Skip to Content Skip to Search

The PostgreSQL adapter works with the native C (github.com/ged/ruby-pg) driver.

Options:

  • :host - Defaults to a Unix-domain socket in /tmp. On machines without Unix-domain sockets, the default is to connect to localhost.

  • :port - Defaults to 5432.

  • :username - Defaults to be the same as the operating system name of the user running the application.

  • :password - Password to be used if the server demands password authentication.

  • :database - Defaults to be the same as the username.

  • :schema_search_path - An optional schema search path for the connection given as a string of comma-separated schema names. This is backward-compatible with the :schema_order option.

  • :encoding - An optional client encoding that is used in a SET client_encoding TO <encoding> call on the connection.

  • :min_messages - An optional client min messages that is used in a SET client_min_messages TO <min_messages> call on the connection.

  • :variables - An optional hash of additional parameters that will be used in SET SESSION key = val calls on the connection.

  • :insert_returning - An optional boolean to control the use of RETURNING for INSERT statements defaults to true.

Any further options are used as connection parameters to libpq. See www.postgresql.org/docs/current/static/libpq-connect.html for the list of parameters.

In addition, default connection parameters of libpq can be set per environment variables. See www.postgresql.org/docs/current/static/libpq-envars.html .

Methods
A
C
D
E
I
M
N
R
S
T
U
Included Modules

Constants

ADAPTER_NAME = "PostgreSQL"
 
DEADLOCK_DETECTED = "40P01"
 
DUPLICATE_DATABASE = "42P04"
 
FOREIGN_KEY_VIOLATION = "23503"
 
LOCK_NOT_AVAILABLE = "55P03"
 
NATIVE_DATABASE_TYPES = { primary_key: "bigserial primary key", string: { name: "character varying" }, text: { name: "text" }, integer: { name: "integer", limit: 4 }, bigint: { name: "bigint" }, float: { name: "float" }, decimal: { name: "decimal" }, datetime: {}, # set dynamically based on datetime_type timestamp: { name: "timestamp" }, timestamptz: { name: "timestamptz" }, time: { name: "time" }, date: { name: "date" }, daterange: { name: "daterange" }, numrange: { name: "numrange" }, tsrange: { name: "tsrange" }, tstzrange: { name: "tstzrange" }, int4range: { name: "int4range" }, int8range: { name: "int8range" }, binary: { name: "bytea" }, boolean: { name: "boolean" }, xml: { name: "xml" }, tsvector: { name: "tsvector" }, hstore: { name: "hstore" }, inet: { name: "inet" }, cidr: { name: "cidr" }, macaddr: { name: "macaddr" }, uuid: { name: "uuid" }, json: { name: "json" }, jsonb: { name: "jsonb" }, ltree: { name: "ltree" }, citext: { name: "citext" }, point: { name: "point" }, line: { name: "line" }, lseg: { name: "lseg" }, box: { name: "box" }, path: { name: "path" }, polygon: { name: "polygon" }, circle: { name: "circle" }, bit: { name: "bit" }, bit_varying: { name: "bit varying" }, money: { name: "money" }, interval: { name: "interval" }, oid: { name: "oid" }, enum: {} # special type https://www.postgresql.org/docs/current/datatype-enum.html }
 
NOT_NULL_VIOLATION = "23502"
 
NUMERIC_VALUE_OUT_OF_RANGE = "22003"
 
QUERY_CANCELED = "57014"
 
SERIALIZATION_FAILURE = "40001"
 
UNIQUE_VIOLATION = "23505"
 
VALUE_LIMIT_VIOLATION = "22001"
 

See www.postgresql.org/docs/current/static/errcodes-appendix.html

Class Public methods

create_unlogged_tables

PostgreSQL allows the creation of “unlogged” tables, which do not record data in the PostgreSQL Write-Ahead Log. This can make the tables faster, but significantly increases the risk of data loss if the database crashes. As a result, this should not be used in production environments. If you would like all created tables to be unlogged in the test environment you can add the following line to your test.rb file:

ActiveRecord::ConnectionAdapters::PostgreSQLAdapter.create_unlogged_tables = true

