path: root/data/featurematrix.yaml

versions:
  min: 8.1
  max: 18
featurematrix:
  Backend:
  - name: 64-bit large objects
    versions:
      '9.3': 'Yes'
    description: Large objects are now 64-bit and can now be up to 4TB where before
      it was up to 2GB.
  - name: Advisory locks
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/explicit-locking.html#ADVISORY-LOCKS
  - name: Custom background workers
    versions:
      '9.3': 'Yes'
    description: Background worker processes can now be added that can attach to PostgreSQL's
      shared memory.
  - name: Disk based FSM
    versions:
      '8.4': 'Yes'
    description: replaces the shared memory fixed size free space map with an on-disk
      implemenation
  - name: Dynamic Background Workers
    versions:
      '9.4': 'Yes'
    description: Allows background workers to be started and stopped dynamically by
      any user process.
  - name: EXPLAIN (BUFFERS) support
    versions:
      '9.0': 'Yes'
    description: adds the ability to show buffer-usage statistics for a query
  - name: EXPLAIN (MEMORY)
    versions:
      '17': 'Yes'
    description: The `EXPLAIN` option `MEMORY` reports optimizer memory usage .
  - name: EXPLAIN (SERIALIZE) support
    versions:
      '17': 'Yes'
    description: The `EXPLAIN` option `SERIALIZE` reports the cost of converting data
      for network transmission.
  - name: EXPLAIN (WAL) support
    versions:
      '13': 'Yes'
    description: "[EXPLAIN](https://www.postgresql.org/docs/13/sql-explain.html) can\
      \ now track WAL usage information along with [auto_explain](https://www.postgresql.org/docs/13/auto-explain.html),\
      \ [autovacuum](https://www.postgresql.org/docs/13/routine-vacuuming.html#AUTOVACUUM),\
      \ and [pg_stat_statements](https://www.postgresql.org/docs/13/pgstatstatements.html).\r\
      \n\r\nFor more information, please see the documentation on [EXPLAIN](https://www.postgresql.org/docs/13/sql-explain.html)."
  - name: '"jsonlog" logging format'
    versions:
      '15': 'Yes'
    description: PostgreSQL logs can now be output in  [JSON](https://www.postgresql.org/docs/15/runtime-config-logging.html#RUNTIME-CONFIG-LOGGING-JSONLOG)
      using the [`jsonlog`](https://www.postgresql.org/docs/15/runtime-config-logging.html#RUNTIME-CONFIG-LOGGING-JSONLOG)
      parameter when setting [`log_destination`](https://www.postgresql.org/docs/15/runtime-config-logging.html#RUNTIME-CONFIG-LOGGING-WHERE).
      This provides a "structured log" that can be used by other utilities for storage
      and analysis.
  - name: Loadable plugin infrastructure for monitoring the planner
    versions:
      '8.3': 'Yes'
    description: allows plugins to augment or even replace the planner (like for providing
      an index advisor)
  - name: Payload support for LISTEN/NOTIFY
    versions:
      '9.0': 'Yes'
    description: '[`NOTIFY`](https://www.postgresql.org/docs/current/sql-notify.html)
      can send payload data, which can be received by [`LISTEN`](https://www.postgresql.org/docs/current/sql-listen.html)'
  - name: pg_stat_checkpointer system view
    versions:
      '17': 'Yes'
    url: https://www.postgresql.org/docs/17/monitoring-stats.html#MONITORING-PG-STAT-CHECKPOINTER-VIEW
  - name: pg_stat_io - I/O metrics view
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/16/monitoring-stats.html#MONITORING-PG-STAT-IO-VIEW
  - name: pg_wait_events system view
    versions:
      '17': 'Yes'
    url: https://www.postgresql.org/docs/17/view-pg-wait-events.html
  - name: Server statistics in shared memory
    versions:
      '15': 'Yes'
    description: '[Server-level statistics](https://www.postgresql.org/docs/current/monitoring-stats.html)
      are stored in shared memory. This eliminates the separate statistics collector
      process.'
  - name: SQL-standard information schema
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/information-schema.html
  - name: Support for anonymous shared memory
    versions:
      '9.3': 'Yes'
    description: Previously sizing shared memory often required adjusting kernel parameters
      to accommodate the change.  This is no longer necessary.
  - name: XML, JSON and YAML output for EXPLAIN
    versions:
      '9.0': 'Yes'
    description: Adds XML, JSON and YAML output to EXPLAIN to help with automatic
      processing of plans
  Data Types, Functions, & Operators:
  - name: Arrays of compound types
    versions:
      '8.3': 'Yes'
    description: allows the creation of arrays of arbitrary complex compound types
  - name: Array support
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/arrays.html
  - name: ENUM data type
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/static/datatype-enum.html
  - name: GUID/UUID data type
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/static/datatype-uuid.html
  - name: macaddr8 data type
    versions:
      '10': 'Yes'
    url: https://www.postgresql.org/docs/10/static/datatype-net-types.html#datatype-macaddr8
  - name: Multiranges
    versions:
      '14': 'Yes'
    url: https://www.postgresql.org/docs/14/rangetypes.html#RANGETYPES-BUILTIN
  - name: NULLs in Array
    versions:
      '8.2': 'Yes'
    description: array elements can contain NULL
  - name: Phrase search
    versions:
      '9.6': 'Yes'
    description: Fulltext search for phrases and word proximity.
  - name: Range types
    versions:
      '9.2': 'Yes'
    url: https://www.postgresql.org/docs/9.2/static/rangetypes.html
  - name: smallserial type
    versions:
      '9.2': 'Yes'
    description: A serial data type that uses smallint (2 byte integer) as its base
      type, giving it a reduced numeric range, but a smaller data footprint.
  - name: Type modifier support
    versions:
      '8.3': 'Yes'
    description: allows user defined datatypes to specify a type modifier
  - name: UUIDv7
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/functions-uuid.html#FUNC_UUID_GEN_TABLE
  - name: XML data type
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/static/datatype-xml.html
  Indexing & Constraints:
  - name: Block-range (BRIN) indexes
    versions:
      '9.5': 'Yes'
    url: https://www.postgresql.org/docs/current/static/brin-intro.html
  - name: B-tree bottom-up index deletion
    versions:
      '14': 'Yes'
    url: https://www.postgresql.org/docs/14/btree-implementation.html#BTREE-DELETION
  - name: B-tree deduplication
    versions:
      '13': 'Yes'
    url: https://www.postgresql.org/docs/13/btree-implementation.html#BTREE-DEDUPLICATION
  - name: Concurrent GiST indexes
    versions:
      '8.1': 'Yes'
    description: Advanced locking for concurrent modification of GiST indexes
  - name: Covering Indexes for B-trees (INCLUDE)
    versions:
      '11': 'Yes'
    description: "Allows additional columns to be included in as \"non-key\" columns\
      \ in an index, and if requirements are met, can return data from these columns\
      \ in an index-only scan.\r\n\r\nFore more information, please see the documentation\
      \ on [Index-Only Scans & Covering Indexes](https://www.postgresql.org/docs/current/indexes-index-only-scans.html)."
  - name: Covering indexes for GiST (INCLUDE)
    versions:
      '12': 'Yes'
    description: "Allows additional columns to be included in as \"non-key\" columns\
      \ in a GiST index, and if requirements are met, can return data from these columns\
      \ in an index-only scan.\r\n\r\nFore more information, please see the documentation\
      \ on [Index-Only Scans & Covering Indexes](https://www.postgresql.org/docs/current/indexes-index-only-scans.html)."
  - name: Deferrable unique constraints
    versions:
      '9.0': 'Yes'
    description: allows unique constraint checking to be deferred until the end of
      the transaction
  - name: Exclusion constraints
    versions:
      '9.0': 'Yes'
    description: generalizes the concept of uniqueness to support any indexable commutative
      operator, not just equality
  - name: GIN (Generalized Inverted Index) Indexes
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/gin.html
  - name: GIN indexes partial match
    versions:
      '8.4': 'Yes'
    url: https://www.postgresql.org/docs/current/gin-implementation.html#GIN-PARTIAL-MATCH
  - name: GIN Index performance and size improvements
    versions:
      '9.4': 'Yes'
    description: GIN Indexes now take up significantly less space and perform a lot
      faster than pre-9.4 releases.
  - name: GiST (Generalized Search Tree) Indexes
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/gist.html
  - name: Indexes on expressions
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/indexes-expressional.html
  - name: Index-only scans
    versions:
      '9.2': 'Yes'
    description: Also known as "covering indexes", allows data in indexes to satisfy
      queries where all columns in a given query are covered by the index.  This removes
      the requirement of having to visit the table pages, which was required previously
      to get visibility information, but now the visibility map is used instead.
  - name: Index-only scans on GiST
    versions:
      '9.5': 'Yes'
    description: GiST indexes now support index-only scans.
  - name: Index support for IS NULL
    versions:
      '8.3': 'Yes'
    description: allows index usage with IS NULL expressions
  - name: In-memory Bitmap Indexes
    versions:
      '8.1': 'Yes'
    description: enabled the planner to combine multiple indexes using an in-memory
      bitmap
  - name: K-nearest neighbor GiST support
    versions:
      '9.1': 'Yes'
    description: allows nearest-neighbor support to contrib/pg_trgm and contrib/btree_gist
  - name: K-nearest neighbor SP-GiST Support
    versions:
      '12': 'Yes'
    description: Add support for K-nearest neighbor (K-NN) searches on SP-GiST indexes
      when the distance operator `<->` is defined.
  - name: Non-blocking CREATE INDEX
    versions:
      '8.2': 'Yes'
    description: online index creation without locking the table
  - name: Parallel B-tree index scans
    versions:
      '10': 'Yes'
    description: B-tree index pages can be searched by separate parallel workers.
  - name: Parallelized CREATE INDEX for BRIN indexes
    versions:
      '17': 'Yes'
    description: ''
  - name: Parallelized CREATE INDEX for B-tree indexes
    versions:
      '11': 'Yes'
    description: ''
  - name: Parallelized CREATE INDEX for GIN indexes
    versions:
      '18': 'Yes'
  - name: Skip scan on multicolumn B-tree indexes
    versions:
      '18': 'Yes'
    description: "Improve execution time for queries that omit an `=` condition\
      \ on one or more prefix index columns."
  - name: Space-Partitioned GiST (SP-GiST) Indexes
    versions:
      '9.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/spgist.html
  - name: SP-GiST indexes for range types
    versions:
      '9.3': 'Yes'
    description: GiST and SP-GiST indexes can be created for table columns of range
      types.
  - name: UNIQUE NULLS NOT DISTINCT
    versions:
      '15': 'Yes'
    description: By default, `NULL` values are treated as distinct entries. Specifying
      `NULLS NOT DISTINCT` on unique indexes / constraints will cause `NULL` to be
      treated as not distinct, or in other words, equivalently.
  - name: WAL support for hash indexes
    versions:
      '10': 'Yes'
    description: Hash indexes have write-ahead log support, which makes them both
      crash safe and replicable.
