilege modifications on newly created tables (Tom)</para></listitem>
5801 <listitem><para>Disallow access to pg_statistic for non-superuser, add user-accessible views (Tom)</para></listitem>
5806 <title>Client Authentication</title>
5808 <listitem><para>Fork postmaster before doing authentication to prevent hangs (Peter E)</para></listitem>
5809 <listitem><para>Add ident authentication over Unix domain sockets on Linux, *BSD (Helge Bahmann, Oliver Elphick, Teodor Sigaev, Bruce)</para></listitem>
5810 <listitem><para>Add a password authentication method that uses MD5 encryption (Bruce)</para></listitem>
5811 <listitem><para>Allow encryption of stored passwords using MD5 (Bruce)</para></listitem>
5812 <listitem><para>PAM authentication (Dominic J. Eidson)</para></listitem>
5813 <listitem><para>Load pg_hba.conf and pg_ident.conf only on startup and SIGHUP (Bruce)</para></listitem>
5818 <title>Server Configuration</title>
5820 <listitem><para>Interpretation of some time zone abbreviations as Australian rather than North American now settable at run time (Bruce)</para></listitem>
5821 <listitem><para>New parameter to set default transaction isolation level (Peter E)</para></listitem>
5822 <listitem><para>New parameter to enable conversion of "expr = NULL" into "expr IS NULL", off by default (Peter E)</para></listitem>
5823 <listitem><para>New parameter to control memory usage by VACUUM (Tom)</para></listitem>
5824 <listitem><para>New parameter to set client authentication timeout (Tom)</para></listitem>
5825 <listitem><para>New parameter to set maximum number of open files (Tom)</para></listitem>
5830 <title>Queries</title>
5832 <listitem><para>Statements added by INSERT rules now execute after the INSERT (Jan)</para></listitem>
5833 <listitem><para>Prevent unadorned relation names in target list (Bruce)</para></listitem>
5834 <listitem><para>NULLs now sort after all normal values in ORDER BY (Tom)</para></listitem>
5835 <listitem><para>New IS UNKNOWN, IS NOT UNKNOWN Boolean tests (Tom)</para></listitem>
5836 <listitem><para>New SHARE UPDATE EXCLUSIVE lock mode (Tom)</para></listitem>
5837 <listitem><para>New EXPLAIN ANALYZE command that shows run times and row counts (Martijn van Oosterhout)</para></listitem>
5838 <listitem><para>Fix problem with LIMIT and subqueries (Tom)</para></listitem>
5839 <listitem><para>Fix for LIMIT, DISTINCT ON pushed into subqueries (Tom)</para></listitem>
5840 <listitem><para>Fix nested EXCEPT/INTERSECT (Tom)</para></listitem>
5845 <title>Schema Manipulation</title>
5847 <listitem><para>Fix SERIAL in temporary tables (Bruce)</para></listitem>
5848 <listitem><para>Allow temporary sequences (Bruce)</para></listitem>
5849 <listitem><para>Sequences now use int8 internally (Tom)</para></listitem>
5850 <listitem><para>New SERIAL8 creates int8 columns with sequences, default still SERIAL4 (Tom)</para></listitem>
5851 <listitem><para>Make OIDs optional using WITHOUT OIDS (Tom)</para></listitem>
5852 <listitem><para>Add %TYPE syntax to CREATE TYPE (Ian Lance Taylor)</para></listitem>
5853 <listitem><para>Add ALTER TABLE / DROP CONSTRAINT for CHECK constraints (Christopher Kings-Lynne)</para></listitem>
5854 <listitem><para>New CREATE OR REPLACE FUNCTION to alter existing function (preserving the function OID) (Gavin Sherry)</para></listitem>
5855 <listitem><para>Add ALTER TABLE / ADD [ UNIQUE | PRIMARY ] (Christopher Kings-Lynne)</para></listitem>
5856 <listitem><para>Allow column renaming in views</para></listitem>
5857 <listitem><para>Make ALTER TABLE / RENAME COLUMN update column names of indexes (Brent Verner)</para></listitem>
5858 <listitem><para>Fix for ALTER TABLE / ADD CONSTRAINT ... CHECK with inherited tables (Stephan Szabo)</para></listitem>
5859 <listitem><para>ALTER TABLE RENAME update foreign-key trigger arguments correctly (Brent Verner)</para></listitem>
5860 <listitem><para>DROP AGGREGATE and COMMENT ON AGGREGATE now accept an aggtype (Tom)</para></listitem>
5861 <listitem><para>Add automatic return type data casting for SQL functions (Tom)</para></listitem>
5862 <listitem><para>Allow GiST indexes to handle NULLs and multikey indexes (Oleg Bartunov, Teodor Sigaev, Tom)</para></listitem>
5863 <listitem><para>Enable partial indexes (Martijn van Oosterhout)</para></listitem>
5868 <title>Utility Commands</title>
5870 <listitem><para>Add RESET ALL, SHOW ALL (Marko Kreen)</para></listitem>
5871 <listitem><para>CREATE/ALTER USER/GROUP now allow options in any order (Vince)</para></listitem>
5872 <listitem><para>Add LOCK A, B, C functionality (Neil Padgett)</para></listitem>
5873 <listitem><para>New ENCRYPTED/UNENCRYPTED option to CREATE/ALTER USER (Bruce)</para></listitem>
5874 <listitem><para>New light-weight VACUUM does not lock table; old semantics are available as VACUUM FULL (Tom)</para></listitem>
5875 <listitem><para>Disable COPY TO/FROM on views (Bruce)</para></listitem>
5876 <listitem><para>COPY DELIMITERS string must be exactly one character (Tom)</para></listitem>
5877 <listitem><para>VACUUM warning about index tuples fewer than heap now only appears when appropriate (Martijn van Oosterhout)</para></listitem>
5878 <listitem><para>Fix privilege checks for CREATE INDEX (Tom)</para></listitem>
5879 <listitem><para>Disallow inappropriate use of CREATE/DROP INDEX/TRIGGER/VIEW (Tom)</para></listitem>
5884 <title>Data Types and Functions</title>
5886 <listitem><para>SUM(), AVG(), COUNT() now uses int8 internally for speed (Tom)</para></listitem>
5887 <listitem><para>Add convert(), convert2() (Tatsuo)</para></listitem>
5888 <listitem><para>New function bit_length() (Peter E)</para></listitem>
5889 <listitem><para>Make the "n" in CHAR(n)/VARCHAR(n) represents letters, not bytes (Tatsuo)</para></listitem>
5890 <listitem><para>CHAR(), VARCHAR() now reject strings that are too long (Peter E)</para></listitem>
5891 <listitem><para>BIT VARYING now rejects bit strings that are too long (Peter E)</para></listitem>
5892 <listitem><para>BIT now rejects bit strings that do not match declared size (Peter E)</para></listitem>
5893 <listitem><para>INET, CIDR text conversion functions (Alex Pilosov)</para></listitem>
5894 <listitem><para>INET, CIDR operators << and <<= indexable (Alex Pilosov)</para></listitem>
5895 <listitem><para>Bytea \### now requires valid three digit octal number</para></listitem>
5896 <listitem><para>Bytea comparison improvements, now supports =, <>, >, >=, <, and <=</para></listitem>
5897 <listitem><para>Bytea now supports B-tree indexes</para></listitem>
5898 <listitem><para>Bytea now supports LIKE, LIKE...ESCAPE, NOT LIKE, NOT LIKE...ESCAPE</para></listitem>
5899 <listitem><para>Bytea now supports concatenation</para></listitem>
5900 <listitem><para>New bytea functions: position, substring, trim, btrim, and length</para></listitem>
5901 <listitem><para>New encode() function mode, "escaped", converts minimally escaped bytea to/from text</para></listitem>
5902 <listitem><para>Add pg_database_encoding_max_length() (Tatsuo)</para></listitem>
5903 <listitem><para>Add pg_client_encoding() function (Tatsuo)</para></listitem>
5904 <listitem><para>now() returns time with millisecond precision (Thomas)</para></listitem>
5905 <listitem><para>New TIMESTAMP WITHOUT TIMEZONE data type (Thomas)</para></listitem>
5906 <listitem><para>Add ISO date/time specification with "T", yyyy-mm-ddThh:mm:ss (Thomas)</para></listitem>
5907 <listitem><para>New xid/int comparison functions (Hiroshi)</para></listitem>
5908 <listitem><para>Add precision to TIME, TIMESTAMP, and INTERVAL data types (Thomas)</para></listitem>
5909 <listitem><para>Modify type coercion logic to attempt binary-compatible functions first (Tom)</para></listitem>
5910 <listitem><para>New encode() function installed by default (Marko Kreen)</para></listitem>
5911 <listitem><para>Improved to_*() conversion functions (Karel Zak)</para></listitem>
5912 <listitem><para>Optimize LIKE/ILIKE when using single-byte encodings (Tatsuo)</para></listitem>
5913 <listitem><para>New functions in contrib/pgcrypto: crypt(), hmac(), encrypt(), gen_salt() (Marko Kreen)</para></listitem>
5914 <listitem><para>Correct description of translate() function (Bruce)</para></listitem>
5915 <listitem><para>Add INTERVAL argument for SET TIME ZONE (Thomas)</para></listitem>
5916 <listitem><para>Add INTERVAL YEAR TO MONTH (etc.) syntax (Thomas)</para></listitem>
5917 <listitem><para>Optimize length functions when using single-byte encodings (Tatsuo)</para></listitem>
5918 <listitem><para>Fix path_inter, path_distance, path_length, dist_ppath to handle closed paths (Curtis Barrett, Tom)</para></listitem>
5919 <listitem><para>octet_length(text) now returns non-compressed length (Tatsuo, Bruce)</para></listitem>
5920 <listitem><para>Handle "July" full name in date/time literals (Greg Sabino Mullane)</para></listitem>
5921 <listitem><para>Some datatype() function calls now evaluated differently</para></listitem>
5922 <listitem><para>Add support for Julian and ISO time specifications (Thomas)</para></listitem>
5927 <title>Internationalization</title>
5929 <listitem><para>National language support in psql, <application>pg_dump</>, libpq, and server (Peter E)</para></listitem>
5930 <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>
5931 <listitem><para>Make trim, ltrim, rtrim, btrim, lpad, rpad, translate multibyte aware (Tatsuo)</para></listitem>
5932 <listitem><para>Add LATIN5,6,7,8,9,10 support (Tatsuo)</para></listitem>
5933 <listitem><para>Add ISO 8859-5,6,7,8 support (Tatsuo)</para></listitem>
5934 <listitem><para>Correct LATIN5 to mean ISO-8859-9, not ISO-8859-5 (Tatsuo)</para></listitem>
5935 <listitem><para>Make mic2ascii() non-ASCII aware (Tatsuo)</para></listitem>
5936 <listitem><para>Reject invalid multibyte character sequences (Tatsuo)</para></listitem>
5941 <title><application>PL/pgSQL</></title>
5943 <listitem><para>Now uses portals for SELECT loops, allowing huge result sets (Jan)</para></listitem>
5944 <listitem><para>CURSOR and REFCURSOR support (Jan)</para></listitem>
5945 <listitem><para>Can now return open cursors (Jan)</para></listitem>
5946 <listitem><para>Add ELSEIF (Klaus Reger)</para></listitem>
5947 <listitem><para>Improve PL/pgSQL error reporting, including location of error (Tom)</para></listitem>
5948 <listitem><para>Allow IS or FOR key words in cursor declaration, for compatibility (Bruce)</para></listitem>
5949 <listitem><para>Fix for SELECT ... FOR UPDATE (Tom)</para></listitem>
5950 <listitem><para>Fix for PERFORM returning multiple rows (Tom)</para></listitem>
5951 <listitem><para>Make PL/pgSQL use the server's type coercion code (Tom)</para></listitem>
5952 <listitem><para>Memory leak fix (Jan, Tom)</para></listitem>
5953 <listitem><para>Make trailing semicolon optional (Tom)</para></listitem>
5958 <title>PL/Perl</title>
5960 <listitem><para>New untrusted PL/Perl (Alex Pilosov)</para></listitem>
5961 <listitem><para>PL/Perl is now built on some platforms even if libperl is not shared (Peter E)</para></listitem>
5966 <title>PL/Tcl</title>
5968 <listitem><para>Now reports errorInfo (Vsevolod Lobko)</para></listitem>
5969 <listitem><para>Add spi_lastoid function (bob@redivi.com)</para></listitem>
5974 <title>PL/Python</title>
5976 <listitem><para>...is new (Andrew Bosma)</para></listitem>
5981 <title><application>psql</></title>
5983 <listitem><para>\d displays indexes in unique, primary groupings (Christopher Kings-Lynne)</para></listitem>
5984 <listitem><para>Allow trailing semicolons in backslash commands (Greg Sabino Mullane)</para></listitem>
5985 <listitem><para>Read password from /dev/tty if possible</para></listitem>
5986 <listitem><para>Force new password prompt when changing user and database (Tatsuo, Tom)</para></listitem>
5987 <listitem><para>Format the correct number of columns for Unicode (Patrice)</para></listitem>
5992 <title><application>libpq</></title>
5994 <listitem><para>New function PQescapeString() to escape quotes in command strings (Florian Weimer)</para></listitem>
5995 <listitem><para>New function PQescapeBytea() escapes binary strings for use as SQL string literals</para></listitem>
6002 <listitem><para>Return OID of INSERT (Ken K)</para></listitem>
6003 <listitem><para>Handle more data types (Ken K)</para></listitem>
6004 <listitem><para>Handle single quotes and newlines in strings (Ken K)</para></listitem>
6005 <listitem><para>Handle NULL variables (Ken K)</para></listitem>
6006 <listitem><para>Fix for time zone handling (Barry Lind)</para></listitem>
6007 <listitem><para>Improved Druid support</para></listitem>
6008 <listitem><para>Allow eight-bit characters with non-multibyte server (Barry Lind)</para></listitem>
6009 <listitem><para>Support BIT, BINARY types (Ned Wolpert)</para></listitem>
6010 <listitem><para>Reduce memory usage (Michael Stephens, Dave Cramer)</para></listitem>
6011 <listitem><para>Update DatabaseMetaData (Peter E)</para></listitem>
6012 <listitem><para>Add DatabaseMetaData.getCatalogs() (Peter E)</para></listitem>
6013 <listitem><para>Encoding fixes (Anders Bengtsson)</para></listitem>
6014 <listitem><para>Get/setCatalog methods (Jason Davies)</para></listitem>
6015 <listitem><para>DatabaseMetaData.getColumns() now returns column defaults (Jason Davies)</para></listitem>
6016 <listitem><para>DatabaseMetaData.getColumns() performance improvement (Jeroen van Vianen)</para></listitem>
6017 <listitem><para>Some JDBC1 and JDBC2 merging (Anders Bengtsson)</para></listitem>
6018 <listitem><para>Transaction performance improvements (Barry Lind)</para></listitem>
6019 <listitem><para>Array fixes (Greg Zoller)</para></listitem>
6020 <listitem><para>Serialize addition </para></listitem>
6021 <listitem><para>Fix batch processing (Rene Pijlman)</para></listitem>
6022 <listitem><para>ExecSQL method reorganization (Anders Bengtsson)</para></listitem>
6023 <listitem><para>GetColumn() fixes (Jeroen van Vianen)</para></listitem>
6024 <listitem><para>Fix isWriteable() function (Rene Pijlman)</para></listitem>
6025 <listitem><para>Improved passage of JDBC2 conformance tests (Rene Pijlman)</para></listitem>
6026 <listitem><para>Add bytea type capability (Barry Lind)</para></listitem>
6027 <listitem><para>Add isNullable() (Rene Pijlman)</para></listitem>
6028 <listitem><para>JDBC date/time test suite fixes (Liam Stewart)</para></listitem>
6029 <listitem><para>Fix for SELECT 'id' AS xxx FROM table (Dave Cramer)</para></listitem>
6030 <listitem><para>Fix DatabaseMetaData to show precision properly (Mark Lillywhite)</para></listitem>
6031 <listitem><para>New getImported/getExported keys (Jason Davies)</para></listitem>
6032 <listitem><para>MD5 password encryption support (Jeremy Wohl)</para></listitem>
6033 <listitem><para>Fix to actually use type cache (Ned Wolpert)</para></listitem>
6040 <listitem><para>Remove query size limit (Hiroshi)</para></listitem>
6041 <listitem><para>Remove text field size limit (Hiroshi)</para></listitem>
6042 <listitem><para>Fix for SQLPrimaryKeys in multibyte mode (Hiroshi)</para></listitem>
6043 <listitem><para>Allow ODBC procedure calls (Hiroshi)</para></listitem>
6044 <listitem><para>Improve boolean handing (Aidan Mountford)</para></listitem>
6045 <listitem><para>Most configuration options now settable via DSN (Hiroshi)</para></listitem>
6046 <listitem><para>Multibyte, performance fixes (Hiroshi)</para></listitem>
6047 <listitem><para>Allow driver to be used with iODBC or unixODBC (Peter E)</para></listitem>
6048 <listitem><para>MD5 password encryption support (Bruce)</para></listitem>
6049 <listitem><para>Add more compatibility functions to odbc.sql (Peter E)</para></listitem>
6054 <title><application>ECPG</></title>
6056 <listitem><para>EXECUTE ... INTO implemented (Christof Petig)</para></listitem>
6057 <listitem><para>Multiple row descriptor support (e.g. CARDINALITY) (Christof Petig)</para></listitem>
6058 <listitem><para>Fix for GRANT parameters (Lee Kindness)</para></listitem>
6059 <listitem><para>Fix INITIALLY DEFERRED bug</para></listitem>
6060 <listitem><para>Various bug fixes (Michael, Christof Petig)</para></listitem>
6061 <listitem><para>Auto allocation for indicator variable arrays (int *ind_p=NULL)</para></listitem>
6062 <listitem><para>Auto allocation for string arrays (char **foo_pp=NULL)</para></listitem>
6063 <listitem><para>ECPGfree_auto_mem fixed</para></listitem>
6064 <listitem><para>All function names with external linkage are now prefixed by ECPG</para></listitem>
6065 <listitem><para>Fixes for arrays of structures (Michael)</para></listitem>
6070 <title>Misc. Interfaces</title>
6072 <listitem><para>Python fix fetchone() (Gerhard Haring)</para></listitem>
6073 <listitem><para>Use UTF, Unicode in Tcl where appropriate (Vsevolod Lobko, Reinhard Max)</para></listitem>
6074 <listitem><para>Add Tcl COPY TO/FROM (ljb)</para></listitem>
6075 <listitem><para>Prevent output of default index op class in <application>pg_dump</> (Tom)</para></listitem>
6076 <listitem><para>Fix libpgeasy memory leak (Bruce)</para></listitem>
6081 <title>Build and Install</title>
6083 <listitem><para>Configure, dynamic loader, and shared library fixes (Peter E)</para></listitem>
6084 <listitem><para>Fixes in QNX 4 port (Bernd Tegge)</para></listitem>
6085 <listitem><para>Fixes in Cygwin and Windows ports (Jason Tishler, Gerhard Haring, Dmitry Yurtaev, Darko Prenosil, Mikhail Terekhov)</para></listitem>
6086 <listitem><para>Fix for Windows socket communication failures (Magnus, Mikhail Terekhov)</para></listitem>
6087 <listitem><para>Hurd compile fix (Oliver Elphick)</para></listitem>
6088 <listitem><para>BeOS fixes (Cyril Velter)</para></listitem>
6089 <listitem><para>Remove configure --enable-unicode-conversion, now enabled by multibyte (Tatsuo)</para></listitem>
6090 <listitem><para>AIX fixes (Tatsuo, Andreas)</para></listitem>
6091 <listitem><para>Fix parallel make (Peter E)</para></listitem>
6092 <listitem><para>Install SQL language manual pages into OS-specific directories (Peter E)</para></listitem>
6093 <listitem><para>Rename config.h to pg_config.h (Peter E)</para></listitem>
6094 <listitem><para>Reorganize installation layout of header files (Peter E)</para></listitem>
6099 <title>Source Code</title>
6101 <listitem><para>Remove SEP_CHAR (Bruce)</para></listitem>
6102 <listitem><para>New GUC hooks (Tom)</para></listitem>
6103 <listitem><para>Merge GUC and command line handling (Marko Kreen)</para></listitem>
6104 <listitem><para>Remove EXTEND INDEX (Martijn van Oosterhout, Tom)</para></listitem>
6105 <listitem><para>New pgjindent utility to indent java code (Bruce)</para></listitem>
6106 <listitem><para>Remove define of true/false when compiling under C++ (Leandro Fanzone, Tom)</para></listitem>
6107 <listitem><para>pgindent fixes (Bruce, Tom)</para></listitem>
6108 <listitem><para>Replace strcasecmp() with strcmp() where appropriate (Peter E)</para></listitem>
6109 <listitem><para>Dynahash portability improvements (Tom)</para></listitem>
6110 <listitem><para>Add 'volatile' usage in spinlock structures</para></listitem>
6111 <listitem><para>Improve signal handling logic (Tom)</para></listitem>
6116 <title>Contrib</title>
6118 <listitem><para>New contrib/rtree_gist (Oleg Bartunov, Teodor Sigaev)</para></listitem>
6119 <listitem><para>New contrib/tsearch full-text indexing (Oleg, Teodor Sigaev)</para></listitem>
6120 <listitem><para>Add contrib/dblink for remote database access (Joe Conway)</para></listitem>
6121 <listitem><para>contrib/ora2pg Oracle conversion utility (Gilles Darold)</para></listitem>
6122 <listitem><para>contrib/xml XML conversion utility (John Gray)</para></listitem>
6123 <listitem><para>contrib/fulltextindex fixes (Christopher Kings-Lynne)</para></listitem>
6124 <listitem><para>New contrib/fuzzystrmatch with levenshtein and metaphone, soundex merged (Joe Conway)</para></listitem>
6125 <listitem><para>Add contrib/intarray boolean queries, binary search, fixes (Oleg Bartunov)</para></listitem>
6126 <listitem><para>New pg_upgrade utility (Bruce)</para></listitem>
6127 <listitem><para>Add new pg_resetxlog options (Bruce, Tom)</para></listitem>
6134 <sect1 id="release-7-1-3">
6135 <title>Release 7.1.3</title>
6138 <title>Release date</title>
6139 <simpara>2001-08-15</simpara>
6143 <title>Migration to version 7.1.3</title>
6146 A dump/restore is <emphasis>not</emphasis> required for those running
6152 <title>Changes</title>
6156 Remove unused WAL segements of large transactions (Tom)
6157 Multiaction rule fix (Tom)
6158 PL/pgSQL memory allocation fix (Jan)
6159 VACUUM buffer fix (Tom)
6160 Regression test fixes (Tom)
6161 pg_dump fixes for GRANT/REVOKE/comments on views, user-defined types (Tom)
