Class: ActiveRecord::ConnectionAdapters::OracleEnhancedAdapter
- Inherits:
-
AbstractAdapter
- Object
- AbstractAdapter
- ActiveRecord::ConnectionAdapters::OracleEnhancedAdapter
- Defined in:
- lib/active_record/connection_adapters/oracle_enhanced_adapter.rb
Overview
Oracle enhanced adapter will work with both Ruby 1.8/1.9 ruby-oci8 gem (which provides interface to Oracle OCI client) or with JRuby and Oracle JDBC driver.
It should work with Oracle 9i, 10g and 11g databases. Limited set of functionality should work on Oracle 8i as well but several features rely on newer functionality in Oracle database.
Usage notes:
-
Key generation assumes a “$table_name_seq” sequence is available for all tables; the sequence name can be changed using ActiveRecord::Base.set_sequence_name. When using Migrations, these sequences are created automatically.
-
Oracle uses DATE or TIMESTAMP datatypes for both dates and times. Consequently some hacks are employed to map data back to Date or Time in Ruby. Timezones and sub-second precision on timestamps are not supported.
-
Default values that are functions (such as “SYSDATE”) are not supported. This is a restriction of the way ActiveRecord supports default values.
Required parameters:
-
:username
-
:password
-
:database
- either TNS alias or connection string for OCI client or database name in JDBC connection string
Optional parameters:
-
:host
- host name for JDBC connection, defaults to “localhost” -
:port
- port number for JDBC connection, defaults to 1521 -
:privilege
- set “SYSDBA” if you want to connect with this privilege -
:allow_concurrency
- set to “true” if non-blocking mode should be enabled (just for OCI client) -
:prefetch_rows
- how many rows should be fetched at one time to increase performance, defaults to 100 -
:cursor_sharing
- cursor sharing mode to minimize amount of unique statements, defaults to “similar” -
:nls_length_semantics
- semantics of size of VARCHAR2 and CHAR columns, defaults to “CHAR” (meaning that size specifies number of characters and not bytes)
Direct Known Subclasses
Constant Summary collapse
- IDENTIFIER_MAX_LENGTH =
maximum length of Oracle identifiers
30
Instance Attribute Summary collapse
-
#auto_retry ⇒ Object
If SQL statement fails due to lost connection then reconnect and retry SQL statement if autocommit mode is enabled.
Class Method Summary collapse
-
.boolean_to_string(bool) ⇒ Object
How boolean value should be quoted to String.
-
.is_boolean_column?(name, field_type, table_name = nil) ⇒ Boolean
Check column name to identify if it is boolean (and not String) column.
-
.is_date_column?(name, table_name = nil) ⇒ Boolean
Check column name to identify if it is Date (and not Time) column.
-
.is_integer_column?(name, table_name = nil) ⇒ Boolean
Check column name to identify if it is Integer (and not Float or BigDecimal) column.
-
.valid_table_name?(name) ⇒ Boolean
unescaped table name should start with letter and contain letters, digits, _, $ or # can be prefixed with schema name CamelCase table names should be quoted.
Instance Method Summary collapse
-
#active? ⇒ Boolean
Returns true if the connection is active.
-
#adapter_name ⇒ Object
:nodoc:.
-
#add_column(table_name, column_name, type, options = {}) ⇒ Object
:nodoc:.
-
#add_column_options!(sql, options) ⇒ Object
:nodoc:.
-
#add_comment(table_name, column_name, comment) ⇒ Object
:nodoc:.
-
#add_index(table_name, column_name, options = {}) ⇒ Object
clear cached indexes when adding new index.
-
#add_limit_offset!(sql, options) ⇒ Object
:nodoc:.
-
#add_order_by_for_association_limiting!(sql, options) ⇒ Object
ORDER BY clause for the passed order option.
-
#add_table_comment(table_name, comment) ⇒ Object
:nodoc:.
-
#begin_db_transaction ⇒ Object
:nodoc:.
-
#change_column(table_name, column_name, type, options = {}) ⇒ Object
:nodoc:.
-
#change_column_default(table_name, column_name, default) ⇒ Object
:nodoc:.
-
#change_column_null(table_name, column_name, null, default = nil) ⇒ Object
:nodoc:.
-
#clear_types_for_columns ⇒ Object
:nodoc:.
-
#column_comment(table_name, column_name) ⇒ Object
:nodoc:.
-
#columns(table_name, name = nil) ⇒ Object
:nodoc:.
-
#commit_db_transaction ⇒ Object
:nodoc:.
-
#create_table(name, options = {}, &block) ⇒ Object
Additional options for
create_table
method in migration files. -
#current_database ⇒ Object
SCHEMA STATEMENTS ========================================.
-
#default_sequence_name(table, column) ⇒ Object
:nodoc:.
-
#default_sequence_start_value ⇒ Object
:singleton-method: Specify default sequence start with value (by default 10000 if not explicitly set), e.g.:.
-
#disconnect! ⇒ Object
Disconnects from the database.
-
#distinct(columns, order_by) ⇒ Object
SELECT DISTINCT clause for a given set of columns and a given ORDER BY clause.
-
#drop_table(name, options = {}) ⇒ Object
:nodoc:.
-
#emulate_booleans ⇒ Object
:singleton-method: By default, the OracleEnhancedAdapter will consider all columns of type
NUMBER(1)
as boolean. -
#emulate_booleans_from_strings ⇒ Object
:singleton-method: If you wish that CHAR(1), VARCHAR2(1) columns or VARCHAR2 columns with FLAG or YN at the end of their name are typecasted to booleans then you can add the following line to your initializer file:.
