2010-??-?? Based on overwhelming user demand, this release of PostgreSQL adds features that have been requested for years, like easy-to-use replication, a mass permission facility, and anonymous blocks. While past major releases have been quite conservative in their scope, this release shows a bold new desire to provide facilities that every new and existing user of PostgreSQL will embrace. This has all been done with few incompatibilities. The major areas of enhancement are: Built-in, binary, log-based replication. This advance consists of two features: Hot Standby allows continuous archive standby database servers to accept read-only queries, and Streaming Replication allows continuous archive (WAL) files to be streamed over a network port to a standby database server. Easier database object permissions management. GRANT/REVOKE IN SCHEMA supports mass permissions changes, and the ALTER DEFAULT PRIVILEGES command controls privileges of all newly-created objects. Large object permissions now support GRANT/REVOKE as well. Broadly enhanced stored procedure support. The DO statement permits ad-hoc or anonymous code blocks. Functions can now be called using named parameters. PL/pgSQL is now installed by default, and PL/Perl and PL/Python have been enhanced in several ways, including support for Python3. Triggers now support two new features, SQL-compliant per-column triggers, and conditional trigger execution. Deferrable unique constraints, now permit mass updates to unique keys. Exclusion constraints let database designers define uniqueness based on complex criteria, including for non-scalar data such as time periods, ranges and arrays. The LISTEN/NOTIFY feature has been overhauled to make it into a high-performance event queuing system. It now stores events in a memory-based queue, and it now allows delivery of a string payload to listeners with each event. As part of our decade-long effort to eliminate the pain of VACUUM, VACUUM FULL is now substantially faster by rewriting the entire table and indexes, rather than moving around single rows to compact space. Add support for compiling on 64-bit Windows and running in 64-bit mode. Add /contrib/pg_upgrade to support in-place upgrades from 8.3 or 8.4 to 9.0. The above items are explained in more detail in the sections below. CURRENT AS OF 2010-07-08 A dump/restore using pg_dump or use of pg_upgrade is required for those wishing to migrate data from any previous release. Version 9.0 contains a number of changes which selectively break backwards compatibility in order to support new features and code quality improvements. Particularly, users who make extensive use of PL/pgSQL and/or PITR and Warm Standby should test their solutions for breakage. Observe the following incompatibilities: Remove server variable add_missing_from, which was defaulted to off for many years (Tom Lane) Remove server variable regex_flavor, which was defaulted to advanced (e.g. Perl-regex compatible) for many years. (Tom Lane) It is now necessary to set wal_level to archive to do continuous archiving. (Heikki Linnakangas) Adjust log_temp_files to use default file sizes units of kilobytes (Robert Haas) When querying a parent table, do not do additional permission checks on child tables returned as part of the query (Peter Eisentraut) The SQL standard specifies this behavior. Have fractional seconds conversion truncate rather than round when using float-based dates/times (Tom Lane) Improve standards compliance of SIMILAR TO patterns and SQL-style substring() patterns (Tom Lane) This includes treating ? and {...} as pattern metacharacters, while they were simple literal characters before; that corresponds to new features added in SQL:2008. Also, ^ and $ are now treated as simple literal characters; formerly they were treated as metacharacters, as if the pattern were following POSIX rather than SQL rules. Also, in SQL-standard substring(), use of parentheses for nesting no longer interferes with capturing of a substring. Also, processing of bracket expressions (character classes) is now more standards-compliant. Reject negative length values in 3-parameter substring() for bit strings, per the SQL standard (Tom Lane) Tighten enforcement of column renaming when a child table inherits the renamed column from an unrelated parent (KaiGai Kohei) No longer rename index names and index column names when table columns are renamed (Tom Lane) Administrators still can rename such columns manually. This change will require an update of the JDBC driver and possibly other drivers so that unique indexes are correctly recognized. CREATE OR REPLACE FUNCTION can no longer change the declared names of function parameters (Pavel