  SQL:
  - name: ANY_VALUE aggregate
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/16/functions-aggregate.html#id-1.5.8.27.5.2.4.1.1.1.1
  - name: FETCH FIRST .. WITH TIES
    versions:
      '13': 'Yes'
    url: https://www.postgresql.org/docs/13/sql-select.html#SQL-LIMIT
  - name: GROUPING SETS, CUBE and ROLLUP support
    versions:
      '9.5': 'Yes'
    url: https://www.postgresql.org/docs/current/static/queries-table-expressions.html#QUERIES-GROUPING-SETS
  - name: INSERT/UPDATE/DELETE RETURNING
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-insert.html
  - name: LATERAL clause
    versions:
      '9.3': 'Yes'
    description: SQL-standard LATERAL option for FROM-clause subqueries and function
      calls.
  - name: MERGE
    versions:
      '15': 'Yes'
    url: https://www.postgresql.org/docs/15/sql-merge.html
  - name: MERGE ... RETURNING
    versions:
      '17': 'Yes'
    description: '[MERGE](https://www.postgresql.org/docs/current/sql-merge.html)
      can now use a `RETURNING` clause to return modified rows and what action was
      used to modify them using the `merge_action()` function.'
  - name: Multirow VALUES
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-values.html
  - name: Non-decimal integer literals
    versions:
      '16': 'Yes'
    description: Support for non-decimal integer literals, for example `0xff`, `0o777`,
      and `0b101010`.
  - name: ORDER BY NULLS FIRST/LAST
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-select.html
  - name: range_agg range type aggregation function
    versions:
      '14': 'Yes'
    description: Range types can be aggregated (e.g. in `GROUP BY` clauses) using
      the [`range_agg`](https://www.postgresql.org/docs/15/functions-aggregate.html)
      function.
  - name: Recursive Queries
    versions:
      '8.4': 'Yes'
    url: https://www.postgresql.org/docs/current/queries-with.html
  - name: regexp_count, regexp_instr, regexp_like
    versions:
      '15': 'Yes'
    description: "- `regexp_count`: counts the number of places that match the regular\
      \ expression in a string\r\n- `regexp_instr `: returns the starting of the Nth\
      \ match of a regular expression in a sting\r\n- `regexp_like`: returns true\
      \ if a regular expression has a match in a string.\r\n\r\n[https://www.postgresql.org/docs/15/functions-matching.html#FUNCTIONS-POSIX-REGEXP](https://www.postgresql.org/docs/15/functions-matching.html#FUNCTIONS-POSIX-REGEXP)"
  - name: Return OLD and NEW values from modified rows
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/dml-returning.html
  - name: Row-wise comparison
    versions:
      '8.2': 'Yes'
    description: ''
  - name: SELECT FOR NO KEY UPDATE/SELECT FOR KEY SHARE lock modes
    versions:
      '9.3': 'Yes'
    description: These prevent non-key-field row updates from locking foreign key
      rows.
  - name: SQL standard interval handling
    versions:
      '8.4': 'Yes'
    description: ''
  - name: SYSTEM_USER
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/16/functions-info.html#id-1.5.8.32.3.4.2.2.24.1.1.1
  - name: TABLE statement
    versions:
      '8.4': 'Yes'
    description: "SQL standard shorthand version for `SELECT * FROM`\r\n\r\nFor more\
      \ information, please see the documentation on [`SELECT`](https://www.postgresql.org/docs/current/sql-select.html)"
  - name: Underscores (_) for thousands separators
    versions:
      '16': 'Yes'
    description: Underscores (\_) can be used as thousand separators, for example,
      `5_432_000`
  - name: unnest/array_agg
    versions:
      '8.4': 'Yes'
    description: "`unnest` allows for arrays to be \"unnested\" and expanded into\
      \ a set of rows.\r\n\r\nConversely, `array_agg` can take the output of a set\
      \ of rows and aggregate them into an array.\r\n\r\nFor more information:\r\n\
      \r\n- [Array Functions](https://www.postgresql.org/docs/current/functions-array.html)\r\
      \n- [Aggregation Functions](https://www.postgresql.org/docs/current/functions-aggregate.html)"
  - name: Upsert (INSERT ... ON CONFLICT DO ...)
    versions:
      '9.5': 'Yes'
    description: "Support for upsert functionality which allows either no action to\
      \ be taken when a conflict is encountered, or convert the INSERT to an UPDATE.\r\
      \n\r\nFor more information, please see the [INSERT](https://www.postgresql.org/docs/current/sql-insert.html)\
      \ documentation."
  - name: Window functions
    versions:
      '8.4': 'Yes'
    url: https://www.postgresql.org/docs/current/tutorial-window.html
  - name: WITHIN GROUP clause
    versions:
      '9.4': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-expressions.html#SYNTAX-AGGREGATES
  - name: WITH ORDINALITY clause
    versions:
      '9.4': 'Yes'
    url: https://www.postgresql.org/docs/current/queries-table-expressions.html#QUERIES-TABLEFUNCTIONS
  - name: WITH Queries (Common Table Expressions)
    versions:
      '8.4': 'Yes'
    url: https://www.postgresql.org/docs/current/queries-with.html
  - name: Writable WITH Queries (Common Table Expressions)
    versions:
      '9.1': 'Yes'
    description: Allows INSERT/UPDATE/DELETE in WITH clauses which can be used in
      conjunction with a RETURNING clause to to pass data up to the containing query.
  Data Definition Language (DDL):
  - name: ALTER object IF EXISTS
    versions:
      '9.2': 'Yes'
    description: Many objects now support IF EXISTS clauses on ALTER statements, such
      as tables, foreign tables, indexes, sequences etc.
  - name: ALTER TABLE ... ADD UNIQUE/PRIMARY KEY USING INDEX
    versions:
      '9.1': 'Yes'
    description: This allows a primary key or unique constraint to be defined using
      an existing unique index, including a concurrently created unique index.
  - name: ALTER TABLE ... SET ACCESS METHOD
    versions:
      '15': 'Yes'
    description: '[`ALTER TABLE`](https://www.postgresql.org/docs/15/sql-altertable.html)
      can specify a different [access method](https://www.postgresql.org/docs/15/tableam.html),
      also known as a storage interface.'
  - name: ALTER TABLE ... SET LOGGED / UNLOGGED
    versions:
      '9.5': 'Yes'
    description: Allows tables to switch between logged and unlogged states.
  - name: Changing column types (ALTER TABLE .. ALTER COLUMN TYPE)
    versions:
      '8.1': 'Yes'
    description: allows online changes of a column datatype to a different one
  - name: CREATE ACCESS METHOD
    versions:
      '12': 'Yes'
    url: https://www.postgresql.org/docs/12/sql-create-access-method.html
  - name: CREATE TABLE ... (LIKE) with foreign tables, views and composite types
    versions:
      '9.2': 'Yes'
    description: ''
  - name: DROP object IF EXISTS
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-commands.html
  - name: ON COMMIT clause for CREATE TEMPORARY TABLE
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-createtable.html
  - name: REINDEX CONCURRENTLY
    versions:
      '12': 'Yes'
    url: https://www.postgresql.org/docs/current/sql-reindex.html#SQL-REINDEX-CONCURRENTLY
  - name: Stored Generated Columns
    versions:
      '12': 'Yes'
    url: https://www.postgresql.org/docs/current/ddl-generated-columns.html
  - name: Temporal constraints
    versions:
      '18': 'Yes'
    description: "Temporal constraints are constraints over ranges. These are\
      \ available for both [`PRIMARY KEY`](https://www.postgresql.org/docs/18/sql-createtable.html#SQL-CREATETABLE-PARMS-PRIMARY-KEY)\
      \ and [`UNIQUE`](https://www.postgresql.org/docs/18/sql-createtable.html#SQL-CREATETABLE-PARMS-UNIQUE)\
      \ constraints using the `WITHOUT OVERLAPS` clause, and\
      \ [`FOREIGN KEY`](https://www.postgresql.org/docs/18/sql-createtable.html#SQL-CREATETABLE-PARMS-REFERENCES)\
      \ constraints using the `PERIOD` clause."
  - name: Typed tables
    versions:
      '9.0': 'Yes'
    description: 'Adds support for syntax: CREATE TABLE name OF type'
  - name: Virtual Generated Columns
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/current/ddl-generated-columns.html
  Performance:
  - name: Abbreviated Keys
    versions:
      '9.5': 'Yes'
    description: The abbreviated keys optimization can be expected to greatly enhance
      the performance of sorts in PostgreSQL, including those used for CREATE INDEX.
      Reportedly, in some cases, CREATE INDEX on text columns can be as much as an
      entire order of magnitude faster (3x is a more typical improvement). Numeric
      sorts also support the optimization.
  - name: Asynchronous Commit
    versions:
      '8.3': 'Yes'
    description: Delays writes to WAL for committed transactions
  - name: Asynchronous I/O (AIO)
    versions:
      '18': 'Yes'
    description: "New I/O subsystem PostgreSQL issue multiple I/O requests\
      \ concurrently instead of waiting for each to finish in sequence. This\
      \ expands existing readahead and improves overall throughput. AIO\
      \ supported in PostgreSQL 18 include sequential scans, bitmap heap scans,\
      \ and vacuum. For more information see [`io_method`](https://www.postgresql.org/docs/18/runtime-config-resource.html#GUC-IO-METHOD)."
  - name: Automatic plan invalidation
    versions:
      '8.3': 'Yes'
    description: automatic invalidation of cached exceutions plans
  - name: Background Checkpointer
    versions:
      '9.2': 'Yes'
    description: Originally the background writer held responsibility for writing
      checkpoints, but it has now been moved into its own separate process and the
      background writer is now only responsible for writing dirty pages to disk.  This
      provides advantages related to performance and power-reduction.
  - name: Background Writer
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/runtime-config-resource.html
  - name: Base backup throttling
    versions:
      '9.4': 'Yes'
    description: pg_basebackup now supports throttling of backups with -r <rate> or
      --max-rate <rate>, allowing control of the data transfer rate of the base backup.
  - name: CREATE STATISTICS - most-common values (MCV) statistics
    versions:
      '12': 'Yes'
    description: CREATE STATISTICS  can collection statistics on the most common value,
      which can improve optimizations for columns that contain nonuniform distributions.
  - name: CREATE STATISTICS - multicolumn
    versions:
      '10': 'Yes'
    description: "CREATE STATISTICS provides the ability to collect statistics over\
      \ multiple columns, which can be used by the optimizer to generate better query\
      \ plans.\r\n\r\nhttps://www.postgresql.org/docs/current/sql-createstatistics.html"
  - name: CREATE STATISTICS - "OR" and "IN/ANY" statistics
    versions:
      '13': 'Yes'
    description: Extend statistics can help the query planner provided better plans
      for queries that have "OR".clauses as well as "IN/ANY" lookups over lists. For
      more information, please see https://www.postgresql.org/docs/13/planner-stats.html#PLANNER-STATS-EXTENDED
  - name: Cross datatype hashing support
    versions:
      '8.3': 'Yes'
    description: allows hashing to be used across compatible but not identical datatypes
  - name: Distributed checkpointing
    versions:
      '8.3': 'Yes'
    description: automatic smoothening of checkpoints to reduce sudden I/O spikes
  - name: Foreign keys marked as NOT VALID
    versions:
      '9.1': 'Yes'
    description: This introduces a new option to the creation of foreign keys called
      NOT VALID, which can later be modified to VALIDATED and validation checks performed.
      Together these allow you to add a foreign key with minimal impact on read and
      write operations.