-
#emulate_dates ⇒ Object
:singleton-method: By default, the OracleEnhancedAdapter will typecast all columns of type
DATE
to Time or DateTime (if value is out of Time value range) value. -
#emulate_dates_by_column_name ⇒ Object
:singleton-method: By default, the OracleEnhancedAdapter will typecast all columns of type
DATE
to Time or DateTime (if value is out of Time value range) value. -
#emulate_integers_by_column_name ⇒ Object
:singleton-method: By default, the OracleEnhancedAdapter will typecast all columns of type
NUMBER
(without precision or scale) to Float or BigDecimal value. -
#execute(sql, name = nil) ⇒ Object
Executes a SQL statement.
-
#get_type_for_column(table_name, column_name) ⇒ Object
:nodoc:.
-
#ignore_table_columns(table_name, *args) ⇒ Object
set ignored columns for table.
-
#ignored_table_columns(table_name) ⇒ Object
:nodoc:.
-
#index_name(table_name, options) ⇒ Object
returned shortened index name if default is too large.
-
#indexes(table_name, name = nil) ⇒ Object
This method selects all indexes at once, and caches them in a class variable.
-
#initialize(connection, logger = nil) ⇒ OracleEnhancedAdapter
constructor
:nodoc:.
-
#insert_fixture(fixture, table_name) ⇒ Object
Inserts the given fixture into the table.
-
#is_date_column?(name, table_name = nil) ⇒ Boolean
instance method uses at first check if column type defined at class level.
-
#lob_order_by_expression(klass, order) ⇒ Object
change LOB column for ORDER BY clause just first 100 characters are taken for ordering.
-
#native_database_types ⇒ Object
:nodoc:.
-
#next_sequence_value(sequence_name) ⇒ Object
Returns the next sequence value from a sequence generator.
-
#pk_and_sequence_for(table_name) ⇒ Object
Find a table’s primary key and sequence.
-
#prefetch_primary_key?(table_name = nil) ⇒ Boolean
Returns true for Oracle adapter (since Oracle requires primary key values to be pre-fetched before insert).
-
#quote(value, column = nil) ⇒ Object
:nodoc:.
-
#quote_column_name(name) ⇒ Object
QUOTING ==================================================.
-
#quote_date_with_to_date(value) ⇒ Object
:nodoc:.
-
#quote_string(s) ⇒ Object
:nodoc:.
-
#quote_table_name(name) ⇒ Object
:nodoc:.
-
#quote_timestamp_with_to_timestamp(value) ⇒ Object
:nodoc:.
-
#quoted_false ⇒ Object
:nodoc:.
-
#quoted_true ⇒ Object
:nodoc:.
-
#raw_connection ⇒ Object
return raw OCI8 or JDBC connection.
-
#reconnect! ⇒ Object
Reconnects to the database.
-
#remove_column(table_name, column_name) ⇒ Object
:nodoc:.
-
#remove_index(table_name, options = {}) ⇒ Object
clear cached indexes when removing index.
-
#rename_column(table_name, column_name, new_column_name) ⇒ Object
:nodoc:.
-
#rename_table(name, new_name) ⇒ Object
:nodoc:.
-
#rollback_db_transaction ⇒ Object
:nodoc:.
-
#select_rows(sql, name = nil) ⇒ Object
Returns an array of arrays containing the field values.
-
#set_type_for_columns(table_name, column_type, *args) ⇒ Object
set explicit type for specified table columns.
-
#structure_drop ⇒ Object
:nodoc:.
-
#structure_dump ⇒ Object
:nodoc:.
-
#structure_dump_db_stored_code ⇒ Object
Extract all stored procedures, packages, synonyms and views.
-
#supports_migrations? ⇒ Boolean
:nodoc:.
-
#table_alias_length ⇒ Object
:nodoc:.
-
#table_comment(table_name) ⇒ Object
:nodoc:.
-
#tables(name = nil) ⇒ Object
:nodoc:.
-
#write_lobs(table_name, klass, attributes) ⇒ Object
Writes LOB values from attributes, as indicated by the LOB columns of klass.
Constructor Details
#initialize(connection, logger = nil) ⇒ OracleEnhancedAdapter
:nodoc:
366 367 368 369 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 366 def initialize(connection, logger = nil) #:nodoc: super @quoted_column_names, @quoted_table_names = {}, {} end |
Instance Attribute Details
#auto_retry ⇒ Object
If SQL statement fails due to lost connection then reconnect and retry SQL statement if autocommit mode is enabled. By default this functionality is disabled.
491 492 493 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 491 def auto_retry @auto_retry end |
Class Method Details
.boolean_to_string(bool) ⇒ Object
How boolean value should be quoted to String. Used if emulate_booleans_from_strings
option is set to true
.
350 351 352 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 350 def self.boolean_to_string(bool) bool ? "Y" : "N" end |
.is_boolean_column?(name, field_type, table_name = nil) ⇒ Boolean
Check column name to identify if it is boolean (and not String) column. Is used if emulate_booleans_from_strings
option is set to true
. Override this method definition in initializer file if different boolean column recognition is needed.
343 344 345 346 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 343 def self.is_boolean_column?(name, field_type, table_name = nil) return true if ["CHAR(1)","VARCHAR2(1)"].include?(field_type) field_type =~ /^VARCHAR2/ && (name =~ /_flag$/i || name =~ /_yn$/i) end |
.is_date_column?(name, table_name = nil) ⇒ Boolean
Check column name to identify if it is Date (and not Time) column. Is used if emulate_dates_by_column_name
option is set to true
. Override this method definition in initializer file if different Date column recognition is needed.