Stehule) In order to support named-parameter calls, it is no longer allowed to change the aliases for input variables in the declaration of an existing function. You now have to DROP and recreate the function. PL/pgSQL now throws an error if a variable name conflicts with a column name used in a query (Tom Lane) The former behavior was to bind to variable names in preference to query column names, which often resulted in surprising misbehavior. Throwing an error allows easy detection of ambiguous situations. Although it's recommended that functions encountering this type of error be modified to remove the conflict, the old behavior can be restored if necessary via the configuration parameter plpgsql.variable_conflict, or via the per-function option #variable_conflict. PL/pgSQL no longer allows variable names that match SQL reserved words (Tom Lane) This is a consequence of aligning the PL/pgSQL parser to match the core SQL parser more closely. If necessary, variable names can be double-quoted to avoid this restriction. PL/pgSQL now requires columns of composite results to match the expected type modifier as well as base type (Pavel Stehule, Tom Lane) For example, if a column of the result type is declared as NUMERIC(30,2), it is no longer acceptable to return a NUMERIC of some other precision in that column. Previous versions neglected to check the type modifier and would thus allow result rows that didn't actually conform to the declared restrictions. Remove PL/pgSQL's RENAME declaration option (Tom Lane) Instead of RENAME, use ALIAS, which can now alias any variable, not just dollar sign parameter names (such as $1). Remove support for platforms that don't have a working 64-bit integer data type (Tom Lane) It is believed all still-supported platforms have working 64-bit integer data types. Version 9.0 has an unprecedented number of new major features, and over 200 enhancements, improvements, new commands, new functions, and other changes. PostgreSQL's native standby capability has been expanded both to support read-only queries on standby slaves and to greatly reduce the lag between master and standby servers. For many users, this will be a useful and low-administration form of replication, either for high availability or for horizontal scalability. Allow a standby system to accept read-only queries (Simon Riggs, Heikki Linnakangas) This feature is called Hot Standby. There are new postgresql.conf and recovery.conf settings to enable this feature, as well as extensive documentation. Allow write-ahead log (WAL) files to be streamed to a standby system (Fujii Masao, Heikki Linnakangas) This feature is called Streaming Replication. Previously WAL files could be sent to standby systems only as 16 megabytes files; this allows master changes to be sent to the standby with very little delay. There are new postgresql.conf and recovery.conf settings to enable this feature, as well as extensive documentation. Add pg_last_xlog_receive_location() and pg_last_xlog_replay_location(), which can be used to monitor standby server WAL activity (Simon Riggs, Fujii Masao, Heikki Linnakangas) Version 9.0 also contains several performance and optimizer enhancements to improve specific uses of PostgreSQL and remedy certain poor-performing cases. Improve performance of finding inherited child tables (Tom Lane) Allow per-tablespace sequential and random page cost variables (seq_page_cost/random_page_cost) via ALTER TABLESPACE ... SET/RESET (Robert Haas) Improve performance of TRUNCATE when used in the same transaction as table creation (Tom Lane) Remove unnecessary outer joins (Robert Haas) Outer joins where the inner side is unique and not referenced in the query are unnecessary and are therefore now removed. This will accelerate many automatically generated queries, such as those created by object-relational mappers. Allow IS NOT NULL restrictions to use indexes (Tom Lane) This is particularly useful for finding MAX()/MIN() values in indexes that also contain NULLs. Improve optimizer equivalence detection of <> boolean tests (Tom Lane) Use the same random seed every time GEQO plans a query (Andres Freund) While the Genetic Query Optimizer (GEQO) still selects random plans, it now selects the same random plans for identical queries. You can modify geqo_seed to randomize the starting value of the random plan generator. This gives GEQO query response times and resource usage repeatability and predictability. Improve GEQO plan selection (Tom Lane). This avoids the rare error, "failed to make a valid plan". Improve ANALYZE to support inheritance-tree statistics (Tom Lane) This is particularly useful for partitioned tables. However, autovacuum