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 113
class_attribute :create_unlogged_tables, default: false

datetime_type

PostgreSQL supports multiple types for DateTimes. By default, if you use datetime in migrations, Rails will translate this to a PostgreSQL “timestamp without time zone”. Change this in an initializer to use another NATIVE_DATABASE_TYPES. For example, to store DateTimes as “timestamp with time zone”:

ActiveRecord::ConnectionAdapters::PostgreSQLAdapter.datetime_type = :timestamptz

Or if you are adding a custom type:

ActiveRecord::ConnectionAdapters::PostgreSQLAdapter::NATIVE_DATABASE_TYPES[:my_custom_type] = { name: "my_custom_type_name" }
ActiveRecord::ConnectionAdapters::PostgreSQLAdapter.datetime_type = :my_custom_type

If you’re using :ruby as your config.active_record.schema_format and you change this setting, you should immediately run bin/rails db:migrate to update the types in your schema.rb.

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 131
class_attribute :datetime_type, default: :timestamp

dbconsole(config, options = {})

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 82
def dbconsole(config, options = {})
  pg_config = config.configuration_hash

  ENV["PGUSER"]         = pg_config[:username] if pg_config[:username]
  ENV["PGHOST"]         = pg_config[:host] if pg_config[:host]
  ENV["PGPORT"]         = pg_config[:port].to_s if pg_config[:port]
  ENV["PGPASSWORD"]     = pg_config[:password].to_s if pg_config[:password] && options[:include_password]
  ENV["PGSSLMODE"]      = pg_config[:sslmode].to_s if pg_config[:sslmode]
  ENV["PGSSLCERT"]      = pg_config[:sslcert].to_s if pg_config[:sslcert]
  ENV["PGSSLKEY"]       = pg_config[:sslkey].to_s if pg_config[:sslkey]
  ENV["PGSSLROOTCERT"]  = pg_config[:sslrootcert].to_s if pg_config[:sslrootcert]
  if pg_config[:variables]
    ENV["PGOPTIONS"] = pg_config[:variables].filter_map do |name, value|
      "-c #{name}=#{value.to_s.gsub(/[ \\]/, '\\\\\0')}" unless value == ":default" || value == :default
    end.join(" ")
  end
  find_cmd_and_exec("psql", config.database)
end

new(...)

Initializes and connects a PostgreSQL adapter.

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 307
def initialize(...)
  super

  conn_params = @config.compact

  # Map ActiveRecords param names to PGs.
  conn_params[:user] = conn_params.delete(:username) if conn_params[:username]
  conn_params[:dbname] = conn_params.delete(:database) if conn_params[:database]

  # Forward only valid config params to PG::Connection.connect.
  valid_conn_param_keys = PG::Connection.conndefaults_hash.keys + [:requiressl]
  conn_params.slice!(*valid_conn_param_keys)

  @connection_parameters = conn_params

  @max_identifier_length = nil
  @type_map = nil
  @raw_connection = nil
  @notice_receiver_sql_warnings = []

  @use_insert_returning = @config.key?(:insert_returning) ? self.class.type_cast_config_to_boolean(@config[:insert_returning]) : true
end

new_client(conn_params)

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 66
def new_client(conn_params)
  PG.connect(**conn_params)
rescue ::PG::Error => error
  if conn_params && conn_params[:dbname] == "postgres"
    raise ActiveRecord::ConnectionNotEstablished, error.message
  elsif conn_params && conn_params[:dbname] && error.message.include?(conn_params[:dbname])
    raise ActiveRecord::NoDatabaseError.db_error(conn_params[:dbname])
  elsif conn_params && conn_params[:user] && error.message.include?(conn_params[:user])
    raise ActiveRecord::DatabaseConnectionError.username_error(conn_params[:user])
  elsif conn_params && conn_params[:host] && error.message.include?(conn_params[:host])
    raise ActiveRecord::DatabaseConnectionError.hostname_error(conn_params[:host])
  else
    raise ActiveRecord::ConnectionNotEstablished, error.message
  end
end

Instance Public methods

active?()

Is this connection alive and ready for queries?

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 331
def active?
  @lock.synchronize do
    return false unless @raw_connection
    @raw_connection.query ";"
  end
  true
rescue PG::Error
  false
end

create_enum(name, values, **options)

Given a name and an array of values, creates an enum type.

