Restructuring directories.

This commit is contained in:
Zach Dennis 2010-06-26 15:37:23 -04:00
parent 5c287f1042
commit 4fcfa7fd14
12 changed files with 326175 additions and 5 deletions

View file

@ -0,0 +1,146 @@
module ActiveRecord # :nodoc:
module ConnectionAdapters # :nodoc:
class AbstractAdapter # :nodoc:
NO_MAX_PACKET = 0
QUERY_OVERHEAD = 8 #This was shown to be true for MySQL, but it's not clear where the overhead is from.
def next_value_for_sequence(sequence_name)
%{#{sequence_name}.nextval}
end
# +sql+ can be a single string or an array. If it is an array all
# elements that are in position >= 1 will be appended to the final SQL.
def insert_many( sql, values, *args ) # :nodoc:
# the number of inserts default
number_of_inserts = 0
base_sql,post_sql = if sql.is_a?( String )
[ sql, '' ]
elsif sql.is_a?( Array )
[ sql.shift, sql.join( ' ' ) ]
end
sql_size = QUERY_OVERHEAD + base_sql.size + post_sql.size
# the number of bytes the requested insert statement values will take up
values_in_bytes = self.class.sum_sizes( *values )
# the number of bytes (commas) it will take to comma separate our values
comma_separated_bytes = values.size-1
# the total number of bytes required if this statement is one statement
total_bytes = sql_size + values_in_bytes + comma_separated_bytes
max = max_allowed_packet
# if we can insert it all as one statement
if NO_MAX_PACKET == max or total_bytes < max
number_of_inserts += 1
sql2insert = base_sql + values.join( ',' ) + post_sql
insert( sql2insert, *args )
else
value_sets = self.class.get_insert_value_sets( values, sql_size, max )
value_sets.each do |values|
number_of_inserts += 1
sql2insert = base_sql + values.join( ',' ) + post_sql
insert( sql2insert, *args )
end
end
number_of_inserts
end
def pre_sql_statements(options)
sql = []
sql << options[:pre_sql] if options[:pre_sql]
sql << options[:command] if options[:command]
sql << "IGNORE" if options[:ignore]
#add keywords like IGNORE or DELAYED
if options[:keywords].is_a?(Array)
sql.concat(options[:keywords])
elsif options[:keywords]
sql << options[:keywords].to_s
end
sql
end
# Synchronizes the passed in ActiveRecord instances with the records in
# the database by calling +reload+ on each instance.
def after_import_synchronize( instances )
instances.each { |e| e.reload }
end
# Returns an array of post SQL statements given the passed in options.
def post_sql_statements( table_name, options ) # :nodoc:
post_sql_statements = []
if options[:on_duplicate_key_update]
post_sql_statements << sql_for_on_duplicate_key_update( table_name, options[:on_duplicate_key_update] )
end
#custom user post_sql
post_sql_statements << options[:post_sql] if options[:post_sql]
#with rollup
post_sql_statements << rollup_sql if options[:rollup]
post_sql_statements
end
# Generates the INSERT statement used in insert multiple value sets.
def multiple_value_sets_insert_sql(table_name, column_names, options) # :nodoc:
"INSERT #{options[:ignore] ? 'IGNORE ':''}INTO #{table_name} (#{column_names.join(',')}) VALUES "
end
# Returns SQL the VALUES for an INSERT statement given the passed in +columns+
# and +array_of_attributes+.
def values_sql_for_column_names_and_attributes( columns, array_of_attributes ) # :nodoc:
values = []
array_of_attributes.each do |arr|
my_values = []
arr.each_with_index do |val,j|
my_values << quote( val, columns[j] )
end
values << my_values
end
values_arr = values.map{ |arr| '(' + arr.join( ',' ) + ')' }
end
# Returns the sum of the sizes of the passed in objects. This should
# probably be moved outside this class, but to where?
def self.sum_sizes( *objects ) # :nodoc:
objects.inject( 0 ){|sum,o| sum += o.size }
end
# Returns the maximum number of bytes that the server will allow
# in a single packet
def max_allowed_packet
NO_MAX_PACKET
end
def self.get_insert_value_sets( values, sql_size, max_bytes ) # :nodoc:
value_sets = []
arr, current_arr_values_size, current_size = [], 0, 0
values.each_with_index do |val,i|
comma_bytes = arr.size
sql_size_thus_far = sql_size + current_size + val.size + comma_bytes
if NO_MAX_PACKET == max_bytes or sql_size_thus_far <= max_bytes
current_size += val.size
arr << val
else
value_sets << arr
arr = [ val ]
current_size = val.size
end
# if we're on the last iteration push whatever we have in arr to value_sets
value_sets << arr if i == (values.size-1)
end
[ *value_sets ]
end
end
end
end