does not yet properly analyze parent tables based on child table changes. Improve autovacuum detection of when re-analyze is necessary (Tom Lane) Improve optimizer statistics for greater/less-than comparisons (Tom Lane) When looking up optimizer statistics for greater/less-than comparisons, if the comparison value is in the first or last histogram bucket, use an index to generate the actual statistics. This is particularly useful for columns that are always increasing, and hence often have inaccurate statistics. Allow setting of distinct statistics using ALTER TABLE (Robert Haas) This allows user-override of the number or percentage of distinct values for a column and optionally its child tables. This value is normally computed by ANALYZE. Database administrators can use this feature to fix some poor statistics, especially on tables with millions or billions of rows. Version 9.0 further extends PostgreSQL's support for multiple authentication methods, including RADIUS and improved LDAP support. Add support for RADIUS (Remote Authentication Dial In User Service) authentication (Magnus Hagander) Allow LDAP (Lightweight Directory Access Protocol) authentication to operate in "search/bind" mode (Robert Fleming, Magnus) This allows the user to be looked up first, then the system uses the DN (Distinguished Name) returned for that user. Add samehost and samenet designations to pg_hba.conf (Stef Walter) These match the server's IP address and network address respectively. Pass trusted SSL root certificate names to the client so the client can return an appropriate client certificate (Craig Ringer) With increased use of PostgreSQL in high-end production systems, users need increased monitoring. PostgresSQL 9.0 continues to add more ways to monitor PostgreSQL applications. Add the ability for clients to set an application name, which is displayed in pg_stat_activity (Dave Page) This allows DBAs to characterize database traffic and troubleshoot problems by source application. Add an SQL state option (%e) to log_line_prefix (Guillaume Smet) This allows users to compile statistics on errors and messages by type. Write to the Windows event log in UTF16 encoding (Itagaki Takahiro) Now there is true multi-lingual support for PostgreSQL log messages on Windows. Add pg_stat_reset_shared('bgwriter') to reset the cluster-wide shared statistics of the bgwriter (Greg Smith) Add pg_stat_reset_single_table_counters() and pg_stat_reset_single_function_counters() to allow the reseting of statistics counters for individual tables and indexes (Magnus Hagander) Allow setting of configuration variables based on database/role combinations (Alvaro Herrera) Previously only per-database and per-role setting were possible, not combinations. All role and database settings are now stored in the new pg_db_role_setting system table. A new psql \drds command shows these settings. Backwards-compatible system views do not show this information. The primary use of this feature is setting schema search_path. Add boolean variable bonjour, which controls whether a Bonjour-enabled binary advertises itself via Bonjour (Tom Lane) The default is off, meaning it does not advertise. Add boolean variable enable_material, which controls the use of materialize nodes in the optimizer (Robert Haas) The default is on. Log changed parameter values when postgresql.conf is reloaded (Peter Eisentraut) This lets DBAs and security staff to audit when database settings were changed. Add proper permissions for custom variables (Tom Lane) Custom variables can now only be created by super-users, but can be modified by ordinary users if variable permissions allow it. This makes custom variables appropriate for security settings. Previously, any user could create or modify custom variables. Adjust log_temp_files to use default file sizes units of kilobytes (Robert Haas) Previously this setting defaulted to bytes if no units were specified. Do SELECT FOR UPDATE/SHARE processing after applying LIMIT, so the number of rows returned is always predictable (Tom Lane) Previously, concurrent transactions could potentially cause SELECT to return fewer rows than specified by LIMIT. FOR UPDATE can still affect ORDER BY ordering, but this can be corrected by using FOR UPDATE in a subquery. Allow mixing of traditional and SQL-standard LIMIT/OFFSET syntax in the same query (Tom Lane) Increase the supported frame options in window functions (Hitoshi Harada) This allows frames (RANGE or ROWS) to start with CURRENT ROW, and to use the ROWS n PRECEDING/FOLLOWING clause. Have SELECT and CREATE TABLE AS return row counts to the client (Boszormenyi Zoltan) For drivers that support this feature, this saves an entire