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 509
      def create_enum(name, values, **options)
        sql_values = values.map { |s| quote(s) }.join(", ")
        scope = quoted_scope(name)
        query = <<~SQL
          DO $$
          BEGIN
              IF NOT EXISTS (
                SELECT 1
                FROM pg_type t
                JOIN pg_namespace n ON t.typnamespace = n.oid
                WHERE t.typname = #{scope[:name]}
                  AND n.nspname = #{scope[:schema]}
              ) THEN
                  CREATE TYPE #{quote_table_name(name)} AS ENUM (#{sql_values});
              END IF;
          END
          $$;
        SQL
        exec_query(query)
      end

disable_extension(name, force: false)

Removes an extension from the database.

:force

Set to :cascade to drop dependent objects as well. Defaults to false.

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 468
def disable_extension(name, force: false)
  exec_query("DROP EXTENSION IF EXISTS \"#{name}\"#{' CASCADE' if force == :cascade}").tap {
    reload_type_map
  }
end

disconnect!()

Disconnects from the database if already connected. Otherwise, this method does nothing.

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 368
def disconnect!
  @lock.synchronize do
    super
    @raw_connection&.close rescue nil
    @raw_connection = nil
  end
end

drop_enum(name, values = nil, **options)

Drops an enum type.

If the if_exists: true option is provided, the enum is dropped only if it exists. Otherwise, if the enum doesn’t exist, an error is raised.

The values parameter will be ignored if present. It can be helpful to provide this in a migration’s change method so it can be reverted. In that case, values will be used by create_enum.

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 539
      def drop_enum(name, values = nil, **options)
        query = <<~SQL
          DROP TYPE#{' IF EXISTS' if options[:if_exists]} #{quote_table_name(name)};
        SQL
        exec_query(query)
      end

enable_extension(name, **)

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 455
def enable_extension(name, **)
  schema, name = name.to_s.split(".").values_at(-2, -1)
  sql = +"CREATE EXTENSION IF NOT EXISTS \"#{name}\""
  sql << " SCHEMA #{schema}" if schema

  exec_query(sql).tap { reload_type_map }
end

enum_types()

Returns a list of defined enum types, and their values.

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 487
      def enum_types
        query = <<~SQL
          SELECT
            type.typname AS name,
            type.OID AS oid,
            n.nspname AS schema,
            string_agg(enum.enumlabel, ',' ORDER BY enum.enumsortorder) AS value
          FROM pg_enum AS enum
          JOIN pg_type AS type ON (type.oid = enum.enumtypid)
          JOIN pg_namespace n ON type.typnamespace = n.oid
          GROUP BY type.OID, n.nspname, type.typname;
        SQL

        exec_query(query, "SCHEMA", allow_retry: true, uses_transaction: false).cast_values.each_with_object({}) do |row, memo|
          name, schema = row[0], row[2]
          schema = nil if schema == current_schema
          full_name = [schema, name].compact.join(".")
          memo[full_name] = row.last
        end.to_a
      end

extension_available?(name)

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 474
def extension_available?(name)
  query_value("SELECT true FROM pg_available_extensions WHERE name = #{quote(name)}", "SCHEMA")
end

extension_enabled?(name)

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 478
def extension_enabled?(name)
  query_value("SELECT installed_version IS NOT NULL FROM pg_available_extensions WHERE name = #{quote(name)}", "SCHEMA")
end

extensions()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 482
def extensions
  exec_query("SELECT extname FROM pg_extension", "SCHEMA", allow_retry: true, uses_transaction: false).cast_values
end

index_algorithms()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 278
def index_algorithms
  { concurrently: "CONCURRENTLY" }
end

max_identifier_length()

Returns the configured supported identifier length supported by PostgreSQL

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 547
def max_identifier_length
  @max_identifier_length ||= query_value("SHOW max_identifier_length", "SCHEMA").to_i
end

reset!()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 353
def reset!
  @lock.synchronize do
    return connect! unless @raw_connection

    unless @raw_connection.transaction_status == ::PG::PQTRANS_IDLE
      @raw_connection.query "ROLLBACK"
    end
    @raw_connection.query "DISCARD ALL"

    super
  end
end

session_auth=(user)