297 298 299 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 297 def self.is_date_column?(name, table_name = nil) name =~ /(^|_)date(_|$)/i end |
.is_integer_column?(name, table_name = nil) ⇒ Boolean
Check column name to identify if it is Integer (and not Float or BigDecimal) column. Is used if emulate_integers_by_column_name
option is set to true
. Override this method definition in initializer file if different Integer column recognition is needed.
327 328 329 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 327 def self.is_integer_column?(name, table_name = nil) name =~ /((^|_)id|_(pk|fk))$/i end |
.valid_table_name?(name) ⇒ Boolean
unescaped table name should start with letter and contain letters, digits, _, $ or # can be prefixed with schema name CamelCase table names should be quoted
421 422 423 424 425 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 421 def self.valid_table_name?(name) #:nodoc: name = name.to_s name =~ /^([A-Za-z_0-9]+\.)?[a-z][a-z_0-9\$#]*$/ || name =~ /^([A-Za-z_0-9]+\.)?[A-Z][A-Z_0-9\$#]*$/ ? true : false end |
Instance Method Details
#active? ⇒ Boolean
Returns true if the connection is active.
505 506 507 508 509 510 511 512 513 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 505 def active? #:nodoc: # Pings the connection to check if it's still good. Note that an # #active? method is also available, but that simply returns the # last known state, which isn't good enough if the connection has # gone stale since the last use. @connection.ping rescue OracleEnhancedConnectionException false end |
#adapter_name ⇒ Object
:nodoc:
371 372 373 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 371 def adapter_name #:nodoc: 'OracleEnhanced' end |
#add_column(table_name, column_name, type, options = {}) ⇒ Object
:nodoc:
888 889 890 891 892 893 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 888 def add_column(table_name, column_name, type, = {}) #:nodoc: add_column_sql = "ALTER TABLE #{quote_table_name(table_name)} ADD #{quote_column_name(column_name)} #{type_to_sql(type, options[:limit], options[:precision], options[:scale])}" [:type] = type (add_column_sql, ) execute(add_column_sql) end |
#add_column_options!(sql, options) ⇒ Object
:nodoc:
1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 1067 def (sql, ) #:nodoc: type = [:type] || ((column = [:column]) && column.type) type = type && type.to_sym # handle case of defaults for CLOB columns, which would otherwise get "quoted" incorrectly if () if type == :text sql << " DEFAULT #{quote(options[:default])}" else # from abstract adapter sql << " DEFAULT #{quote(options[:default], options[:column])}" end end # must explicitly add NULL or NOT NULL to allow change_column to work on migrations if [:null] == false sql << " NOT NULL" elsif [:null] == true sql << " NULL" unless type == :primary_key end end |
#add_comment(table_name, column_name, comment) ⇒ Object
:nodoc:
932 933 934 935 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 932 def add_comment(table_name, column_name, comment) #:nodoc: return if comment.blank? execute "COMMENT ON COLUMN #{quote_table_name(table_name)}.#{column_name} IS '#{comment}'" end |
#add_index(table_name, column_name, options = {}) ⇒ Object
clear cached indexes when adding new index
856 857 858 859 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 856 def add_index(table_name, column_name, = {}) #:nodoc: self.all_schema_indexes = nil super end |
#add_limit_offset!(sql, options) ⇒ Object
:nodoc:
568 569 570 571 572 573 574 575 576 577 578 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 568 def add_limit_offset!(sql, ) #:nodoc: # added to_i for limit and offset to protect from SQL injection offset = ([:offset] || 0).to_i if limit = [:limit] limit = limit.to_i sql.replace "select * from (select raw_sql_.*, rownum raw_rnum_ from (#{sql}) raw_sql_ where rownum <= #{offset+limit}) where raw_rnum_ > #{offset}" elsif offset > 0 sql.replace "select * from (select raw_sql_.*, rownum raw_rnum_ from (#{sql}) raw_sql_) where raw_rnum_ > #{offset}" end end |
#add_order_by_for_association_limiting!(sql, options) ⇒ Object
ORDER BY clause for the passed order option.
Uses column aliases as defined by #distinct.