round-trip to the client, allowing result counts and pagination to be calculated without a second COUNT query. psql does not display these counts. Add Unicode surrogate pair (dual 16-bit) support to U& strings and identifiers (Peter Eisentraut) Allow Unicode escapes in E'...' strings (Marko Kreen) Speed up CREATE DATABASE by deferring flushes to disk (Andres Freund, Greg Stark) Allow comments on columns only of tables, views, and composite types, not other objects like indexes and TOAST tables (Tom Lane) Allow the creation of enumerate types with no labels (Bruce Momjian) This is useful for supporting binary upgrades. Have columns defined with storage type MAIN remain on the main heap page unless it cannot fit (Kevin Grittner) Previously MAIN values were forced to TOAST tables until the row size was one-quarter of the page size. Add ALTER DEFAULT PRIVILEGES command to control privileges of all newly-created objects (Petr Jelinek) This greatly simplifies the assignment of object privileges in a complex database application. Defaults currently support tables views, sequences, and functions. Defaults may be assigned on a per-schema basis or database-wide. Implement IF EXISTS for DROP COLUMN and DROP CONSTRAINT (Andres Freund) Allow ALTER TABLE commands which rewrite tables to skip WAL logging (Itagaki Takahiro) Such operations either complete fully or are rolled back, so WAL archiving can be skipped, unless running in continuous archiving mode. This reduces I/O overhead and improves performance. Add support for copying COMMENTS and STORAGE to the CREATE TABLE ... LIKE INCLUDING command (Itagaki Takahiro) Add support for copying all attributes to the CREATE TABLE LIKE INCLUDING command (Itagaki Takahiro) Add the SQL-standard CREATE TABLE ... OF type command (Peter Eisentraut) This allows the creation of a table to match an existing composite type. Additional constraints and defaults can be specified in the command. Add deferrable unique constraints (Dean Rasheed) This allows UPDATE tab SET col = col + 1 to work on columns that have a unique indexes or are marked as primary keys, but DEFERRABLE INITIALLY DEFERRED must be used to mark the constraint as deferred. Generalize uniqueness constraints by allowing arbitrary operator comparisons, not just equality (Jeff Davis) This is enabled with the CREATE TABLE CONSTRAINT ... EXCLUDE clause. While uniqueness checks could be specified using this syntax, the real value of this feature is in using complex operators that do not have built-in constraints. The primary use of exclusion constraints is to allow defining non-overlapping uniqueness, such as for time periods, arrays or ranges of values. This supports data integrity at the table level for calendaring, time-management, and scientific applications. Improve the constraint violation error message to report the values causing the failure (Itagaki Takahiro) For example, a uniqueness constraint violation now reports Key (x)=(2) already exists. Add the ability to make mass permission changes per schema using the new GRANT/REVOKE IN SCHEMA clause (Petr Jelinek) This simplifies the assignment of object permissions and makes it easier to utilize database roles for application data security. Add the ability to control large object permissions with GRANT/REVOKE (KaiGai Kohei) Have LISTEN/NOTIFY store events in a memory queue, rather than a system table (Joachim Wieland) LISTEN/NOTIFY may now be used as a full-featured, high-performance event queue system for PostgreSQL, with transactional support and guaranteed delivery. Allow NOTIFY to pass an optional string to listeners (Joachim Wieland) Allow CLUSTER on all system tables (Tom Lane) Global system tables still cannot be clustered. Allow * as a parameter in FORCE QUOTE for COPY CSV (Itagaki Takahiro) This forces quotes for all CSV output columns. Add new COPY syntax that allows parameters to be specified in parentheses (Robert Haas, Emmanuel Cecchet) This allows greater flexibility for future COPY options. The old syntax is still supported. Allow EXPLAIN output in XML, JSON, and YAML formats (Robert Haas, Greg Sabino Mullane) The new output formats will support the development of new tools for analysis of EXPLAIN output. Add new EXPLAIN (BUFFERS) to report query buffer activity (Itagaki Takahiro) This allows better query profiling for individual queries. log_*_stats log output, e.g. log_statement_stats, no longer shows this information. Add hash usage information to EXPLAIN output (Robert Haas) Allow EXPLAIN options to be specified inside parentheses (Robert Haas) This allows for the expansion of EXPLAIN options. The old syntax is still supported. Change VACUUM FULL to rewrite the entire