  - name: Frozen page map
    versions:
      '9.6': 'Yes'
    description: Frozen page data in visibility map to prevent re-vacuuming already-frozen
      pages during wraparound.
  - name: Full Text Search
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/static/textsearch.html
  - name: Hash aggregation can use disk
    versions:
      '13': 'Yes'
    description: "The query planner can now consider using hash aggregation if it\
      \ expects the aggregate to exceed working memory.\r\n\r\nThis can be controlled\
      \ by two configuration parameters:\r\n\r\n- [enable_hashagg](https://www.postgresql.org/docs/13/runtime-config-query.html#GUC-ENABLE-HASHAGG)\r\
      \n- [hash_mem_multiplier](https://www.postgresql.org/docs/13/runtime-config-resource.html#GUC-HASH-MEM-MULTIPLIER)"
  - name: Hashing support for DISTINCT/UNION/INTERSECT/EXCEPT
    versions:
      '8.4': 'Yes'
    description: allows hashing to be used in query plans involving DISTINCT/UNION/EXCEPT/INTERSECT
  - name: Hashing support for FULL OUTER JOIN, LEFT OUTER JOIN and RIGHT OUTER JOIN
    versions:
      '9.1': 'Yes'
    description: allows FULL OUTER JOIN to be implemented as a hash join, and allow
      either side of a LEFT OUTER JOIN or RIGHT OUTER JOIN to be hashed
  - name: Heap Only Tuples (HOT)
    versions:
      '8.3': 'Yes'
    description: optimization for heavily updated tables
  - name: Improved performance for sorts exceeding working memory
    versions:
      '15': 'Yes'
    description: Sorts for data sets that exceed [`work_mem`](https://www.postgresql.org/docs/current/runtime-config-resource.html#GUC-WORK-MEM)
      now use a batch sorting algorithm that uses more output streams and improves
      performance.
  - name: Improved window function performance
    versions:
      '15': 'Yes'
    description: Performance optimizations for executing [window functions](https://www.postgresql.org/docs/current/functions-window.html)
      `row_number()`, `rank()`, and `count()`.
  - name: Incremental sort
    versions:
      '13': 'Yes'
    url: https://www.postgresql.org/docs/13/runtime-config-query.html#GUC-ENABLE-INCREMENTAL-SORT
  - name: Incremental sort for SELECT DISTINCT
    versions:
      '16': 'Yes'
    description: Queries with `SELECT DISTINCT` can use [incremental sort](https://www.postgresql.org/docs/16/runtime-config-query.html#GUC-ENABLE-INCREMENTAL-SORT).
  - name: Incremental sort for window functions
    versions:
      '14': 'Yes'
    description: '[Window functions](https://www.postgresql.org/docs/14/functions-window.html)
      can use incremental sorting'
  - name: Inlined WITH Queries (Common Table Expressions)
    versions:
      '12': 'Yes'
    description: "A WITH query that is neither recursive nor has any side-effects\
      \ (e.g. an INSERT/UPDATE/DELETE) can be executed inline, which can lead to performance\
      \ improvements. This behavior can be forced on a query by using the \"NOT MATERIALIZED\"\
      \ clause, e.g.\r\n\r\n```\r\nWITH cte AS NOT MATERIALIZED (\r\n    SELECT *\
      \ FROM a\r\n)\r\nSELECT * FROM cte\r\nJOIN b ON b.id = cte.id;\r\n```\r\n\r\n\
      For more information, please visit [https://www.postgresql.org/docs/12/queries-with.html](https://www.postgresql.org/docs/12/queries-with.html)"
  - name: Inlining of SQL-functions
    versions:
      '8.1': 'Yes'
    description: allows the planner to inline most SQL functions into the whole query
  - name: Just-in-Time (JIT) compilation for expression evaluation and tuple deforming
    versions:
      '11': 'Yes'
    url: https://www.postgresql.org/docs/current/static/jit.html
  - name: Load balancing for libpq / psql
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/16/libpq-connect.html#LIBPQ-CONNECT-LOAD-BALANCE-HOSTS
  - name: LZ4 compression for TOAST tables
    versions:
      '14': 'Yes'
    description: The compression method for [TOAST](https://www.postgresql.org/docs/14/storage-toast.html)
      tables is now selectable and adds support for lz4 compression.
  - name: Multi-core scalability for read-only workloads
    versions:
      '9.2': 'Yes'
    description: Improve concurrent read-only workloads on multi-core servers.  Previously
      this had limited scalability offering performance benefits only up to a few
      cores, but now performance gains are seen utilising all cores in 32 and 64 core
      servers.
  - name: Multiple temporary tablespaces
    versions:
      '8.3': 'Yes'
    description: allows multiple tablespaces to be defined for temporary operations
  - name: Outer Join reordering
    versions:
      '8.2': 'Yes'
    description: enables the query planner to reorder out joins
  - name: Parallel bitmap heap scans
    versions:
      '10': 'Yes'
    description: A single index scan can dispatch parallel workers to process different
      areas of the heap.
  - name: Parallel FULL and RIGHT joins
    versions:
      '16': 'Yes'
    description: Ability to execute in parallell `FULL` and `RIGHT OUTER` joins.
  - name: Parallel full table scans (sequential scans)
    versions:
      '9.6': 'Yes'
    url: https://www.postgresql.org/docs/current/parallel-plans.html#PARALLEL-SCANS
  - name: Parallel hash joins
    versions:
      '11': 'Yes'
    url: https://www.postgresql.org/docs/current/parallel-plans.html#PARALLEL-JOINS
  - name: Parallel JOIN, aggregate
    versions:
      '9.6': 'Yes'
    description: Ability to execute some kinds of JOINs, as well as aggregates, using
      multiple parallel processes.
  - name: Parallel merge joins
    versions:
      '10': 'Yes'
    description: Merge joins can be performed in parallel.
  - name: Parallel query
    versions:
      '9.6': 'Yes'
    description: Ability to execute multiple query nodes in parallel processes.
  - name: Parallel "SELECT DISTINCT"
    versions:
      '15': 'Yes'
    description: A query using [`SELECT DISTINCT`](https://www.postgresql.org/docs/current/queries-select-lists.html#QUERIES-DISTINCT)
      can now be executed using parallelism.
  - name: Partial sort capability (top-n sorting)
    versions:
      '8.3': 'Yes'
    description: optimizes sort operations for ORDER BY x LIMIT ionsy like operat
  - name: Query pipelining
    versions:
      '14': 'Yes'
    url: https://www.postgresql.org/docs/14/libpq-pipeline-mode.html
  - name: Reduced lock levels for ALTER TABLE commands
    versions:
      '9.4': 'Yes'
    description: "The following commands now have reduced lock levels for reduced\
      \ lock contention:\r\n\r\nVALIDATE CONSTRAINT\r\n\r\nCLUSTER ON\r\nSET WITHOUT\
      \ CLUSTER\r\n\r\nALTER COLUMN SET STATISTICS\r\nALTER COLUMN SET ()\r\nALTER\
      \ COLUMN RESET ()"
  - name: SELECT ... FOR UPDATE/SHARE NOWAIT
    versions:
      '8.1': 'Yes'
    description: ''
  - name: Set costs specific to TABLESPACEs
    versions:
      '9.0': 'Yes'
    description: Allows for setting seq_page_cost and random_page_cost on a per tablespace
      level
  - name: Shared row level locking
    versions:
      '8.1': 'Yes'
    description: reduces deadlock scenarios with foreign keys
  - name: SIMD support for ARM
    versions:
      '16': 'Yes'
    description: Allows for SIMD execution of vector operations on ARM architecture,
      including JSON string parsing, including subtransaction ID searches.
  - name: SIMD support for x86
    versions:
      '16': 'Yes'
    description: Allows for SIMD execution of vector operations on x86 architecture,
      including JSON string parsing, ASCII string detection, and subtransaction ID
      searches.
  - name: SKIP LOCKED clause
    versions:
      '9.5': 'Yes'
    description: Skips over rows that are currently locked without waiting.
  - name: Synchronized sequential scanning
    versions:
      '8.3': 'Yes'
    description: enabled sequential scans to piggyback on concurrently running scans
  - name: TABLESAMPLE clause
    versions:
      '9.5': 'Yes'
    description: Specifies that a sample of the table is to be used, which is useful
      for use-cases such as approximate averages.
  - name: Tablespaces
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-createtablespace.html
  - name: Unlogged tables
    versions:
      '9.1': 'Yes'
    description: 'Such tables provide better update performance than regular tables,
      but are not crash-safe: their contents are automatically cleared in case of
      a server crash. Their contents do not propagate to replication slaves, either.'
  - name: WAL Buffer auto-tuning
    versions:
      '9.1': 'Yes'
    description: wal_buffers is now auto-tuned by default based on the size of shared_buffers
  JSON:
  - name: Improved set of JSON functions and operators
    versions:
      '9.3': 'Yes'
    description: New operators and functions to extract values from JSON data strings,
      JSON data strings can now to be converted into records and new functions to
      convert values, records, and hstore data to JSON.
  - name: JSONB data type
    versions:
      '9.4': 'Yes'
    url: https://www.postgresql.org/docs/current/static/datatype-json.html
  - name: JSONB-modifying operators and functions
    versions:
      '9.5': 'Yes'
    description: New operators and functions which allow deleting, modifying or inserting
      values into JSONB values, including at specific path locations.
  - name: JSONB Subscripting
    versions:
      '14': 'Yes'
    url: https://www.postgresql.org/docs/14/datatype-json.html#JSONB-SUBSCRIPTING
  - name: JSON data type
    versions:
      '9.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/datatype-json.html
  - name: SQL/JSON constructors
    versions:
      '16': 'Yes'
    description: Adds SQL/JSON constructors, including `JSON_ARRAY()`, `JSON_ARRAYAGG()`,
      `JSON_OBJECT()`, and `JSON_OBJECTAGG()`.
  - name: 'SQL/JSON: datetime()'
    versions:
      '13': 'Yes'
    description: "The SQL/JSON path `datetime()` function in JSON path expressions\
      \ converts valid date and time formats (e.g. ISO 8601) to their corresponding\
      \ PostgreSQL type.\r\n\r\nFor more information, see the [JSON path documentation](https://www.postgresql.org/docs/13/functions-json.html#FUNCTIONS-SQLJSON-OP-TABLE)"
  - name: SQL/JSON IS JSON
    versions:
      '16': 'Yes'
    description: "SQL/JSON syntax for testing the type of JSON object. For example:\r\
      \n\r\n```\r\nSELECT js,\r\n  js IS JSON \"json?\",\r\n  js IS JSON SCALAR \"\
      scalar?\",\r\n  js IS JSON OBJECT \"object?\",\r\n  js IS JSON ARRAY \"array?\"\
      \r\nFROM (VALUES\r\n      ('123'), ('\"abc\"'), ('{\"a\": \"b\"}'), ('[1,2]'),('abc'))\
      \ foo(js);\r\n```"
  - name: SQL/JSON JSON_TABLE
    versions:
      '17': 'Yes'
    url: https://www.postgresql.org/docs/17/functions-json.html#FUNCTIONS-SQLJSON-TABLE
  - name: SQL/JSON path expressions
    versions:
      '12': 'Yes'
    url: https://www.postgresql.org/docs/current/functions-json.html#FUNCTIONS-SQLJSON-PATH
  - name: SQL/JSON query functions
    versions:
      '17': 'Yes'
    url: https://www.postgresql.org/docs/17/functions-json.html#FUNCTIONS-SQLJSON-QUERYING
  Partitioning & Inheritance:
  - name: Accelerated partition pruning
    versions:
      '12': 'Yes'
    description: Performance improvements for queries on tables with thousands of
      partitions when only a limited subset of partitions need to be accessed.
