Movatterモバイル変換

E.10. Release 17#

• Add newCOPY optionON_ERROR ignore to discard error rows (Damir Belyalov, Atsushi Torikoshi, Alex Shulgin, Jian He, Yugo Nagata)§§§§

  The default behavior isON_ERROR stop.

• Add newCOPY optionLOG_VERBOSITY which reportsCOPY FROM ignored error rows (Bharath Rupireddy)§

• AllowCOPY FROM to report the number of skipped rows during processing (Atsushi Torikoshi)§

  This appears in system view columnpg_stat_progress_copy.tuples_skipped.

• InCOPY FROM, allow easy specification that all columns should be forced null or not null (Zhang Mingli)§

• Allow partitioned tables to have identity columns (Ashutosh Bapat)§

• Allowexclusion constraints on partitioned tables (Paul A. Jungwirth)§

  As long as exclusion constraints compare partition key columns for equality, other columns can use exclusion constraint-specific comparisons.

• Add clearerALTER TABLE method to set a column to the default statistics target (Peter Eisentraut)§

  The new syntax isALTER TABLE ... SET STATISTICS DEFAULT; usingSET STATISTICS -1 is still supported.

• AllowALTER TABLE to change a column's generation expression (Amul Sul)§

  The syntax isALTER TABLE ... ALTER COLUMN ... SET EXPRESSION.

• Allow specification oftable access methods on partitioned tables (Justin Pryzby, Soumyadeep Chakraborty, Michael Paquier)§§

• AddDEFAULT setting forALTER TABLE .. SET ACCESS METHOD (Michael Paquier)§

• Add support forevent triggers that fire at connection time (Konstantin Knizhnik, Mikhail Gribkov)§

• Add event trigger support forREINDEX (Garrett Thornburg, Jian He)§

• Allow parenthesized syntax forCLUSTER options if a table name is not specified (Nathan Bossart)§

• Allow theinterval data type to support+/-infinity values (Joseph Koshakow, Jian He, Ashutosh Bapat)§

• Allow the use of anENUM added viaALTER TYPE if the type was created in the same transaction (Tom Lane)§

  This was previously disallowed.

• AllowMERGE to modify updatable views (Dean Rasheed)§

• AddWHEN NOT MATCHED BY SOURCE toMERGE (Dean Rasheed)§

  WHEN NOT MATCHED on target rows was already supported.

• AllowMERGE to use theRETURNING clause (Dean Rasheed)§

  The newRETURNING functionmerge_action() reports on theDML that generated the row.

• Add functionJSON_TABLE() to convertJSON data to a table representation (Nikita Glukhov, Teodor Sigaev, Oleg Bartunov, Alexander Korotkov, Andrew Dunstan, Amit Langote, Jian He)§§

  This function can be used in theFROM clause ofSELECT queries as a tuple source.

• AddSQL/JSON constructor functionsJSON(),JSON_SCALAR(), andJSON_SERIALIZE() (Nikita Glukhov, Teodor Sigaev, Oleg Bartunov, Alexander Korotkov, Andrew Dunstan, Amit Langote)§

• AddSQL/JSON query functionsJSON_EXISTS(),JSON_QUERY(), andJSON_VALUE() (Nikita Glukhov, Teodor Sigaev, Oleg Bartunov, Alexander Korotkov, Andrew Dunstan, Amit Langote, Peter Eisentraut, Jian He)§§§§§

• Addjsonpath methods to convertJSON values to otherJSON data types (Jeevan Chalke)§

  The jsonpath methods are.bigint(),.boolean(),.date(),.decimal([precision [, scale]]),.integer(),.number(),.string(),.time(),.time_tz(),.timestamp(), and.timestamp_tz().

• Addto_timestamp() time zone format specifiers (Tom Lane)§

  TZ accepts time zone abbreviations or numeric offsets, whileOF accepts only numeric offsets.

• Allow the sessiontime zone to be specified byAT LOCAL (Vik Fearing)§

  This is useful when converting adding and removing time zones from time stamps values, rather than specifying the literal session time zone.

• Add functionsuuid_extract_timestamp() anduuid_extract_version() to returnUUID information (Andrey Borodin)§

• Add functions to generate random numbers in a specified range (Dean Rasheed)§

  The functions arerandom(min, max) and they take values of typeinteger,bigint, andnumeric.