View file

@ -0,0 +1,11 @@
require "active_record/connection_adapters/postgresql_adapter"
module ActiveRecord # :nodoc:
module ConnectionAdapters # :nodoc:
class PostgreSQLAdapter # :nodoc:
def next_value_for_sequence(sequence_name)
%{nextval('#{sequence_name}')}
end
end
end
end

View file

@ -0,0 +1,15 @@
require "pathname"
require "active_record"
require "active_record/version"
module ActiveRecord::Extensions
AdapterPath = File.join File.expand_path(File.dirname(__FILE__)), "/../active_record/adapters"
def self.require_adapter(adapter)
require File.join(AdapterPath,"/#{adapter}_adapter")
end
end
this_dir = Pathname.new File.dirname(__FILE__)
require this_dir.join("import")
require this_dir.join("active_record/adapters/abstract_adapter")

View file

@ -0,0 +1,348 @@
require "ostruct"
module ActiveRecord::Extensions::ConnectionAdapters ; end
module ActiveRecord::Extensions::Import #:nodoc:
module ImportSupport #:nodoc:
def supports_import? #:nodoc:
true
end
end
module OnDuplicateKeyUpdateSupport #:nodoc:
def supports_on_duplicate_key_update? #:nodoc:
true
end
end
end
class ActiveRecord::Base
class << self
# use tz as set in ActiveRecord::Base
tproc = lambda do
ActiveRecord::Base.default_timezone == :utc ? Time.now.utc : Time.now
end
AREXT_RAILS_COLUMNS = {
:create => { "created_on" => tproc ,
"created_at" => tproc },
:update => { "updated_on" => tproc ,
"updated_at" => tproc }
}
AREXT_RAILS_COLUMN_NAMES = AREXT_RAILS_COLUMNS[:create].keys + AREXT_RAILS_COLUMNS[:update].keys
# Returns true if the current database connection adapter
# supports import functionality, otherwise returns false.
def supports_import?
connection.supports_import?
rescue NoMethodError
false
end
# Returns true if the current database connection adapter
# supports on duplicate key update functionality, otherwise
# returns false.
def supports_on_duplicate_key_update?
connection.supports_on_duplicate_key_update?
rescue NoMethodError
false
end
# Imports a collection of values to the database.
#
# This is more efficient than using ActiveRecord::Base#create or
# ActiveRecord::Base#save multiple times. This method works well if
# you want to create more than one record at a time and do not care
# about having ActiveRecord objects returned for each record
# inserted.
#
# This can be used with or without validations. It does not utilize
# the ActiveRecord::Callbacks during creation/modification while
# performing the import.
#
# == Usage
# Model.import array_of_models
# Model.import column_names, array_of_values
# Model.import column_names, array_of_values, options
#
# ==== Model.import array_of_models
#
# With this form you can call _import_ passing in an array of model
# objects that you want updated.
#
# ==== Model.import column_names, array_of_values
#
# The first parameter +column_names+ is an array of symbols or
# strings which specify the columns that you want to update.
#
# The second parameter, +array_of_values+, is an array of
# arrays. Each subarray is a single set of values for a new
# record. The order of values in each subarray should match up to
# the order of the +column_names+.
#
# ==== Model.import column_names, array_of_values, options
#
# The first two parameters are the same as the above form. The third
# parameter, +options+, is a hash. This is optional. Please see
# below for what +options+ are available.
#
# == Options
# * +validate+ - true|false, tells import whether or not to use \
# ActiveRecord validations. Validations are enforced by default.
# * +on_duplicate_key_update+ - an Array or Hash, tells import to \
# use MySQL's ON DUPLICATE KEY UPDATE ability. See On Duplicate\
# Key Update below.