  - name: Declarative table partitioning
    versions:
      '10': 'Yes'
    url: https://www.postgresql.org/docs/10/static/ddl-partitioning.html#ddl-partitioning-declarative
  - name: Default Partition
    versions:
      '11': 'Yes'
    description: A default partition stores data that does not match the partition
      key for any other partition, i.e. it is a "catch all" partition.
  - name: Foreign Key references for partitioned tables
    versions:
      '12': 'Yes'
    description: A foreign key can reference a partitioned table.
  - name: Foreign table inheritance
    versions:
      '9.5': 'Yes'
    description: Foreign tables can now inherit from other foreign tables or other
      regular tables.  The reverse also applies.
  - name: Partitioning by a hash key
    versions:
      '11': 'Yes'
    description: ''
  - name: Partition pruning during query execution
    versions:
      '11': 'Yes'
    description: Queries on partitioned tables can eliminate partitions during query
      execution time, instead of only at query planning time. This allows PostgreSQL
      to use information that becomes available during query execution to further
      prune the number of partitions it needs to use, leading to improved query performance.
  - name: Support for PRIMARY KEY, FOREIGN KEY, indexes, and triggers on partitioned
      tables
    versions:
      '11': 'Yes'
    description: ''
  - name: Table Partitioning
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/ddl-partitioning.html
  - name: UPDATE on a partition key
    versions:
      '11': 'Yes'
    description: When a partition key is updated on a row, the row is moved to the
      appropriate partition.
  Views & Materialized Views:
  - name: Materialized Views
    versions:
      '9.3': 'Yes'
    url: https://www.postgresql.org/docs/current/static/rules-materializedviews.html
  - name: Materialized views with concurrent refresh
    versions:
      '9.4': 'Yes'
    description: Materialized views which can be refreshed with REFRESH MATERIALIZED
      VIEW CONCURRENTLY name;
  - name: SECURITY INVOKER views
    versions:
      '15': 'Yes'
    description: '[Views](https://www.postgresql.org/docs/15/sql-createview.html)
      can be created using the `SECURITY INVOKER` parameter, which executes the view
      with the privileges of the caller instead of the privileges of the owner.'
  - name: Temporary VIEWs
    versions:
      '8.1': 'Yes'
    description: allows the creation of temporary VIEWs
  - name: Updatable views
    versions:
      '9.3': 'Yes'
    description: Simple views can now have UPDATE/INSERT/DELETE performed against
      them.
  - name: WITH CHECK clause
    versions:
      '9.4': 'Yes'
    description: Auto-updatable views can now specify whether an INSERT or UPDATE
      would change the state of the row so that it would no longer be visible in the
      view.  Using WITH CHECK OPTION will prevent any such changes from occuring.
  Replication:
  - name: ALTER SUBSCRIPTION ... SKIP
    versions:
      '15': 'Yes'
    description: '[`ALTER SUBSCRIPTION ... SKIP`] allows a subscriber to skip replaying
      a transaction. For more information, please read the section on handling [conflicts](https://www.postgresql.org/docs/15/logical-replication-conflicts.html)
      with logical replication.'
  - name: Cascading streaming replication
    versions:
      '9.2': 'Yes'
    description: Standbys can now stream to other standbys, reducing replication load
      on the primary server.
  - name: Configure max WAL retention for replication slots
    versions:
      '13': 'Yes'
    url: https://www.postgresql.org/docs/13/runtime-config-replication.html#GUC-MAX-SLOT-WAL-KEEP-SIZE
  - name: Failover support for logical replication slots
    versions:
      '17': 'Yes'
    url: https://www.postgresql.org/docs/17/logical-replication-failover.html
  - name: Logical replication
    versions:
      '10': 'Yes'
    url: https://www.postgresql.org/docs/current/static/logical-replication.html
  - name: Logical replication avoids replication loops
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/16/sql-createsubscription.html#SQL-CREATESUBSCRIPTION-WITH-ORIGIN
  - name: Logical replication column lists
    versions:
      '15': 'Yes'
    description: '[Publications](https://www.postgresql.org/docs/15/logical-replication-publication.html)
      for logical replication can now specify which columns to publish. Previously,
      a publication would publish all columns in a table.'
  - name: Logical replication for partitioned tables
    versions:
      '13': 'Yes'
    description: Fully supports logical replication of partitioned tables.
  - name: Logical replication from standbys
    versions:
      '16': 'Yes'
    description: Logical replication publishers can be created from standby instances.
  - name: Logical replication of stored generated columns
    versions:
      '18': 'Yes'
    description: "Stored generated columns are logically replicated."
  - name: Logical replication initial sync using binary protocol
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/16/sql-createsubscription.html#SQL-CREATESUBSCRIPTION-WITH-BINARY
  - name: Logical replication lookups with additional indexes
    versions:
      '16': 'Yes'
    description: "A logical replication subscriber can use indexes other than the\
      \ `PRIMARY KEY` to perform lookups during\r\n`UPDATE` or `DELETE` operations,"
  - name: Logical replication parallel apply of transactions
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/16/sql-createsubscription.html#SQL-CREATESUBSCRIPTION-WITH-STREAMING
  - name: Logical replication publish all tables in schema
    versions:
      '15': 'Yes'
    description: When creating a logical replication publication, a user can specify
      to publish all the tables within a schema , e.g. [`CREATE PUBLICATION ... FOR
      TABLES IN SCHEMA ...`](https://www.postgresql.org/docs/15/sql-createpublication.html).
  - name: Logical replication row filtering
    versions:
      '15': 'Yes'
    url: https://www.postgresql.org/docs/15/logical-replication-row-filter.html
  - name: Logical replication slots migrate through pg_upgrade migrate
    versions:
      '17': 'Yes'
    description: This allows logical replication to continue quickly after the upgrade.
      This only works for old PostgreSQL clusters that are version 17 or later.
  - name: Logical replication stream in-progress transactions
    versions:
      '14': 'Yes'
    description: In-progress transactions can be streamed when using [logical replication](https://www.postgresql.org/docs/current/logical-replication.html),
      versus having to wait for the transaction to be completed first.
  - name: Logical replication subscriber can disable on error
    versions:
      '15': 'Yes'
    description: Logical replication subscribers can be automatically disabled using
      the [`disable_on_error`](https://www.postgresql.org/docs/15/logical-replication-conflicts.html)
      setting.
  - name: Logical replication write conflict logging
    versions:
      '18': 'Yes'
    description: "Write conflicts that occur during ogical replication are\
      \ reported both in logs and in the\
      \ [`pg_stat_subscription_stats`](https://www.postgresql.org/docs/18/monitoring-stats.html#MONITORING-PG-STAT-SUBSCRIPTION-STATS)\
      \ view."
  - name: Quorum commit for synchronous replication
    versions:
      '10': 'Yes'
    description: Using the synchronous_standby_names configuration parameter, synchronous
      replication can be setup to allow for any number of standbys to confirm writes
      have been committed, irrespective of their ordering. This is also known as "quorum
      commit."
  - name: Replication Slots
    versions:
      '9.4': 'Yes'
    url: https://www.postgresql.org/docs/9.4/static/warm-standby.html#STREAMING-REPLICATION-SLOTS
  - name: Streaming-only cascading replication
    versions:
      '9.3': 'Yes'
    description: Cascading replication previously required that an xlog archive be
      available in order for a standbys of the new primary to correctly switch to
      the new timeline.  This change removes this dependency.
  - name: Streaming Replication
    versions:
      '9.0': 'Yes'
    description: Integrated replication support based on streaming WAL data
  - name: Synchronous replication
    versions:
      '9.1': 'Yes'
    description: This allows the primary to wait for a standby to write the transaction
      information to disk before acknowledging the commit. One standby at a time can
      take the role of the synchronous standby, as controlled by the synchronous_standby_names
      setting. Synchronous replication can be enabled or disabled on a per-transaction
      basis using the synchronous_commit setting.
  Backup, Restore, & Data Integrity:
  - name: Archive modules
    versions:
      '15': 'Yes'
    url: https://www.postgresql.org/docs/15/archive-modules.html
  - name: Checksum on data pages
    versions:
      '9.3': 'Yes'
    description: Clusters can now be created with checksums on pages to detect and
      report page corruption.
  - name: Enable/Disable page checksums in an offline cluster
    versions:
      '12': 'Yes'
    description: "Page checksums can be enabled or disabled in an offline PostgreSQL\
      \ cluster via the pg_checksums command:\r\n\r\nhttps://www.postgresql.org/docs/12/app-pgchecksums.html"
  - name: Generic WAL facility
    versions:
      '9.6': 'Yes'
    description: Consistent, simple API for WAL-logging of database objects, supporting
      new index access methods and other storage objects.
  - name: Hot Standby
    versions:
      '9.0': 'Yes'
    description: allow read only connections during recovery
  - name: lz4 and Zstandard (zstd) compression for WAL full page writes
    versions:
      '15': 'Yes'
    url: https://www.postgresql.org/docs/15/runtime-config-wal.html#GUC-WAL-COMPRESSION
  - name: min_wal_size / max_wal_size
    versions:
      '9.5': 'Yes'
    description: "Replaces `checkpoint_segments` parameter.\r\n\r\nFor more information,\
      \ please see the [WAL configuration](https://www.postgresql.org/docs/current/wal-configuration.html)\
      \ documentation."
  - name: Multiple synchronous standbys
    versions:
      '9.6': 'Yes'
    description: Ability to require synch messages from multiple prioritized standbys
      when committing a synchronous transaction.
  - name: Named restore points
    versions:
      '9.1': 'Yes'
    description: Create a named point for performing restore using pg_create_restore_point
  - name: Parallel pg_dump
    versions:
      '9.3': 'Yes'
    description: pg_dump now has a --jobs option to dump tables in parallel.
  - name: Parallel restore
    versions:
      '8.4': 'Yes'
    description: allows pg_restore to restore data from custom format dumps in parallel
  - name: pg_basebackup client decompression
    versions:
      '15': 'Yes'
    description: '[`pg_basebackup`](https://www.postgresql.org/docs/15/app-pgbasebackup.html)
      can now decompress backups using `lz4` and `zstd` (Zstandard) compression.'
  - name: pg_basebackup incremental backup
    versions:
      '17': 'Yes'
    description: Incremental backups can be created using the `--incremental` option
      for [`pg_basebackup`](https://www.postgresql.org/docs/current/app-pgbasebackup.html).
      The new application [pg_combinebackup](https://www.postgresql.org/docs/17/app-pgcombinebackup.html)
      allows manipulation of base and incremental file system backups.
  - name: pg_basebackup server-side compression
    versions:
      '15': 'Yes'
    description: '[`pg_basebackup`](https://www.postgresql.org/docs/current/app-pgbasebackup.html)
      now supports server-side compression when creating a backup using the `--compress`
      flag. This includes `gzip`, `lz4`, and `zstd` (Zstandard) compression.'