• Add functions to convert integers to binary and octal strings (Eric Radman, Nathan Bossart)§

  The functions areto_bin() andto_oct().

• Add Unicode informational functions (Jeff Davis)§

  Functionunicode_version() returns the Unicode version,icu_unicode_version() returns theICU version, andunicode_assigned() returns if the characters are assigned Unicode codepoints.

• Add functionxmltext() to convert text to a singleXML text node (Jim Jones)§

• Add functionto_regtypemod() to return the type modifier of a type specification (David Wheeler, Erik Wienhold)§

• Addpg_basetype() function to return a domain's base type (Steve Chavez)§

• Add functionpg_column_toast_chunk_id() to return a value'sTOAST identifier (Yugo Nagata)§

  This returnsNULL if the value is not stored inTOAST.

• Allow plpgsql%TYPE and%ROWTYPE specifications to represent arrays of non-array types (Quan Zongliang, Pavel Stehule)§

• Allow plpgsql%TYPE specification to reference composite column (Tom Lane)§

• Add libpq function to change role passwords (Joe Conway)§

  The new function,PQchangePassword(), hashes the new password before sending it to the server.

• Add libpq functions to close portals and prepared statements (Jelte Fennema-Nio)§

  The functions arePQclosePrepared(),PQclosePortal(),PQsendClosePrepared(), andPQsendClosePortal().

• Add libpqAPI which allows for blocking and non-blockingcancel requests, with encryption if already in use (Jelte Fennema-Nio)§

  Previously only blocking, unencrypted cancel requests were supported.

• Add libpq functionPQsocketPoll() to allow polling of network sockets (Tristan Partin, Tom Lane)§§

• Add libpq functionPQsendPipelineSync() to send a pipeline synchronization point (Anton Kirilov)§

  This is similar toPQpipelineSync() but it does not flush to the server unless the size threshold of the output buffer is reached.

• Add libpq functionPQsetChunkedRowsMode() to allow retrieval of results in chunks (Daniel Vérité)§

• AllowTLS connections without requiring a network round-trip negotiation (Greg Stark, Heikki Linnakangas, Peter Eisentraut, Michael Paquier, Daniel Gustafsson)§§§§§§§§

  This is enabled with the client-side optionsslnegotiation=direct, requiresALPN, and only works onPostgreSQL 17 and later servers.

• Improvepsql display of default and empty privileges (Erik Wienhold, Laurenz Albe)§

  Command\dp now displays(none) for empty privileges; default still displays as empty.

• Have backslash commands honor\pset null (Erik Wienhold, Laurenz Albe)§

  Previously\pset null was ignored.

• Allowpsql's\watch to stop after a minimum number of rows returned (Greg Sabino Mullane)§

  The parameter ismin_rows.

• Allowpsql connection attempts to be canceled with control-C (Tristan Partin)§

• Allowpsql to honorFETCH_COUNT for non-SELECT queries (Daniel Vérité)§

• Improvepsql tab completion (Dagfinn Ilmari Mannsåker, Gilles Darold, Christoph Heiss, Steve Chavez, Vignesh C, Pavel Borisov, Jian He)§§§§§§§§

• Add applicationpg_walsummary to dumpWAL summary files (Robert Haas)§

• Allowpg_dump's large objects to be restorable in batches (Tom Lane)§

  This allows the restoration of many large objects to avoid transaction limits and to be restored in parallel.

• Addpg_dump option--exclude-extension (Ayush Vatsa)§

• Allowpg_dump,pg_dumpall, andpg_restore to specify include/exclude objects in a file (Pavel Stehule, Daniel Gustafsson)§

  The option is called--filter.

• Add the--sync-method parameter to several client applications (Justin Pryzby, Nathan Bossart)§

  The applications areinitdb,pg_basebackup,pg_checksums,pg_dump,pg_rewind, andpg_upgrade.

• Addpg_restore option--transaction-size to allow object restores in transaction batches (Tom Lane)§

  This allows the performance benefits of transaction batches without the problems of excessively large transaction blocks.

• Changepgbench debug mode option from-d to--debug (Greg Sabino Mullane)§

  Option-d is now used for the database name, and the new--dbname option can be used as well.

• Add pgbench option--exit-on-abort to exit after any client aborts (Yugo Nagata)§

• Add pgbench command\syncpipeline to allow sending of sync messages (Anthonin Bonnefoy)§

• Allowpg_archivecleanup to remove backup history files (Atsushi Torikoshi)§

  The option is--clean-backup-history.