table and indexes, rather than moving around single rows to compact space (Itagaki Takahiro, Tom Lane) The previous method was usually slower and caused index bloat. Note that the new method may use more disk space during VACUUM FULL. Add new VACUUM syntax that allows parameters to be specified in parentheses (Itagaki Takahiro) This allows greater flexibility for future VACUUM options. The old syntax is still supported. Allow an index to be auto-named by not supplying an index name to CREATE INDEX (Tom Lane) Allow REINDEX on system indexes (Tom Lane) WAS THIS POSSIBLE ON ANY SYSTEM TABLE BEFORE? NON-HARDWIRED ONES? Add point_ops opclass for GiST (Teodor Sigaev) This feature supports GiST indexing of point operations on polygons, circles, and other points, such as "point is in polygon". Previously indexing only worked for bounding boxes. This should make many PostGIS queries faster. Use red-black trees for GIN index creation (Teodor Sigaev) Red-black trees are self-balanced. This means much faster GIN index creation. Allow hex values to be specified in bytea strings (Peter Eisentraut) The variable bytea_output controls if hex (default) or octal escapes are used for bytea output. (SWITCH DEFAULT FOR BETA? PETER) Libpq's PQescapeByteaConn() now uses the hex format for PostgreSQL 9.0 servers. The new hex format will be directly compatible with more applications which use binary data, allowing them to store and retrieve it without conversion. Allow extra_float_digits to be increased to 3 (Tom Lane) The previous maximum extra_float_digits was 2. Add prefix support for the full text search synonym dictionary (Teodor Sigaev) Add full text search filtering dictionaries (Teodor Sigaev) Filtering dictionaries allow tokens to be modified and passed to subsequent dictionaries. Allow underscores in full text email addresses (Teodor Sigaev) Use more standards-compliant rules for URL parsing (Tom Lane) Allow case-insensitive regular expression matching with UTF-8 server encodings. Previously, only ASCII characters and single-byte encodings worked properly. Other multi-byte, non-UTF-8 encodings are still broken for case-insensitive regular expression matching. Allow function calls to supply parameter names and match them to named parameters in the function definition (Pavel Stehule) For example, if a function is defined to take parameters a and b, it can be called with func(a := 7, b := 12) or func(b := 12, a := 7). Add support for to_char() scientific notation output ('EEEE') (Pavel Stehule, Brendan Jurd) Have to_char() honor 'FM' (fill mode) in 'Y', 'YY', and 'YYY' specifications (Bruce Momjian, Tom Lane) It was already honored by 'YYYY'. Fix to_char() to output the proper localized numeric and monetary characters on Windows (Hiroshi Inoue, Itagaki Takahir, Bruce Momjian) Correct calculations of "overlap" and "contains" operations over polygons (Teodor Sigaev) Allow aggregate functions to use ORDER BY (Andrew Gierth) For example, this is now supported, array_agg(a ORDER BY b). This is useful for aggregates where the order of values is significant, and eliminates the need to have a subquery for the ordering. Add the string_agg() aggregate function which aggregates values into a single string (Pavel Stehule) An optional second argument allows specification of a delimiter. Aggregate functions that are called with DISTINCT are now passed NULL values if the aggregate transition function is not marked as STRICT (Andrew Gierth) For example, agg(DISTINCT x) might pass NULL x values to agg(). Add get_bit() and set_bit() functions for bit strings, mirroring those for bytea (Leonardo F) Implement OVERLAY() (replace) for bit strings and bytea (Leonardo F) Add pg_table_size() and pg_indexes_size() to provide a more user-friendly interface to the pg_relation_size() function (Bernd Helmle) Add has_sequence_privilege() for sequence permission checking (Abhijit Menon-Sen) Have information schema properly display date type octet lengths (Peter Eisentraut) The reported length is now the maximum octet length; previously, a huge value was reported. Speed up information schema privilege views (Joachim Wieland) Implement anonymous functions using the DO statement (Petr Jelinek, Joshua Tolley, Hannu Valtonen) This allows execution of server-side code without the need to create a new function and execute it. Allow SQL-compliant per-column triggers (Itagaki Takahiro) Such triggers are fired only if the specified columns are affected by the query, e.g. in UPDATE's SET list. information_schema now also shows this information. Add WHEN clause to CREATE TRIGGER to allow control over whether a trigger is fired (Takahiro Itagaki) While a check can be performed inside