  - name: pg_basebackup tool
    versions:
      '9.1': 'Yes'
    description: pg_basebackup is a tool to take a base backup of a PostgreSQL cluster
  - name: pg_dump, pg_dumpall, pg_restore --filter
    versions:
      '17': 'Yes'
    description: Allow [pg_dump](https://www.postgresql.org/docs/current/app-pgdump.html),
      [pg_dumpall](https://www.postgresql.org/docs/current/app-pg-dumpall.html), and
      [pg_restore](https://www.postgresql.org/docs/current/app-pgrestore.html) to
      specify include/exclude objects in a file using the `--filter` option.
  - name: pg_receivewal (formerly pg_receivexlog)
    versions:
      '9.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/app-pgreceivewal.html
  - name: Point-in-Time Recovery
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/continuous-archiving.html
  - name: Pre-fetch WAL during recovery
    versions:
      '15': 'Yes'
    url: https://www.postgresql.org/docs/15/runtime-config-wal.html#GUC-RECOVERY-PREFETCH
  - name: remote_apply mode
    versions:
      '9.6': 'Yes'
    description: Ability to require that synchronous replicas be caught up with the
      master in applying changes before commit.  Supports "consistent clustering".
  - name: Time-delayed Standbys
    versions:
      '9.4': 'Yes'
    description: A new setting called recovery_min_apply_delay will allow a standby
      to lag behind the primary by a specified amount of time.
  - name: Verify backup integrity (pg_verifybackup)
    versions:
      '13': 'Yes'
    url: https://www.postgresql.org/docs/13/app-pgverifybackup.html
  - name: Warm Standby
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/warm-standby.html
  Upgrade:
  - name: pg_upgrade --swap
    versions:
      '18': 'Yes'
    description: "The `--swap` option moves the data directories from the old\
      \ cluster to the new cluster and then replaces catalog files with those\
      \ generated for the new cluster. See\
      [`pg_upgrade`](https://www.postgresql.org/docs/18/pgupgrade.html)\
      \ for more information."
  - name: Planner statistics preserved on major version upgrade
    versions:
      '18': 'Yes'
    description: "PostgreSQL now keeps planner statistics\
      \ through a major version upgrade, which helps an upgraded cluster\
      \ reach expected performance more quickly after the upgrade. See the\
      \ statistics section in [`pg_upgrade`](https://www.postgresql.org/docs/18/pgupgrade.html)\
      \ for more information."
  Data Import & Export:
  - name: COPY from/to STDIN/STDOUT
    versions:
      '9.3': 'Yes'
    description: COPY and psql's \copy now support piping from and to STDIN and STDOUT
      respectively.
  - name: COPY FROM ... WHERE
    versions:
      '12': 'Yes'
    description: Specify which rows are loaded from a data source via a WHERE clause
      when using COPY FROM
  - name: COPY ... ON_ERROR
    versions:
      '17': 'Yes'
    description: "The new `ON_ERROR` option for COPY provides the option to ignore\
      \ and discard error rows. The default behavior is ON_ERROR stop the import.\r\
      \n\r\nSee [`COPY`](https://www.postgresql.org/docs/current/sql-copy.html)."
  - name: COPY with arbitrary SELECT
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-copy.html
  - name: CSV support for COPY
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-copy.html
  Configuration Management:
  - name: ALTER SYSTEM
    versions:
      '9.4': 'Yes'
    url: https://www.postgresql.org/docs/9.4/static/sql-altersystem.html
  - name: Fractional input for "integer" values
    versions:
      '12': 'Yes'
    description: "Fractional input for \"integer\" server variables is now accepted\
      \ e.g.\r\n\r\nSET work_mem = '24.2MB'"
  - name: Include directives for pg_hba.conf and pg_ident.conf
    versions:
      '16': 'Yes'
    description: 'Both [`pg_hba.conf`](https://www.postgresql.org/docs/16/auth-pg-hba-conf.html)
      and [`pg_ident.conf`](https://www.postgresql.org/docs/16/auth-username-maps.html)
      support include directives that let you specify additional files for loading
      rules. These include: `include`, `include_if_exists`, and `include_dir`.'
  - name: Per user/database server configuration settings
    versions:
      '9.0': 'Yes'
    description: provides for per user/database server configuration settings
  - name: pg_config system view
    versions:
      '9.6': 'Yes'
    description: New system view to display information about the PostgreSQL build,
      storage, and build configuration options.
  - name: Regular expression matching in pg_hba.conf and pg_ident.conf
    versions:
      '16': 'Yes'
    description: Both [`pg_hba.conf`](https://www.postgresql.org/docs/16/auth-pg-hba-conf.html)
      and [`pg_ident.conf`](https://www.postgresql.org/docs/16/auth-username-maps.html)
      allow for regular expression matching on usernames. `pg_hba.conf` also supports
      regular expression matching on database names.
  Security:
  - name: Channel binding for SCRAM authentication
    versions:
      '11': 'Yes'
    url: https://www.postgresql.org/docs/current/sasl-authentication.html#SASL-SCRAM-SHA-256
  - name: Client can require SCRAM channel binding
    versions:
      '13': 'Yes'
    url: https://www.postgresql.org/docs/13/libpq-connect.html#LIBPQ-CONNECT-CHANNEL-BINDING
  - name: Client-specified requirements for authentication
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/devel/libpq-connect.html#LIBPQ-CONNECT-REQUIRE-AUTH
  - name: Column level permissions
    versions:
      '8.4': 'Yes'
    description: allows specification of granular column-level permissions in addition
      to table level grants
  - name: Default permissions
    versions:
      '9.0': 'Yes'
    description: allows the specification of default permissions for objects
  - name: Direct TLS negotiation ("sslnegotiation")
    versions:
      '17': 'Yes'
    url: https://www.postgresql.org/docs/17/libpq-connect.html#LIBPQ-CONNECT-SSLNEGOTIATION
  - name: FIPS mode validation
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/pgcrypto.html#PGCRYPTO-OPENSSL-SUPPORT-FUNCS
  - name: GRANT/REVOKE ON ALL TABLES/SEQUENCES/FUNCTIONS
    versions:
      '9.0': 'Yes'
    description: simplifies permission management by allowing mass changes of GRANTS
  - name: GSSAPI client and server-side encryption
    versions:
      '12': 'Yes'
    description: "Support for client and server-side connection encryption when using\
      \ GSSAPI for authentication.\r\n\r\nhttps://www.postgresql.org/docs/12/gssapi-auth.html"
  - name: GSSAPI support
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/gssapi-auth.html
  - name: Kerberos credential delegation
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/16/runtime-config-connection.html#GUC-GSS-ACCEPT-DELEGATION
  - name: krb5 authentication (without gssapi)
    versions:
      '8.1': 'Yes'
      '9.4': Obsolete
    description: native Kerberos 5 wire authentication (replaced with GSSAPI)
  - name: Large object access controls
    versions:
      '9.0': 'Yes'
    description: Access controls on large objects
  - name: LDAP server discovery
    versions:
      '12': 'Yes'
    description: "LDAP servers with DNS SRV can be discovered if PostgreSQL is built\
      \ with OpenLDAP\r\n\r\nhttps://www.postgresql.org/docs/12/auth-ldap.html"
  - name: Multifactor authentication via valid client SSL/TLS certificate
    versions:
      '12': 'Yes'
    description: "If an authentication entry in the pg_hba.conf specifies the `clientcert=verify-full`,\
      \ then the client must present a valid SSL certificate that matches the login\
      \ name, or the client name based on a map.\r\n\r\n[https://www.postgresql.org/docs/12/ssl-tcp.html#SSL-CLIENT-CERTIFICATES](https://www.postgresql.org/docs/12/ssl-tcp.html#SSL-CLIENT-CERTIFICATES)"
  - name: Native LDAP authentication
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/auth-methods.html#AUTH-LDAP
  - name: Native RADIUS authentication
    versions:
      '9.0': 'Yes'
    description: adds support for authentictaing against a RADIUS server
  - name: OAuth Authentication / Authorization
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/auth-oauth.html
  - name: Per user/database connection limits
    versions:
      '8.1': 'Yes'
    description: allows restrictions on the number of connections a given user or
      database will be allowed to make or accept
  - name: Predefined roles
    versions:
      '9.6': 'Yes'
    url: https://www.postgresql.org/docs/current/predefined-roles.html
  - name: Privileges for setting configuration parameters
    versions:
      '15': 'Yes'
    description: Unprivileged users can be [granted permission](https://www.postgresql.org/docs/15/sql-grant.html)
      to change server variables via [`SET`](https://www.postgresql.org/docs/current/sql-set.html)
      and [`ALTER SYSTEM`](https://www.postgresql.org/docs/current/sql-altersystem.html)
  - name: ROLES
    versions:
      '8.1': 'Yes'
    description: replaces user and group concept by a more generalizes role based
      concept
  - name: Row-Level Security
    versions:
      '9.5': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-createpolicy.html
  - name: SCRAM-SHA-256 Authentication
    versions:
      '10': 'Yes'
    description: "As described in RFC 7677, SCRAM-SHA-256 authentication provides\
      \ challenge-response scheme, that prevents password sniffing on untrusted connections.\
      \ It is more secure than the md5 method, but might not be supported by older\
      \ clients.\r\n\r\nSCRAM-SHA-256 authentication can be enabled in the host-based\
      \ authentication configuration file."
  - name: Search+bind mode operation for LDAP authentication
    versions:
      '9.0': 'Yes'
    description: Allows searching for the user in the directory first, and then binds
      with the DN found for this user
  - name: security_barrier option on views
    versions:
      '9.2': 'Yes'
    description: This prevents security_barrier views from being moved into other
      scopes, preventing possible leakage of view-prohibited data.
  - name: Security Service Provider Interface (SSPI)
    versions:
      '8.3': 'Yes'
    description: full integration with the native authentication framework in Microsoft
      Windows
  - name: SHA-2 encryption for password hashing
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/pgcrypto.html#PGCRYPTO-CRYPT-ALGORITHMS
  - name: SSL certificate validation in libpq
    versions:
      '8.4': 'Yes'
    description: certificate validation in libpq
  - name: SSL client certificate authentication
    versions:
      '8.4': 'Yes'
    description: allows authentication of clients by the use of an SSL certificate
  - name: SSPI authentication via GSSAPI
    versions:
      '9.1': 'Yes'
    description: allows Unix-based GSSAPI clients to do SSPI authentication with Windows
      servers.
  - name: Support using the client's OS trusted CA.
    versions:
      '16': 'Yes'
    description: Specifying `sslroot=system` instructs PostgreSQL to use the trusted
      certificate authority (CA) store provided by the client's operating system.
  - name: TLS v1.3 cipher suite allowlisting
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/runtime-config-connection.html#GUC-SSL-TLS13-CIPHERS
  Transactions and Visibility:
  - name: Cursors
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-declare.html
  - name: Savepoints
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-savepoint.html
  - name: Serializable Snapshot Isolation
    versions:
      '9.1': 'Yes'
    description: This implements true serializable snapshot isolation. Previously,
      asking for serializable isolation guaranteed only that a single MVCC snapshot
      would be used for the entire transaction, which allowed certain documented anomalies.
