support redshift (see activerecord-redshift-adapter)

Author: Keith Gabryelski

lib/monkey_patch_postgres.rb

@@ -29,6 +29,9 @@ def check_constraint(constraint)
   # to take advantage of these SQL builders.
   #
   class PostgreSQLAdapter < AbstractAdapter
+    def partitioned_sql_adapter(model)
+      return Partitioned::PartitionedBase::SqlAdapter.new(model)
+    end
 
     #
     # Returns the sequence name for a table's primary key or some other specified key.

lib/monkey_patch_redshift.rb

@@ -0,0 +1,114 @@
+require 'active_record'
+require 'active_record/base'
+require 'active_record/connection_adapters/abstract_adapter'
+require 'active_record/connection_adapters/redshift_adapter'
+
+#
+# Patching {ActiveRecord::ConnectionAdapters::TableDefinition} and
+# {ActiveRecord::ConnectionAdapters::PostgreSQLAdapter} to add functionality
+# needed to abstract partition specific SQL statements.
+#
+module ActiveRecord::ConnectionAdapters
+  #
+  # Patches extending the postgres adapter with new operations for managing
+  # sequences (and sets of sequence values), schemas and foreign keys.
+  # These should go into AbstractAdapter allowing any database adapter
+  # to take advantage of these SQL builders.
+  #
+  class RedshiftAdapter < AbstractAdapter
+    def partitioned_sql_adapter(model)
+      return Partitioned::PartitionedBase::RedshiftSqlAdapter.new(model)
+    end
+
+    #
+    # Returns the sequence name for a table's primary key or some other specified key.
+    #
+    # the default version strips off the schema name on the table (if it exists), as:
+    #   serial_sequence(table_name, pk || 'id').split('.').last
+    # i can't see any good reason for that -- in fact, it seems completely
+    # broken -- if you have a table public.foos and other.foos, you'll fail to
+    # get the correct schema if you fetch the default schema name from model
+    # associated with other.foos
+    #
+    def default_sequence_name(table_name, pk = nil) #:nodoc:
+      serial_sequence(table_name, pk || 'id')
+    rescue ActiveRecord::StatementInvalid => e
+      "#{table_name}_#{pk || 'id'}_seq"
+    end
+
+    #
+    # Get the next value in a sequence. Used on INSERT operation for
+    # partitioning like by_id because the ID is required before the insert
+    # so that the specific child table is known ahead of time.
+    #
+    # @param [String] sequence_name the name of the sequence to fetch the next value from
+    # @return [Integer] the value from the sequence
+    def next_sequence_value(sequence_name)
+      return execute("select nextval('#{sequence_name}')").field_values("nextval").first.to_i
+    end
+
+    #
+    # Get the some next values in a sequence.
+    #
+    # @param [String] sequence_name the name of the sequence to fetch the next values from
+    # @param [Integer] batch_size count of values.
+    # @return [Array<Integer>] an array of values from the sequence
+    def next_sequence_values(sequence_name, batch_size)
+      result = execute("select nextval('#{sequence_name}') from generate_series(1, #{batch_size})")
+      return result.field_values("nextval").map(&:to_i)
+    end
+
+    #
+    # Causes active resource to fetch the primary key for the table (using next_sequence_value())
+    # just before an insert. We need the prefetch to happen but we don't have enough information
+    # here to determine if it should happen, so Relation::insert has been modified to request of
+    # the ActiveRecord::Base derived class if it requires a prefetch.
+    #
+    # @param [String] table_name the table name to query
+    # @return [Boolean] returns true if the table should have its primary key prefetched.
+    def prefetch_primary_key?(table_name)
+      return false
+    end
+
+    #
+    # Creates a schema given a name.
+    #
+    # @param [String] name the name of the schema.
+    # @param [Hash] options ({}) options for creating a schema
+    # @option options [Boolean] :unless_exists (false) check if schema exists.
+    # @return [optional] undefined
+    def create_schema(name, options = {})
+      if options[:unless_exists]
+        return if execute("select count(*) from pg_namespace where nspname = '#{name}'").getvalue(0,0).to_i > 0
+      end
+      execute("CREATE SCHEMA #{name}")
+    end
+
+    #
+    # Drop a schema given a name.
+    #
+    # @param [String] name the name of the schema.
+    # @param [Hash] options ({}) options for dropping a schema
+    # @option options [Boolean] :if_exists (false) check if schema exists.
+    # @option options [Boolean] :cascade (false) drop dependant objects
+    # @return [optional] undefined
+    def drop_schema(name, options = {})
+      if options[:if_exists]
+        return if execute("select count(*) from pg_namespace where nspname = '#{name}'").getvalue(0,0).to_i == 0
+      end
+      execute("DROP SCHEMA #{name}#{' cascade' if options[:cascade]}")
+    end
+
+    #
+    # Add foreign key constraint to table.
+    #
+    # @param [String] referencing_table_name the name of the table containing the foreign key
+    # @param [String] referencing_field_name the name of foreign key column
+    # @param [String] referenced_table_name the name of the table referenced by the foreign key
+    # @param [String] referenced_field_name (:id) the name of the column referenced by the foreign key
+    # @return [optional] undefined
+    def add_foreign_key(referencing_table_name, referencing_field_name, referenced_table_name, referenced_field_name = :id)
+      execute("ALTER TABLE #{referencing_table_name} add foreign key (#{referencing_field_name}) references #{referenced_table_name}(#{referenced_field_name})")
+    end
+  end
+end

lib/partitioned.rb

@@ -1,5 +1,6 @@
 require 'monkey_patch_activerecord'
 require 'monkey_patch_postgres'
+require 'monkey_patch_redshift'
 
 require 'partitioned/active_record_overrides'
 require 'partitioned/partitioned_base/configurator.rb'
@@ -9,6 +10,7 @@
 require 'partitioned/partitioned_base/configurator/reader'
 require 'partitioned/partitioned_base/partition_manager'
 require 'partitioned/partitioned_base/sql_adapter'
+require 'partitioned/partitioned_base/redshift_sql_adapter'
 
 require 'partitioned/by_time_field'
 require 'partitioned/by_yearly_time_field'

lib/partitioned/partitioned_base.rb

@@ -103,7 +103,7 @@ def self.partition_manager
     #
     # @return [{SqlAdapter}] the object used to create sql statements for this partitioned model
     def self.sql_adapter
-      @sql_adapter = self::SqlAdapter.new(self) unless @sql_adapter.present?
+      @sql_adapter ||= connection.partitioned_sql_adapter(self)
       return @sql_adapter
     end