1113 1114 1115 1116 1117 1118 1119 1120 1121 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 1113 def add_order_by_for_association_limiting!(sql, ) #:nodoc: return sql if [:order].blank? order = [:order].split(',').collect { |s| s.strip }.reject(&:blank?) order.map! {|s| $1 if s =~ / (.*)/} order = order.zip((0...order.size).to_a).map { |s,i| "alias_#{i}__ #{s}" }.join(', ') sql << " ORDER BY #{order}" end |
#add_table_comment(table_name, comment) ⇒ Object
:nodoc:
937 938 939 940 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 937 def add_table_comment(table_name, comment) #:nodoc: return if comment.blank? execute "COMMENT ON TABLE #{quote_table_name(table_name)} IS '#{comment}'" end |
#begin_db_transaction ⇒ Object
:nodoc:
552 553 554 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 552 def begin_db_transaction #:nodoc: @connection.autocommit = false end |
#change_column(table_name, column_name, type, options = {}) ⇒ Object
:nodoc:
909 910 911 912 913 914 915 916 917 918 919 920 921 922 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 909 def change_column(table_name, column_name, type, = {}) #:nodoc: column = column_for(table_name, column_name) # remove :null option if its value is the same as current column definition # otherwise Oracle will raise error if .has_key?(:null) && [:null] == column.null [:null] = nil end change_column_sql = "ALTER TABLE #{quote_table_name(table_name)} MODIFY #{quote_column_name(column_name)} #{type_to_sql(type, options[:limit], options[:precision], options[:scale])}" [:type] = type (change_column_sql, ) execute(change_column_sql) end |
#change_column_default(table_name, column_name, default) ⇒ Object
:nodoc:
895 896 897 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 895 def change_column_default(table_name, column_name, default) #:nodoc: execute "ALTER TABLE #{quote_table_name(table_name)} MODIFY #{quote_column_name(column_name)} DEFAULT #{quote(default)}" end |
#change_column_null(table_name, column_name, null, default = nil) ⇒ Object
:nodoc:
899 900 901 902 903 904 905 906 907 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 899 def change_column_null(table_name, column_name, null, default = nil) #:nodoc: column = column_for(table_name, column_name) unless null || default.nil? execute("UPDATE #{quote_table_name(table_name)} SET #{quote_column_name(column_name)}=#{quote(default)} WHERE #{quote_column_name(column_name)} IS NULL") end change_column table_name, column_name, column.sql_type, :null => null end |
#clear_types_for_columns ⇒ Object
:nodoc:
722 723 724 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 722 def clear_types_for_columns #:nodoc: @table_column_type = nil end |
#column_comment(table_name, column_name) ⇒ Object
:nodoc:
951 952 953 954 955 956 957 958 959 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 951 def column_comment(table_name, column_name) #:nodoc: (owner, table_name) = @connection.describe(table_name) select_value "SELECT comments FROM all_col_comments\nWHERE owner = '\#{owner}'\nAND table_name = '\#{table_name}'\nAND column_name = '\#{column_name.upcase}'\n" end |
#columns(table_name, name = nil) ⇒ Object
:nodoc:
726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 726 def columns(table_name, name = nil) #:nodoc: # get ignored_columns by original table name ignored_columns = ignored_table_columns(table_name) (owner, desc_table_name) = @connection.describe(table_name) table_cols = "select column_name as name, data_type as sql_type, data_default, nullable,\ndecode(data_type, 'NUMBER', data_precision,\n'FLOAT', data_precision,\n'VARCHAR2', decode(char_used, 'C', char_length, data_length),\n'CHAR', decode(char_used, 'C', char_length, data_length),\nnull) as limit,\ndecode(data_type, 'NUMBER', data_scale, null) as scale\nfrom all_tab_columns\nwhere owner = '\#{owner}'\nand table_name = '\#{desc_table_name}'\norder by column_id\n" # added deletion of ignored columns select_all(table_cols, name).delete_if do |row| ignored_columns && ignored_columns.include?(row['name'].downcase) end.map do |row| limit, scale = row['limit'], row['scale'] if limit || scale row['sql_type'] << "(#{(limit || 38).to_i}" + ((scale = scale.to_i) > 0 ? ",#{scale})" : ")") end # clean up odd default spacing from Oracle if row['data_default'] row['data_default'].sub!(/^(.*?)\s*$/, '\1') row['data_default'].sub!(/^'(.*)'$/, '\1') row['data_default'] = nil if row['data_default'] =~ /^(null|empty_[bc]lob\(\))$/i end OracleEnhancedColumn.new(oracle_downcase(row['name']), row['data_default'], row['sql_type'], row['nullable'] == 'Y', # pass table name for table specific column definitions table_name, # pass column type if specified in class definition get_type_for_column(table_name, oracle_downcase(row['name']))) end end |
#commit_db_transaction ⇒ Object
:nodoc:
556 557 558 559 560 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 556 def commit_db_transaction #:nodoc: @connection.commit ensure @connection.autocommit = true end |
#create_table(name, options = {}, &block) ⇒ Object
Additional options for create_table
method in migration files.
You can specify individual starting value in table creation migration file, e.g.:
create_table :users, :sequence_start_value => 100 do |t|
# ...
end
You can also specify other sequence definition additional parameters, e.g.:
create_table :users, :sequence_start_value =>
It is possible to add table and column comments in table creation migration files:
create_table :employees, :comment =>
801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 801 def create_table(name, = {}, &block) create_sequence = [:id] != false column_comments = {} super(name, ) do |t| # store that primary key was defined in create_table block unless create_sequence class <<t attr_accessor :create_sequence def primary_key(*args) self.create_sequence = true super(*args) end end end # store column comments class <<t attr_accessor :column_comments def column(name, type, = {}) if [:comment] self.column_comments ||= {} self.column_comments[name] = [:comment] end super(name, type, ) end end result = block.call(t) if block create_sequence = create_sequence || t.create_sequence column_comments = t.column_comments if t.column_comments end seq_name = [:sequence_name] || quote_table_name("#{name}_seq") seq_start_value = [:sequence_start_value] || default_sequence_start_value execute "CREATE SEQUENCE #{seq_name} START WITH #{seq_start_value}" if create_sequence add_table_comment name, [:comment] column_comments.each do |column_name, comment| add_comment name, column_name, comment end end |
#current_database ⇒ Object
SCHEMA STATEMENTS ========================================
see: abstract/schema_statements.rb
649 650 651 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 649 def current_database #:nodoc: select_one("select sys_context('userenv','db_name') db from dual")["db"] end |
#default_sequence_name(table, column) ⇒ Object
:nodoc:
586 587 588 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 586 def default_sequence_name(table, column) #:nodoc: quote_table_name("#{table}_seq") end |
#default_sequence_start_value ⇒ Object
:singleton-method: Specify default sequence start with value (by default 10000 if not explicitly set), e.g.:
ActiveRecord::ConnectionAdapters::OracleEnhancedAdapter.default_sequence_start_value = 1
778 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 778 cattr_accessor :default_sequence_start_value |
#disconnect! ⇒ Object
Disconnects from the database.