  - name: Two Phase commit
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-prepare-transaction.html
  - name: Updatable cursors
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-declare.html
  VACUUM and Maintenance:
  - name: Inserted data can trigger autovacuum
    versions:
      '13': 'Yes'
    description: This new behavior reduces the work necessary when the table needs
      to be frozen and allows pages to be set as all-visible. All-visible pages allow
      index-only scans to access fewer heap rows.
  - name: Integrated autovacuum daemon
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/routine-vacuuming.html#AUTOVACUUM
  - name: Page freezing optimizations
    versions:
      '16': 'Yes'
    description: Improvements to the page freezing strategy helps the general performance
      of vacuuming and other maintenance operations.
  - name: Parallelized VACUUM for Indexes
    versions:
      '13': 'Yes'
    description: "The VACUUM process can be parallelized for indexes. The number of\
      \ VACUUM jobs for an index can be specified.\r\n\r\nFor more information, please\
      \ see the [VACUUM documentation](https://www.postgresql.org/docs/current/sql-vacuum.html)"
  - name: Parallel vacuumdb jobs
    versions:
      '9.5': 'Yes'
    description: The vacuumdb tool now supports an option to define the number of
      jobs in which to run VACUUM.
  - name: Radix tree memory structure for vacuum
    versions:
      '17': 'Yes'
    description: The internal structure for vacuum now uses a radix tree structure,
      which reduces the amount of memory required for the vacuum process. Additionally,
      vacuum is no longer limited to 1GB of memory even when `maintenance_work_mem`
      or `autovacuum_work_mem` are set to higher values.
  - name: Vacuum "emergency mode"
    versions:
      '14': 'Yes'
    description: PostgreSQL will vacuum aggressively when it detects the possibility
      of transaction ID wraparound. This is controlled with the [`vacuum_failsafe_age
      `](https://www.postgresql.org/docs/14/runtime-config-client.html#GUC-VACUUM-FAILSAFE-AGE)
      and [`vacuum_multixact_failsafe_age `](https://www.postgresql.org/docs/14/runtime-config-client.html#GUC-MULTIXACT-FAILSAFE-AGE)
      configuration parameters.
  - name: Visibility Map for Vacuuming
    versions:
      '8.4': 'Yes'
    description: This reduces the overhead of vacuuming as the visibility map tracks
      only the pages that need to be vacuumed.
  Foreign Data Wrappers:
  - name: Certificate authentication with postgres_fdw
    versions:
      '13': 'Yes'
    description: "The [postgres_fdw](https://www.postgresql.org/docs/13/postgres-fdw.html)\
      \ can now use the built-in [certificate authentication](https://www.postgresql.org/docs/current/auth-cert.html)\
      \ mechanism.\r\n\r\nFor more information, please visit the [postgres_fdw documentation](https://www.postgresql.org/docs/13/postgres-fdw.html#id-1.11.7.42.10)"
  - name: CREATE FOREIGN TABLE ... LIKE
    versions:
      '18': 'Yes'
    description: "Specify a source table from which the new table automatically\
      \ copies all column names, their data types, and their not-null\
      \ constraints."
  - name: Foreign data wrapper query parallelism
    versions:
      '14': 'Yes'
    description: Queries that reference multiple foreign tables can perform table
      scans in parallel. This is implemented in the [`postgres_fdw`](https://www.postgresql.org/docs/14/postgres-fdw.html)
  - name: Foreign data wrappers
    versions:
      '9.1': 'Yes'
    description: Foreign data wrappers allow access to external data for querying.
  - name: Foreign Tables
    versions:
      '9.1': 'Yes'
    description: This allows data stored outside the database to be used like native
      PostgreSQL-stored data.
  - name: IMPORT FOREIGN SCHEMA
    versions:
      '9.5': 'Yes'
    url: https://www.postgresql.org/docs/9.5/static/sql-importforeignschema.html
  - name: Import foreign table partitions
    versions:
      '14': 'Yes'
    description: '[`IMPORT FOREIGN SCHEMA`](https://www.postgresql.org/docs/14/sql-importforeignschema.html)
      can import the schema of partitioned tables.'
  - name: Parallel query execution on remote databases
    versions:
      '14': 'Yes'
    description: Support for leveraging query parallelism on remote databases. This
      is implemented in the [`postgres_fdw`](https://www.postgresql.org/docs/current/postgres-fdw.html)
      available with PostgreSQL.
  - name: postgres_fdw parallel commit
    versions:
      '15': 'Yes'
    url: https://www.postgresql.org/docs/15/postgres-fdw.html#id-1.11.7.47.11.7
  - name: postgres_fdw pushdown
    versions:
      '9.6': 'Yes'
    description: Ability to push down JOIN, Sorts, UPDATEs and DELETEs to the remote
      database in postgres_fdw driver, and theoretically in other drivers.  Also some
      generic operator/function pushdown.
  - name: postgres_fdw SCRAM authentication passthrough
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/postgres-fdw.html#POSTGRES-FDW-OPTION-USE-SCRAM-PASSTHROUGH
  - name: PostgreSQL Foreign Data Wrapper
    versions:
      '9.3': 'Yes'
    url: https://www.postgresql.org/docs/current/static/postgres-fdw.html
  - name: Writable Foreign Data Wrappers
    versions:
      '9.3': 'Yes'
    description: The foreign data wrapper infrastructure now supports INSERT/DELETE/UPDATE
      operations which individual FDWs can now add support for.
  Custom Functions, Stored Procedures, & Triggers:
  - name: ALTER TABLE ENABLE/DISABLE TRIGGER
    versions:
      '8.1': 'Yes'
    description: allows for disabling and enabling a specific or all triggers on a
      given table
  - name: ALTER TABLE / ENABLE REPLICA TRIGGER/RULE
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/static/sql-altertable.html
  - name: BEGIN ATOMIC function bodies
    versions:
      '14': 'Yes'
    description: "PostgreSQL functions and stored procedures can now be created using\
      \ the SQL-standard `BEGIN ATOMIC` definition. Using `BEGIN ATOMIC` allows function\
      \ bodies to be parsed on creation and also provides dependency. For example:\r\
      \n\r\n```\r\nCREATE FUNCTION add(int, int)\r\nRETURNS int\r\nIMMUTABLE PARALLEL\
      \ SAFE\r\nBEGIN ATOMIC;\r\n  SELECT $1 + $2;\r\nEND;\r\n```"
  - name: CALL syntax for executing procedures
    versions:
      '11': 'Yes'
    url: https://www.postgresql.org/docs/11/static/sql-call.html
  - name: Column level triggers
    versions:
      '9.0': 'Yes'
    description: adds support for SQL standard column level trigger support, by allowing
      triggers to only fire if specific columns are mentioned in the UPDATEs SET list
  - name: CREATE PROCEDURE syntax for SQL stored procedures
    versions:
      '11': 'Yes'
    url: https://www.postgresql.org/docs/11/static/sql-createprocedure.html
  - name: Event triggers
    versions:
      '9.3': 'Yes'
    description: Support for creating triggers for DDL events.
  - name: FILTER clause for aggregate functions
    versions:
      '9.4': 'Yes'
    description: 'Aggregate functions can now be filtered with the FILTER clause.  For
      example: count(col_a) FILTER (WHERE col_b > 5)'
  - name: ORDER BY support within aggregates
    versions:
      '9.0': 'Yes'
    description: allows controlling the order of values fed into an aggregate function
  - name: Per function GUC settings
    versions:
      '8.3': 'Yes'
    description: Server configuration parameters can now be set on a per-function
      basis
  - name: Per function statistics
    versions:
      '8.4': 'Yes'
    description: runtime statistics for user-defined functions
  - name: RETURN QUERY EXECUTE
    versions:
      '8.4': 'Yes'
    description: allows for returning the result of a dynamically generated query
  - name: RETURNS TABLE
    versions:
      '8.4': 'Yes'
    description: SQL standard RETURNS TABLE clause for CREATE FUNCTION
  - name: Statement level triggers
    versions:
      '8.1': 'Yes'
    description: Triggers that fire on a per statement level instead of a per row
      level
  - name: Statement level TRUNCATE triggers
    versions:
      '8.4': 'Yes'
    description: Trigger support for TRUNCATE
  - name: Triggers on views
    versions:
      '9.1': 'Yes'
    description: This adds support for INSTEAD OF triggers on views which can be used
      to implement fully updatable views.
  - name: Variadic functions
    versions:
      '8.4': 'Yes'
    description: Allows a variable number of function parameters
  - name: WHEN clause for CREATE TRIGGER
    versions:
      '9.0': 'Yes'
    description: adds support for a boolean expression to be checked if a trigger
      should be fired or not
  Procedural Languages:
  - name: CASE in pl/pgsql
    versions:
      '8.4': 'Yes'
    description: allows the use of CASE as a conditional expression in addition to
      IF-THEN
  - name: CONTINUE statement for PL/pgSQL
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/plpgsql-control-structures.html#id-1.8.8.8.7.6
  - name: CREATE TRANSFORM
    versions:
      '9.5': 'Yes'
    url: https://www.postgresql.org/docs/9.5/static/sql-createtransform.html
  - name: DO statement for pl/perl
    versions:
      '9.0': 'Yes'
    description: allows the execution of anonymous code blocks in pl/perl
  - name: DO statement for pl/pgsql
    versions:
      '9.0': 'Yes'
    description: allows the execution of anonymous code blocks in pl/pgsql
  - name: EXCEPTION support in PL/pgSQL
    versions:
      '8.1': 'Yes'
    description: full exception handling support in pl/pgsql
  - name: EXECUTE USING in PL/pgSQL
    versions:
      '8.4': 'Yes'
    description: ''
  - name: FOREACH IN ARRAY in pl/pgsql
    versions:
      '9.1': 'Yes'
    description: This is more efficient and readable than previous methods of iterating
      through the elements of an array value.
  - name: IN/OUT/INOUT parameters for pl/pgsql and PL/SQL
    versions:
      '8.1': 'Yes'
    description: ''
  - name: Named parameters
    versions:
      '8.1': 'Yes'
    description: allows the use of named parameters in pl/pgsql
  - name: Non-superuser language creation
    versions:
      '8.3': 'Yes'
    description: database owners can now create trusted languages
  - name: pl/pgsql installed by default
    versions:
      '9.0': 'Yes'
    description: pl/pgsql is now installed by default in all newly created database
  - name: Polymorphic functions
    versions:
      '8.1': 'Yes'
    description: ''
  - name: Python 3 support for pl/python
    versions:
      '9.0': 'Yes'
    description: support for Python 3 in pl/python
  - name: 'Qualified function parameters '
    versions:
      '8.3': 'Yes'
    description: ''
  - name: Query parallelism for RETURN QUERY
    versions:
      '14': 'Yes'
    description: The `RETURN QUERY` directive in PL/pgSQL supports executing queries
      with parallelism.