the trigger, doing it in an external WHEN clause has performance benefits. Add the OR REPLACE clause to CREATE LANGUAGE (Tom Lane) This is helpful to optionally install a language if it does not already exist, and is particularly helpful now that PL/pgSQL is installed by default. Install PL/pgSQL by default (Bruce Momjian) Improve PL/pgSQL's ability to handle row types with dropped columns (Pavel Stehule) Allow input parameters to be assigned values within PL/pgSQL functions (Steve Prentice) Formerly, input parameters were treated as being declared CONST. This restriction has been removed to simplify porting of functions from other DBMSes that do not impose the equivalent restriction. An input parameter now acts like a local variable initialized to the passed-in value. Improve error location reporting in PL/pgSQL (Tom Lane) Make PL/pgSQL use the main lexer, rather than its own version (Tom Lane) This ensures accurate tracking of the main system's behavior for details such as string escaping. Add count and ALL options to MOVE FORWARD/BACKWARD in PL/pgSQL (Pavel Stehule) Allow PL/pgSQL's WHERE CURRENT OF to use a cursor variable (Tom Lane) Allow PL/pgSQL's OPEN cursor FOR EXECUTE to use parameters (Pavel Stehule, Itagaki Takahiro) This is accomplished with a new USING clause. Add new PL/Perl functions: quote_literal(), quote_nullable(), quote_ident(), encode_bytea(), decode_bytea(), looks_like_number(), encode_array_literal(), encode_array_constructor() (Tim Bunce) Add server variable plperl.on_init to specify a PL/Perl Perl initialization function (Tim Bunce) plperl.on_plperl_init and plperl.on_plperlu_init are also available for trusted/untrusted-specific initialization. Improve error context support in PL/Perl (Alexey Klyukin) Support END blocks in PL/Perl (Tim Bunce) END blocks do not currently allow database access. Allow use strict in PL/Perl (Tim Bunce) This can be enabled with the server variable plperl.use_strict. Allow require in PL/Perl (Tim Bunce) This basically tests to see if the module is loaded, and if not, generates an error. Allow use feature in PL/Perl if Perl version 5.10 or later is used (Tim Bunce) Verify that PL/Perl return values are valid in the server encoding (Andrew Dunstan) Improve PL/Perl code structure (Tim Bunce) Add Unicode support in PL/Python (Peter Eisentraut) Strings are automatically converted from/to the server encoding as necessary. Improve bytea support in PL/Python (Caleb Welton) Bytea values passed into PL/Python now are represented as binary, rather than the Postgres bytea text format. Null bytes are now also output properly from PL/Python. Boolean and numeric value passing in PL/Python was also improved. Add array parameter/return support to PL/Python (Peter Eisentraut) Improve mapping of domains to Python base types in PL/Python (Peter Eisentraut) Add Python 3 support to PL/Python (Peter Eisentraut) The new server-side language is called plpython3u. This cannot be used in the same backend with the usual Python 2 server-side language. Improve error location and exception reporting in PL/Python (Peter Eisentraut) Add vacuumdb --analyze-only option to only analyze (Bruce Momjian) Properly escape psql variables and identifiers (Pavel Stehule, Robert Haas) For example, :'var' will be escaped as a literal string, and :"var" will be escaped as an SQL identifier. Ignore leading UTF-8-encoded Unicode byte-order marker in psql (Itagaki Takahiro) This is enabled when the client encoding is UTF-8. Fix psql --file - to properly honor --single-transaction (Bruce Momjian) Prevent overwriting of psql's command-line history if two psql sessions are run simultaneously (Tom Lane) Improve psql's tab completion support (Itagaki Takahiro) Allow psql to use fancy Unicode line-drawing characters via \pset linestyle unicode (Roger Leigh) Improve display of wrapped columns in psql (Roger Leigh) The previous format is available by using \pset linestyle old-ascii. Have \d show child tables that inherit from the specified parent (Damien Clochard) \d shows only the number of child tables, while \d+ shows the names of all child tables. Show definition of indexes in \d index_name (Khee Chin) The definition is useful for expression indexes. In psql, show the view definition only with \d+, not with \d (Peter Eisentraut) Have pg_dump/pg_restore --clean also remove large objects (Itagaki Takahiro) Fix pg_dump to properly dump large objects if standard_conforming_strings is enabled (Tom Lane) Large objects dumps now use hex format for output. (SWITCH DEFAULT FOR BETA? TOM) Allow pg_dump to dump comments attached to columns of composite types (Taro Minowa (Higepon)) Have pg_dump --verbose output the pg_dump and