523 524 525 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 523 def disconnect! #:nodoc: @connection.logoff rescue nil end |
#distinct(columns, order_by) ⇒ Object
SELECT DISTINCT clause for a given set of columns and a given ORDER BY clause.
Oracle requires the ORDER BY columns to be in the SELECT list for DISTINCT queries. However, with those columns included in the SELECT DISTINCT list, you won’t actually get a distinct list of the column you want (presuming the column has duplicates with multiple values for the ordered-by columns. So we use the FIRST_VALUE function to get a single (first) value for each column, effectively making every row the same.
distinct("posts.id", "posts.created_at desc")
1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 1097 def distinct(columns, order_by) #:nodoc: return "DISTINCT #{columns}" if order_by.blank? # construct a valid DISTINCT clause, ie. one that includes the ORDER BY columns, using # FIRST_VALUE such that the inclusion of these columns doesn't invalidate the DISTINCT order_columns = order_by.split(',').map { |s| s.strip }.reject(&:blank?) order_columns = order_columns.zip((0...order_columns.size).to_a).map do |c, i| "FIRST_VALUE(#{c.split.first}) OVER (PARTITION BY #{columns} ORDER BY #{c}) AS alias_#{i}__" end sql = "DISTINCT #{columns}, " sql << order_columns * ", " end |
#drop_table(name, options = {}) ⇒ Object
:nodoc:
849 850 851 852 853 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 849 def drop_table(name, = {}) #:nodoc: super(name) seq_name = [:sequence_name] || quote_table_name("#{name}_seq") execute "DROP SEQUENCE #{seq_name}" rescue nil end |
#emulate_booleans ⇒ Object
:singleton-method: By default, the OracleEnhancedAdapter will consider all columns of type NUMBER(1)
as boolean. If you wish to disable this emulation you can add the following line to your initializer file:
ActiveRecord::ConnectionAdapters::OracleEnhancedAdapter.emulate_booleans = false
263 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 263 cattr_accessor :emulate_booleans |
#emulate_booleans_from_strings ⇒ Object
:singleton-method: If you wish that CHAR(1), VARCHAR2(1) columns or VARCHAR2 columns with FLAG or YN at the end of their name are typecasted to booleans then you can add the following line to your initializer file:
ActiveRecord::ConnectionAdapters::OracleEnhancedAdapter.emulate_booleans_from_strings = true
337 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 337 cattr_accessor :emulate_booleans_from_strings |
#emulate_dates ⇒ Object
:singleton-method: By default, the OracleEnhancedAdapter will typecast all columns of type DATE
to Time or DateTime (if value is out of Time value range) value. If you wish that DATE values with hour, minutes and seconds equal to 0 are typecasted to Date then you can add the following line to your initializer file:
ActiveRecord::ConnectionAdapters::OracleEnhancedAdapter.emulate_dates = true
As this option can have side effects when unnecessary typecasting is done it is recommended that Date columns are explicily defined with set_date_columns
method.
277 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 277 cattr_accessor :emulate_dates |
#emulate_dates_by_column_name ⇒ Object
:singleton-method: By default, the OracleEnhancedAdapter will typecast all columns of type DATE
to Time or DateTime (if value is out of Time value range) value. If you wish that DATE columns with “date” in their name (e.g. “creation_date”) are typecasted to Date then you can add the following line to your initializer file:
ActiveRecord::ConnectionAdapters::OracleEnhancedAdapter.emulate_dates_by_column_name = true
As this option can have side effects when unnecessary typecasting is done it is recommended that Date columns are explicily defined with set_date_columns
method.
291 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 291 cattr_accessor :emulate_dates_by_column_name |
#emulate_integers_by_column_name ⇒ Object
:singleton-method: By default, the OracleEnhancedAdapter will typecast all columns of type NUMBER
(without precision or scale) to Float or BigDecimal value. If you wish that NUMBER columns with name “id” or that end with “_id” are typecasted to Integer then you can add the following line to your initializer file:
ActiveRecord::ConnectionAdapters::OracleEnhancedAdapter.emulate_integers_by_column_name = true
321 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 321 cattr_accessor :emulate_integers_by_column_name |
#execute(sql, name = nil) ⇒ Object
Executes a SQL statement
533 534 535 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 533 def execute(sql, name = nil) log(sql, name) { @connection.exec sql } end |
#get_type_for_column(table_name, column_name) ⇒ Object
:nodoc:
718 719 720 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 718 def get_type_for_column(table_name, column_name) #:nodoc: @table_column_type && @table_column_type[table_name] && @table_column_type[table_name][column_name.to_s.downcase] end |
#ignore_table_columns(table_name, *args) ⇒ Object
set ignored columns for table
697 698 699 700 701 702 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 697 def ignore_table_columns(table_name, *args) #:nodoc: @ignore_table_columns ||= {} @ignore_table_columns[table_name] ||= [] @ignore_table_columns[table_name] += args.map{|a| a.to_s.downcase} @ignore_table_columns[table_name].uniq! end |
#ignored_table_columns(table_name) ⇒ Object
:nodoc:
704 705 706 707 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 704 def ignored_table_columns(table_name) #:nodoc: @ignore_table_columns ||= {} @ignore_table_columns[table_name] end |
#index_name(table_name, options) ⇒ Object
returned shortened index name if default is too large
868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 868 def index_name(table_name, ) #:nodoc: default_name = super(table_name, ) return default_name if default_name.length <= IDENTIFIER_MAX_LENGTH # remove 'index', 'on' and 'and' keywords shortened_name = "i_#{table_name}_#{Array(options[:column]) * '_'}" # leave just first three letters from each word if shortened_name.length > IDENTIFIER_MAX_LENGTH shortened_name = shortened_name.split('_').map{|w| w[0,3]}.join('_') end if shortened_name.length <= IDENTIFIER_MAX_LENGTH @logger.warn "#{adapter_name} shortened index name #{default_name} to #{shortened_name}" if @logger return shortened_name else raise ArgumentError, "#{adapter_name} cannot shorten index name #{default_name}, please use add_index with :name option" end end |
#indexes(table_name, name = nil) ⇒ Object
This method selects all indexes at once, and caches them in a class variable. Subsequent index calls get them from the variable, without going to the DB.