  - name: RETURN QUERY in pl/pgsql
    versions:
      '8.3': 'Yes'
    description: simply way to return the result of an arbitrary query in pl/pgsql
  - name: ROWS and COST specification for functions
    versions:
      '8.3': 'Yes'
    description: CPU costs and the number of expected rows can now be specified for
      procedural functions to give better hints to the planner
  - name: Scrollable and updatable cursor support for pl/pgsql
    versions:
      '8.3': 'Yes'
    description: support for scroll- and updatable cursors in pl/pgsql functions
  - name: SQLERRM/SQLSTATE for pl/pgsql
    versions:
      '8.1': 'Yes'
    description: allows simple access to the SQL error state and the associated error
      text if an exception happened
  - name: Unicode object support in PL/python
    versions:
      '9.0': 'Yes'
    description: ''
  - name: User defined exceptions
    versions:
      '8.4': 'Yes'
    description: allows to generate user defined exceptions through RAISE
  - name: Validator function for pl/perl
    versions:
      '8.1': 'Yes'
    description: compile time validation of pl/pgsql code
  Extensions:
  - name: CREATE EXTENSION .. CASCADE
    versions:
      '9.6': 'Yes'
    description: Ability to automatically pull in required additional extensions for
      an extension dependency tree.
  - name: Extension Installation
    versions:
      '9.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/external-extensions.html
  - name: Trusted Extensions
    versions:
      '13': 'Yes'
    description: "Extensions that are marked as \"trusted\" by a superuser can subsequently\
      \ be installed via `CREATE EXTENSION` by unprivileged users. Some functions\
      \ are marked as \"trusted\" by default.\r\n\r\nFor more information, please\
      \ see the [`CREATE EXTENSION`](https://www.postgresql.org/docs/current/sql-createextension.html)\
      \ documentation."
  Internationalisation:
  - name: Built-in, platform independent immutable collation
    versions:
      '17': 'Yes'
    description: PostgreSQL has a built-in, platform independent immutable collation
      provider that supports C and C.UTF-8 collations. For more information, see [standard
      collations](https://www.postgresql.org/docs/17/collation.html#COLLATION-MANAGING-STANDARD).
  - name: casefold
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/functions-string.html#id-1.5.8.10.7.2.2.8.1.1.1
  - name: Column-level collation support
    versions:
      '9.1': 'Yes'
    description: Previously collation could only be set at database creation. Collation
      can now be set per column, domain, index, or expression, via the SQL-standard
      COLLATE clause.
  - name: Database level Collation
    versions:
      '8.4': 'Yes'
    description: Support for different collation order and character categories on
      a per database level
  - name: Default ICU collations for clusters/databases
    versions:
      '15': 'Yes'
    description: ICU collations can be set as the default collation type for an entire
      PostgreSQL clusters or for individual databases.
  - name: EUC_JIS_2004/ SHIFT_JIS_2004 support
    versions:
      '8.3': 'Yes'
    description: ''
  - name: ICU collations
    versions:
      '10': 'Yes'
    url: https://www.postgresql.org/docs/current/static/collation.html#id-1.6.10.4.5.7.5
  - name: LIKE comparisons for nondeterministic collations
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/collation.html#COLLATION-NONDETERMINISTIC
  - name: Multibyte encoding support, incl. UTF8
    versions:
      '8.1': 'Yes'
    description: full support for using UTF8 and other multibyte encodings
  - name: Multiple language support
    versions:
      '8.1': 'Yes'
    url: https://babel.postgresql.org/
  - name: Nondeterministic collations
    versions:
      '12': 'Yes'
    url: https://www.postgresql.org/docs/12/collation.html#COLLATION-NONDETERMINISTIC
  - name: pg_unicode_fast collation
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/collation.html#COLLATION-MANAGING-STANDARD
  - name: Unicode string literals and identifiers
    versions:
      '8.4': 'Yes'
    description: allows specification of unicode string literals and identifiers using
      code points
  - name: UTF8 support on Windows
    versions:
      '8.1': 'Yes'
    description: full support for using UTF8 on microsoft windows
  Client Applications:
  - name: pgbench
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/pgbench.html
  - name: pg_combinebackup
    versions:
      '17': 'Yes'
    url: https://www.postgresql.org/docs/17/app-pgcombinebackup.html
  - name: pg_createsubscriber
    versions:
      '17': 'Yes'
    url: https://www.postgresql.org/docs/17/app-pgcreatesubscriber.html
  - name: pg_prewarm
    versions:
      '9.4': 'Yes'
    url: https://www.postgresql.org/docs/9.4/static/pgprewarm.html
  - name: pg_rewind
    versions:
      '9.5': 'Yes'
    url: https://www.postgresql.org/docs/current/static/app-pgrewind.html
  - name: pg_standby
    versions:
      '8.3': 'Yes'
      '14': Obsolete
    description: Was used to initialize standbys. Removed in PostgreSQL 14.
  - name: pg_upgrade
    versions:
      '9.0': 'Yes'
    url: https://www.postgresql.org/docs/current/pgupgrade.html
  - name: pg_waldump
    versions:
      '10': 'Yes'
    url: https://www.postgresql.org/docs/current/pgwaldump.html
  - name: pg_walsummary
    versions:
      '17': 'Yes'
    url: https://www.postgresql.org/docs/17/app-pgwalsummary.html
  - name: pg_xlogdump
    versions:
      '9.3': 'Yes'
      '10': Obsolete
    description: Renamed to [`pg_waldump`](https://www.postgresql.org/docs/current/pgwaldump.html)
      in PostgreSQL 10.
  - name: psql \bind
    versions:
      '16': 'Yes'
    url: https://www.postgresql.org/docs/16/app-psql.html#APP-PSQL-META-COMMAND-BIND
  - name: psql \dconfig
    versions:
      '15': 'Yes'
    description: The [`psql`](https://www.postgresql.org/docs/15/app-psql.html) client
      includes a `\dconfig` command for inspecting and finding the values of configuration
      parameters.
  - name: psql pipeline queries
    versions:
      '18': 'Yes'
    description: "psql can issue pipeline queries using the commands\
      \ [`\\startpipeline`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-PIPELINE),\
      \ [`\\syncpipeline`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-PIPELINE),\
      \ [`\\sendpipeline`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-PIPELINE),\
      \ [`\\endpipeline`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-PIPELINE),\
      \ [`\\flushrequest`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-PIPELINE),\
      \ [`\\flush`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-PIPELINE), and\
      \ [`\\getresults`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-PIPELINE)."
  - name: psql named prepared statements
    versions:
      '18': 'Yes'
    description: "psql can parse ([`\\parse`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-PARSE)),\
      \ bind ([`\\bind_named`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-BIND-NAMED)),\
      \ and close ([`\\close_prepared`](https://www.postgresql.org/docs/18/app-psql.html#APP-PSQL-META-COMMAND-CLOSE-PREPARED))\
      \ named prepared statements."
  - name: Version aware psql
    versions:
      '8.4': 'Yes'
    description: Allows [psql](https://www.postgresql.org/docs/current/app-psql.html)
      to work with older versions of PostgreSQL.
  Additional Modules (contrib):
  - name: adminpack
    versions:
      '8.2': 'Yes'
      '17': 'No'
    url: https://www.postgresql.org/docs/current/static/adminpack.html
  - name: auth_delay
    versions:
      '9.1': 'Yes'
    url: https://www.postgresql.org/docs/current/auth-delay.html
  - name: auto_explain
    versions:
      '8.4': 'Yes'
    url: https://www.postgresql.org/docs/current/auto-explain.html
  - name: btree_gin
    versions:
      '8.4': 'Yes'
    url: https://www.postgresql.org/docs/current/static/btree-gin.html
  - name: btree_gist
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/8.3/static/btree-gist.html
  - name: citext
    versions:
      '8.4': 'Yes'
    url: https://www.postgresql.org/docs/current/citext.html
  - name: dblink
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/dblink.html
  - name: dblink asyncronous notification support
    versions:
      '9.0': 'Yes'
    description: Add the ability to retrieve asynchronous notifications using dblink
  - name: file_fdw
    versions:
      '9.1': 'Yes'
    url: https://www.postgresql.org/docs/current/file-fdw.html
  - name: fuzzystrmatch
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/fuzzystrmatch.html
  - name: hstore
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/hstore.html
  - name: intarray
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/intarray.html
  - name: isn (ISBN)
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/isn.html
  - name: KNN support for CUBE
    versions:
      '9.6': 'Yes'
    description: KNN indexing support for the CUBE module.  Supports n-dimensional
      nearest neighbor searches.
  - name: ltree
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/ltree.html
  - name: pageinspect
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/pageinspect.html
  - name: passwordcheck
    versions:
      '9.0': 'Yes'
    url: https://www.postgresql.org/docs/current/static/passwordcheck.html
  - name: pg_buffercache
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/pgbuffercache.html
  - name: pg_freespacemap
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/pgfreespacemap.html
  - name: pg_logicalinspect
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/pglogicalinspect.html
  - name: pg_overexplain
    versions:
      '18': 'Yes'
    url: https://www.postgresql.org/docs/18/pgoverexplain.html
  - name: pg_stat_statements
    versions:
      '8.4': 'Yes'
    url: https://www.postgresql.org/docs/current/static/pgstatstatements.html
  - name: pg_stat_statements improvements
    versions:
      '9.2': 'Yes'
    description: pg_stat_statements now aggregates similar queries so that such data
      is no longer confined to just prepared queries.
  - name: pgstattuple
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/pgstattuple.html
  - name: pg_trgm
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/pgtrgm.html
  - name: pg_trgm regular expressions indexing
    versions:
      '9.3': 'Yes'
    description: Regular expressions can now be indexed using GIN in [`pg_trgm`](https://www.postgresql.org/docs/current/pgtrgm.html).
  - name: pg_walinspect
    versions:
      '15': 'Yes'
    description: '[`pg_walinspect`](https://www.postgresql.org/docs/15/pgwalinspect.html)
      allows for the contents of write-ahead logs (WAL) files to be inspected from
      SQL.'