6162 Fix subselects with DISTINCT ON or LIMIT (Tom)
6164 Disable COPY TO/FROM a view (Tom)
6165 Cygwin build (Jason Tishler)
6172 <sect1 id="release-7-1-2">
6173 <title>Release 7.1.2</title>
6176 <title>Release date</title>
6177 <simpara>2001-05-11</simpara>
6181 This has one fix from 7.1.1.
6186 <title>Migration to version 7.1.2</title>
6189 A dump/restore is <emphasis>not</emphasis> required for those running
6195 <title>Changes</title>
6199 Fix PL/pgSQL SELECTs when returning no rows
6200 Fix for psql backslash core dump
6201 Referential integrity privilege fix
6210 <sect1 id="release-7-1-1">
6211 <title>Release 7.1.1</title>
6214 <title>Release date</title>
6215 <simpara>2001-05-05</simpara>
6219 This has a variety of fixes from 7.1.
6224 <title>Migration to version 7.1.1</title>
6227 A dump/restore is <emphasis>not</emphasis> required for those running
6233 <title>Changes</title>
6237 Fix for numeric MODULO operator (Tom)
6238 pg_dump fixes (Philip)
6239 pg_dump can dump 7.0 databases (Philip)
6240 readline 4.2 fixes (Peter E)
6242 AIX, MSWIN, VAX, N32K fixes (Tom)
6243 Multibytes fixes (Tom)
6244 Unicode fixes (Tatsuo)
6245 Optimizer improvements (Tom)
6246 Fix for whole rows in functions (Tom)
6247 Fix for pg_ctl and option strings with spaces (Peter E)
6248 ODBC fixes (Hiroshi)
6249 EXTRACT can now take string argument (Thomas)
6250 Python fixes (Darcy)
6257 <sect1 id="release-7-1">
6258 <title>Release 7.1</title>
6261 <title>Release date</title>
6262 <simpara>2001-04-13</simpara>
6266 This release focuses on removing limitations that have existed in the
6267 <productname>PostgreSQL</productname> code for many years.
6271 Major changes in this release:
6276 Write-ahead Log (WAL)
6280 To maintain database consistency in case of an operating system crash,
6281 previous releases of <productname>PostgreSQL</productname> have forced
6282 all data modifications to disk before each transaction commit. With
6283 WAL, only one log file must be flushed to disk, greatly improving
6284 performance. If you have been using -F in previous releases to
6285 disable disk flushes, you may want to consider discontinuing its use.
6296 TOAST - Previous releases had a compiled-in row length limit,
6297 typically 8k - 32k. This limit made storage of long text fields
6298 difficult. With TOAST, long rows of any length can be stored with good
6310 We now support outer joins. The UNION/NOT IN
6311 workaround for outer joins is no longer required. We use the SQL92
6323 The previous C function manager did not
6324 handle null values properly, nor did it support 64-bit <acronym>CPU</acronym>'s (Alpha). The new
6325 function manager does. You can continue using your old custom
6326 functions, but you may want to rewrite them in the future to use the new
6327 function manager call interface.
6338 A large number of complex queries that were
6339 unsupported in previous releases now work. Many combinations of views,
6340 aggregates, UNION, LIMIT, cursors, subqueries, and inherited tables
6341 now work properly. Inherited tables are now accessed by default.
6342 Subqueries in FROM are now supported.
6351 <title>Migration to version 7.1</title>
6354 A dump/restore using pg_dump is required for those wishing to migrate
6355 data from any previous release.
6360 <title>Changes</title>
6366 Many multibyte/Unicode/locale fixes (Tatsuo and others)
6367 More reliable ALTER TABLE RENAME (Tom)
6368 Kerberos V fixes (David Wragg)
6369 Fix for INSERT INTO...SELECT where targetlist has subqueries (Tom)
6370 Prompt username/password on standard error (Bruce)
6371 Large objects inv_read/inv_write fixes (Tom)
6372 Fixes for to_char(), to_date(), to_ascii(), and to_timestamp() (Karel,
6374 Prevent query expressions from leaking memory (Tom)
6375 Allow UPDATE of arrays elements (Tom)
6376 Wake up lock waiters during cancel (Hiroshi)
6377 Fix rare cursor crash when using hash join (Tom)
6378 Fix for DROP TABLE/INDEX in rolled-back transaction (Hiroshi)
6379 Fix psql crash from \l+ if MULTIBYTE enabled (Peter E)
6380 Fix truncation of rule names during CREATE VIEW (Ross Reedstrom)
6381 Fix PL/perl (Alex Kapranoff)
6382 Disallow LOCK on views (Mark Hollomon)
6383 Disallow INSERT/UPDATE/DELETE on views (Mark Hollomon)
6384 Disallow DROP RULE, CREATE INDEX, TRUNCATE on views (Mark Hollomon)
6385 Allow PL/pgSQL accept non-ASCII identifiers (Tatsuo)
6386 Allow views to proper handle GROUP BY, aggregates, DISTINCT (Tom)
6387 Fix rare failure with TRUNCATE command (Tom)
6388 Allow UNION/INTERSECT/EXCEPT to be used with ALL, subqueries, views,
6389 DISTINCT, ORDER BY, SELECT...INTO (Tom)
6390 Fix parser failures during aborted transactions (Tom)
6391 Allow temporary relations to properly clean up indexes (Bruce)
6392 Fix VACUUM problem with moving rows in same page (Tom)
6393 Modify pg_dump to better handle user-defined items in template1 (Philip)
6394 Allow LIMIT in VIEW (Tom)
6395 Require cursor FETCH to honor LIMIT (Tom)
6396 Allow PRIMARY/FOREIGN Key definitions on inherited columns (Stephan)
6397 Allow ORDER BY, LIMIT in subqueries (Tom)
6398 Allow UNION in CREATE RULE (Tom)
6399 Make ALTER/DROP TABLE rollback-able (Vadim, Tom)
6400 Store initdb collation in pg_control so collation cannot be changed (Tom)
6401 Fix INSERT...SELECT with rules (Tom)
6402 Fix FOR UPDATE inside views and subselects (Tom)
6403 Fix OVERLAPS operators conform to SQL92 spec regarding NULLs (Tom)
6404 Fix lpad() and rpad() to handle length less than input string (Tom)
6405 Fix use of NOTIFY in some rules (Tom)
6406 Overhaul btree code (Tom)
6407 Fix NOT NULL use in Pl/pgSQL variables (Tom)
6408 Overhaul GIST code (Oleg)
6409 Fix CLUSTER to preserve constraints and column default (Tom)
6410 Improved deadlock detection handling (Tom)
6411 Allow multiple SERIAL columns in a table (Tom)
6412 Prevent occasional index corruption (Vadim)
6416 Add OUTER JOINs (Tom)
6417 Function manager overhaul (Tom)
6418 Allow ALTER TABLE RENAME on indexes (Tom)
6419 Improve CLUSTER (Tom)
6420 Improve ps status display for more platforms (Peter E, Marc)
6421 Improve CREATE FUNCTION failure message (Ross)
6422 JDBC improvements (Peter, Travis Bauer, Christopher Cain, William Webber,
6424 Grand Unified Configuration scheme/GUC. Many options can now be set in
6425 data/postgresql.conf, postmaster/postgres flags, or SET commands (Peter E)
6426 Improved handling of file descriptor cache (Tom)
6427 New warning code about auto-created table alias entries (Bruce)
6428 Overhaul initdb process (Tom, Peter E)
6429 Overhaul of inherited tables; inherited tables now accessed by default;
6430 new ONLY key word prevents it (Chris Bitmead, Tom)
6431 ODBC cleanups/improvements (Nick Gorham, Stephan Szabo, Zoltan Kovacs,
6433 Allow renaming of temp tables (Tom)
6434 Overhaul memory manager contexts (Tom)
6435 pg_dumpall uses CREATE USER or CREATE GROUP rather using COPY (Peter E)
6436 Overhaul pg_dump (Philip Warner)
6437 Allow pg_hba.conf secondary password file to specify only username (Peter E)
6438 Allow TEMPORARY or TEMP key word when creating temporary tables (Bruce)
6439 New memory leak checker (Karel)
6440 New SET SESSION CHARACTERISTICS (Thomas)
6441 Allow nested block comments (Thomas)
6442 Add WITHOUT TIME ZONE type qualifier (Thomas)
6443 New ALTER TABLE ADD CONSTRAINT (Stephan)
6444 Use NUMERIC accumulators for INTEGER aggregates (Tom)
6445 Overhaul aggregate code (Tom)
6446 New VARIANCE and STDDEV() aggregates
6447 Improve dependency ordering of pg_dump (Philip)
6448 New pg_restore command (Philip)
6449 New pg_dump tar output option (Philip)
6450 New pg_dump of large objects (Philip)
6451 New ESCAPE option to LIKE (Thomas)
6452 New case-insensitive LIKE - ILIKE (Thomas)
6453 Allow functional indexes to use binary-compatible type (Tom)
6454 Allow SQL functions to be used in more contexts (Tom)
6455 New pg_config utility (Peter E)
6456 New PL/pgSQL EXECUTE command which allows dynamic SQL and utility statements
6458 New PL/pgSQL GET DIAGNOSTICS statement for SPI value access (Jan)
6459 New quote_identifiers() and quote_literal() functions (Jan)
6460 New ALTER TABLE table OWNER TO user command (Mark Hollomon)
6461 Allow subselects in FROM, i.e. FROM (SELECT ...) [AS] alias (Tom)
6462 Update PyGreSQL to version 3.1 (D'Arcy)
6463 Store tables as files named by OID (Vadim)
6464 New SQL function setval(seq,val,bool) for use in pg_dump (Philip)
6465 Require DROP VIEW to remove views, no DROP TABLE (Mark)
6466 Allow DROP VIEW view1, view2 (Mark)
6467 Allow multiple objects in DROP INDEX, DROP RULE, and DROP TYPE (Tom)
6468 Allow automatic conversion to/from Unicode (Tatsuo, Eiji)
6469 New /contrib/pgcrypto hashing functions (Marko Kreen)
6470 New pg_dumpall --globals-only option (Peter E)
6471 New CHECKPOINT command for WAL which creates new WAL log file (Vadim)
6472 New AT TIME ZONE syntax (Thomas)
6473 Allow location of Unix domain socket to be configurable (David J. MacKenzie)
6474 Allow postmaster to listen on a specific IP address (David J. MacKenzie)
6475 Allow socket path name to be specified in hostname by using leading slash
6476 (David J. MacKenzie)
6477 Allow CREATE DATABASE to specify template database (Tom)
6478 New utility to convert MySQL schema dumps to SQL92 and PostgreSQL (Thomas)
6479 New /contrib/rserv replication toolkit (Vadim)
6480 New file format for COPY BINARY (Tom)
6481 New /contrib/oid2name to map numeric files to table names (B Palmer)
6482 New "idle in transaction" ps status message (Marc)
6483 Update to pgaccess 0.98.7 (Constantin Teodorescu)
6484 pg_ctl now defaults to -w (wait) on shutdown, new -l (log) option
6485 Add rudimentary dependency checking to pg_dump (Philip)
6489 Fix INET/CIDR type ordering and add new functions (Tom)
6490 Make OID behave as an unsigned type (Tom)
6491 Allow BIGINT as synonym for INT8 (Peter E)
6492 New int2 and int8 comparison operators (Tom)
6493 New BIT and BIT VARYING types (Adriaan Joubert, Tom, Peter E)
6494 CHAR() no longer faster than VARCHAR() because of TOAST (Tom)
6495 New GIST seg/cube examples (Gene Selkov)
6496 Improved round(numeric) handling (Tom)
6497 Fix CIDR output formatting (Tom)
6498 New CIDR abbrev() function (Tom)
6502 Write-Ahead Log (WAL) to provide crash recovery with less performance
6504 ANALYZE stage of VACUUM no longer exclusively locks table (Bruce)
6505 Reduced file seeks (Denis Perchine)
6506 Improve BTREE code for duplicate keys (Tom)
6507 Store all large objects in a single table (Denis Perchine, Tom)
6508 Improve memory allocation performance (Karel, Tom)
6512 New function manager call conventions (Tom)
6513 SGI portability fixes (David Kaelbling)
6514 New configure --enable-syslog option (Peter E)
6515 New BSDI README (Bruce)
6516 configure script moved to top level, not /src (Peter E)
6517 Makefile/configuration/compilation overhaul (Peter E)
6518 New configure --with-python option (Peter E)
6519 Solaris cleanups (Peter E)
6520 Overhaul /contrib Makefiles (Karel)
6521 New OpenSSL configuration option (Magnus, Peter E)
6523 QNX fixes (Maurizio)
6524 New heap_open(), heap_openr() API (Tom)
6525 Remove colon and semi-colon operators (Thomas)
6526 New pg_class.relkind value for views (Mark Hollomon)
6527 Rename ichar() to chr() (Karel)
6528 New documentation for btrim(), ascii(), chr(), repeat() (Karel)
6529 Fixes for NT/Cygwin (Pete Forman)
6530 AIX port fixes (Andreas)
6531 New BeOS port (David Reid, Cyril Velter)
6532 Add proofreader's changes to docs (Addison-Wesley, Bruce)
6533 New Alpha spinlock code (Adriaan Joubert, Compaq)
6534 UnixWare port overhaul (Peter E)
6535 New Darwin/MacOS X port (Peter Bierman, Bruce Hartzler)
6536 New FreeBSD Alpha port (Alfred)
6537 Overhaul shared memory segments (Tom)
6538 Add IBM S/390 support (Neale Ferguson)
6539 Moved macmanuf to /contrib (Larry Rosenman)
6540 Syslog improvements (Larry Rosenman)
6541 New template0 database that contains no user additions (Tom)
6542 New /contrib/cube and /contrib/seg GIST sample code (Gene Selkov)
6543 Allow NetBSD's libedit instead of readline (Peter)
6544 Improved assembly language source code format (Bruce)
6545 New contrib/pg_logger
6546 New --template option to createdb
6547 New contrib/pg_control utility (Oliver)
6548 New FreeBSD tools ipc_check, start-scripts/freebsd
6555 <sect1 id="release-7-0-3">
6556 <title>Release 7.0.3</title>
6559 <title>Release date</title>
6560 <simpara>2000-11-11</simpara>
6564 This has a variety of fixes from 7.0.2.
6569 <title>Migration to version 7.0.3</title>
6572 A dump/restore is <emphasis>not</emphasis> required for those running
6578 <title>Changes</title>
6583 Large object fix (Tom)
6584 Fix lean in COPY WITH OIDS leak (Tom)
6585 Fix backwards-index-scan (Tom)
6586 Fix SELECT ... FOR UPDATE so it checks for duplicate keys (Hiroshi)
6587 Add --enable-syslog to configure (Marc)
6588 Fix abort transaction at backend exit in rare cases (Tom)
6589 Fix for psql \l+ when multibyte enabled (Tatsuo)
6590 Allow PL/pgSQL to accept non ascii identifiers (Tatsuo)
6591 Make vacuum always flush buffers (Tom)
6592 Fix to allow cancel while waiting for a lock (Hiroshi)
6593 Fix for memory aloocation problem in user authentication code (Tom)
6594 Remove bogus use of int4out() (Tom)
6595 Fixes for multiple subqueries in COALESCE or BETWEEN (Tom)
6596 Fix for failure of triggers on heap open in certain cases (Jeroen van
6598 Fix for erroneous selectivity of not-equals (Tom)
6599 Fix for erroneous use of strcmp() (Tom)
6600 Fix for bug where storage manager accesses items beyond end of file
6602 Fix to include kernel errno message in all smgr elog messages (Tom)
6603 Fix for '.' not in PATH at build time (SL Baur)
6604 Fix for out-of-file-descriptors error (Tom)
6605 Fix to make pg_dump dump 'iscachable' flag for functions (Tom)
6606 Fix for subselect in targetlist of Append node (Tom)
6607 Fix for mergejoin plans (Tom)
6608 Fix TRUNCATE failure on relations with indexes (Tom)
6609 Avoid database-wide restart on write error (Hiroshi)
6610 Fix nodeMaterial to honor chgParam by recomputing its output (Tom)
6611 Fix VACUUM problem with moving chain of update row versions when source
6612 and destination of a row version lie on the same page (Tom)
6613 Fix user.c CommandCounterIncrement (Tom)
6614 Fix for AM/PM boundary problem in to_char() (Karel Zak)
6615 Fix TIME aggregate handling (Tom)
6616 Fix to_char() to avoid coredump on NULL input (Tom)
6618 Fix for inserting/copying longer multibyte strings into char() data
6620 Fix for crash of backend, on abort (Tom)
6627 <sect1 id="release-7-0-2">
6628 <title>Release 7.0.2</title>
6631 <title>Release date</title>
6632 <simpara>2000-06-05</simpara>
6636 This is a repackaging of 7.0.1 with added documentation.
6641 <title>Migration to version 7.0.2</title>
6644 A dump/restore is <emphasis>not</emphasis> required for those running
6650 <title>Changes</title>
6654 Added documentation to tarball.
6661 <sect1 id="release-7-0-1">
6662 <title>Release 7.0.1</title>
6665 <title>Release date</title>
6666 <simpara>2000-06-01</simpara>
6670 This is a cleanup release for 7.0.
6674 <title>Migration to version 7.0.1</title>
6677 A dump/restore is <emphasis>not</emphasis> required for those running
6683 <title>Changes</title>
6687 Fix many CLUSTER failures (Tom)
6688 Allow ALTER TABLE RENAME works on indexes (Tom)
6689 Fix plpgsql to handle datetime->timestamp and timespan->interval (Bruce)
6690 New configure --with-setproctitle switch to use setproctitle() (Marc, Bruce)
6691 Fix the off by one errors in ResultSet from 6.5.3, and more.
6692 jdbc ResultSet fixes (Joseph Shraibman)
6693 optimizer tunings (Tom)
6694 Fix create user for pgaccess
6695 Fix for UNLISTEN failure
6696 IRIX fixes (David Kaelbling)
6697 QNX fixes (Andreas Kardos)
6698 Reduce COPY IN lock level (Tom)
6699 Change libpqeasy to use PQconnectdb() style parameters (Bruce)
6700 Fix pg_dump to handle OID indexes (Tom)
6701 Fix small memory leak (Tom)
6702 Solaris fix for createdb/dropdb (Tatsuo)
6703 Fix for non-blocking connections (Alfred Perlstein)
6704 Fix improper recovery after RENAME TABLE failures (Tom)
6705 Copy pg_ident.conf.sample into /lib directory in install (Bruce)
6706 Add SJIS UDC (NEC selection IBM kanji) support (Eiji Tokuya)
6707 Fix too long syslog message (Tatsuo)
6708 Fix problem with quoted indexes that are too long (Tom)
6709 JDBC ResultSet.getTimestamp() fix (Gregory Krasnow & Floyd Marinescu)
6710 ecpg changes (Michael)
6716 <sect1 id="release-7-0">
6717 <title>Release 7.0</title>
6720 <title>Release date</title>
6721 <simpara>2000-05-08</simpara>
6725 This release contains improvements in many areas, demonstrating
6726 the continued growth of <productname>PostgreSQL</productname>.