662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 662 def indexes(table_name, name = nil) #:nodoc: (owner, table_name) = @connection.describe(table_name) unless all_schema_indexes result = select_all("SELECT lower(i.table_name) as table_name, lower(i.index_name) as index_name, i.uniqueness, lower(c.column_name) as column_name\nFROM all_indexes i, all_ind_columns c\nWHERE i.owner = '\#{owner}'\nAND i.table_owner = '\#{owner}'\nAND c.index_name = i.index_name\nAND c.index_owner = i.owner\nAND NOT EXISTS (SELECT uc.index_name FROM all_constraints uc WHERE uc.index_name = i.index_name AND uc.owner = i.owner AND uc.constraint_type = 'P')\nORDER BY i.index_name, c.column_position\n") current_index = nil self.all_schema_indexes = [] result.each do |row| # have to keep track of indexes because above query returns dups # there is probably a better query we could figure out if current_index != row['index_name'] self.all_schema_indexes << ::ActiveRecord::ConnectionAdapters::IndexDefinition.new(row['table_name'], row['index_name'], row['uniqueness'] == "UNIQUE", []) current_index = row['index_name'] end self.all_schema_indexes.last.columns << row['column_name'] end end # Return the indexes just for the requested table, since AR is structured that way table_name = table_name.downcase all_schema_indexes.select{|i| i.table == table_name} end |
#insert_fixture(fixture, table_name) ⇒ Object
Inserts the given fixture into the table. Overridden to properly handle lobs.
592 593 594 595 596 597 598 599 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 592 def insert_fixture(fixture, table_name) #:nodoc: super klass = fixture.class_name.constantize rescue nil if klass.respond_to?(:ancestors) && klass.ancestors.include?(ActiveRecord::Base) write_lobs(table_name, klass, fixture) end end |
#is_date_column?(name, table_name = nil) ⇒ Boolean
instance method uses at first check if column type defined at class level
302 303 304 305 306 307 308 309 310 311 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 302 def is_date_column?(name, table_name = nil) #:nodoc: case get_type_for_column(table_name, name) when nil self.class.is_date_column?(name, table_name) when :date true else false end end |
#lob_order_by_expression(klass, order) ⇒ Object
change LOB column for ORDER BY clause just first 100 characters are taken for ordering
630 631 632 633 634 635 636 637 638 639 640 641 642 643 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 630 def lob_order_by_expression(klass, order) #:nodoc: return order if order.nil? changed = false new_order = order.to_s.strip.split(/, */).map do |order_by_col| column_name, asc_desc = order_by_col.split(/ +/) if column = klass.columns.detect { |col| col.name == column_name && col.sql_type =~ /LOB$/i} changed = true "DBMS_LOB.SUBSTR(#{column_name},100,1) #{asc_desc}" else order_by_col end end.join(', ') changed ? new_order : order end |
#native_database_types ⇒ Object
:nodoc:
379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 379 def native_database_types #:nodoc: { :primary_key => "NUMBER(38) NOT NULL PRIMARY KEY", :string => { :name => "VARCHAR2", :limit => 255 }, :text => { :name => "CLOB" }, :integer => { :name => "NUMBER", :limit => 38 }, :float => { :name => "NUMBER" }, :decimal => { :name => "DECIMAL" }, :datetime => { :name => "DATE" }, # changed to native TIMESTAMP type # :timestamp => { :name => "DATE" }, :timestamp => { :name => "TIMESTAMP" }, :time => { :name => "DATE" }, :date => { :name => "DATE" }, :binary => { :name => "BLOB" }, # if emulate_booleans_from_strings then store booleans in VARCHAR2 :boolean => emulate_booleans_from_strings ? { :name => "VARCHAR2", :limit => 1 } : { :name => "NUMBER", :limit => 1 } } end |
#next_sequence_value(sequence_name) ⇒ Object
Returns the next sequence value from a sequence generator. Not generally called directly; used by ActiveRecord to get the next primary key value when inserting a new database record (see #prefetch_primary_key?).
548 549 550 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 548 def next_sequence_value(sequence_name) select_one("select #{sequence_name}.nextval id from dual")['id'] end |
#pk_and_sequence_for(table_name) ⇒ Object
Find a table’s primary key and sequence. Note: Only primary key is implemented - sequence will be nil.
963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 963 def pk_and_sequence_for(table_name) #:nodoc: (owner, table_name) = @connection.describe(table_name) # changed select from all_constraints to user_constraints - much faster in large data dictionaries pks = select_values("select cc.column_name\nfrom user_constraints c, user_cons_columns cc\nwhere c.owner = '\#{owner}'\nand c.table_name = '\#{table_name}'\nand c.constraint_type = 'P'\nand cc.owner = c.owner\nand cc.constraint_name = c.constraint_name\n", 'Primary Key') # only support single column keys pks.size == 1 ? [oracle_downcase(pks.first), nil] : nil end |
#prefetch_primary_key?(table_name = nil) ⇒ Boolean
Returns true for Oracle adapter (since Oracle requires primary key values to be pre-fetched before insert). See also #next_sequence_value.