  - name: seg
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/seg.html
  - name: sepgsql
    versions:
      '9.1': 'Yes'
    url: https://www.postgresql.org/docs/current/sepgsql.html
  - name: sslinfo
    versions:
      '8.2': 'Yes'
    url: https://www.postgresql.org/docs/current/sslinfo.html
  - name: tablefunc
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/tablefunc.html
  - name: tcn
    versions:
      '9.2': 'Yes'
    url: https://www.postgresql.org/docs/current/static/tcn.html
  - name: tsearch2 compatibility wrapper
    versions:
      '8.3': 'Yes'
      '10': Obsolete
    description: compatibility wrapper for migrating from contrib/tsearch2 to the
      integrated full text search in 8.3
  - name: unaccent
    versions:
      '9.0': 'Yes'
    url: https://www.postgresql.org/docs/current/static/unaccent.html
  - name: uuid-ossp
    versions:
      '8.3': 'Yes'
    url: https://www.postgresql.org/docs/current/uuid-ossp.html
  - name: xml2
    versions:
      '8.1': 'Yes'
      '8.4': Obsolete
    url: https://www.postgresql.org/docs/current/xml2.html
  Network:
  - name: Full SSL support
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/ssl-tcp.html
  - name: IPv6 Support
    versions:
      '8.1': 'Yes'
    description: Full Support for IPv6
  - name: V2 client protocol
    versions:
      '8.1': Obsolete
      '14': 'No'
    description: obsoleted client protocol interface
  - name: V3 client protocol
    versions:
      '8.1': 'Yes'
    url: https://www.postgresql.org/docs/current/static/protocol.html
  Platforms:
  - name: Microsoft Visual C++ Support
    versions:
      '8.3': 'Yes'
    description: allows the backend to be compiled with Microsoft Visual Studio
  - name: Native Windows Port
    versions:
      '8.1': 'Yes'
    description: Native Port to the Microsoft Windows Platform
  - name: Spinlock support for the SuperH hardware platform
    versions:
      '9.0': 'Yes'
    description: s_lock support for the SuperH CPU
  - name: Sun Studio compiler on Linux
    versions:
      '8.4': 'Yes'
    description: support for compiling PostgreSQL using the Sun Studio compiler on
      Linux
  - name: Windows x64 support
    versions:
      '9.0': 'Yes'
    description: support for compiling PostgreSQL as a 64bit binary on Microsoft Windows
legacymap:  # This maps old id based URLs to the new titles and will generate redirects to the new page
  4: Heap Only Tuples (HOT)
  8: Asynchronous Commit
  11: Distributed checkpointing
  13: Scrollable and updatable cursor support for pl/pgsql
  14: ROWS and COST specification for functions
  15: SQLERRM/SQLSTATE for pl/pgsql
  19: EUC_JIS_2004/ SHIFT_JIS_2004 support
  20: Per function GUC settings
  24: Partial sort capability (top-n sorting)
  26: Synchronized sequential scanning
  27: Loadable plugin infrastructure for monitoring the planner
  28: Multiple temporary tablespaces
  29: Arrays of compound types
  34: Automatic plan invalidation
  35: Microsoft Visual C++ Support
  38: NULLs in Array
  39: Non-blocking CREATE INDEX
  42: Outer Join reordering
  45: Row-wise comparison
  49: ROLES
  50: In-memory Bitmap Indexes
  53: Shared row level locking
  55: Per user/database connection limits
  56: Temporary VIEWs
  57: SELECT ... FOR UPDATE/SHARE NOWAIT
  58: ALTER TABLE ENABLE/DISABLE TRIGGER
  59: Native Windows Port
  63: Changing column types (ALTER TABLE .. ALTER COLUMN TYPE)
  67: IPv6 Support
  68: Inlining of SQL-functions
  73: Statement level triggers
  78: RETURN QUERY in pl/pgsql
  79: 'Qualified function parameters '
  80: IN/OUT/INOUT parameters for pl/pgsql and PL/SQL
  82: Validator function for pl/perl
  83: Named parameters
  84: EXCEPTION support in PL/pgSQL
  85: Polymorphic functions
  89: pg_standby
  100: UTF8 support on Windows
  108: Security Service Provider Interface (SSPI)
  110: Index support for IS NULL
  111: Cross datatype hashing support
  112: Type modifier support
  113: Non-superuser language creation
  117: V2 client protocol
  118: Concurrent GiST indexes
  119: Multibyte encoding support, incl. UTF8
  127: tsearch2 compatibility wrapper
  128: Statement level TRUNCATE triggers
  129: EXECUTE USING in PL/pgSQL
  130: CASE in pl/pgsql
  133: Database level Collation
  135: Hashing support for DISTINCT/UNION/INTERSECT/EXCEPT
  136: RETURNS TABLE
  137: Variadic functions
  138: User defined exceptions
  139: unnest/array_agg
  140: Disk based FSM
  141: Per function statistics
  142: RETURN QUERY EXECUTE
  143: SQL standard interval handling
  144: SSL certificate validation in libpq
  146: TABLE statement
  147: SSL client certificate authentication
  151: Column level permissions
  152: Parallel restore
  154: Version aware psql
  157: Unicode string literals and identifiers
  158: Visibility Map for Vacuuming
  161: Sun Studio compiler on Linux
  162: Column level triggers
  163: Deferrable unique constraints
  165: XML, JSON and YAML output for EXPLAIN
  166: Default permissions
  167: DO statement for pl/pgsql
  168: GRANT/REVOKE ON ALL TABLES/SEQUENCES/FUNCTIONS
  170: Per user/database server configuration settings
  171: Unicode object support in PL/python
  175: dblink asyncronous notification support
  176: Spinlock support for the SuperH hardware platform
  178: WHEN clause for CREATE TRIGGER
  179: Exclusion constraints
  180: ORDER BY support within aggregates
  181: EXPLAIN (BUFFERS) support
  183: Large object access controls
  184: Search+bind mode operation for LDAP authentication
  185: pl/pgsql installed by default
  186: DO statement for pl/perl
  187: Python 3 support for pl/python
  188: Hot Standby
  189: Typed tables
  190: Payload support for LISTEN/NOTIFY
  191: Set costs specific to TABLESPACEs
  192: Streaming Replication
  193: Native RADIUS authentication
  195: Windows x64 support
  200: K-nearest neighbor GiST support
  201: Unlogged tables
  202: Hashing support for FULL OUTER JOIN, LEFT OUTER JOIN and RIGHT OUTER JOIN
  203: SSPI authentication via GSSAPI
  204: WAL Buffer auto-tuning
  205: Synchronous replication
  206: pg_basebackup tool
  207: Named restore points
  208: Writable WITH Queries (Common Table Expressions)
  209: Serializable Snapshot Isolation
  210: Column-level collation support
  212: Foreign Tables
  213: ALTER TABLE ... ADD UNIQUE/PRIMARY KEY USING INDEX
  214: Foreign keys marked as NOT VALID
  215: Triggers on views
  216: FOREACH IN ARRAY in pl/pgsql
  217: Index-only scans
  218: Multi-core scalability for read-only workloads
  219: Cascading streaming replication
  223: smallserial type
  226: pg_stat_statements improvements
  227: Background Checkpointer
  228: ALTER object IF EXISTS
  229: CREATE TABLE ... (LIKE) with foreign tables, views and composite types
  230: security_barrier option on views
  231: LATERAL clause
  232: 64-bit large objects
  233: Event triggers
  234: Updatable views
  235: Parallel pg_dump
  237: Writable Foreign Data Wrappers
  238: Foreign data wrappers
  240: Support for anonymous shared memory
  241: Custom background workers
  242: SELECT FOR NO KEY UPDATE/SELECT FOR KEY SHARE lock modes
  243: Checksum on data pages
  244: Streaming-only cascading replication
  245: pg_xlogdump
  246: pg_trgm regular expressions indexing
  247: SP-GiST indexes for range types
  248: COPY from/to STDIN/STDOUT
  249: Improved set of JSON functions and operators
  251: Materialized views with concurrent refresh
  253: krb5 authentication (without gssapi)
  254: Dynamic Background Workers
  256: GIN Index performance and size improvements
  257: Time-delayed Standbys
  258: Reduced lock levels for ALTER TABLE commands
  260: Base backup throttling
  263: FILTER clause for aggregate functions
  264: WITH CHECK clause
  266: ALTER TABLE ... SET LOGGED / UNLOGGED
  268: SKIP LOCKED clause
  270: Parallel vacuumdb jobs
  271: Foreign table inheritance
  273: Index-only scans on GiST
  275: Upsert (INSERT ... ON CONFLICT DO ...)
  276: JSONB-modifying operators and functions
  277: TABLESAMPLE clause
  279: Abbreviated Keys
  280: min_wal_size / max_wal_size
  281: Parallel query
  283: Parallel JOIN, aggregate
  284: postgres_fdw pushdown
  285: Multiple synchronous standbys
  286: remote_apply mode
  287: Frozen page map
  288: KNN support for CUBE
  289: Phrase search
  290: pg_config system view
  291: Generic WAL facility
  292: CREATE EXTENSION .. CASCADE
  295: SCRAM-SHA-256 Authentication
  296: Quorum commit for synchronous replication
  297: Parallel B-tree index scans
  298: Parallel bitmap heap scans
  299: Parallel merge joins
  300: WAL support for hash indexes
  302: Partitioning by a hash key
  303: Parallelized CREATE INDEX for B-tree indexes
  304: Support for PRIMARY KEY, FOREIGN KEY, indexes, and triggers on partitioned
    tables
  311: Default Partition
  312: UPDATE on a partition key
  313: Covering Indexes for B-trees (INCLUDE)
  314: Covering indexes for GiST (INCLUDE)
  316: K-nearest neighbor SP-GiST Support
  317: CREATE STATISTICS - multicolumn
  318: CREATE STATISTICS - most-common values (MCV) statistics
  319: Foreign Key references for partitioned tables
  320: Accelerated partition pruning
  321: Inlined WITH Queries (Common Table Expressions)
  324: GSSAPI client and server-side encryption
  325: LDAP server discovery
  326: Multifactor authentication via valid client SSL/TLS certificate
  328: Enable/Disable page checksums in an offline cluster
  329: Fractional input for "integer" values
  330: COPY FROM ... WHERE
  332: Partition pruning during query execution
  335: CREATE STATISTICS - "OR" and "IN/ANY" statistics
  336: Hash aggregation can use disk
  339: Certificate authentication with postgres_fdw
  340: Parallelized VACUUM for Indexes
  341: Inserted data can trigger autovacuum
  345: Logical replication for partitioned tables
  346: 'SQL/JSON: datetime()'
  347: Trusted Extensions
  348: EXPLAIN (WAL) support
  349: Query parallelism for RETURN QUERY
  350: Parallel query execution on remote databases
  351: Incremental sort for window functions
  352: LZ4 compression for TOAST tables
  357: Logical replication stream in-progress transactions
  358: Foreign data wrapper query parallelism
  359: Import foreign table partitions
  360: Vacuum "emergency mode"
  366: range_agg range type aggregation function
  367: regexp_count, regexp_instr, regexp_like
  368: Improved performance for sorts exceeding working memory
  369: Parallel "SELECT DISTINCT"
  370: Improved window function performance
  373: pg_basebackup server-side compression
  374: pg_basebackup client decompression
  376: pg_walinspect
  380: Logical replication column lists
  381: ALTER SUBSCRIPTION ... SKIP
  382: Logical replication subscriber can disable on error
  383: '"jsonlog" logging format'
  384: psql \dconfig
  386: Server statistics in shared memory
  387: ALTER TABLE ... SET ACCESS METHOD
  389: SECURITY INVOKER views
  390: Privileges for setting configuration parameters
  391: Logical replication publish all tables in schema
  392: UNIQUE NULLS NOT DISTINCT
  393: BEGIN ATOMIC function bodies
  394: Default ICU collations for clusters/databases
  395: SQL/JSON constructors
  396: SQL/JSON IS JSON
  397: Parallel FULL and RIGHT joins
  398: Incremental sort for SELECT DISTINCT
  399: SIMD support for x86
  400: SIMD support for ARM
  402: Logical replication from standbys
  404: Logical replication lookups with additional indexes
  407: Non-decimal integer literals
  408: Underscores (_) for thousands separators
  413: Include directives for pg_hba.conf and pg_ident.conf
  414: Regular expression matching in pg_hba.conf and pg_ident.conf
  415: Support using the client's OS trusted CA.
  417: Page freezing optimizations
  419: Radix tree memory structure for vacuum
  421: Parallelized CREATE INDEX for BRIN indexes
  425: MERGE ... RETURNING
  426: Built-in, platform independent immutable collation
  429: pg_basebackup incremental backup
  432: Logical replication slots migrate through pg_upgrade migrate
  433: COPY ... ON_ERROR
  434: EXPLAIN (SERIALIZE) support
  435: EXPLAIN (MEMORY)
  437: pg_dump, pg_dumpall, pg_restore --filter