6727 There are more improvements and fixes in 7.0 than in any previous
6728 release. The developers have confidence that this is the best
6729 release yet; we do our best to put out only solid releases, and
6730 this one is no exception.
6734 Major changes in this release:
6743 Foreign keys are now implemented, with the exception of PARTIAL MATCH
6744 foreign keys. Many users have been asking for this feature, and we are
6745 pleased to offer it.
6756 Continuing on work started a year ago, the optimizer has been
6757 improved, allowing better query plan selection and faster performance
6758 with less memory usage.
6765 Updated <application>psql</application>
6769 <application>psql</application>, our interactive terminal monitor, has been
6770 updated with a variety of new features. See the <application>psql</application> manual page for details.
6781 SQL92 join syntax is now supported, though only as
6782 <literal>INNER JOIN</> for this release. <literal>JOIN</>,
6783 <literal>NATURAL JOIN</>, <literal>JOIN</>/<literal>USING</>,
6784 and <literal>JOIN</>/<literal>ON</> are available, as are
6785 column correlation names.
6793 <title>Migration to version 7.0</title>
6796 A dump/restore using <application>pg_dump</application>
6797 is required for those wishing to migrate data from any
6798 previous release of <productname>PostgreSQL</productname>.
6799 For those upgrading from 6.5.*, you may instead use
6800 <application>pg_upgrade</application> to upgrade to this
6801 release; however, a full dump/reload installation is always the
6802 most robust method for upgrades.
6806 Interface and compatibility issues to consider for the new
6812 The date/time types <type>datetime</type> and
6813 <type>timespan</type> have been superseded by the
6814 SQL92-defined types <type>timestamp</type> and
6815 <type>interval</type>. Although there has been some effort to
6816 ease the transition by allowing
6817 <productname>PostgreSQL</productname> to recognize
6818 the deprecated type names and translate them to the new type
6819 names, this mechanism may not be completely transparent to
6820 your existing application.
6826 The optimizer has been substantially improved in the area of
6827 query cost estimation. In some cases, this will result in
6828 decreased query times as the optimizer makes a better choice
6829 for the preferred plan. However, in a small number of cases,
6830 usually involving pathological distributions of data, your
6831 query times may go up. If you are dealing with large amounts
6832 of data, you may want to check your queries to verify
6839 The <acronym>JDBC</acronym> and <acronym>ODBC</acronym>
6840 interfaces have been upgraded and extended.
6846 The string function <function>CHAR_LENGTH</function> is now a
6847 native function. Previous versions translated this into a call
6848 to <function>LENGTH</function>, which could result in
6849 ambiguity with other types implementing
6850 <function>LENGTH</function> such as the geometric types.
6859 <title>Changes</title>
6865 Prevent function calls exceeding maximum number of arguments (Tom)
6866 Improve CASE construct (Tom)
6867 Fix SELECT coalesce(f1,0) FROM int4_tbl GROUP BY f1 (Tom)
6868 Fix SELECT sentence.words[0] FROM sentence GROUP BY sentence.words[0] (Tom)
6869 Fix GROUP BY scan bug (Tom)
6870 Improvements in SQL grammar processing (Tom)
6871 Fix for views involved in INSERT ... SELECT ... (Tom)
6872 Fix for SELECT a/2, a/2 FROM test_missing_target GROUP BY a/2 (Tom)
6873 Fix for subselects in INSERT ... SELECT (Tom)
6874 Prevent INSERT ... SELECT ... ORDER BY (Tom)
6875 Fixes for relations greater than 2GB, including vacuum
6876 Improve propagating system table changes to other backends (Tom)
6877 Improve propagating user table changes to other backends (Tom)
6878 Fix handling of temp tables in complex situations (Bruce, Tom)
6879 Allow table locking at table open, improving concurrent reliability (Tom)
6880 Properly quote sequence names in pg_dump (Ross J. Reedstrom)
6881 Prevent DROP DATABASE while others accessing
6882 Prevent any rows from being returned by GROUP BY if no rows processed (Tom)
6883 Fix SELECT COUNT(1) FROM table WHERE ...' if no rows matching WHERE (Tom)
6884 Fix pg_upgrade so it works for MVCC (Tom)
6885 Fix for SELECT ... WHERE x IN (SELECT ... HAVING SUM(x) > 1) (Tom)
6886 Fix for "f1 datetime DEFAULT 'now'" (Tom)
6887 Fix problems with CURRENT_DATE used in DEFAULT (Tom)
6888 Allow comment-only lines, and ;;; lines too. (Tom)
6889 Improve recovery after failed disk writes, disk full (Hiroshi)
6890 Fix cases where table is mentioned in FROM but not joined (Tom)
6891 Allow HAVING clause without aggregate functions (Tom)
6892 Fix for "--" comment and no trailing newline, as seen in perl interface
6893 Improve pg_dump failure error reports (Bruce)
6894 Allow sorts and hashes to exceed 2GB file sizes (Tom)
6895 Fix for pg_dump dumping of inherited rules (Tom)
6896 Fix for NULL handling comparisons (Tom)
6897 Fix inconsistent state caused by failed CREATE/DROP commands (Hiroshi)
6898 Fix for dbname with dash
6899 Prevent DROP INDEX from interfering with other backends (Tom)
6900 Fix file descriptor leak in verify_password()
6901 Fix for "Unable to identify an operator =$" problem
6902 Fix ODBC so no segfault if CommLog and Debug enabled (Dirk Niggemann)
6903 Fix for recursive exit call (Massimo)
6904 Fix for extra-long timezones (Jeroen van Vianen)
6905 Make pg_dump preserve primary key information (Peter E)
6906 Prevent databases with single quotes (Peter E)
6907 Prevent DROP DATABASE inside transaction (Peter E)
6908 ecpg memory leak fixes (Stephen Birch)
6909 Fix for SELECT null::text, SELECT int4fac(null) and SELECT 2 + (null) (Tom)
6910 Y2K timestamp fix (Massimo)
6911 Fix for VACUUM 'HEAP_MOVED_IN was not expected' errors (Tom)
6912 Fix for views with tables/columns containing spaces (Tom)
6913 Prevent privileges on indexes (Peter E)
6914 Fix for spinlock stuck problem when error is generated (Hiroshi)
6915 Fix ipcclean on Linux
6916 Fix handling of NULL constraint conditions (Tom)
6917 Fix memory leak in odbc driver (Nick Gorham)
6918 Fix for privilege check on UNION tables (Tom)
6919 Fix to allow SELECT 'a' LIKE 'a' (Tom)
6920 Fix for SELECT 1 + NULL (Tom)
6922 Fix log() on numeric type (Tom)
6923 Deprecate ':' and ';' operators
6924 Allow vacuum of temporary tables
6925 Disallow inherited columns with the same name as new columns
6926 Recover or force failure when disk space is exhausted (Hiroshi)
6927 Fix INSERT INTO ... SELECT with AS columns matching result columns
6928 Fix INSERT ... SELECT ... GROUP BY groups by target columns not source columns (Tom)
6929 Fix CREATE TABLE test (a char(5) DEFAULT text '', b int4) with INSERT (Tom)
6930 Fix UNION with LIMIT
6931 Fix CREATE TABLE x AS SELECT 1 UNION SELECT 2
6932 Fix CREATE TABLE test(col char(2) DEFAULT user)
6933 Fix mismatched types in CREATE TABLE ... DEFAULT
6934 Fix SELECT * FROM pg_class where oid in (0,-1)
6935 Fix SELECT COUNT('asdf') FROM pg_class WHERE oid=12
6936 Prevent user who can create databases can modifying pg_database table (Peter E)
6937 Fix btree to give a useful elog when key > 1/2 (page - overhead) (Tom)
6938 Fix INSERT of 0.0 into DECIMAL(4,4) field (Tom)
6942 New CLI interface include file sqlcli.h, based on SQL3/SQL98
6943 Remove all limits on query length, row length limit still exists (Tom)
6944 Update jdbc protocol to 2.0 (Jens Glaser <email>jens@jens.de</email>)
6945 Add TRUNCATE command to quickly truncate relation (Mike Mascari)
6946 Fix to give super user and createdb user proper update catalog rights (Peter E)
6947 Allow ecpg bool variables to have NULL values (Christof)
6948 Issue ecpg error if NULL value for variable with no NULL indicator (Christof)
6949 Allow ^C to cancel COPY command (Massimo)
6950 Add SET FSYNC and SHOW PG_OPTIONS commands(Massimo)
6951 Function name overloading for dynamically-loaded C functions (Frankpitt)
6952 Add CmdTuples() to libpq++(Vince)
6953 New CREATE CONSTRAINT TRIGGER and SET CONSTRAINTS commands(Jan)
6954 Allow CREATE FUNCTION/WITH clause to be used for all language types
6955 configure --enable-debug adds -g (Peter E)
6956 configure --disable-debug removes -g (Peter E)
6957 Allow more complex default expressions (Tom)
6958 First real FOREIGN KEY constraint trigger functionality (Jan)
6959 Add FOREIGN KEY ... MATCH FULL ... ON DELETE CASCADE (Jan)
6960 Add FOREIGN KEY ... MATCH <unspecified> referential actions (Don Baccus)
6961 Allow WHERE restriction on ctid (physical heap location) (Hiroshi)
6962 Move pginterface from contrib to interface directory, rename to pgeasy (Bruce)
6963 Change pgeasy connectdb() parameter ordering (Bruce)
6964 Require SELECT DISTINCT target list to have all ORDER BY columns (Tom)
6965 Add Oracle's COMMENT ON command (Mike Mascari <email>mascarim@yahoo.com</email>)
6966 libpq's PQsetNoticeProcessor function now returns previous hook(Peter E)
6967 Prevent PQsetNoticeProcessor from being set to NULL (Peter E)
6968 Make USING in COPY optional (Bruce)
6969 Allow subselects in the target list (Tom)
6970 Allow subselects on the left side of comparison operators (Tom)
6971 New parallel regression test (Jan)
6972 Change backend-side COPY to write files with permissions 644 not 666 (Tom)
6973 Force permissions on PGDATA directory to be secure, even if it exists (Tom)
6974 Added psql LASTOID variable to return last inserted oid (Peter E)
6975 Allow concurrent vacuum and remove pg_vlock vacuum lock file (Tom)
6976 Add privilege check for vacuum (Peter E)
6977 New libpq functions to allow asynchronous connections: PQconnectStart(),
6978 PQconnectPoll(), PQresetStart(), PQresetPoll(), PQsetenvStart(),
6979 PQsetenvPoll(), PQsetenvAbort (Ewan Mellor)
6980 New libpq PQsetenv() function (Ewan Mellor)
6981 create/alter user extension (Peter E)
6982 New postmaster.pid and postmaster.opts under $PGDATA (Tatsuo)
6983 New scripts for create/drop user/db (Peter E)
6984 Major psql overhaul (Peter E)
6985 Add const to libpq interface (Peter E)
6986 New libpq function PQoidValue (Peter E)
6987 Show specific non-aggregate causing problem with GROUP BY (Tom)
6988 Make changes to pg_shadow recreate pg_pwd file (Peter E)
6989 Add aggregate(DISTINCT ...) (Tom)
6990 Allow flag to control COPY input/output of NULLs (Peter E)
6991 Make postgres user have a password by default (Peter E)
6992 Add CREATE/ALTER/DROP GROUP (Peter E)
6993 All administration scripts now support --long options (Peter E, Karel)
6994 Vacuumdb script now supports --all option (Peter E)
6995 ecpg new portable FETCH syntax
6996 Add ecpg EXEC SQL IFDEF, EXEC SQL IFNDEF, EXEC SQL ELSE, EXEC SQL ELIF
6997 and EXEC SQL ENDIF directives
6998 Add pg_ctl script to control backend start-up (Tatsuo)
6999 Add postmaster.opts.default file to store start-up flags (Tatsuo)
7000 Allow --with-mb=SQL_ASCII
7001 Increase maximum number of index keys to 16 (Bruce)
7002 Increase maximum number of function arguments to 16 (Bruce)
7003 Allow configuration of maximum number of index keys and arguments (Bruce)
7004 Allow unprivileged users to change their passwords (Peter E)
7005 Password authentication enabled; required for new users (Peter E)
7006 Disallow dropping a user who owns a database (Peter E)
7007 Change initdb option --with-mb to --enable-multibyte
7008 Add option for initdb to prompts for superuser password (Peter E)
7009 Allow complex type casts like col::numeric(9,2) and col::int2::float8 (Tom)
7010 Updated user interfaces on initdb, initlocation, pg_dump, ipcclean (Peter E)
7011 New pg_char_to_encoding() and pg_encoding_to_char() functions (Tatsuo)
7012 libpq non-blocking mode (Alfred Perlstein)
7013 Improve conversion of types in casts that don't specify a length
7014 New plperl internal programming language (Mark Hollomon)
7015 Allow COPY IN to read file that do not end with a newline (Tom)
7016 Indicate when long identifiers are truncated (Tom)
7017 Allow aggregates to use type equivalency (Peter E)
7018 Add Oracle's to_char(), to_date(), to_datetime(), to_timestamp(), to_number()
7019 conversion functions (Karel Zak <zakkr@zf.jcu.cz>)
7020 Add SELECT DISTINCT ON (expr [, expr ...]) targetlist ... (Tom)
7021 Check to be sure ORDER BY is compatible with the DISTINCT operation (Tom)
7022 Add NUMERIC and int8 types to ODBC
7023 Improve EXPLAIN results for Append, Group, Agg, Unique (Tom)
7024 Add ALTER TABLE ... ADD FOREIGN KEY (Stephan Szabo)
7025 Allow SELECT .. FOR UPDATE in PL/pgSQL (Hiroshi)
7026 Enable backward sequential scan even after reaching EOF (Hiroshi)
7027 Add btree indexing of boolean values, >= and <= (Don Baccus)
7028 Print current line number when COPY FROM fails (Massimo)
7029 Recognize POSIX time zone e.g. "PST+8" and "GMT-8" (Thomas)
7030 Add DEC as synonym for DECIMAL (Thomas)
7031 Add SESSION_USER as SQL92 key word, same as CURRENT_USER (Thomas)
7032 Implement SQL92 column aliases (aka correlation names) (Thomas)
7033 Implement SQL92 join syntax (Thomas)
7034 Make INTERVAL reserved word allowed as a column identifier (Thomas)
7035 Implement REINDEX command (Hiroshi)
7036 Accept ALL in aggregate function SUM(ALL col) (Tom)
7037 Prevent GROUP BY from using column aliases (Tom)
7038 New psql \encoding option (Tatsuo)
7039 Allow PQrequestCancel() to terminate when in waiting-for-lock state (Hiroshi)
7040 Allow negation of a negative number in all cases
7041 Add ecpg descriptors (Christof, Michael)
7042 Allow CREATE VIEW v AS SELECT f1::char(8) FROM tbl
7043 Allow casts with length, like foo::char(8)
7044 New libpq functions PQsetClientEncoding(), PQclientEncoding() (Tatsuo)
7045 Add support for SJIS user defined characters (Tatsuo)
7046 Larger views/rules supported
7047 Make libpq's PQconndefaults() thread-safe (Tom)
7048 Disable // as comment to be ANSI conforming, should use -- (Tom)
7049 Allow column aliases on views CREATE VIEW name (collist)
7050 Fixes for views with subqueries (Tom)
7051 Allow UPDATE table SET fld = (SELECT ...) (Tom)
7052 SET command options no longer require quotes
7053 Update pgaccess to 0.98.6
7054 New SET SEED command
7055 New pg_options.sample file
7056 New SET FSYNC command (Massimo)
7057 Allow pg_descriptions when creating tables
7058 Allow pg_descriptions when creating types, columns, and functions
7059 Allow psql \copy to allow delimiters (Peter E)
7060 Allow psql to print nulls as distinct from "" [null] (Peter E)
7064 Many array fixes (Tom)
7065 Allow bare column names to be subscripted as arrays (Tom)
7066 Improve type casting of int and float constants (Tom)
7067 Cleanups for int8 inputs, range checking, and type conversion (Tom)
7068 Fix for SELECT timespan('21:11:26'::time) (Tom)
7069 netmask('x.x.x.x/0') is 255.255.255.255 instead of 0.0.0.0 (Oleg Sharoiko)