582 583 584 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 582 def prefetch_primary_key?(table_name = nil) true end |
#quote(value, column = nil) ⇒ Object
:nodoc:
440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 440 def quote(value, column = nil) #:nodoc: if value && column case column.type when :text, :binary %Q{empty_#{ column.sql_type.downcase rescue 'blob' }()} # NLS_DATE_FORMAT independent TIMESTAMP support when :timestamp (value) # NLS_DATE_FORMAT independent DATE support when :date, :time, :datetime quote_date_with_to_date(value) else super end elsif value.acts_like?(:date) quote_date_with_to_date(value) elsif value.acts_like?(:time) value.to_i == value.to_f ? quote_date_with_to_date(value) : (value) else super end end |
#quote_column_name(name) ⇒ Object
QUOTING ==================================================
see: abstract/quoting.rb
411 412 413 414 415 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 411 def quote_column_name(name) #:nodoc: # camelCase column names need to be quoted; not that anyone using Oracle # would really do this, but handling this case means we pass the test... @quoted_column_names[name] = name.to_s =~ /[A-Z]/ ? "\"#{name}\"" : quote_oracle_reserved_words(name) end |
#quote_date_with_to_date(value) ⇒ Object
:nodoc:
473 474 475 476 477 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 473 def quote_date_with_to_date(value) #:nodoc: # should support that composite_primary_keys gem will pass date as string value = quoted_date(value) if value.acts_like?(:date) || value.acts_like?(:time) "TO_DATE('#{value}','YYYY-MM-DD HH24:MI:SS')" end |
#quote_string(s) ⇒ Object
:nodoc:
436 437 438 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 436 def quote_string(s) #:nodoc: s.gsub(/'/, "''") end |
#quote_table_name(name) ⇒ Object
:nodoc:
427 428 429 430 431 432 433 434 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 427 def quote_table_name(name) #:nodoc: # abstract_adapter calls quote_column_name from quote_table_name, so prevent that @quoted_table_names[name] ||= if self.class.valid_table_name?(name) name else "\"#{name}\"" end end |
#quote_timestamp_with_to_timestamp(value) ⇒ Object
:nodoc:
479 480 481 482 483 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 479 def (value) #:nodoc: # add up to 9 digits of fractional seconds to inserted time value = "#{quoted_date(value)}.#{("%.6f"%value.to_f).split('.')[1]}" if value.acts_like?(:time) "TO_TIMESTAMP('#{value}','YYYY-MM-DD HH24:MI:SS.FF6')" end |
#quoted_false ⇒ Object
:nodoc:
468 469 470 471 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 468 def quoted_false #:nodoc: return "'#{self.class.boolean_to_string(false)}'" if emulate_booleans_from_strings "0" end |
#quoted_true ⇒ Object
:nodoc:
463 464 465 466 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 463 def quoted_true #:nodoc: return "'#{self.class.boolean_to_string(true)}'" if emulate_booleans_from_strings "1" end |
#raw_connection ⇒ Object
return raw OCI8 or JDBC connection
500 501 502 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 500 def raw_connection @connection.raw_connection end |
#reconnect! ⇒ Object
Reconnects to the database.
516 517 518 519 520 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 516 def reconnect! #:nodoc: @connection.reset! rescue OracleEnhancedConnectionException => e @logger.warn "#{adapter_name} automatic reconnection failed: #{e.message}" if @logger end |
#remove_column(table_name, column_name) ⇒ Object
:nodoc:
928 929 930 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 928 def remove_column(table_name, column_name) #:nodoc: execute "ALTER TABLE #{quote_table_name(table_name)} DROP COLUMN #{quote_column_name(column_name)}" end |
#remove_index(table_name, options = {}) ⇒ Object
clear cached indexes when removing index
862 863 864 865 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 862 def remove_index(table_name, = {}) #:nodoc: self.all_schema_indexes = nil execute "DROP INDEX #{index_name(table_name, options)}" end |
#rename_column(table_name, column_name, new_column_name) ⇒ Object
:nodoc:
924 925 926 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 924 def rename_column(table_name, column_name, new_column_name) #:nodoc: execute "ALTER TABLE #{quote_table_name(table_name)} RENAME COLUMN #{quote_column_name(column_name)} to #{quote_column_name(new_column_name)}" end |
#rename_table(name, new_name) ⇒ Object
:nodoc:
844 845 846 847 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 844 def rename_table(name, new_name) #:nodoc: execute "RENAME #{quote_table_name(name)} TO #{quote_table_name(new_name)}" execute "RENAME #{quote_table_name("#{name}_seq")} TO #{quote_table_name("#{new_name}_seq")}" rescue nil end |
#rollback_db_transaction ⇒ Object
:nodoc:
562 563 564 565 566 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 562 def rollback_db_transaction #:nodoc: @connection.rollback ensure @connection.autocommit = true end |
#select_rows(sql, name = nil) ⇒ Object
Returns an array of arrays containing the field values. Order is the same as that returned by #columns.