• Add some long options topg_archivecleanup (Atsushi Torikoshi)§

  The long options are--debug,--dry-run, and--strip-extension.

• Allowpg_basebackup andpg_receivewal to use dbname in their connection specification (Jelte Fennema-Nio)§

  This is useful for connection poolers that are sensitive to the database name.

• Addpg_upgrade option--copy-file-range (Thomas Munro)§

  This is supported onLinux andFreeBSD.

• Allowreindexdb--index to process indexes from different tables in parallel (Maxim Orlov, Svetlana Derevyanko, Alexander Korotkov)§

• Allowreindexdb,vacuumdb, andclusterdb to process objects in all databases matching a pattern (Nathan Bossart)§§§

  The new option--all controls this behavior.

• Remove support forOpenSSL 1.0.1 (Michael Paquier)§

• Allow tests to pass inOpenSSLFIPS mode (Peter Eisentraut)§§

• UseCPU AVX-512 instructions for bit counting (Paul Amonson, Nathan Bossart, Ants Aasma)§§

• RequireLLVM version 10 or later (Thomas Munro)§

• Use nativeCRC instructions on 64-bitLoongArch CPUs (Xudong Yang)§

• RemoveAIX support (Heikki Linnakangas)§

• Remove theMicrosoft Visual Studio-specificPostgreSQL build option (Michael Paquier)§

  Meson is now the only available method forVisual Studio builds.

• Remove configure option--disable-thread-safety (Thomas Munro, Heikki Linnakangas)§§

  We now assume all supported platforms have sufficient thread support.

• Removeconfigure option--with-CC (Heikki Linnakangas)§

  Setting theCC environment variable is now the only supported method for specifying the compiler.

• User-defined data type receive functions will no longer receive their data null-terminated (David Rowley)§

• Add incrementalJSON parser for use with hugeJSON documents (Andrew Dunstan)§

• Convert top-levelREADME file toMarkdown (Nathan Bossart)§

• Remove no longer needed top-levelINSTALL file (Tom Lane)§

• Removemake'sdistprep option (Peter Eisentraut)§

• Addmake support forAndroid shared libraries (Peter Eisentraut)§

• Add backend support for injection points (Michael Paquier)§§§§

  This is used for server debugging and they must be enabled at server compile time.

• Add dynamic shared memory registry (Nathan Bossart)§

  This allows shared libraries which are not initialized at startup to coordinate dynamic shared memory access.

• Fixemit_log_hook to use the same time value as other log records for the same query (Kambam Vinay, Michael Paquier)§

• Improve documentation for usingjsonpath for predicate checks (David Wheeler)§

• Allow joins with non-join qualifications to be pushed down to foreign servers and custom scans (Richard Guo, Etsuro Fujita)§

  Foreign data wrappers and custom scans will need to be modified to handle these cases.

• Allow pushdown ofEXISTS andIN subqueries topostgres_fdw foreign servers (Alexander Pyhalov)§

• Increase the default foreign data wrapper tuple cost (David Rowley, Umair Shahid)§§

  This value is used by the optimizer.

• Allowdblink database operations to be interrupted (Noah Misch)§

• Allow the creation of hash indexes onltree columns (Tommy Pavlicek)§

  This also enables hash join and hash aggregation onltree columns.

• Allowunaccent character translation rules to contain whitespace and quotes (Michael Paquier)§

  The syntax for theunaccent.rules file has changed.

• Allowamcheck to check for unique constraint violations using new option--checkunique (Anastasia Lubennikova, Pavel Borisov, Maxim Orlov)§

• Allowcitext tests to pass in OpenSSLFIPS mode (Peter Eisentraut)§

• Allowpgcrypto tests to pass in OpenSSLFIPS mode (Peter Eisentraut)§

• Remove some unusedSPI macros (Bharath Rupireddy)§

• AllowALTER OPERATOR to set more optimization attributes (Tommy Pavlicek)§

  This is useful for extensions.

• Allow extensions to definecustom wait events (Masahiro Ikeda)§§§§

  Custom wait events have been added topostgres_fdw anddblink.

• Addpg_buffercache functionpg_buffercache_evict() to allow shared buffer eviction (Palak Chaturvedi, Thomas Munro)§

  This is useful for testing.