Set the authorized user for this session

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 560
def session_auth=(user)
  clear_cache!
  internal_execute("SET SESSION AUTHORIZATION #{user}", nil, uses_transaction: true)
end

set_standard_conforming_strings()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 394
def set_standard_conforming_strings
  internal_execute("SET standard_conforming_strings = on")
end

supports_advisory_locks?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 402
def supports_advisory_locks?
  true
end

supports_bulk_alter?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 187
def supports_bulk_alter?
  true
end

supports_check_constraints?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 219
def supports_check_constraints?
  true
end

supports_comments?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 251
def supports_comments?
  true
end

supports_common_table_expressions?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 433
def supports_common_table_expressions?
  true
end

supports_datetime_with_precision?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 243
def supports_datetime_with_precision?
  true
end

supports_ddl_transactions?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 398
def supports_ddl_transactions?
  true
end

supports_deferrable_constraints?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 235
def supports_deferrable_constraints?
  true
end

supports_exclusion_constraints?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 223
def supports_exclusion_constraints?
  true
end

supports_explain?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 406
def supports_explain?
  true
end

supports_expression_index?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 207
def supports_expression_index?
  true
end

supports_extensions?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 410
def supports_extensions?
  true
end

supports_foreign_keys?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 215
def supports_foreign_keys?
  true
end

supports_foreign_tables?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 418
def supports_foreign_tables?
  true
end

supports_index_include?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 203
def supports_index_include?
  database_version >= 11_00_00 # >= 11.0
end

supports_index_sort_order?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 191
def supports_index_sort_order?
  true
end

supports_insert_conflict_target?()

Alias for: supports_insert_on_conflict?

supports_insert_on_conflict?()

Also aliased as: supports_insert_on_duplicate_skip?, supports_insert_on_duplicate_update?, supports_insert_conflict_target?

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 267
def supports_insert_on_conflict?
  database_version >= 9_05_00 # >= 9.5
end

supports_insert_on_duplicate_skip?()

Alias for: supports_insert_on_conflict?

supports_insert_on_duplicate_update?()

Alias for: supports_insert_on_conflict?

supports_insert_returning?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 263
def supports_insert_returning?
  true
end

supports_json?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 247
def supports_json?
  true
end

supports_lazy_transactions?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 437
def supports_lazy_transactions?
  true
end

supports_materialized_views?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 414
def supports_materialized_views?
  true
end

supports_optimizer_hints?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 426
def supports_optimizer_hints?
  unless defined?(@has_pg_hint_plan)
    @has_pg_hint_plan = extension_available?("pg_hint_plan")
  end
  @has_pg_hint_plan
end

supports_partial_index?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 199
def supports_partial_index?
  true
end

supports_partitioned_indexes?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 195
def supports_partitioned_indexes?
  database_version >= 11_00_00 # >= 11.0
end

supports_pgcrypto_uuid?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 422
def supports_pgcrypto_uuid?
  database_version >= 9_04_00 # >= 9.4
end

supports_restart_db_transaction?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 259
def supports_restart_db_transaction?
  database_version >= 12_00_00 # >= 12.0
end

supports_savepoints?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 255
def supports_savepoints?
  true
end

supports_transaction_isolation?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 211
def supports_transaction_isolation?
  true
end

supports_unique_keys?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 227
def supports_unique_keys?
  true
end

supports_validate_constraints?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 231
def supports_validate_constraints?
  true
end

supports_views?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 239
def supports_views?
  true
end

supports_virtual_columns?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 274
def supports_virtual_columns?
  database_version >= 12_00_00 # >= 12.0
end

table_name_length()

Returns the maximum length of a table name.

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 552
def table_name_length
  # PostgreSQL automatically creates an index for PRIMARY KEY with name consisting of
  # truncated table name and "_pkey" suffix fitting into max_identifier_length number of characters.
  # We allow smaller table names to be able to correctly rename this index when renaming the table.
  max_identifier_length - "_pkey".length
end

use_insert_returning?()

Source: show | on GitHub 

# File activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb, line 565
def use_insert_returning?
  @use_insert_returning
end