539 540 541 542 543 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 539 def select_rows(sql, name = nil) # last parameter indicates to return also column list result, columns = select(sql, name, true) result.map{ |v| columns.map{|c| v[c]} } end |
#set_type_for_columns(table_name, column_type, *args) ⇒ Object
set explicit type for specified table columns
710 711 712 713 714 715 716 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 710 def set_type_for_columns(table_name, column_type, *args) #:nodoc: @table_column_type ||= {} @table_column_type[table_name] ||= {} args.each do |col| @table_column_type[table_name][col.to_s.downcase] = column_type end end |
#structure_drop ⇒ Object
:nodoc:
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 1055 def structure_drop #:nodoc: s = select_all("select sequence_name from user_sequences order by 1").inject("") do |drop, seq| drop << "drop sequence #{seq.to_a.first.last};\n\n" end # changed select from user_tables to all_tables - much faster in large data dictionaries select_all("select table_name from all_tables where owner = sys_context('userenv','session_user') order by 1").inject(s) do |drop, table| drop << "drop table #{table.to_a.first.last} cascade constraints;\n\n" end end |
#structure_dump ⇒ Object
:nodoc:
981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 981 def structure_dump #:nodoc: s = select_all("select sequence_name from user_sequences order by 1").inject("") do |structure, seq| structure << "create sequence #{seq.to_a.first.last};\n\n" end # changed select from user_tables to all_tables - much faster in large data dictionaries select_all("select table_name from all_tables where owner = sys_context('userenv','session_user') order by 1").inject(s) do |structure, table| ddl = "create table #{table.to_a.first.last} (\n " cols = select_all(%Q{ select column_name, data_type, data_length, char_used, char_length, data_precision, data_scale, data_default, nullable from user_tab_columns where table_name = '#{table.to_a.first.last}' order by column_id }).map do |row| col = "#{row['column_name'].downcase} #{row['data_type'].downcase}" if row['data_type'] =='NUMBER' and !row['data_precision'].nil? col << "(#{row['data_precision'].to_i}" col << ",#{row['data_scale'].to_i}" if !row['data_scale'].nil? col << ')' elsif row['data_type'].include?('CHAR') length = row['char_used'] == 'C' ? row['char_length'].to_i : row['data_length'].to_i col << "(#{length})" end col << " default #{row['data_default']}" if !row['data_default'].nil? col << ' not null' if row['nullable'] == 'N' col end ddl << cols.join(",\n ") ddl << ");\n\n" structure << ddl end end |
#structure_dump_db_stored_code ⇒ Object
Extract all stored procedures, packages, synonyms and views.
1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 1015 def structure_dump_db_stored_code #:nodoc: structure = "\n" select_all("select distinct name, type from all_source where type in ('PROCEDURE', 'PACKAGE', 'PACKAGE BODY', 'FUNCTION') and owner = sys_context('userenv','session_user')").inject("\n\n") do |structure, source| ddl = "create or replace \n " lines = select_all(%Q{ select text from all_source where name = '#{source['name']}' and type = '#{source['type']}' and owner = sys_context('userenv','session_user') order by line }).map do |row| ddl << row['text'] if row['text'].size > 1 end ddl << ";" structure << ddl << "\n" end # export views select_all("select view_name, text from user_views").inject(structure) do |structure, view| ddl = "create or replace view #{view['view_name']} AS\n " # any views with empty lines will cause OCI to barf when loading. remove blank lines =/ ddl << view['text'].gsub(/^\n/, '') ddl << ";\n\n" structure << ddl end # export synonyms select_all("select owner, synonym_name, table_name, table_owner from all_synonyms where table_owner = sys_context('userenv','session_user') ").inject(structure) do |structure, synonym| ddl = "create or replace #{synonym['owner'] == 'PUBLIC' ? 'PUBLIC' : '' } SYNONYM #{synonym['synonym_name']} for #{synonym['table_owner']}.#{synonym['table_name']};\n\n" structure << ddl; end end |
#supports_migrations? ⇒ Boolean
:nodoc:
375 376 377 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 375 def supports_migrations? #:nodoc: true end |
#table_alias_length ⇒ Object
:nodoc:
403 404 405 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 403 def table_alias_length #:nodoc: IDENTIFIER_MAX_LENGTH end |
#table_comment(table_name) ⇒ Object
:nodoc:
942 943 944 945 946 947 948 949 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 942 def table_comment(table_name) #:nodoc: (owner, table_name) = @connection.describe(table_name) select_value "SELECT comments FROM all_tab_comments\nWHERE owner = '\#{owner}'\nAND table_name = '\#{table_name}'\n" end |
#tables(name = nil) ⇒ Object
:nodoc:
653 654 655 656 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 653 def tables(name = nil) #:nodoc: # changed select from user_tables to all_tables - much faster in large data dictionaries select_all("select decode(table_name,upper(table_name),lower(table_name),table_name) name from all_tables where owner = sys_context('userenv','session_user')").map {|t| t['name']} end |
#write_lobs(table_name, klass, attributes) ⇒ Object
Writes LOB values from attributes, as indicated by the LOB columns of klass.
602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 |
# File 'lib/active_record/connection_adapters/oracle_enhanced_adapter.rb', line 602 def write_lobs(table_name, klass, attributes) #:nodoc: # is class with composite primary key> is_with_cpk = klass.respond_to?(:composite?) && klass.composite? if is_with_cpk id = klass.primary_key.map {|pk| attributes[pk.to_s] } else id = quote(attributes[klass.primary_key]) end klass.columns.select { |col| col.sql_type =~ /LOB$/i }.each do |col| value = attributes[col.name] # changed sequence of next two lines - should check if value is nil before converting to yaml next if value.nil? || (value == '') value = value.to_yaml if col.text? && klass.serialized_attributes[col.name] uncached do if is_with_cpk lob = select_one("SELECT #{col.name} FROM #{table_name} WHERE #{klass.composite_where_clause(id)} FOR UPDATE", 'Writable Large Object')[col.name] else lob = select_one("SELECT #{col.name} FROM #{table_name} WHERE #{klass.primary_key} = #{id} FOR UPDATE", 'Writable Large Object')[col.name] end @connection.write_lob(lob, value.to_s, col.type == :binary) end end end |