7070 Add btree index on NUMERIC (Jan)
7071 Perl fix for large objects containing NUL characters (Douglas Thomson)
7072 ODBC fix for for large objects (free)
7073 Fix indexing of cidr data type
7074 Fix for Ethernet MAC addresses (macaddr type) comparisons
7075 Fix for date/time types when overflows happened in computations (Tom)
7076 Allow array on int8 (Peter E)
7077 Fix for rounding/overflow of NUMERIC type, like NUMERIC(4,4) (Tom)
7078 Allow NUMERIC arrays
7079 Fix bugs in NUMERIC ceil() and floor() functions (Tom)
7080 Make char_length()/octet_length including trailing blanks (Tom)
7081 Made abstime/reltime use int4 instead of time_t (Peter E)
7082 New lztext data type for compressed text fields
7083 Revise code to handle coercion of int and float constants (Tom)
7084 Start at new code to implement a BIT and BIT VARYING type (Adriaan Joubert)
7085 NUMERIC now accepts scientific notation (Tom)
7086 NUMERIC to int4 rounds (Tom)
7087 Convert float4/8 to NUMERIC properly (Tom)
7088 Allow type conversion with NUMERIC (Thomas)
7089 Make ISO date style (2000-02-16 09:33) the default (Thomas)
7090 Add NATIONAL CHAR [ VARYING ] (Thomas)
7091 Allow NUMERIC round and trunc to accept negative scales (Tom)
7092 New TIME WITH TIME ZONE type (Thomas)
7093 Add MAX()/MIN() on time type (Thomas)
7094 Add abs(), mod(), fac() for int8 (Thomas)
7095 Rename functions to round(), sqrt(), cbrt(), pow() for float8 (Thomas)
7096 Add transcendental math functions (e.g. sin(), acos()) for float8 (Thomas)
7097 Add exp() and ln() for NUMERIC type
7098 Rename NUMERIC power() to pow() (Thomas)
7099 Improved TRANSLATE() function (Edwin Ramirez, Tom)
7100 Allow X=-Y operators (Tom)
7101 Allow SELECT float8(COUNT(*))/(SELECT COUNT(*) FROM t) FROM t GROUP BY f1; (Tom)
7102 Allow LOCALE to use indexes in regular expression searches (Tom)
7103 Allow creation of functional indexes to use default types
7107 Prevent exponential space consumption with many AND's and OR's (Tom)
7108 Collect attribute selectivity values for system columns (Tom)
7109 Reduce memory usage of aggregates (Tom)
7110 Fix for LIKE optimization to use indexes with multibyte encodings (Tom)
7111 Fix r-tree index optimizer selectivity (Thomas)
7112 Improve optimizer selectivity computations and functions (Tom)
7113 Optimize btree searching for cases where many equal keys exist (Tom)
7114 Enable fast LIKE index processing only if index present (Tom)
7115 Re-use free space on index pages with duplicates (Tom)
7116 Improve hash join processing (Tom)
7117 Prevent descending sort if result is already sorted(Hiroshi)
7118 Allow commuting of index scan query qualifications (Tom)
7119 Prefer index scans in cases where ORDER BY/GROUP BY is required (Tom)
7120 Allocate large memory requests in fix-sized chunks for performance (Tom)
7121 Fix vacuum's performance by reducing memory allocation requests (Tom)
7122 Implement constant-expression simplification (Bernard Frankpitt, Tom)
7123 Use secondary columns to be used to determine start of index scan (Hiroshi)
7124 Prevent quadruple use of disk space when doing internal sorting (Tom)
7125 Faster sorting by calling fewer functions (Tom)
7126 Create system indexes to match all system caches (Bruce, Hiroshi)
7127 Make system caches use system indexes (Bruce)
7128 Make all system indexes unique (Bruce)
7129 Improve pg_statistics management for VACUUM speed improvement (Tom)
7130 Flush backend cache less frequently (Tom, Hiroshi)
7131 COPY now reuses previous memory allocation, improving performance (Tom)
7132 Improve optimization cost estimation (Tom)
7133 Improve optimizer estimate of range queries x > lowbound AND x < highbound (Tom)
7134 Use DNF instead of CNF where appropriate (Tom, Taral)
7135 Further cleanup for OR-of-AND WHERE-clauses (Tom)
7136 Make use of index in OR clauses (x = 1 AND y = 2) OR (x = 2 AND y = 4) (Tom)
7137 Smarter optimizer computations for random index page access (Tom)
7138 New SET variable to control optimizer costs (Tom)
7139 Optimizer queries based on LIMIT, OFFSET, and EXISTS qualifications (Tom)
7140 Reduce optimizer internal housekeeping of join paths for speedup (Tom)
7141 Major subquery speedup (Tom)
7142 Fewer fsync writes when fsync is not disabled (Tom)
7143 Improved LIKE optimizer estimates (Tom)
7144 Prevent fsync in SELECT-only queries (Vadim)
7145 Make index creation use psort code, because it is now faster (Tom)
7146 Allow creation of sort temp tables > 1 Gig
7150 Fix for linux PPC compile
7151 New generic expression-tree-walker subroutine (Tom)
7152 Change form() to varargform() to prevent portability problems
7153 Improved range checking for large integers on Alphas
7154 Clean up #include in /include directory (Bruce)
7155 Add scripts for checking includes (Bruce)
7156 Remove un-needed #include's from *.c files (Bruce)
7157 Change #include's to use <> and "" as appropriate (Bruce)
7158 Enable Windows compilation of libpq
7159 Alpha spinlock fix from Uncle George <email>gatgul@voicenet.com</email>
7160 Overhaul of optimizer data structures (Tom)
7161 Fix to cygipc library (Yutaka Tanida)
7162 Allow pgsql to work on newer Cygwin snapshots (Dan)
7163 New catalog version number (Tom)
7165 Rename heap_replace to heap_update
7166 Update for QNX (Dr. Andreas Kardos)
7167 New platform-specific regression handling (Tom)
7168 Rename oid8 -> oidvector and int28 -> int2vector (Bruce)
7169 Included all yacc and lex files into the distribution (Peter E.)
7170 Remove lextest, no longer needed (Peter E)
7171 Fix for libpq and psql on Windows (Magnus)
7172 Internally change datetime and timespan into timestamp and interval (Thomas)
7173 Fix for plpgsql on BSD/OS
7174 Add SQL_ASCII test case to the regression test (Tatsuo)
7175 configure --with-mb now deprecated (Tatsuo)
7177 NetBSD fixes (Johnny C. Lam <email>lamj@stat.cmu.edu</email>)
7178 Fixes for Alpha compiles
7179 New multibyte encodings
7185 <sect1 id="release-6-5-3">
7186 <title>Release 6.5.3</title>
7189 <title>Release date</title>
7190 <simpara>1999-10-13</simpara>
7194 This is basically a cleanup release for 6.5.2. We have added a new
7195 <application>PgAccess</> that was missing in 6.5.2, and installed an NT-specific fix.
7200 <title>Migration to version 6.5.3</title>
7203 A dump/restore is <emphasis>not</emphasis> required for those running
7208 <title>Changes</title>
7212 Updated version of pgaccess 0.98
7214 Fix dumping rules on inherited tables
7221 <sect1 id="release-6-5-2">
7222 <title>Release 6.5.2</title>
7225 <title>Release date</title>
7226 <simpara>1999-09-15</simpara>
7230 This is basically a cleanup release for 6.5.1. We have fixed a variety of
7231 problems reported by 6.5.1 users.
7236 <title>Migration to version 6.5.2</title>
7239 A dump/restore is <emphasis>not</emphasis> required for those running
7245 <title>Changes</title>
7249 subselect+CASE fixes(Tom)
7250 Add SHLIB_LINK setting for solaris_i386 and solaris_sparc ports(Daren Sefcik)
7251 Fixes for CASE in WHERE join clauses(Tom)
7252 Fix BTScan abort(Tom)
7253 Repair the check for redundant UNIQUE and PRIMARY KEY indexes(Thomas)
7254 Improve it so that it checks for multicolumn constraints(Thomas)
7255 Fix for Windows making problem with MB enabled(Hiroki Kataoka)
7256 Allow BSD yacc and bison to compile pl code(Bruce)
7257 Fix SET NAMES working
7259 Fix vacuum's memory consumption(Hiroshi,Tatsuo)
7260 Reduce the total memory consumption of vacuum(Tom)
7261 Fix for timestamp(datetime)
7262 Rule deparsing bugfixes(Tom)
7263 Fix quoting problems in mkMakefile.tcldefs.sh.in and mkMakefile.tkdefs.sh.in(Tom)
7264 This is to re-use space on index pages freed by vacuum(Vadim)
7265 document -x for pg_dump(Bruce)
7266 Fix for unary operators in rule deparser(Tom)
7267 Comment out FileUnlink of excess segments during mdtruncate()(Tom)
7268 IRIX linking fix from Yu Cao >yucao@falcon.kla-tencor.com<
7269 Repair logic error in LIKE: should not return LIKE_ABORT
7270 when reach end of pattern before end of text(Tom)
7271 Repair incorrect cleanup of heap memory allocation during transaction abort(Tom)
7272 Updated version of pgaccess 0.98
7278 <sect1 id="release-6-5-1">
7279 <title>Release 6.5.1</title>
7282 <title>Release date</title>
7283 <simpara>1999-07-15</simpara>
7287 This is basically a cleanup release for 6.5. We have fixed a variety of
7288 problems reported by 6.5 users.
7292 <title>Migration to version 6.5.1</title>
7295 A dump/restore is <emphasis>not</emphasis> required for those running
7301 <title>Changes</title>
7306 Portability fixes for linux_ppc, IRIX, linux_alpha, OpenBSD, alpha
7307 Remove QUERY_LIMIT, use SELECT...LIMIT
7308 Fix for EXPLAIN on inheritance(Tom)
7309 Patch to allow vacuum on multisegment tables(Hiroshi)
7310 R-Tree optimizer selectivity fix(Tom)
7311 ACL file descriptor leak fix(Atsushi Ogawa)
7312 New expresssion subtree code(Tom)
7313 Avoid disk writes for read-only transactions(Vadim)
7314 Fix for removal of temp tables if last transaction was aborted(Bruce)
7315 Fix to prevent too large row from being created(Bruce)
7317 Allow port numbers 32k - 64k(Bruce)
7318 Add ^ precidence(Bruce)
7319 Rename sort files called pg_temp to pg_sorttemp(Bruce)
7320 Fix for microseconds in time values(Tom)
7321 Tutorial source cleanup
7323 Fix for sorting of NULL's in some cases(Tom)
7324 Shared library dependencies fixed (Tom)
7325 Fixed glitches affecting GROUP BY in subselects(Tom)
7326 Fix some compiler warnings (Tomoaki Nishiyama)
7327 Add Win1250 (Czech) support (Pavel Behal)
7333 <sect1 id="release-6-5">
7334 <title>Release 6.5</title>
7337 <title>Release date</title>
7338 <simpara>1999-06-09</simpara>
7342 This release marks a major step in the development team's mastery of the source
7343 code we inherited from Berkeley. You will see we are now easily adding
7344 major features, thanks to the increasing size and experience of our
7345 world-wide development team.
7349 Here is a brief summary of the more notable changes:
7354 Multiversion concurrency control(MVCC)
7358 This removes our old table-level locking, and replaces it with
7359 a locking system that is superior to most commercial database
7360 systems. In a traditional system, each row that is modified
7361 is locked until committed, preventing reads by other users.
7362 MVCC uses the natural multiversion nature of
7363 <productname>PostgreSQL</productname> to allow readers to
7364 continue reading consistent data during writer activity.
7365 Writers continue to use the compact pg_log transaction system.
7366 This is all performed without having to allocate a lock for
7367 every row like traditional database systems. So, basically,
7368 we no longer are restricted by simple table-level locking; we
7369 have something better than row-level locking.
7376 Hot backups from <application>pg_dump</application>
7380 <application>pg_dump</application> takes advantage of the new
7381 MVCC features to give a consistent database dump/backup while
7382 the database stays online and available for queries.
7393 We now have a true numeric data type, with
7394 user-specified precision.
7405 Temporary tables are guaranteed to have unique names
7406 within a database session, and are destroyed on session exit.
7417 We now have CASE, INTERSECT, and EXCEPT statement
7418 support. We have new LIMIT/OFFSET, SET TRANSACTION ISOLATION LEVEL,
7419 SELECT ... FOR UPDATE, and an improved LOCK TABLE command.
7430 We continue to speed up <productname>PostgreSQL</productname>,
7431 thanks to the variety of talents within our team. We have
7432 sped up memory allocation, optimization, table joins, and row
7444 We continue to expand our port list, this time including
7445 <systemitem class="osname">Windows NT</>/<systemitem>ix86</> and <systemitem class="osname">NetBSD</>/<systemitem>arm32</>.
7456 Most interfaces have new versions, and existing functionality
7468 New and updated material is present throughout the
7469 documentation. New <acronym>FAQ</acronym>s have been
7470 contributed for <systemitem class="osname">SGI</> and <systemitem class="osname">AIX</> platforms.
7471 The <citetitle>Tutorial</citetitle> has introductory information
7472 on <acronym>SQL</acronym> from Stefan Simkovics.
7473 For the <citetitle>User's Guide</citetitle>, there are
7474 reference pages covering the postmaster and more utility
7475 programs, and a new appendix
7476 contains details on date/time behavior.
7477 The <citetitle>Administrator's Guide</citetitle> has a new
7478 chapter on troubleshooting from Tom Lane.
7479 And the <citetitle>Programmer's Guide</citetitle> has a
7480 description of query processing, also from Stefan, and details
7481 on obtaining the <productname>PostgreSQL</productname> source
7482 tree via anonymous <productname>CVS</productname> and
7483 <productname>CVSup</productname>.
7491 <title>Migration to version 6.5</title>
7494 A dump/restore using <application>pg_dump</application>
7495 is required for those wishing to migrate data from any
7496 previous release of <productname>PostgreSQL</productname>.
7497 <application>pg_upgrade</application> can <emphasis>not</emphasis>
7498 be used to upgrade to this release because the on-disk structure
7499 of the tables has changed compared to previous releases.
7503 The new Multiversion Concurrency Control (MVCC) features can
7504 give somewhat different behaviors in multiuser
7505 environments. <emphasis>Read and understand the following section
7506 to ensure that your existing applications will give you the
7507 behavior you need.</emphasis>
7511 <title>Multiversion Concurrency Control</title>
7514 Because readers in 6.5 don't lock data, regardless of transaction
7515 isolation level, data read by one transaction can be overwritten by
7516 another. In other words, if a row is returned by
7517 <command>SELECT</command> it doesn't mean that this row really exists
7518 at the time it is returned (i.e. sometime after the statement or
7519 transaction began) nor that the row is protected from being deleted or
7520 updated by concurrent transactions before the current transaction does
7521 a commit or rollback.
7525 To ensure the actual existence of a row and protect it against
7526 concurrent updates one must use <command>SELECT FOR UPDATE</command> or
7527 an appropriate <command>LOCK TABLE</command> statement. This should be
7528 taken into account when porting applications from previous releases of
7529 <productname>PostgreSQL</productname> and other environments.
7533 Keep the above in mind if you are using
7534 <filename>contrib/refint.*</filename> triggers for
7535 referential integrity. Additional techniques are required now. One way is
7536 to use <command>LOCK parent_table IN SHARE ROW EXCLUSIVE MODE</command>
7537 command if a transaction is going to update/delete a primary key and
7538 use <command>LOCK parent_table IN SHARE MODE</command> command if a
7539 transaction is going to update/insert a foreign key.
7543 Note that if you run a transaction in SERIALIZABLE mode then you must
7544 execute the <command>LOCK</command> commands above before execution of any
7545 <acronym>DML</acronym> statement
7546 (<command>SELECT/INSERT/DELETE/UPDATE/FETCH/COPY_TO</command>) in the
7553 These inconveniences will disappear in the future
7554 when the ability to read dirty
7555 (uncommitted) data (regardless of isolation level) and true referential
7556 integrity will be implemented.
7562 <title>Changes</title>
7568 Fix text<->float8 and text<->float4 conversion functions(Thomas)
7569 Fix for creating tables with mixed-case constraints(Billy)
7570 Change exp()/pow() behavior to generate error on underflow/overflow(Jan)
7571 Fix bug in pg_dump -z
7572 Memory overrun cleanups(Tatsuo)
7573 Fix for lo_import crash(Tatsuo)
7574 Adjust handling of data type names to suppress double quotes(Thomas)
7575 Use type coercion for matching columns and DEFAULT(Thomas)
7576 Fix deadlock so it only checks once after one second of sleep(Bruce)
7577 Fixes for aggregates and PL/pgsql(Hiroshi)
7578 Fix for subquery crash(Vadim)
7579 Fix for libpq function PQfnumber and case-insensitive names(Bahman Rafatjoo)
7580 Fix for large object write-in-middle, no extra block, memory consumption(Tatsuo)
7581 Fix for pg_dump -d or -D and quote special characters in INSERT
7582 Repair serious problems with dynahash(Tom)
7583 Fix INET/CIDR portability problems
7584 Fix problem with selectivity error in ALTER TABLE ADD COLUMN(Bruce)
7585 Fix executor so mergejoin of different column types works(Tom)
7586 Fix for Alpha OR selectivity bug
7587 Fix OR index selectivity problem(Bruce)
7588 Fix so \d shows proper length for char()/varchar()(Ryan)
7589 Fix tutorial code(Clark)
7590 Improve destroyuser checking(Oliver)
7591 Fix for Kerberos(Rodney McDuff)
7592 Fix for dropping database while dirty buffers(Bruce)
7593 Fix so sequence nextval() can be case-sensitive(Bruce)
7595 Drop buffers before destroying database files(Bruce)
7596 Fix case where executor evaluates functions twice(Tatsuo)
7597 Allow sequence nextval actions to be case-sensitive(Bruce)
7598 Fix optimizer indexing not working for negative numbers(Bruce)
7599 Fix for memory leak in executor with fjIsNull
7600 Fix for aggregate memory leaks(Erik Riedel)
7601 Allow user name containing a dash to grant privileges
7602 Cleanup of NULL in inet types
7603 Clean up system table bugs(Tom)
7604 Fix problems of PAGER and \? command(Masaaki Sakaida)
7605 Reduce default multisegment file size limit to 1GB(Peter)
7606 Fix for dumping of CREATE OPERATOR(Tom)
7607 Fix for backward scanning of cursors(Hiroshi Inoue)
7608 Fix for COPY FROM STDIN when using \i(Tom)
7609 Fix for subselect is compared inside an expression(Jan)
7610 Fix handling of error reporting while returning rows(Tom)
7611 Fix problems with reference to array types(Tom,Jan)
7612 Prevent UPDATE SET oid(Jan)
7613 Fix pg_dump so -t option can handle case-sensitive tablenames
7614 Fixes for GROUP BY in special cases(Tom, Jan)
7615 Fix for memory leak in failed queries(Tom)
7616 DEFAULT now supports mixed-case identifiers(Tom)
7617 Fix for multisegment uses of DROP/RENAME table, indexes(Ole Gjerde)
7618 Disable use of pg_dump with both -o and -d options(Bruce)
7619 Allow pg_dump to properly dump group privileges(Bruce)
7620 Fix GROUP BY in INSERT INTO table SELECT * FROM table2(Jan)
7621 Fix for computations in views(Jan)
7622 Fix for aggregates on array indexes(Tom)
7623 Fix for DEFAULT handles single quotes in value requiring too many quotes
7624 Fix security problem with non-super users importing/exporting large objects(Tom)
7625 Rollback of transaction that creates table cleaned up properly(Tom)
7626 Fix to allow long table and column names to generate proper serial names(Tom)
7630 Add "vacuumdb" utility
7631 Speed up libpq by allocating memory better(Tom)
7632 EXPLAIN all indexes used(Tom)
7633 Implement CASE, COALESCE, NULLIF expression(Thomas)
7634 New pg_dump table output format(Constantin)
7635 Add string min()/max() functions(Thomas)
7636 Extend new type coercion techniques to aggregates(Thomas)
7637 New moddatetime contrib(Terry)
7638 Update to pgaccess 0.96(Constantin)
7639 Add routines for single-byte "char" type(Thomas)
7640 Improved substr() function(Thomas)
7641 Improved multibyte handling(Tatsuo)
7642 Multiversion concurrency control/MVCC(Vadim)
7643 New Serialized mode(Vadim)
7644 Fix for tables over 2gigs(Peter)
7645 New SET TRANSACTION ISOLATION LEVEL(Vadim)
7646 New LOCK TABLE IN ... MODE(Vadim)
7647 Update ODBC driver(Byron)
7648 New NUMERIC data type(Jan)
7649 New SELECT FOR UPDATE(Vadim)
7650 Handle "NaN" and "Infinity" for input values(Jan)
7651 Improved date/year handling(Thomas)
7652 Improved handling of backend connections(Magnus)
7653 New options ELOG_TIMESTAMPS and USE_SYSLOG options for log files(Massimo)
7654 New TCL_ARRAYS option(Massimo)
7655 New INTERSECT and EXCEPT(Stefan)
7656 New pg_index.indisprimary for primary key tracking(D'Arcy)
7657 New pg_dump option to allow dropping of tables before creation(Brook)
7658 Speedup of row output routines(Tom)
7659 New READ COMMITTED isolation level(Vadim)
7660 New TEMP tables/indexes(Bruce)
7661 Prevent sorting if result is already sorted(Jan)
7662 New memory allocation optimization(Jan)
7663 Allow psql to do \p\g(Bruce)
7664 Allow multiple rule actions(Jan)
7665 Added LIMIT/OFFSET functionality(Jan)
7666 Improve optimizer when joining a large number of tables(Bruce)
7667 New intro to SQL from S. Simkovics' Master's Thesis (Stefan, Thomas)
7668 New intro to backend processing from S. Simkovics' Master's Thesis (Stefan)
7669 Improved int8 support(Ryan Bradetich, Thomas, Tom)
7670 New routines to convert between int8 and text/varchar types(Thomas)
7671 New bushy plans, where meta-tables are joined(Bruce)
7672 Enable right-hand queries by default(Bruce)
7673 Allow reliable maximum number of backends to be set at configure time
7674 (--with-maxbackends and postmaster switch (-N backends))(Tom)
7675 GEQO default now 10 tables because of optimizer speedups(Tom)
7676 Allow NULL=Var for MS-SQL portability(Michael, Bruce)
7677 Modify contrib check_primary_key() so either "automatic" or "dependent"(Anand)
7678 Allow psql \d on a view show query(Ryan)
7679 Speedup for LIKE(Bruce)
7680 Ecpg fixes/features, see src/interfaces/ecpg/ChangeLog file(Michael)
7681 JDBC fixes/features, see src/interfaces/jdbc/CHANGELOG(Peter)
7682 Make % operator have precedence like /(Bruce)
7683 Add new postgres -O option to allow system table structure changes(Bruce)
7684 Update contrib/pginterface/findoidjoins script(Tom)
7685 Major speedup in vacuum of deleted rows with indexes(Vadim)
7686 Allow non-SQL functions to run different versions based on arguments(Tom)
7687 Add -E option that shows actual queries sent by \dt and friends(Masaaki Sakaida)
7688 Add version number in start-up banners for psql(Masaaki Sakaida)
7689 New contrib/vacuumlo removes large objects not referenced(Peter)
7690 New initialization for table sizes so non-vacuumed tables perform better(Tom)
7691 Improve error messages when a connection is rejected(Tom)
7692 Support for arrays of char() and varchar() fields(Massimo)
7693 Overhaul of hash code to increase reliability and performance(Tom)
7694 Update to PyGreSQL 2.4(D'Arcy)
7695 Changed debug options so -d4 and -d5 produce different node displays(Jan)
7696 New pg_options: pretty_plan, pretty_parse, pretty_rewritten(Jan)
7697 Better optimization statistics for system table access(Tom)
7698 Better handling of non-default block sizes(Massimo)
7699 Improve GEQO optimizer memory consumption(Tom)
7700 UNION now suppports ORDER BY of columns not in target list(Jan)
7701 Major libpq++ improvements(Vince Vielhaber)
7702 pg_dump now uses -z(ACL's) as default(Bruce)
7703 backend cache, memory speedups(Tom)
7704 have pg_dump do everything in one snapshot transaction(Vadim)
7705 fix for large object memory leakage, fix for pg_dumping(Tom)
7706 INET type now respects netmask for comparisons
7707 Make VACUUM ANALYZE only use a readlock(Vadim)
7708 Allow VIEWs on UNIONS(Jan)
7709 pg_dump now can generate consistent snapshots on active databases(Vadim)
7713 Improve port matching(Tom)
7714 Portability fixes for SunOS
7715 Add Windows NT backend port and enable dynamic loading(Magnus and Daniel Horak)
7716 New port to Cobalt Qube(Mips) running Linux(Tatsuo)
7717 Port to NetBSD/m68k(Mr. Mutsuki Nakajima)
7718 Port to NetBSD/sun3(Mr. Mutsuki Nakajima)
7719 Port to NetBSD/macppc(Toshimi Aoki)
7720 Fix for tcl/tk configuration(Vince)
7721 Removed CURRENT key word for rule queries(Jan)
7722 NT dynamic loading now works(Daniel Horak)
7723 Add ARM32 support(Andrew McMurry)
7724 Better support for HP-UX 11 and UnixWare
7725 Improve file handling to be more uniform, prevent file descriptor leak(Tom)
7726 New install commands for plpgsql(Jan)
7733 <sect1 id="release-6-4-2">
7734 <title>Release 6.4.2</title>
7737 <title>Release date</title>
7738 <simpara>1998-12-20</simpara>
7742 The 6.4.1 release was improperly packaged. This also has one additional
7748 <title>Migration to version 6.4.2</title>
7751 A dump/restore is <emphasis>not</emphasis> required for those running
7756 <title>Changes</title>
7760 Fix for datetime constant problem on some platforms(Thomas)
7768 <sect1 id="release-6-4-1">
7769 <title>Release 6.4.1</title>
7772 <title>Release date</title>
7773 <simpara>1998-12-18</simpara>
7777 This is basically a cleanup release for 6.4. We have fixed a variety of
7778 problems reported by 6.4 users.