server versions in text output mode (Jim Cox, Tom Lane) These were already present in custom output mode. Allow pg_ctl to be safely used to start the postmaster at boot-time (Tom Lane) Previously the pg_ctl process could have been mistakenly identified as a running postmaster based on a stale postmaster lock file. Give pg_ctl the ability to initialize the database (like initdb) (Zdenek Kotala) Add new libpq functions PQconnectdbParams() and PQconnectStartParams() (Guillaume Lelarge) These functions are similar to PQconnectdb() and PQconnectStart() except they allow a null-terminated array of connection options, rather than requiring all options to be sent in a single string. Add libpq functions PQescapeLiteral() and PQescapeIdentifier() (Robert Haas) These functions return appropriately quoted and escaped literal strings and identifiers. The caller is not required to pre-allocate the string result, as is required by PQescapeStringConn(). Add checking for a per-user service file (.pg_service.conf), which is checked before the site-wide service file (Peter Eisentraut) The file .pg_service.conf is assumed to be in the user's home directory. Add keepalive settings in libpq (Tollef Fog Heen, Fujii Masao, Robert Haas) Keepalive settings were already supported on the server end of TCP connections. Properly report an error if the specified libpq service cannot be found (Peter Eisentraut) Issue a warning if the .pgpass-retrieved password fails (Bruce Momjian) Load SSL certificate chains (Tom Lane) This improves handling of indirectly-signed SSL client certificates. Add SQLDA (SQL Descriptor Area) support to ecpg (Boszormenyi Zoltan) Add the DESCRIBE [OUTPUT] statement to ecpg (Boszormenyi Zoltan) Add an ecpg function to return the current transaction status (Bernd Helmle) Add the string data type in ecpg Informix-compatibility mode (Zoltan Boszormenyi) Allow ecpg to use new and old variable names without restriction (Michael Meskes) Allow ecpg to use variable names in free()(Michael Meskes) Have ecpg return zero for non-SQL3 data types (Michael Meskes)) Previously it returned the negative of the data type oid. Support long long types on platforms that already have 64-bit longs (Michael Meskes)) Add ecpg out-of-scope cursor support in native mode (Boszormenyi Zoltan) This allows DECLARE to use variables that are not in scope when OPEN is called. This facility already existed in ecpg Informix-compatibility mode. Allow dynamic ecpg cursor names (Boszormenyi Zoltan) Allow ecpg to use noise words FROM and IN in FETCH and MOVE (Zoltan Boszormenyi) Enable client thread safety by default (Bruce Momjian) Thread-safe builds can be disabled with configure --disable-thread-safety. Add a compile-time option to allow the Linux out-of-memory killer to kill backends (Alex Hunsaker, Tom Lane) Now that /proc/self/oom_adj allows disabling of the Linux out-of-memory (OOM) killer for the postmaster and its children, the new compile-time option -DLINUX_OOM_ADJ=0 allows the killer to be enabled for postmaster children. pg_config shows if this flag was used during compilation. Use DocBook XSL stylesheets for man page building (Peter Eisentraut) New Makefile targets world, install-world, and installcheck-world (Andrew Dunstan) These are similar to the existing all, install, and installcheck targets, but they build HTML documentation, build and test /contrib, and test server-side languages and ecpg. Add data and documentation location control to PGXS Makefiles (Mark Cave-Ayland) Restructure the HTML documentation build Makefile rules (Peter Eisentraut) Add a Makefile rule to build documentation as a single text file (Bruce Momjian) Restructure use of LDFLAGS to be more consistent across platforms (Tom Lane) Require Autoconf 2.63 for building from source (Peter Eisentraut) Require Flex 2.5.31 or later to build from source (Tom Lane) Require Perl version 5.8 or greater to build the server from a CVS copy (John Naylor, Andrew Dunstan) Add support for compiling on 64-bit Windows and running in 64-bit mode (Tsutomu Yamada, Magnus) This allows for large shared memory sizes on Windows. Allow server builds using Visual Studio 2008 (Magnus Hagander) Allow multi-processor compilation using Microsoft Visual C (Magnus Hagander) Distribute documentation in a proper directory tree, rather than as tar archive files inside the main distribution tarball (Peter Eisentraut) For example, the HTML documentation is now in doc/src/sgml/html; the manual pages are packaged similarly. Enable the server's lexer to be reentrant (Tom Lane) This was needed for use of the lexer by PL/pgSQL. Improve speed of memory allocation (Tom Lane, Greg Stark) Add system columns to better document