7783 <title>Migration to version 6.4.1</title>
7786 A dump/restore is <emphasis>not</emphasis> required for those running
7791 <title>Changes</title>
7795 Add pg_dump -N flag to force double quotes around identifiers. This is
7797 Fix for NOT in where clause causing crash(Bruce)
7798 EXPLAIN VERBOSE coredump fix(Vadim)
7799 Fix shared-library problems on Linux
7800 Fix test for table existence to allow mixed-case and whitespace in
7801 the table name(Thomas)
7802 Fix a couple of pg_dump bugs
7803 Configure matches template/.similar entries better(Tom)
7804 Change builtin function names from SPI_* to spi_*
7805 OR WHERE clause fix(Vadim)
7806 Fixes for mixed-case table names(Billy)
7807 contrib/linux/postgres.init.csh/sh fix(Thomas)
7808 libpq memory overrun fix
7810 Change exp() behavior to generate error on underflow(Thomas)
7811 pg_dump fixes for memory leak, inheritance constraints, layout change
7812 update pgaccess to 0.93
7813 Fix prototype for 64-bit platforms
7814 Multibyte fixes(Tatsuo)
7816 Fix memory overruns(Tatsuo)
7817 Fix for lo_import() crash(Bruce)
7818 Better search for install program(Tom)
7821 Use implicit type coercion for matching DEFAULT values(Thomas)
7822 Add routines to help with single-byte (internal) character type(Thomas)
7823 Compilation of libpq for Windows fixes(Magnus)
7824 Upgrade to PyGreSQL 2.2(D'Arcy)
7832 <sect1 id="release-6-4">
7833 <title>Release 6.4</title>
7836 <title>Release date</title>
7837 <simpara>1998-10-30</simpara>
7841 There are <emphasis>many</emphasis> new features and improvements in this release.
7842 Thanks to our developers and maintainers, nearly every aspect of the system
7843 has received some attention since the previous release.
7844 Here is a brief, incomplete summary:
7849 Views and rules are now functional thanks to extensive new code in the
7850 rewrite rules system from Jan Wieck. He also wrote a chapter on it
7851 for the <citetitle>Programmer's Guide</citetitle>.
7856 Jan also contributed a second procedural language, <application>PL/pgSQL</application>, to go with the
7857 original <application>PL/pgTCL</application> procedural language he contributed last release.
7863 We have optional multiple-byte character set support from Tatsuo Ishii
7864 to complement our existing locale support.
7870 Client/server communications has been cleaned up, with better support for
7871 asynchronous messages and interrupts thanks to Tom Lane.
7877 The parser will now perform automatic type coercion to match arguments
7878 to available operators and functions, and to match columns and expressions
7879 with target columns. This uses a generic mechanism which supports
7880 the type extensibility features of <productname>PostgreSQL</productname>.
7881 There is a new chapter in the <citetitle>User's Guide</citetitle>
7882 which covers this topic.
7888 Three new data types have been added.
7889 Two types, <type>inet</type> and <type>cidr</type>, support various forms
7890 of IP network, subnet, and machine addressing. There is now an 8-byte integer
7891 type available on some platforms. See the chapter on data types
7892 in the <citetitle>User's Guide</citetitle> for details.
7893 A fourth type, <type>serial</type>, is now supported by the parser as an
7894 amalgam of the <type>int4</type> type, a sequence, and a unique index.
7900 Several more <acronym>SQL92</acronym>-compatible syntax features have been
7901 added, including <command>INSERT DEFAULT VALUES</command>
7907 The automatic configuration and installation system has received some
7908 attention, and should be more robust for more platforms than it has ever
7917 <title>Migration to version 6.4</title>
7920 A dump/restore using <application>pg_dump</application>
7921 or <application>pg_dumpall</application>
7922 is required for those wishing to migrate data from any
7923 previous release of <productname>PostgreSQL</productname>.
7928 <title>Changes</title>
7934 Fix for a tiny memory leak in PQsetdb/PQfinish(Bryan)
7935 Remove char2-16 data types, use char/varchar(Darren)
7936 Pqfn not handles a NOTICE message(Anders)
7937 Reduced busywaiting overhead for spinlocks with many backends (dg)
7938 Stuck spinlock detection (dg)
7939 Fix up "ISO-style" timespan decoding and encoding(Thomas)
7940 Fix problem with table drop after rollback of transaction(Vadim)
7941 Change error message and remove non-functional update message(Vadim)
7942 Fix for COPY array checking
7943 Fix for SELECT 1 UNION SELECT NULL
7944 Fix for buffer leaks in large object calls(Pascal)
7945 Change owner from oid to int4 type(Bruce)
7946 Fix a bug in the oracle compatibility functions btrim() ltrim() and rtrim()
7947 Fix for shared invalidation cache overflow(Massimo)
7948 Prevent file descriptor leaks in failed COPY's(Bruce)
7949 Fix memory leak in libpgtcl's pg_select(Constantin)
7950 Fix problems with username/passwords over 8 characters(Tom)
7951 Fix problems with handling of asynchronous NOTIFY in backend(Tom)
7952 Fix of many bad system table entries(Tom)
7956 Upgrade ecpg and ecpglib,see src/interfaces/ecpc/ChangeLog(Michael)
7957 Show the index used in an EXPLAIN(Zeugswetter)
7958 EXPLAIN invokes rule system and shows plan(s) for rewritten queries(Jan)
7959 Multibyte awareness of many data types and functions, via configure(Tatsuo)
7960 New configure --with-mb option(Tatsuo)
7961 New initdb --pgencoding option(Tatsuo)
7962 New createdb -E multibyte option(Tatsuo)
7963 Select version(); now returns PostgreSQL version(Jeroen)
7964 libpq now allows asynchronous clients(Tom)
7965 Allow cancel from client of backend query(Tom)
7966 psql now cancels query with Control-C(Tom)
7967 libpq users need not issue dummy queries to get NOTIFY messages(Tom)
7968 NOTIFY now sends sender's PID, so you can tell whether it was your own(Tom)
7969 PGresult struct now includes associated error message, if any(Tom)
7970 Define "tz_hour" and "tz_minute" arguments to date_part()(Thomas)
7971 Add routines to convert between varchar and bpchar(Thomas)
7972 Add routines to allow sizing of varchar and bpchar into target columns(Thomas)
7973 Add bit flags to support timezonehour and minute in data retrieval(Thomas)
7974 Allow more variations on valid floating point numbers (e.g. ".1", "1e6")(Thomas)
7975 Fixes for unary minus parsing with leading spaces(Thomas)
7976 Implement TIMEZONE_HOUR, TIMEZONE_MINUTE per SQL92 specs(Thomas)
7977 Check for and properly ignore FOREIGN KEY column constraints(Thomas)
7978 Define USER as synonym for CURRENT_USER per SQL92 specs(Thomas)
7979 Enable HAVING clause but no fixes elsewhere yet.
7980 Make "char" type a synonym for "char(1)" (actually implemented as bpchar)(Thomas)
7981 Save string type if specified for DEFAULT clause handling(Thomas)
7982 Coerce operations involving different data types(Thomas)
7983 Allow some index use for columns of different types(Thomas)
7984 Add capabilities for automatic type conversion(Thomas)
7985 Cleanups for large objects, so file is truncated on open(Peter)
7986 Readline cleanups(Tom)
7987 Allow psql \f \ to make spaces as delimiter(Bruce)
7988 Pass pg_attribute.atttypmod to the frontend for column field lengths(Tom,Bruce)
7989 Msql compatibility library in /contrib(Aldrin)
7990 Remove the requirement that ORDER/GROUP BY clause identifiers be
7991 included in the target list(David)
7992 Convert columns to match columns in UNION clauses(Thomas)
7993 Remove fork()/exec() and only do fork()(Bruce)
7994 Jdbc cleanups(Peter)
7995 Show backend status on ps command line(only works on some platforms)(Bruce)
7996 Pg_hba.conf now has a sameuser option in the database field
7997 Make lo_unlink take oid param, not int4
7998 New DISABLE_COMPLEX_MACRO for compilers that can't handle our macros(Bruce)
7999 Libpgtcl now handles NOTIFY as a Tcl event, need not send dummy queries(Tom)
8000 libpgtcl cleanups(Tom)
8001 Add -error option to libpgtcl's pg_result command(Tom)
8002 New locale patch, see docs/README/locale(Oleg)
8003 Fix for pg_dump so CONSTRAINT and CHECK syntax is correct(ccb)
8004 New contrib/lo code for large object orphan removal(Peter)
8005 New psql command "SET CLIENT_ENCODING TO 'encoding'" for multibytes
8006 feature, see /doc/README.mb(Tatsuo)
8007 contrib/noupdate code to revoke update permission on a column
8008 libpq can now be compiled on Windows(Magnus)
8009 Add PQsetdbLogin() in libpq
8010 New 8-byte integer type, checked by configure for OS support(Thomas)
8011 Better support for quoted table/column names(Thomas)
8012 Surround table and column names with double-quotes in pg_dump(Thomas)
8013 PQreset() now works with passwords(Tom)
8014 Handle case of GROUP BY target list column number out of range(David)
8015 Allow UNION in subselects
8016 Add auto-size to screen to \d? commands(Bruce)
8017 Use UNION to show all \d? results in one query(Bruce)
8018 Add \d? field search feature(Bruce)
8019 Pg_dump issues fewer \connect requests(Tom)
8020 Make pg_dump -z flag work better, document it in manual page(Tom)
8021 Add HAVING clause with full support for subselects and unions(Stephan)
8022 Full text indexing routines in contrib/fulltextindex(Maarten)
8023 Transaction ids now stored in shared memory(Vadim)
8024 New PGCLIENTENCODING when issuing COPY command(Tatsuo)
8025 Support for SQL92 syntax "SET NAMES"(Tatsuo)
8026 Support for LATIN2-5(Tatsuo)
8027 Add UNICODE regression test case(Tatsuo)
8028 Lock manager cleanup, new locking modes for LLL(Vadim)
8029 Allow index use with OR clauses(Bruce)
8030 Allows "SELECT NULL ORDER BY 1;"
8031 Explain VERBOSE prints the plan, and now pretty-prints the plan to
8032 the postmaster log file(Bruce)
8033 Add indexes display to \d command(Bruce)
8034 Allow GROUP BY on functions(David)
8035 New pg_class.relkind for large objects(Bruce)
8036 New way to send libpq NOTICE messages to a different location(Tom)
8037 New \w write command to psql(Bruce)
8038 New /contrib/findoidjoins scans oid columns to find join relationships(Bruce)
8039 Allow binary-compatible indexes to be considered when checking for valid
8040 Indexes for restriction clauses containing a constant(Thomas)
8041 New ISBN/ISSN code in /contrib/isbn_issn
8042 Allow NOT LIKE, IN, NOT IN, BETWEEN, and NOT BETWEEN constraint(Thomas)
8043 New rewrite system fixes many problems with rules and views(Jan)
8044 * Rules on relations work
8045 * Event qualifications on insert/update/delete work
8046 * New OLD variable to reference CURRENT, CURRENT will be remove in future
8047 * Update rules can reference NEW and OLD in rule qualifications/actions
8048 * Insert/update/delete rules on views work
8049 * Multiple rule actions are now supported, surrounded by parentheses
8050 * Regular users can create views/rules on tables they have RULE permits
8051 * Rules and views inherit the privileges of the creator
8052 * No rules at the column level
8053 * No UPDATE NEW/OLD rules
8054 * New pg_tables, pg_indexes, pg_rules and pg_views system views
8055 * Only a single action on SELECT rules
8056 * Total rewrite overhaul, perhaps for 6.5
8058 * handle aggregates on views
8059 * handle insert into select from view works
8060 System indexes are now multikey(Bruce)
8061 Oidint2, oidint4, and oidname types are removed(Bruce)
8062 Use system cache for more system table lookups(Bruce)
8063 New backend programming language PL/pgSQL in backend/pl(Jan)
8064 New SERIAL data type, auto-creates sequence/index(Thomas)
8065 Enable assert checking without a recompile(Massimo)
8066 User lock enhancements(Massimo)
8067 New setval() command to set sequence value(Massimo)
8068 Auto-remove unix socket file on start-up if no postmaster running(Massimo)
8069 Conditional trace package(Massimo)
8070 New UNLISTEN command(Massimo)
8071 psql and libpq now compile under Windows using win32.mak(Magnus)
8072 Lo_read no longer stores trailing NULL(Bruce)
8073 Identifiers are now truncated to 31 characters internally(Bruce)
8074 Createuser options now availble on the command line
8075 Code for 64-bit integer supported added, configure tested, int8 type(Thomas)
8076 Prevent file descriptor leaf from failed COPY(Bruce)
8077 New pg_upgrade command(Bruce)
8078 Updated /contrib directories(Massimo)
8079 New CREATE TABLE DEFAULT VALUES statement available(Thomas)
8080 New INSERT INTO TABLE DEFAULT VALUES statement available(Thomas)
8081 New DECLARE and FETCH feature(Thomas)
8082 libpq's internal structures now not exported(Tom)
8083 Allow up to 8 key indexes(Bruce)
8084 Remove ARCHIVE key word, that is no longer used(Thomas)
8085 pg_dump -n flag to supress quotes around indentifiers
8086 disable system columns for views(Jan)
8087 new INET and CIDR types for network addresses(TomH, Paul)
8088 no more double quotes in psql output
8089 pg_dump now dumps views(Terry)
8090 new SET QUERY_LIMIT(Tatsuo,Jan)
8094 /contrib cleanup(Jun)
8095 Inline some small functions called for every row(Bruce)
8098 Multibyte regression tests(Soonmyung.)
8099 Remove --disabled options from configure
8100 Define PGDOC to use POSTGRESDIR by default
8101 Make regression optional
8102 Remove extra braces code to pgindent(Bruce)
8103 Add bsdi shared library support(Bruce)
8104 New --without-CXX support configure option(Brook)
8106 Update backend flowchart in tools/backend(Bruce)
8107 Change atttypmod from int16 to int32(Bruce, Tom)
8108 Getrusage() fix for platforms that do not have it(Tom)
8109 Add PQconnectdb, PGUSER, PGPASSWORD to libpq man page
8110 NS32K platform fixes(Phil Nelson, John Buller)
8111 SCO 7/UnixWare 2.x fixes(Billy,others)
8112 Sparc/Solaris 2.5 fixes(Ryan)
8113 Pgbuiltin.3 is obsolete, move to doc files(Thomas)
8114 Even more documention(Thomas)
8115 Nextstep support(Jacek)
8117 pginterface manual page(Bruce)
8118 shared libraries all have version numbers
8119 merged all OS-specific shared library defines into one file
8120 smarter TCL/TK configuration checking(Billy)
8121 smarter perl configuration(Brook)
8122 configure uses supplied install-sh if no install script found(Tom)
8123 new Makefile.shlib for shared library configuration(Tom)
8129 <sect1 id="release-6-3-2">
8130 <title>Release 6.3.2</title>
8133 <title>Release date</title>
8134 <simpara>1998-04-07</simpara>
8138 This is a bug-fix release for 6.3.x.
8139 Refer to the release notes for version 6.3 for a more complete summary of new features.
8147 Repairs automatic configuration support for some platforms, including Linux,
8148 from breakage inadvertently introduced in version 6.3.1.
8154 Correctly handles function calls on the left side of BETWEEN and LIKE clauses.
8161 A dump/restore is NOT required for those running 6.3 or 6.3.1. A
8162 <literal>make distclean</>, <literal>make</>, and <literal>make install</> is all that is required.
8163 This last step should be performed while the postmaster is not running.
8164 You should re-link any custom applications that use <productname>PostgreSQL</productname> libraries.
8167 For upgrades from pre-6.3 installations,
8168 refer to the installation and migration instructions for version 6.3.
8172 <title>Changes</title>
8176 Configure detection improvements for tcl/tk(Brook Milligan, Alvin)
8177 Manual page improvements(Bruce)
8178 BETWEEN and LIKE fix(Thomas)
8179 fix for psql \connect used by pg_dump(Oliver Elphick)
8181 pgaccess, version 0.86
8182 qsort removed, now uses libc version, cleanups(Jeroen)
8183 fix for buffer over-runs detected(Maurice Gittens)
8184 fix for buffer overrun in libpgtcl(Randy Kunkee)
8185 fix for UNION with DISTINCT or ORDER BY(Bruce)
8186 gettimeofday configure check(Doug Winterburn)
8187 Fix "indexes not used" bug(Vadim)
8188 docs additions(Thomas)
8189 Fix for backend memory leak(Bruce)
8190 libreadline cleanup(Erwan MAS)
8191 Remove DISTDIR(Bruce)
8192 Makefile dependency cleanup(Jeroen van Vianen)
8199 <sect1 id="release-6-3-1">
8200 <title>Release 6.3.1</title>
8203 <title>Release date</title>
8204 <simpara>1998-03-23</simpara>
8213 Additional support for multibyte character sets.
8219 Repair byte ordering for mixed-endian clients and servers.
8225 Minor updates to allowed SQL syntax.
8231 Improvements to the configuration autodetection for installation.
8238 A dump/restore is NOT required for those running 6.3. A
8239 <literal>make distclean</>, <literal>make</>, and <literal>make install</> is all that is required.
8240 This last step should be performed while the postmaster is not running.
8241 You should re-link any custom applications that use <productname>PostgreSQL</productname> libraries.
8244 For upgrades from pre-6.3 installations,
8245 refer to the installation and migration instructions for version 6.3.
8249 <title>Changes</title>
8253 ecpg cleanup/fixes, now version 1.1(Michael Meskes)
8254 pg_user cleanup(Bruce)
8255 large object fix for pg_dump and tclsh (alvin)
8256 LIKE fix for multiple adjacent underscores
8257 fix for redefining builtin functions(Thomas)
8259 upgrade to pg_access 0.83
8260 updated CLUSTER manual page
8261 multibyte character set support, see doc/README.mb(Tatsuo)
8262 configure --with-pgport fix
8264 big-endian fix for backend communications(Kataoka)
8265 SUBSTR() and substring() fix(Jan)
8266 several jdbc fixes(Peter)
8267 libpgtcl improvements, see libptcl/README(Randy Kunkee)
8268 Fix for "Datasize = 0" error(Vadim)
8269 Prevent \do from wrapping(Bruce)
8270 Remove duplicate Russian character set entries
8272 Allow optional TABLE key word in LOCK and SELECT INTO(Thomas)
8273 CREATE SEQUENCE options to allow a negative integer(Thomas)
8274 Add "PASSWORD" as an allowed column identifier(Thomas)
8275 Add checks for UNION target fields(Bruce)
8276 Fix Alpha port(Dwayne Bailey)
8277 Fix for text arrays containing quotes(Doug Gibson)
8278 Solaris compile fix(Albert Chin-A-Young)
8279 Better identify tcl and tk libs and includes(Bruce)
8285 <sect1 id="release-6-3">
8286 <title>Release 6.3</title>
8289 <title>Release date</title>
8290 <simpara>1998-03-01</simpara>
8294 There are <emphasis>many</emphasis> new features and improvements in this release.
8295 Here is a brief, incomplete summary:
8300 Many new SQL features, including
8301 full <acronym>SQL92</acronym> subselect capability
8302 (everything is here but target-list subselects).
8308 Support for client-side environment variables to specify time zone and date style.
8314 Socket interface for client/server connection. This is the default now
8315 so you may need to start <application>postmaster</application> with the
8316 <option>-i</option> flag.
8322 Better password authorization mechanisms. Default table privileges have changed.
8328 Old-style <firstterm>time travel</firstterm>
8329 has been removed. Performance has been improved.
8337 Bruce Momjian wrote the following notes to introduce the new release.
8342 There are some general 6.3 issues that I want to mention. These are
8343 only the big items that can not be described in one sentence. A review
8344 of the detailed changes list is still needed.
8347 First, we now have subselects. Now that we have them, I would like to
8348 mention that without subselects, SQL is a very limited language.
8349 Subselects are a major feature, and you should review your code for
8350 places where subselects provide a better solution for your queries. I
8351 think you will find that there are more uses for subselects than you may
8352 think. Vadim has put us on the big SQL map with subselects, and fully
8353 functional ones too. The only thing you can't do with subselects is to
8354 use them in the target list.
8357 Second, 6.3 uses Unix domain sockets rather than TCP/IP by default. To
8358 enable connections from other machines, you have to use the new
8359 postmaster -i option, and of course edit <filename>pg_hba.conf</filename>. Also, for this
8360 reason, the format of <filename>pg_hba.conf</filename> has changed.
8363 Third, <type>char()</type> fields will now allow faster access than <type>varchar()</type> or
8364 <type>text</type>. Specifically, the <type>text</> and <type>varchar()</type> have a penalty for access to
8365 any columns after the first column of this type. <type>char()</type> used to also
8366 have this access penalty, but it no longer does. This may suggest that
8367 you redesign some of your tables, especially if you have short character
8368 columns that you have defined as <type>varchar()</type> or <type>text</type>. This and other
8369 changes make 6.3 even faster than earlier releases.
8372 We now have passwords definable independent of any Unix file. There are
8373 new SQL USER commands.
8374 See the <citetitle>Administrator's Guide</citetitle> for more
8375 information. There is a new table, pg_shadow, which is used to store
8376 user information and user passwords, and it by default only SELECT-able
8377 by the <systemitem>postgres</systemitem> super-user. pg_user is now a view of pg_shadow, and is
8378 SELECT-able by PUBLIC. You should keep using pg_user in your
8379 application without changes.
8382 User-created tables now no longer have SELECT privilege to PUBLIC by
8383 default. This was done because the ANSI standard requires it. You can
8384 of course GRANT any privileges you want after the table is created.
8385 System tables continue to be SELECT-able by PUBLIC.
8388 We also have real deadlock detection code. No more sixty-second
8389 timeouts. And the new locking code implements a <acronym>FIFO</acronym> better, so there
8390 should be less resource starvation during heavy use.
8393 Many complaints have been made about inadequate documentation in previous
8394 releases. Thomas has put much effort into many new manuals for this
8395 release. Check out the doc/ directory.
8398 For performance reasons, time travel is gone, but can be implemented
8399 using triggers (see <filename>pgsql/contrib/spi/README</filename>). Please check out the new
8400 \d command for types, operators, etc. Also, views have their own
8401 privileges now, not based on the underlying tables, so privileges on
8402 them have to be set separately. Check <filename>/pgsql/interfaces</filename> for some new
8403 ways to talk to <productname>PostgreSQL</productname>.
8406 This is the first release that really required an explanation for
8407 existing users. In many ways, this was necessary because the new
8408 release removes many limitations, and the work-arounds people were using
8409 are no longer needed.
8413 <title>Migration to version 6.3</title>
8416 A dump/restore using <application>pg_dump</application>
8417 or <application>pg_dumpall</application>
8418 is required for those wishing to migrate data from any
8419 previous release of <productname>PostgreSQL</productname>.
8424 <title>Changes</title>
8430 Fix binary cursors broken by MOVE implementation(Vadim)
8431 Fix for tcl library crash(Jan)
8432 Fix for array handling, from Gerhard Hintermayer
8433 Fix acl error, and remove duplicate pqtrace(Bruce)
8434 Fix psql \e for empty file(Bruce)
8435 Fix for textcat on varchar() fields(Bruce)
8436 Fix for DBT Sendproc (Zeugswetter Andres)
8437 Fix vacuum analyze syntax problem(Bruce)
8438 Fix for international identifiers(Tatsuo)
8439 Fix aggregates on inherited tables(Bruce)
8440 Fix substr() for out-of-bounds data
8441 Fix for select 1=1 or 2=2, select 1=1 and 2=2, and select sum(2+2)(Bruce)
8442 Fix notty output to show status result. -q option still turns it off(Bruce)
8443 Fix for count(*), aggs with views and multiple tables and sum(3)(Bruce)
8445 Fix for PQtrace start/stop several times(Bruce)
8446 Fix a variety of locking problems like newer lock waiters getting
8447 lock before older waiters, and having readlock people not share
8448 locks if a writer is waiting for a lock, and waiting writers not
8449 getting priority over waiting readers(Bruce)
8450 Fix crashes in psql when executing queries from external files(James)
8451 Fix problem with multiple order by columns, with the first one having
8453 Use correct hash table support functions for float8 and int4(Thomas)
8454 Re-enable JOIN= option in CREATE OPERATOR statement (Thomas)
8455 Change precedence for boolean operators to match expected behavior(Thomas)
8456 Generate elog(ERROR) on over-large integer(Bruce)
8457 Allow multiple-argument functions in constraint clauses(Thomas)
8458 Check boolean input literals for 'true','false','yes','no','1','0'
8459 and throw elog(ERROR) if unrecognized(Thomas)
8460 Major large objects fix
8461 Fix for GROUP BY showing duplicates(Vadim)
8462 Fix for index scans in MergeJion(Vadim)
8466 Subselects with EXISTS, IN, ALL, ANY key words (Vadim, Bruce, Thomas)
8467 New User Manual(Thomas, others)
8468 Speedup by inlining some frequently-called functions
8469 Real deadlock detection, no more timeouts(Bruce)
8470 Add SQL92 "constants" CURRENT_DATE, CURRENT_TIME, CURRENT_TIMESTAMP,
8471 CURRENT_USER(Thomas)
8472 Modify constraint syntax to be SQL92-compliant(Thomas)
8473 Implement SQL92 PRIMARY KEY and UNIQUE clauses using indexes(Thomas)
8474 Recognize SQL92 syntax for FOREIGN KEY. Throw elog notice(Thomas)
8475 Allow NOT NULL UNIQUE constraint clause (each allowed separately before)(Thomas)
8476 Allow PostgreSQL-style casting ("::") of non-constants(Thomas)
8477 Add support for SQL3 TRUE and FALSE boolean constants(Thomas)
8478 Support SQL92 syntax for IS TRUE/IS FALSE/IS NOT TRUE/IS NOT FALSE(Thomas)
8479 Allow shorter strings for boolean literals (e.g. "t", "tr", "tru")(Thomas)
8480 Allow SQL92 delimited identifiers(Thomas)
8481 Implement SQL92 binary and hexadecimal string decoding (b'10' and x'1F')(Thomas)
8482 Support SQL92 syntax for type coercion of literal strings
8483 (e.g. "DATETIME 'now'")(Thomas)
8484 Add conversions for int2, int4, and OID types to and from text(Thomas)
8485 Use shared lock when building indexes(Vadim)
8486 Free memory allocated for an user query inside transaction block after
8487 this query is done, was turned off in <= 6.2.1(Vadim)
8488 New SQL statement CREATE PROCEDURAL LANGUAGE(Jan)
8489 New <productname>PostgreSQL</productname> Procedural Language (PL) backend interface(Jan)
8490 Rename pg_dump -H option to -h(Bruce)
8491 Add Java support for passwords, European dates(Peter)
8492 Use indexes for LIKE and ~, !~ operations(Bruce)
8493 Add hash functions for datetime and timespan(Thomas)
8494 Time Travel removed(Vadim, Bruce)
8495 Add paging for \d and \z, and fix \i(Bruce)
8496 Add Unix domain socket support to backend and to frontend library(Goran)
8497 Implement CREATE DATABASE/WITH LOCATION and initlocation utility(Thomas)
8498 Allow more SQL92 and/or <productname>PostgreSQL</productname> reserved words as column identifiers(Thomas)
8499 Augment support for SQL92 SET TIME ZONE...(Thomas)
8500 SET/SHOW/RESET TIME ZONE uses TZ backend environment variable(Thomas)
8501 Implement SET keyword = DEFAULT and SET TIME ZONE DEFAULT(Thomas)
8502 Enable SET TIME ZONE using TZ environment variable(Thomas)
8503 Add PGDATESTYLE environment variable to frontend and backend initialization(Thomas)
8504 Add PGTZ, PGCOSTHEAP, PGCOSTINDEX, PGRPLANS, PGGEQO
8505 frontend library initialization environment variables(Thomas)
8506 Regression tests time zone automatically set with "setenv PGTZ PST8PDT"(Thomas)
8507 Add pg_description table for info on tables, columns, operators, types, and
8509 Increase 16 char limit on system table/index names to 32 characters(Bruce)
8510 Rename system indexes(Bruce)
8511 Add 'GERMAN' option to SET DATESTYLE(Thomas)
8512 Define an "ISO-style" timespan output format with "hh:mm:ss" fields(Thomas)
8513 Allow fractional values for delta times (e.g. '2.5 days')(Thomas)
8514 Validate numeric input more carefully for delta times(Thomas)
8515 Implement day of year as possible input to date_part()(Thomas)
8516 Define timespan_finite() and text_timespan() functions(Thomas)
8517 Remove archive stuff(Bruce)
8518 Allow for a pg_password authentication database that is separate from
8519 the system password file(Todd)
8520 Dump ACLs, GRANT, REVOKE privileges(Matt)
8521 Define text, varchar, and bpchar string length functions(Thomas)
8522 Fix Query handling for inheritance, and cost computations(Bruce)
8523 Implement CREATE TABLE/AS SELECT (alternative to SELECT/INTO)(Thomas)
8524 Allow NOT, IS NULL, IS NOT NULL in constraints(Thomas)
8525 Implement UNIONs for SELECT(Bruce)
8526 Add UNION, GROUP, DISTINCT to INSERT(Bruce)
8527 varchar() stores only necessary bytes on disk(Bruce)
8528 Fix for BLOBs(Peter)
8529 Mega-Patch for JDBC...see README_6.3 for list of changes(Peter)
8530 Remove unused "option" from PQconnectdb()
8531 New LOCK command and lock manual page describing deadlocks(Bruce)
8532 Add new psql \da, \dd, \df, \do, \dS, and \dT commands(Bruce)
8533 Enhance psql \z to show sequences(Bruce)
8534 Show NOT NULL and DEFAULT in psql \d table(Bruce)
8535 New psql .psqlrc file start-up(Andrew)
8536 Modify sample start-up script in contrib/linux to show syslog(Thomas)
8537 New types for IP and MAC addresses in contrib/ip_and_mac(TomH)
8538 Unix system time conversions with date/time types in contrib/unixdate(Thomas)
8539 Update of contrib stuff(Massimo)
8540 Add Unix socket support to DBD::Pg(Goran)
8541 New python interface (PyGreSQL 2.0)(D'Arcy)
8542 New frontend/backend protocol has a version number, network byte order(Phil)
8543 Security features in pg_hba.conf enhanced and documented, many cleanups(Phil)
8544 CHAR() now faster access than VARCHAR() or TEXT
8545 ecpg embedded SQL preprocessor
8546 Reduce system column overhead(Vadmin)
8547 Remove pg_time table(Vadim)
8548 Add pg_type attribute to identify types that need length (bpchar, varchar)
8549 Add report of offending line when COPY command fails
8550 Allow VIEW privileges to be set separately from the underlying tables.
8551 For security, use GRANT/REVOKE on views as appropriate(Jan)
8552 Tables now have no default GRANT SELECT TO PUBLIC. You must
8553 explicitly grant such privileges.
8554 Clean up tutorial examples(Darren)
8558 Add new html development tools, and flow chart in /tools/backend
8559 Fix for SCO compiles
8560 Stratus computer port Robert Gillies
8561 Added support for shlib for BSD44_derived & i386_solaris
8562 Make configure more automated(Brook)
8563 Add script to check regression test results
8564 Break parser functions into smaller files, group together(Bruce)
8565 Rename heap_create to heap_create_and_catalog, rename heap_creatr
8566 to heap_create()(Bruce)
8567 Sparc/Linux patch for locking(TomS)
8568 Remove PORTNAME and reorganize port-specific stuff(Marc)
8569 Add optimizer README file(Bruce)
8570 Remove some recursion in optimizer and clean up some code there(Bruce)
8571 Fix for NetBSD locking(Henry)
8572 Fix for libptcl make(Tatsuo)
8574 Change IS TRUE, IS FALSE, ... to expressions using "=" rather than
8575 function calls to istrue() or isfalse() to allow optimization(Thomas)
8576 Various fixes NetBSD/Sparc related(TomH)
8577 Alpha linux locking(Travis,Ryan)
8578 Change elog(WARN) to elog(ERROR)(Bruce)
8579 FAQ for FreeBSD(Marc)
8580 Bring in the PostODBC source tree as part of our standard distribution(Marc)
8581 A minor patch for HP/UX 10 vs 9(Stan)
8582 New pg_attribute.atttypmod for type-specific info like varchar length(Bruce)
8583 UnixWare patches(Billy)
8584 New i386 'lock' for spinlock asm(Billy)
8585 Support for multiplexed backends is removed
8586 Start an OpenBSD port
8589 Add string functions to regression suite(Thomas)
8590 Expand a few function names formerly truncated to 16 characters(Thomas)
8591 Remove un-needed malloc() calls and replace with palloc()(Bruce)
8597 <sect1 id="release-6-2-1">
8598 <title>Release 6.2.1</title>
8601 <title>Release date</title>
8602 <simpara>1997-10-17</simpara>
8606 6.2.1 is a bug-fix and usability release on 6.2.
8614 Allow strings to span lines, per <acronym>SQL92</acronym>.
8620 Include example trigger function for inserting user names on table updates.
8627 This is a minor bug-fix release on 6.2.
8628 For upgrades from pre-6.2 systems, a full dump/reload is required.
8629 Refer to the 6.2 release notes for instructions.
8633 <title>Migration from version 6.2 to version 6.2.1</title>
8636 This is a minor bug-fix release. A dump/reload is not required from version 6.2,
8637 but is required from any release prior to 6.2.
8640 In upgrading from version 6.2, if you choose to dump/reload you will find that
8641 avg(money) is now calculated correctly. All other bug fixes take effect
8642 upon updating the executables.
8645 Another way to avoid dump/reload is to use the following SQL command
8646 from <command>psql</command> to update the existing system table:
8649 update pg_aggregate set aggfinalfn = 'cash_div_flt8'
8650 where aggname = 'avg' and aggbasetype = 790;
8654 This will need to be done to every existing database, including template1.
8659 <title>Changes</title>
8663 Allow TIME and TYPE column names(Thomas)
8664 Allow larger range of true/false as boolean values(Thomas)
8665 Support output of "now" and "current"(Thomas)
8666 Handle DEFAULT with INSERT of NULL properly(Vadim)
8667 Fix for relation reference counts problem in buffer manager(Vadim)
8668 Allow strings to span lines, like ANSI(Thomas)
8669 Fix for backward cursor with ORDER BY(Vadim)
8670 Fix avg(cash) computation(Thomas)
8671 Fix for specifying a column twice in ORDER/GROUP BY(Vadim)
8672 Documented new libpq function to return affected rows, PQcmdTuples(Bruce)
8673 Trigger function for inserting user names for INSERT/UPDATE(Brook Milligan)
8679 <sect1 id="release-6-2">
8680 <title>Release 6.2</title>
8683 <title>Release date</title>
8684 <simpara>1997-10-02</simpara>
8688 A dump/restore is required for those wishing to migrate data from
8689 previous releases of <productname>PostgreSQL</productname>.
8693 <title>Migration from version 6.1 to version 6.2</title>
8696 This migration requires a complete dump of the 6.1 database and a
8697 restore of the database in 6.2.
8700 Note that the <command>pg_dump</command> and <command>pg_dumpall</command> utility from 6.2 should be used
8701 to dump the 6.1 database.
8706 <title>Migration from version 1.<replaceable>x</> to version 6.2</title>
8709 Those migrating from earlier 1.* releases should first upgrade to 1.09
8710 because the COPY output format was improved from the 1.02 release.