the use of indexes for constraint enforcement (Tom Lane) Allow multiple actions to be communicated using the same operating system signal (Fujii Masao) This allows improved backend communication as new features are added. Improve source code test coverage, including /contrib, PL/Python, and PL/Perl (Peter Eisentraut, Andrew Dunstan) Remove the use of flat files for system table bootstrapping (Tom Lane, Alvaro Herrera) This also improves performance when using millions of users and databases. Improve the ability to translate psql strings (Peter Eisentraut) Reduce the length of some file names so file paths are less than 100 characters (Tom Lane) Some decompression programs have problems with long file names. Tighten input requirements for int2 vector input (Caleb Welton) Add a new ERRCODE_INVALID_PASSWORD SQLSTATE error code (Bruce Momjian) Properly remove the few remaining personal source code copyright entries (Bruce Momjian) The personal copyright notices were insignificant but the community occasionally had to answer questions about them. New documentation section about running PostgreSQL in non-durable mode, to improve performance (Bruce Momjian) Use a more modern API for Bonjour (Tom Lane) Bonjour now requires OS X 10.3 or later. Add processor test-and-test lock support for the SuperH architecture (Nobuhiro Iwamatsu) Allow non-GCC compilers to use inline functions if they support them (Kurt Harriman) Remove support for platforms that don't have a working 64-bit integer data types (Tom Lane) It is believed all supported platforms have working 64-bit integer data types. Simplify use of C++ functions in backend code (Kurt Harriman, Peter Eisentraut) While this removes keyword conflicts that previously made C++ usage difficult in backend code, there are still other complexities when using C++ for backend functions. extern "C" { } is still necessary in appropriate places. Add AggCheckCallContext() for use in detecting if a C function is being called as an aggregate (Hitoshi Harada) Require fastgetattr() and heap_getattr() backend macros to use a non-NULL fourth argument (Robert Haas) KEEP? Add parser hooks to access column and parameter references in queries (Tom Lane) Add a hook so loadable modules can control utility commands (Itagaki Takahiro) Allow the calling of parser hooks from SPI and cached plans (Tom Lane) Add /contrib/pg_upgrade to support in-place upgrades (Bruce Momjian) This avoids the requirement of dumping/reloading the database when upgrading to a new major release of PostgreSQL and speeds up offline upgrades by orders of magnitude. It supports upgrades to 9.0 from PostgreSQL 8.3 and 8.4. Add support for preservation of all relfilenodes, for use during binary upgrades (Bruce Momjian) Add support for binary upgrades to preserve pg_type and pg_enum oids (Bruce Momjian) This is needed to allow binary upgrades of user-defined composite types, arrays, and enums (enumerated types). Move tablespace data directories into their own PostgreSQL version-specific subdirectory (Bruce Momjian) This simplifies binary upgrades. Add multi-threaded option (-j) to /contrib/pgbench (Itagaki Takahiro) This allows multiple CPUs to be used for pgbench tests. Add \shell and \setshell meta commands to /contrib/pgbench (Michael Paquier) New features for /contrib/dict_xsyn (Sergey Karpov) The new options are matchorig, matchsynonyms, and keepsynonyms. Add full text dictionary /contrib/unaccent (Teodor Sigaev) This filter dictionary removes accents from letters, which makes full-text searches over multiple languages much easier. Add dblink_get_notify() to /contrib/dblink (Marcus Kempe) This allows async notifications in dblink. Improve /contrib/dblinks handling of dropped columns (Tom Lane) This affects dblink_build_sql_insert() and related functions. Greatly increase /contrib/hstore's length limit and add btree and hash abilities so GROUP BY and DISTINCT operations are possible (Andrew Gierth) New functions and operators were also added. These improvements make HStore a full-functional key-value store embedded in PostgreSQL. Add /contrib/passwordcheck which can check the strength of assigned passwords (Laurenz Albe) The source code of this module should be modified to implement site-specific password policies. Add /contrib/pg_archivecleanup tool (Simon Riggs) This is designed to be used by the archive_cleanup_command setting to remove unnecessary archive files. Add query text to /contrib/auto_explain output (Andrew Dunstan) Add buffer access counters to /contrib/pg_stat_statements (Itagaki Takahiro) Update /contrib/start-scripts/linux to use /proc/self/oom_adj to disable the Linux out-of-memory (OOM) killer (Alex Hunsaker, Tom Lane)