8715 <title>Changes</title>
8721 Fix problems with pg_dump for inheritance, sequences, archive tables(Bruce)
8722 Fix compile errors on overflow due to shifts, unsigned, and bad prototypes
8723 from Solaris(Diab Jerius)
8724 Fix bugs in geometric line arithmetic (bad intersection calculations)(Thomas)
8725 Check for geometric intersections at endpoints to avoid rounding ugliness(Thomas)
8726 Catch non-functional delete attempts(Vadim)
8727 Change time function names to be more consistent(Michael Reifenberg)
8728 Check for zero divides(Michael Reifenberg)
8729 Fix very old bug which made rows changed/inserted by a command
8730 visible to the command itself (so we had multiple update of
8731 updated rows, etc.)(Vadim)
8732 Fix for SELECT null, 'fail' FROM pg_am (Patrick)
8733 SELECT NULL as EMPTY_FIELD now allowed(Patrick)
8734 Remove un-needed signal stuff from contrib/pginterface
8735 Fix OR (where x != 1 or x isnull didn't return rows with x NULL) (Vadim)
8736 Fix time_cmp function (Vadim)
8737 Fix handling of functions with non-attribute first argument in
8738 WHERE clauses (Vadim)
8739 Fix GROUP BY when order of entries is different from order
8740 in target list (Vadim)
8741 Fix pg_dump for aggregates without sfunc1 (Vadim)
8745 Default genetic optimizer GEQO parameter is now 8(Bruce)
8746 Allow use parameters in target list having aggregates in functions(Vadim)
8747 Added JDBC driver as an interface(Adrian & Peter)
8749 Return number of rows inserted/affected by INSERT/UPDATE/DELETE etc.(Vadim)
8750 Triggers implemented with CREATE TRIGGER (SQL3)(Vadim)
8751 SPI (Server Programming Interface) allows execution of queries inside
8753 NOT NULL implemented (SQL92)(Robson Paniago de Miranda)
8754 Include reserved words for string handling, outer joins, and unions(Thomas)
8755 Implement extended comments ("/* ... */") using exclusive states(Thomas)
8756 Add "//" single-line comments(Bruce)
8757 Remove some restrictions on characters in operator names(Thomas)
8758 DEFAULT and CONSTRAINT for tables implemented (SQL92)(Vadim & Thomas)
8759 Add text concatenation operator and function (SQL92)(Thomas)
8760 Support WITH TIME ZONE syntax (SQL92)(Thomas)
8761 Support INTERVAL unit TO unit syntax (SQL92)(Thomas)
8762 Define types DOUBLE PRECISION, INTERVAL, CHARACTER,
8763 and CHARACTER VARYING (SQL92)(Thomas)
8764 Define type FLOAT(p) and rudimentary DECIMAL(p,s), NUMERIC(p,s) (SQL92)(Thomas)
8765 Define EXTRACT(), POSITION(), SUBSTRING(), and TRIM() (SQL92)(Thomas)
8766 Define CURRENT_DATE, CURRENT_TIME, CURRENT_TIMESTAMP (SQL92)(Thomas)
8767 Add syntax and warnings for UNION, HAVING, INNER and OUTER JOIN (SQL92)(Thomas)
8768 Add more reserved words, mostly for SQL92 compliance(Thomas)
8769 Allow hh:mm:ss time entry for timespan/reltime types(Thomas)
8770 Add center() routines for lseg, path, polygon(Thomas)
8771 Add distance() routines for circle-polygon, polygon-polygon(Thomas)
8772 Check explicitly for points and polygons contained within polygons
8773 using an axis-crossing algorithm(Thomas)
8774 Add routine to convert circle-box(Thomas)
8775 Merge conflicting operators for different geometric data types(Thomas)
8776 Replace distance operator "<===>" with "<->"(Thomas)
8777 Replace "above" operator "!^" with ">^" and "below" operator "!|" with "<^"(Thomas)
8778 Add routines for text trimming on both ends, substring, and string position(Thomas)
8779 Added conversion routines circle(box) and poly(circle)(Thomas)
8780 Allow internal sorts to be stored in memory rather than in files(Bruce & Vadim)
8781 Allow functions and operators on internally-identical types to succeed(Bruce)
8782 Speed up backend start-up after profiling analysis(Bruce)
8783 Inline frequently called functions for performance(Bruce)
8784 Reduce open() calls(Bruce)
8785 psql: Add PAGER for \h and \?,\C fix
8786 Fix for psql pager when no tty(Bruce)
8787 New entab utility(Bruce)
8788 General trigger functions for referential integrity (Vadim)
8789 General trigger functions for time travel (Vadim)
8790 General trigger functions for AUTOINCREMENT/IDENTITY feature (Vadim)
8791 MOVE implementation (Vadim)
8795 HP-UX 10 patches (Vladimir Turin)
8796 Added SCO support, (Daniel Harris)
8797 MkLinux patches (Tatsuo Ishii)
8798 Change geometric box terminology from "length" to "width"(Thomas)
8799 Deprecate temporary unstored slope fields in geometric code(Thomas)
8800 Remove restart instructions from INSTALL(Bruce)
8801 Look in /usr/ucb first for install(Bruce)
8802 Fix c++ copy example code(Thomas)
8803 Add -o to psql manual page(Bruce)
8804 Prevent relname unallocated string length from being copied into database(Bruce)
8805 Cleanup for NAMEDATALEN use(Bruce)
8806 Fix pg_proc names over 15 chars in output(Bruce)
8807 Add strNcpy() function(Bruce)
8808 remove some (void) casts that are unnecessary(Bruce)
8809 new interfaces directory(Marc)
8810 Replace fopen() calls with calls to fd.c functions(Bruce)
8811 Make functions static where possible(Bruce)
8812 enclose unused functions in #ifdef NOT_USED(Bruce)
8813 Remove call to difftime() in timestamp support to fix SunOS(Bruce & Thomas)
8814 Changes for Digital Unix
8815 Portability fix for pg_dumpall(Bruce)
8816 Rename pg_attribute.attnvals to attdispersion(Bruce)
8817 "intro/unix" manual page now "pgintro"(Bruce)
8818 "built-in" manual page now "pgbuiltin"(Bruce)
8819 "drop" manual page now "drop_table"(Bruce)
8820 Add "create_trigger", "drop_trigger" manual pages(Thomas)
8821 Add constraints regression test(Vadim & Thomas)
8822 Add comments syntax regression test(Thomas)
8823 Add PGINDENT and support program(Bruce)
8824 Massive commit to run PGINDENT on all *.c and *.h files(Bruce)
8825 Files moved to /src/tools directory(Bruce)
8826 SPI and Trigger programming guides (Vadim & D'Arcy)
8832 <sect1 id="release-6-1-1">
8833 <title>Release 6.1.1</title>
8836 <title>Release date</title>
8837 <simpara>1997-07-22</simpara>
8841 <title>Migration from version 6.1 to version 6.1.1</title>
8844 This is a minor bug-fix release. A dump/reload is not required from version 6.1,
8845 but is required from any release prior to 6.1.
8846 Refer to the release notes for 6.1 for more details.
8851 <title>Changes</title>
8855 fix for SET with options (Thomas)
8856 allow pg_dump/pg_dumpall to preserve ownership of all tables/objects(Bruce)
8857 new psql \connect option allows changing usernames without changing databases
8858 fix for initdb --debug option(Yoshihiko Ichikawa))
8859 lextest cleanup(Bruce)
8861 fix date/time month boundary arithmetic(Thomas)
8862 fix timezone daylight handling for some ports(Thomas, Bruce, Tatsuo)
8863 timestamp overhauled to use standard functions(Thomas)
8864 other code cleanup in date/time routines(Thomas)
8865 psql's \d now case-insensitive(Bruce)
8866 psql's backslash commands can now have trailing semicolon(Bruce)
8867 fix memory leak in psql when using \g(Bruce)
8868 major fix for endian handling of communication to server(Thomas, Tatsuo)
8869 Fix for Solaris assembler and include files(Yoshihiko Ichikawa)
8870 allow underscores in usernames(Bruce)
8871 pg_dumpall now returns proper status, portability fix(Bruce)
8877 <sect1 id="release-6-1">
8878 <title>Release 6.1</title>
8881 <title>Release date</title>
8882 <simpara>1997-06-08</simpara>
8886 The regression tests have been adapted and extensively modified for the
8887 6.1 release of <productname>PostgreSQL</productname>.
8891 Three new data types (<type>datetime</type>, <type>timespan</type>, and <type>circle</type>) have been added to
8892 the native set of <productname>PostgreSQL</productname> types. Points, boxes, paths, and polygons
8893 have had their output formats made consistent across the data types.
8894 The polygon output in misc.out has only been spot-checked for correctness
8895 relative to the original regression output.
8899 <productname>PostgreSQL</productname> 6.1 introduces a new, alternate
8900 optimizer which uses <firstterm>genetic</firstterm>
8901 algorithms. These algorithms introduce a random behavior in the ordering
8902 of query results when the query contains multiple qualifiers or multiple
8903 tables (giving the optimizer a choice on order of evaluation). Several
8904 regression tests have been modified to explicitly order the results, and
8905 hence are insensitive to optimizer choices. A few regression tests are
8906 for data types which are inherently unordered (e.g. points and time
8907 intervals) and tests involving those types are explicitly bracketed with
8908 <command>set geqo to 'off'</command> and <command>reset geqo</command>.
8912 The interpretation of array specifiers (the curly braces around atomic
8913 values) appears to have changed sometime after the original regression
8914 tests were generated. The current <filename>./expected/*.out</filename> files reflect this
8915 new interpretation, which may not be correct!
8919 The float8 regression test fails on at least some platforms. This is due
8920 to differences in implementations of <function>pow()</function> and <function>exp()</function> and the signaling
8921 mechanisms used for overflow and underflow conditions.
8925 The <quote>random</> results in the random test should cause the
8926 <quote>random</quote> test to be <quote>failed</quote>, since the
8927 regression tests are evaluated using a simple diff. However,
8928 <quote>random</> does not seem to produce random results on my test
8929 machine (Linux/<application>gcc</>/i686).
8933 <title>Migration to version 6.1</title>
8936 This migration requires a complete dump of the 6.0 database and a
8937 restore of the database in 6.1.
8940 Those migrating from earlier 1.* releases should first upgrade to 1.09
8941 because the COPY output format was improved from the 1.02 release.
8946 <title>Changes</title>
8952 packet length checking in library routines
8953 lock manager priority patch
8954 check for under/over flow of float8(Bruce)
8955 multitable join fix(Vadim)
8956 SIGPIPE crash fix(Darren)
8957 large object fixes(Sven)
8958 allow btree indexes to handle NULLs(Vadim)
8959 timezone fixes(D'Arcy)
8960 select SUM(x) can return NULL on no rows(Thomas)
8961 internal optimizer, executor bug fixes(Vadim)
8962 fix problem where inner loop in < or <= has no rows(Vadim)
8963 prevent re-commuting join index clauses(Vadim)
8964 fix join clauses for multiple tables(Vadim)
8965 fix hash, hashjoin for arrays(Vadim)
8966 fix btree for abstime type(Vadim)
8967 large object fixes(Raymond)
8968 fix buffer leak in hash indexes (Vadim)
8969 fix rtree for use in inner scan (Vadim)
8970 fix gist for use in inner scan, cleanups (Vadim, Andrea)
8971 avoid unnecessary local buffers allocation (Vadim, Massimo)
8972 fix local buffers leak in transaction aborts (Vadim)
8973 fix file manager memmory leaks, cleanups (Vadim, Massimo)
8974 fix storage manager memmory leaks (Vadim)
8975 fix btree duplicates handling (Vadim)
8976 fix deleted rows reincarnation caused by vacuum (Vadim)
8977 fix SELECT varchar()/char() INTO TABLE made zero-length fields(Bruce)
8978 many psql, pg_dump, and libpq memory leaks fixed using Purify (Igor)
8982 attribute optimization statistics(Bruce)
8983 much faster new btree bulk load code(Paul)
8984 BTREE UNIQUE added to bulk load code(Vadim)
8985 new lock debug code(Massimo)
8986 massive changes to libpg++(Leo)
8987 new GEQO optimizer speeds table multitable optimization(Martin)
8988 new WARN message for non-unique insert into unique key(Marc)
8989 update x=-3, no spaces, now valid(Bruce)
8990 remove case-sensitive identifier handling(Bruce,Thomas,Dan)
8991 debug backend now pretty-prints tree(Darren)
8992 new Oracle character functions(Edmund)
8993 new plaintext password functions(Dan)
8994 no such class or insufficient privilege changed to distinct messages(Dan)
8995 new ANSI timestamp function(Dan)
8996 new ANSI Time and Date types (Thomas)
8997 move large chunks of data in backend(Martin)
8998 multicolumn btree indexes(Vadim)
8999 new SET var TO value command(Martin)
9000 update transaction status on reads(Dan)
9001 new locale settings for character types(Oleg)
9002 new SEQUENCE serial number generator(Vadim)
9003 GROUP BY function now possible(Vadim)
9004 re-organize regression test(Thomas,Marc)
9005 new optimizer operation weights(Vadim)
9006 new psql \z grant/permit option(Marc)
9007 new MONEY data type(D'Arcy,Thomas)
9008 tcp socket communication speed improved(Vadim)
9009 new VACUUM option for attribute statistics, and for certain columns (Vadim)
9010 many geometric type improvements(Thomas,Keith)
9011 additional regression tests(Thomas)
9012 new datestyle variable(Thomas,Vadim,Martin)
9013 more comparison operators for sorting types(Thomas)
9014 new conversion functions(Thomas)
9015 new more compact btree format(Vadim)
9016 allow pg_dumpall to preserve database ownership(Bruce)
9017 new SET GEQO=# and R_PLANS variable(Vadim)
9018 old (!GEQO) optimizer can use right-sided plans (Vadim)
9019 typechecking improvement in SQL parser(Bruce)
9020 new SET, SHOW, RESET commands(Thomas,Vadim)
9021 new \connect database USER option
9022 new destroydb -i option (Igor)
9023 new \dt and \di psql commands (Darren)
9024 SELECT "\n" now escapes newline (A. Duursma)
9025 new geometry conversion functions from old format (Thomas)
9029 new configuration script(Marc)
9030 readline configuration option added(Marc)
9031 OS-specific configuration options removed(Marc)
9032 new OS-specific template files(Marc)
9033 no more need to edit Makefile.global(Marc)
9034 re-arrange include files(Marc)
9035 nextstep patches (Gregor Hoffleit)
9036 removed Windows-specific code(Bruce)
9037 removed postmaster -e option, now only postgres -e option (Bruce)
9038 merge duplicate library code in front/backends(Martin)
9039 now works with eBones, international Kerberos(Jun)
9040 more shared library support
9041 c++ include file cleanup(Bruce)
9042 warn about buggy flex(Bruce)
9043 DG/UX, Ultrix, IRIX, AIX portability fixes
9049 <sect1 id="release-6-0">
9050 <title>Release 6.0</title>
9053 <title>Release date</title>
9054 <simpara>1997-01-29</simpara>
9058 A dump/restore is required for those wishing to migrate data from
9059 previous releases of <productname>PostgreSQL</productname>.
9063 <title>Migration from version 1.09 to version 6.0</title>
9066 This migration requires a complete dump of the 1.09 database and a
9067 restore of the database in 6.0.
9072 <title>Migration from pre-1.09 to version 6.0</title>
9075 Those migrating from earlier 1.* releases should first upgrade to 1.09
9076 because the COPY output format was improved from the 1.02 release.
9081 <title>Changes</title>
9087 ALTER TABLE bug - running postgress process needs to re-read table definition
9088 Allow vacuum to be run on one table or entire database(Bruce)
9090 Fix array over-runs of memory writes(Kurt)
9091 Fix elusive btree range/non-range bug(Dan)
9092 Fix for hash indexes on some types like time and date
9093 Fix for pg_log size explosion
9094 Fix permissions on lo_export()(Bruce)
9095 Fix unitialized reads of memory(Kurt)
9096 Fixed ALTER TABLE ... char(3) bug(Bruce)
9097 Fixed a few small memory leaks
9098 Fixed EXPLAIN handling of options and changed full_path option name
9099 Fixed output of group acl privileges
9100 Memory leaks (hunt and destroy with tools like Purify(Kurt)
9101 Minor improvements to rules system
9103 New asserts for run-checking
9104 Overhauled parser/analyze code to properly report errors and increase speed
9105 Pg_dump -d now handles NULL's properly(Bruce)
9106 Prevent SELECT NULL from crashing server (Bruce)
9107 Properly report errors when INSERT ... SELECT columns did not match
9108 Properly report errors when insert column names were not correct
9109 psql \g filename now works(Bruce)
9110 psql fixed problem with multiple statements on one line with multiple outputs
9111 Removed duplicate system OIDs
9112 SELECT * INTO TABLE . GROUP/ORDER BY gives unlink error if table exists(Bruce)
9113 Several fixes for queries that crashed the backend
9114 Starting quote in insert string errors(Bruce)
9115 Submitting an empty query now returns empty status, not just " " query(Bruce)
9119 Add EXPLAIN manual page(Bruce)
9120 Add UNIQUE index capability(Dan)
9121 Add hostname/user level access control rather than just hostname and user
9122 Add synonym of != for <>(Bruce)
9123 Allow "select oid,* from table"
9124 Allow BY,ORDER BY to specify columns by number, or by non-alias table.column(Bruce)
9125 Allow COPY from the frontend(Bryan)
9126 Allow GROUP BY to use alias column name(Bruce)
9127 Allow actual compression, not just reuse on the same page(Vadim)
9128 Allow installation-configuration option to auto-add all local users(Bryan)
9129 Allow libpq to distinguish between text value '' and null(Bruce)
9130 Allow non-postgres users with createdb privs to destroydb's
9131 Allow restriction on who can create C functions(Bryan)
9132 Allow restriction on who can do backend COPY(Bryan)
9133 Can shrink tables, pg_time and pg_log(Vadim & Erich)
9134 Change debug level 2 to print queries only, changed debug heading layout(Bruce)
9135 Change default decimal constant representation from float4 to float8(Bruce)
9136 European date format now set when postmaster is started
9137 Execute lowercase function names if not found with exact case
9138 Fixes for aggregate/GROUP processing, allow 'select sum(func(x),sum(x+y) from z'
9139 Gist now included in the distrubution(Marc)
9140 Idend authentication of local users(Bryan)
9141 Implement BETWEEN qualifier(Bruce)
9142 Implement IN qualifier(Bruce)
9143 libpq has PQgetisnull()(Bruce)
9144 libpq++ improvements
9145 New options to initdb(Bryan)
9146 Pg_dump allow dump of OIDs(Bruce)
9147 Pg_dump create indexes after tables are loaded for speed(Bruce)
9148 Pg_dumpall dumps all databases, and the user table
9149 Pginterface additions for NULL values(Bruce)
9150 Prevent postmaster from being run as root
9151 psql \h and \? is now readable(Bruce)
9152 psql allow backslashed, semicolons anywhere on the line(Bruce)
9153 psql changed command prompt for lines in query or in quotes(Bruce)
9154 psql char(3) now displays as (bp)char in \d output(Bruce)
9155 psql return code now more accurate(Bryan?)
9156 psql updated help syntax(Bruce)
9157 Re-visit and fix vacuum(Vadim)
9158 Reduce size of regression diffs, remove timezone name difference(Bruce)
9159 Remove compile-time parameters to enable binary distributions(Bryan)
9160 Reverse meaning of HBA masks(Bryan)
9161 Secure Authentication of local users(Bryan)
9162 Speed up vacuum(Vadim)
9163 Vacuum now had VERBOSE option(Bruce)
9167 All functions now have prototypes that are compared against the calls
9168 Allow asserts to be disabled easly from Makefile.global(Bruce)
9169 Change oid constants used in code to #define names
9170 Decoupled sparc and solaris defines(Kurt)
9171 Gcc -Wall compiles cleanly with warnings only from unfixable constructs
9172 Major include file reorganization/reduction(Marc)
9173 Make now stops on compile failure(Bryan)
9174 Makefile restructuring(Bryan, Marc)
9175 Merge bsdi_2_1 to bsdi(Bruce)
9176 Monitor program removed
9177 Name change from Postgres95 to PostgreSQL
9178 New config.h file(Marc, Bryan)
9179 PG_VERSION now set to 6.0 and used by postmaster
9180 Portability additions, including Ultrix, DG/UX, AIX, and Solaris
9181 Reduced the number of #define's, centeralized #define's
9182 Remove duplicate OIDS in system tables(Dan)
9183 Remove duplicate system catalog info or report mismatches(Dan)
9184 Removed many os-specific #define's
9185 Restructured object file generation/location(Bryan, Marc)
9186 Restructured port-specific file locations(Bryan, Marc)
9187 Unused/uninialized variables corrected
9193 <sect1 id="release-1-09">
9194 <title>Release 1.09</title>
9197 <title>Release date</title>
9198 <simpara>1996-11-04</simpara>
9202 Sorry, we didn't keep track of changes from 1.02 to 1.09. Some of
9203 the changes listed in 6.0 were actually included in the 1.02.1 to 1.09
9208 <sect1 id="release-1-02">
9209 <title>Release 1.02</title>
9212 <title>Release date</title>
9213 <simpara>1996-08-01</simpara>
9217 <title>Migration from version 1.02 to version 1.02.1</title>
9220 Here is a new migration file for 1.02.1. It includes the 'copy' change
9221 and a script to convert old <acronym>ASCII</acronym> files.
9225 The following notes are for the benefit of users who want to migrate
9226 databases from <productname>Postgres95</> 1.01 and 1.02 to <productname>Postgres95</> 1.02.1.
9229 If you are starting afresh with <productname>Postgres95</> 1.02.1 and do not need
9230 to migrate old databases, you do not need to read any further.
9235 In order to upgrade older <productname>Postgres95</> version 1.01 or 1.02 databases to
9236 version 1.02.1, the following steps are required:
9241 Start up a new 1.02.1 postmaster
9246 Add the new built-in functions and operators of 1.02.1 to 1.01 or 1.02
9247 databases. This is done by running the new 1.02.1 server against
9248 your own 1.01 or 1.02 database and applying the queries attached at
9249 the end of the file. This can be done easily through <command>psql</>. If your
9250 1.01 or 1.02 database is named <literal>testdb</literal> and you have cut the commands
9251 from the end of this file and saved them in <filename>addfunc.sql</filename>:
9253 % psql testdb -f addfunc.sql
9256 Those upgrading 1.02 databases will get a warning when executing the
9257 last two statements in the file because they are already present in 1.02. This is
9258 not a cause for concern.
9265 <title>Dump/Reload Procedure</title>
9268 If you are trying to reload a pg_dump or text-mode, <literal>copy tablename to
9269 stdout</literal> generated with a previous version, you will need to run the
9270 attached <command>sed</command> script on the ASCII file before loading it into the
9271 database. The old format used '.' as end-of-data, while '\.' is now the
9272 end-of-data marker. Also, empty strings are now loaded in as '' rather
9273 than NULL. See the copy manual page for full details.
9276 sed 's/^\.$/\\./g' <in_file >out_file
9280 If you are loading an older binary copy or non-<systemitem>stdout</> copy, there is no
9281 end-of-data character, and hence no conversion necessary.
9284 -- following lines added by agc to reflect the case-insensitive
9285 -- regexp searching for varchar (in 1.02), and bpchar (in 1.02.1)
9286 create operator ~* (leftarg = bpchar, rightarg = text, procedure = texticregexeq);
9287 create operator !~* (leftarg = bpchar, rightarg = text, procedure = texticregexne);
9288 create operator ~* (leftarg = varchar, rightarg = text, procedure = texticregexeq);
9289 create operator !~* (leftarg = varchar, rightarg = text, procedure = texticregexne);
9295 <title>Changes</title>
9299 Source code maintenance and development
9300 * worldwide team of volunteers
9301 * the source tree now in CVS at ftp.ki.net
9304 * psql (and underlying libpq library) now has many more options for
9305 formatting output, including HTML
9306 * pg_dump now output the schema and/or the data, with many fixes to
9307 enhance completeness.
9308 * psql used in place of monitor in administration shell scripts.
9309 monitor to be deprecated in next release.
9310 * date/time functions enhanced
9311 * NULL insert/update/comparison fixed/enhanced
9312 * TCL/TK lib and shell fixed to work with both tck7.4/tk4.0 and tcl7.5/tk4.1
9314 Bug Fixes (almost too numerous to mention)
9316 * storage management
9317 * check for NULL pointer before dereferencing
9321 * added SolarisX86 port
9322 * added BSD/OS 2.1 port
9327 Contributors (appologies to any missed)
9328 * Kurt J. Lidl <lidl@va.pubnix.com>
9329 (missed in first run, but no less important)
9330 * Erich Stamberger <eberger@gewi.kfunigraz.ac.at>
9331 * Jason Wright <jason@shiloh.vnet.net>
9332 * Cees de Groot <C.deGroot@inter.NL.net>
9333 * ernst.molitor@uni-bonn.de
9334 * michael.siebenborn@ae3.Hypo.DE (Michael Siebenborn (6929))
9335 * Brian E. Gallew <geek+@cmu.edu>
9336 * Vadim B. Mikheev <vadim@sable.krasnoyarsk.su>
9337 * Adam Sussman <myddryn@vidya.com>
9338 * Chris Dunlop <chris@onthe.net.au>
9339 * Marc G. Fournier <scrappy@ki.net>
9340 * Dan McGuirk <mcguirk@indirect.com>
9341 * Dr_George_D_Detlefsen <drgeorge@ilt.com>
9342 * Erich Stamberger <eberger@gewi.kfunigraz.ac.at>
9343 * Massimo Dal Zotto <dz@cs.unitn.it>
9344 * Randy Kunkee <kunkee@Starbase.NeoSoft.COM>
9345 * Rick Weldon <rick@wisetech.com>
9346 * Thomas van Reimersdahl <reimersd@dali.techinfo.rwth-aachen.de>
9347 * david bennett <dave@bensoft.com>
9348 * ernst.molitor@uni-bonn.de
9349 * Julian Assange <proff@suburbia.net>
9350 * Bruce Momjian <pgman@candle.pha.pa.us>
9351 * Paul "Shag" Walmsley <ccshag@cclabs.missouri.edu>
9352 * "Alistair G. Crooks" <azcb0@sde.uts.amdahl.com>
9357 <sect1 id="release-1-01">
9358 <title>Release 1.01</title>
9361 <title>Release date</title>
9362 <simpara>1996-02-23</simpara>
9367 <title>Migration from version 1.0 to version 1.01</title>
9370 The following notes are for the benefit of users who want to migrate
9371 databases from <productname>Postgres95</> 1.0 to <productname>Postgres95</> 1.01.
9374 If you are starting afresh with <productname>Postgres95</> 1.01 and do not need
9375 to migrate old databases, you do not need to read any further.
9378 In order to <productname>Postgres95</> version 1.01 with databases created with
9379 <productname>Postgres95</> version 1.0, the following steps are required:
9384 Set the definition of <symbol>NAMEDATALEN</symbol> in <filename>src/Makefile.global</filename> to 16
9385 and <symbol>OIDNAMELEN</symbol> to 20.
9390 Decide whether you want to use Host based authentication.
9395 If you do, you must create a file name <literal>pg_hba</literal> in your top-level data
9396 directory (typically the value of your <envar>$PGDATA</envar>). <filename>src/libpq/pg_hba</filename>
9397 shows an example syntax.
9402 If you do not want host-based authentication, you can comment out
9407 in <filename>src/Makefile.global</filename>
9410 Note that host-based authentication is turned on by default, and if
9411 you do not take steps A or B above, the out-of-the-box 1.01 will
9412 not allow you to connect to 1.0 databases.
9420 Compile and install 1.01, but DO NOT do the <command>initdb</command> step.
9425 Before doing anything else, terminate your 1.0 postmaster, and
9426 backup your existing <envar>$PGDATA</envar> directory.
9431 Set your <envar>PGDATA</envar> environment variable to your 1.0 databases, but set up
9432 path up so that 1.01 binaries are being used.
9437 Modify the file <filename><envar>$PGDATA</envar>/PG_VERSION</filename> from 5.0 to 5.1
9442 Start up a new 1.01 postmaster
9447 Add the new built-in functions and operators of 1.01 to 1.0
9448 databases. This is done by running the new 1.01 server against
9449 your own 1.0 database and applying the queries attached and saving
9450 in the file 1.0_to_1.01.sql. This can be done easily through <command>psql</command>.
9451 If your 1.0 database is name <literal>testdb</literal>:
9454 % psql testdb -f 1.0_to_1.01.sql
9457 and then execute the following commands (cut and paste from here):
9460 -- add builtin functions that are new to 1.01
9462 create function int4eqoid (int4, oid) returns bool as 'foo'
9463 language 'internal';
9464 create function oideqint4 (oid, int4) returns bool as 'foo'
9465 language 'internal';
9466 create function char2icregexeq (char2, text) returns bool as 'foo'
9467 language 'internal';
9468 create function char2icregexne (char2, text) returns bool as 'foo'
9469 language 'internal';
9470 create function char4icregexeq (char4, text) returns bool as 'foo'
9471 language 'internal';
9472 create function char4icregexne (char4, text) returns bool as 'foo'
9473 language 'internal';
9474 create function char8icregexeq (char8, text) returns bool as 'foo'
9475 language 'internal';
9476 create function char8icregexne (char8, text) returns bool as 'foo'
9477 language 'internal';
9478 create function char16icregexeq (char16, text) returns bool as 'foo'
9479 language 'internal';
9480 create function char16icregexne (char16, text) returns bool as 'foo'
9481 language 'internal';
9482 create function texticregexeq (text, text) returns bool as 'foo'
9483 language 'internal';
9484 create function texticregexne (text, text) returns bool as 'foo'
9485 language 'internal';
9487 -- add builtin functions that are new to 1.01
9489 create operator = (leftarg = int4, rightarg = oid, procedure = int4eqoid);
9490 create operator = (leftarg = oid, rightarg = int4, procedure = oideqint4);
9491 create operator ~* (leftarg = char2, rightarg = text, procedure = char2icregexeq);
9492 create operator !~* (leftarg = char2, rightarg = text, procedure = char2icregexne);
9493 create operator ~* (leftarg = char4, rightarg = text, procedure = char4icregexeq);
9494 create operator !~* (leftarg = char4, rightarg = text, procedure = char4icregexne);
9495 create operator ~* (leftarg = char8, rightarg = text, procedure = char8icregexeq);
9496 create operator !~* (leftarg = char8, rightarg = text, procedure = char8icregexne);
9497 create operator ~* (leftarg = char16, rightarg = text, procedure = char16icregexeq);
9498 create operator !~* (leftarg = char16, rightarg = text, procedure = char16icregexne);
9499 create operator ~* (leftarg = text, rightarg = text, procedure = texticregexeq);
9500 create operator !~* (leftarg = text, rightarg = text, procedure = texticregexne);
9508 <title>Changes</title>
9513 * 1.01 is backwards compatible with 1.0 database provided the user
9514 follow the steps outlined in the MIGRATION_from_1.0_to_1.01 file.
9515 If those steps are not taken, 1.01 is not compatible with 1.0 database.
9518 * added PQdisplayTuples() to libpq and changed monitor and psql to use it
9519 * added NeXT port (requires SysVIPC implementation)
9520 * added CAST .. AS ... syntax
9521 * added ASC and DESC key words
9522 * added 'internal' as a possible language for CREATE FUNCTION
9523 internal functions are C functions which have been statically linked
9524 into the postgres backend.
9525 * a new type "name" has been added for system identifiers (table names,
9526 attribute names, etc.) This replaces the old char16 type. The
9527 of name is set by the NAMEDATALEN #define in src/Makefile.global
9528 * a readable reference manual that describes the query language.
9529 * added host-based access control. A configuration file ($PGDATA/pg_hba)
9530 is used to hold the configuration data. If host-based access control
9531 is not desired, comment out HBA=1 in src/Makefile.global.
9532 * changed regex handling to be uniform use of Henry Spencer's regex code
9533 regardless of platform. The regex code is included in the distribution
9534 * added functions and operators for case-insensitive regular expressions.
9535 The operators are ~* and !~*.
9536 * pg_dump uses COPY instead of SELECT loop for better performance
9539 * fixed an optimizer bug that was causing core dumps when
9540 functions calls were used in comparisons in the WHERE clause
9541 * changed all uses of getuid to geteuid so that effective uids are used
9542 * psql now returns non-zero status on errors when using -c
9543 * applied public patches 1-14
9549 <sect1 id="release-1-0">
9550 <title>Release 1.0</title>
9553 <title>Release date</title>
9554 <simpara>1995-09-05</simpara>
9558 <title>Changes</title>
9563 * The copyright of <productname>Postgres</productname> 1.0 has been loosened to be freely modifiable
9564 and modifiable for any purpose. Please read the COPYRIGHT file.
9565 Thanks to Professor Michael Stonebraker for making this possible.
9568 * date formats have to be MM-DD-YYYY (or DD-MM-YYYY if you're using
9569 EUROPEAN STYLE). This follows SQL-92 specs.
9570 * "delimiters" is now a key word
9573 * sql LIKE syntax has been added
9574 * copy command now takes an optional USING DELIMITER specification.
9575 delimiters can be any single-character string.
9576 * IRIX 5.3 port has been added.
9577 Thanks to Paul Walmsley and others.
9578 * updated pg_dump to work with new libpq
9579 * \d has been added psql
9580 Thanks to Keith Parks
9581 * regexp performance for architectures that use POSIX regex has been
9582 improved due to caching of precompiled patterns.
9583 Thanks to Alistair Crooks
9584 * a new version of libpq++
9585 Thanks to William Wanders
9588 * arbitrary userids can be specified in the createuser script
9589 * \c to connect to other databases in psql now works.
9590 * bad pg_proc entry for float4inc() is fixed
9591 * users with usecreatedb field set can now create databases without
9592 having to be usesuper
9593 * remove access control entries when the entry no longer has any
9595 * fixed non-portable datetimes implementation
9596 * added kerberos flags to the src/backend/Makefile
9597 * libpq now works with kerberos
9598 * typographic errors in the user manual have been corrected.
9599 * btrees with multiple index never worked, now we tell you they don't
9600 work when you try to use them
9606 <sect1 id="release-0-03">
9607 <title><productname>Postgres95</productname> Release 0.03</title>
9610 <title>Release date</title>
9611 <simpara>1995-07-21</simpara>
9615 <title>Changes</title>
9618 Incompatible changes:
9619 * BETA-0.3 IS INCOMPATIBLE WITH DATABASES CREATED WITH PREVIOUS VERSIONS
9620 (due to system catalog changes and indexing structure changes).
9621 * double-quote (") is deprecated as a quoting character for string literals;
9622 you need to convert them to single quotes ('). <!-- " -->
9623 * name of aggregates (eg. int4sum) are renamed in accordance with the
9624 SQL standard (eg. sum).
9625 * CHANGE ACL syntax is replaced by GRANT/REVOKE syntax.
9626 * float literals (eg. 3.14) are now of type float4 (instead of float8 in
9627 previous releases); you might have to do typecasting if you depend on it
9628 being of type float8. If you neglect to do the typecasting and you assign
9629 a float literal to a field of type float8, you may get incorrect values
9631 * LIBPQ has been totally revamped so that frontend applications
9632 can connect to multiple backends
9633 * the usesysid field in pg_user has been changed from int2 to int4 to
9634 allow wider range of Unix user ids.
9635 * the netbsd/freebsd/bsd o/s ports have been consolidated into a
9636 single BSD44_derived port. (thanks to Alistair Crooks)
9638 SQL standard-compliance (the following details changes that makes postgres95
9639 more compliant to the SQL-92 standard):
9640 * the following SQL types are now built-in: smallint, int(eger), float, real,
9641 char(N), varchar(N), date and time.
9643 The following are aliases to existing postgres types:
9645 integer, int -> int4
9646 float, real -> float4
9647 char(N) and varchar(N) are implemented as truncated text types. In
9648 addition, char(N) does blank-padding.
9649 * single-quote (') is used for quoting string literals; '' (in addition to
9650 \') is supported as means of inserting a single quote in a string
9651 * SQL standard aggregate names (MAX, MIN, AVG, SUM, COUNT) are used
9652 (Also, aggregates can now be overloaded, i.e. you can define your
9653 own MAX aggregate to take in a user-defined type.)
9654 * CHANGE ACL removed. GRANT/REVOKE syntax added.
9655 - Privileges can be given to a group using the "GROUP" key word.
9657 GRANT SELECT ON foobar TO GROUP my_group;
9658 The key word 'PUBLIC' is also supported to mean all users.
9660 Privileges can only be granted or revoked to one user or group
9663 "WITH GRANT OPTION" is not supported. Only class owners can change
9665 - The default access control is to to grant users readonly access.
9666 You must explicitly grant insert/update access to users. To change
9667 this, modify the line in
9668 src/backend/utils/acl.h
9669 that defines ACL_WORLD_DEFAULT
9672 * the bug where aggregates of empty tables were not run has been fixed. Now,
9673 aggregates run on empty tables will return the initial conditions of the
9674 aggregates. Thus, COUNT of an empty table will now properly return 0.
9675 MAX/MIN of an empty table will return a row of value NULL.
9676 * allow the use of \; inside the monitor
9677 * the LISTEN/NOTIFY asynchronous notification mechanism now work
9678 * NOTIFY in rule action bodies now work
9679 * hash indexes work, and access methods in general should perform better.
9680 creation of large btree indexes should be much faster. (thanks to Paul
9683 Other changes and enhancements:
9684 * addition of an EXPLAIN statement used for explaining the query execution
9685 plan (eg. "EXPLAIN SELECT * FROM EMP" prints out the execution plan for
9687 * WARN and NOTICE messages no longer have timestamps on them. To turn on
9688 timestamps of error messages, uncomment the line in
9689 src/backend/utils/elog.h:
9690 /* define ELOG_TIMESTAMPS */
9691 * On an access control violation, the message
9692 "Either no such class or insufficient privilege"
9693 will be given. This is the same message that is returned when
9694 a class is not found. This dissuades non-privileged users from
9695 guessing the existence of privileged classes.
9696 * some additional system catalog changes have been made that are not
9697 visible to the user.
9700 * The -oid option has been added to the "pg_result" tcl command.
9701 pg_result -oid returns oid of the last row inserted. If the
9702 last command was not an INSERT, then pg_result -oid returns "".
9703 * the large object interface is available as pg_lo* tcl commands:
9704 pg_lo_open, pg_lo_close, pg_lo_creat, etc.
9706 Portability enhancements and New Ports:
9707 * flex/lex problems have been cleared up. Now, you should be able to use
9708 flex instead of lex on any platforms. We no longer make assumptions of
9709 what lexer you use based on the platform you use.
9710 * The Linux-ELF port is now supported. Various configuration have been
9711 tested: The following configuration is known to work:
9712 kernel 1.2.10, gcc 2.6.3, libc 4.7.2, flex 2.5.2, bison 1.24
9713 with everything in ELF format,
9716 * ipcclean added to the distribution
9717 ipcclean usually does not need to be run, but if your backend crashes
9718 and leaves shared memory segments hanging around, ipcclean will
9719 clean them up for you.
9722 * the user manual has been revised and libpq documentation added.
9728 <sect1 id="release-0-02">
9729 <title><productname>Postgres95</productname> Release 0.02</title>
9732 <title>Release date</title>
9733 <simpara>1995-05-25</simpara>
9737 <title>Changes</title>
9741 Incompatible changes:
9742 * The SQL statement for creating a database is 'CREATE DATABASE' instead
9743 of 'CREATEDB'. Similarly, dropping a database is 'DROP DATABASE' instead
9744 of 'DESTROYDB'. However, the names of the executables 'createdb' and
9745 'destroydb' remain the same.
9748 * pgperl - a Perl (4.036) interface to Postgres95
9749 * pg_dump - a utility for dumping out a postgres database into a
9750 script file containing query commands. The script files are in a ASCII
9751 format and can be used to reconstruct the database, even on other
9752 machines and other architectures. (Also good for converting
9753 a Postgres 4.2 database to Postgres95 database.)
9755 The following ports have been incorporated into postgres95-beta-0.02:
9756 * the NetBSD port by Alistair Crooks
9757 * the AIX port by Mike Tung
9758 * the Windows NT port by Jon Forrest (more stuff but not done yet)
9759 * the Linux ELF port by Brian Gallew
9761 The following bugs have been fixed in postgres95-beta-0.02:
9762 * new lines not escaped in COPY OUT and problem with COPY OUT when first
9764 * cannot type return to use the default user id in createuser
9765 * SELECT DISTINCT on big tables crashes
9766 * Linux installation problems
9767 * monitor doesn't allow use of 'localhost' as PGHOST
9768 * psql core dumps when doing \c or \l
9769 * the "pgtclsh" target missing from src/bin/pgtclsh/Makefile
9770 * libpgtcl has a hard-wired default port number
9771 * SELECT DISTINCT INTO TABLE hangs
9772 * CREATE TYPE doesn't accept 'variable' as the internallength
9773 * wrong result using more than 1 aggregate in a SELECT
9779 <sect1 id="release-0-01">
9780 <title><productname>Postgres95</productname> Release 0.01</title>
9783 <title>Release date</title>
9784 <simpara>1995-05-01</simpara>
9793 <sect1 id="timing-results">
9794 <title>Timing Results</title>
9797 These timing results are from running the regression test with the commands
9800 % cd src/test/regress
9806 Timing under Linux 2.0.27 seems to have a roughly 5% variation from run
9807 to run, presumably due to the scheduling vagaries of multitasking systems.
9811 <title>Version 6.5</title>
9814 As has been the case for previous releases, timing between
9815 releases is not directly comparable since new regression tests
9816 have been added. In general, 6.5 is faster than previous
9821 Timing with <function>fsync()</function> disabled:
9825 02:00 Dual Pentium Pro 180, 224MB, UW-SCSI, Linux 2.0.36, gcc 2.7.2.3 -O2 -m486
9826 04:38 Sparc Ultra 1 143MHz, 64MB, Solaris 2.6
9831 Timing with <function>fsync()</function> enabled:
9835 04:21 Dual Pentium Pro 180, 224MB, UW-SCSI, Linux 2.0.36, gcc 2.7.2.3 -O2 -m486
9838 For the <systemitem class="osname">Linux</systemitem> system above, using <acronym>UW-SCSI</acronym> disks rather than (older) <acronym>IDE</acronym>
9839 disks leads to a 50% improvement in speed on the regression test.
9844 <title>Version 6.4beta</title>
9847 The times for this release are not directly comparable to those for previous releases
9848 since some additional regression tests have been included.
9849 In general, however, 6.4 should be slightly faster than the previous release (thanks, Bruce!).
9854 02:26 Dual Pentium Pro 180, 96MB, UW-SCSI, Linux 2.0.30, gcc 2.7.2.1 -O2 -m486
9860 <title>Version 6.3</title>
9863 The times for this release are not directly comparable to those for previous releases
9864 since some additional regression tests have been included and some obsolete tests involving
9865 time travel have been removed.
9866 In general, however, 6.3 is substantially faster than previous releases (thanks, Bruce!).
9871 02:30 Dual Pentium Pro 180, 96MB, UW-SCSI, Linux 2.0.30, gcc 2.7.2.1 -O2 -m486
9872 04:12 Dual Pentium Pro 180, 96MB, EIDE, Linux 2.0.30, gcc 2.7.2.1 -O2 -m486
9878 <title>Version 6.1</title>
9883 06:12 Pentium Pro 180, 32MB, EIDE, Linux 2.0.30, gcc 2.7.2 -O2 -m486
9884 12:06 P-100, 48MB, Linux 2.0.29, gcc
9885 39:58 Sparc IPC 32MB, Solaris 2.5, gcc 2.7.2.1 -O -g
9893 <!-- Keep this comment at the end of the file
9898 sgml-minimize-attributes:nil
9899 sgml-always-quote-attributes:t
9902 sgml-parent-document:nil
9903 sgml-default-dtd-file:"./reference.ced"
9904 sgml-exposed-tags:nil
9905 sgml-local-catalogs:("/usr/lib/sgml/catalog")
9906 sgml-local-ecat-files:nil
OSDN Git Service