# * +synchronize+ - an array of ActiveRecord instances for the model
# that you are currently importing data into. This synchronizes
# existing model instances in memory with updates from the import.
# * +timestamps+ - true|false, tells import to not add timestamps \
# (if false) even if record timestamps is disabled in ActiveRecord::Base
#
# == Examples
# class BlogPost < ActiveRecord::Base ; end
#
# # Example using array of model objects
# posts = [ BlogPost.new :author_name=>'Zach Dennis', :title=>'AREXT',
# BlogPost.new :author_name=>'Zach Dennis', :title=>'AREXT2',
# BlogPost.new :author_name=>'Zach Dennis', :title=>'AREXT3' ]
# BlogPost.import posts
#
# # Example using column_names and array_of_values
# columns = [ :author_name, :title ]
# values = [ [ 'zdennis', 'test post' ], [ 'jdoe', 'another test post' ] ]
# BlogPost.import columns, values
#
# # Example using column_names, array_of_value and options
# columns = [ :author_name, :title ]
# values = [ [ 'zdennis', 'test post' ], [ 'jdoe', 'another test post' ] ]
# BlogPost.import( columns, values, :validate => false )
#
# # Example synchronizing existing instances in memory
# post = BlogPost.find_by_author_name( 'zdennis' )
# puts post.author_name # => 'zdennis'
# columns = [ :author_name, :title ]
# values = [ [ 'yoda', 'test post' ] ]
# BlogPost.import posts, :synchronize=>[ post ]
# puts post.author_name # => 'yoda'
#
# == On Duplicate Key Update (MySQL only)
#
# The :on_duplicate_key_update option can be either an Array or a Hash.
#
# ==== Using an Array
#
# The :on_duplicate_key_update option can be an array of column
# names. The column names are the only fields that are updated if
# a duplicate record is found. Below is an example:
#
# BlogPost.import columns, values, :on_duplicate_key_update=>[ :date_modified, :content, :author ]
#
# ==== Using A Hash
#
# The :on_duplicate_key_update option can be a hash of column name
# to model attribute name mappings. This gives you finer grained
# control over what fields are updated with what attributes on your
# model. Below is an example:
#
# BlogPost.import columns, attributes, :on_duplicate_key_update=>{ :title => :title }
#
# = Returns
# This returns an object which responds to +failed_instances+ and +num_inserts+.
# * failed_instances - an array of objects that fails validation and were not committed to the database. An empty array if no validation is performed.
# * num_inserts - the number of insert statements it took to import the data
def import( *args )
options = { :validate=>true, :timestamps=>true }
options.merge!( args.pop ) if args.last.is_a? Hash
# assume array of model objects
if args.last.is_a?( Array ) and args.last.first.is_a? ActiveRecord::Base
if args.length == 2
models = args.last
column_names = args.first
else
models = args.first
column_names = self.column_names.dup
end
array_of_attributes = []
models.each do |model|
# this next line breaks sqlite.so with a segmentation fault
# if model.new_record? || options[:on_duplicate_key_update]
attributes = []
column_names.each do |name|
attributes << model.send( "#{name}_before_type_cast" )
end
array_of_attributes << attributes
# end
end
# supports 2-element array and array
elsif args.size == 2 and args.first.is_a?( Array ) and args.last.is_a?( Array )
column_names, array_of_attributes = args
else
raise ArgumentError.new( "Invalid arguments!" )
end
# Force the primary key col into the insert if it's not
# on the list and we are using a sequence and stuff a nil
# value for it into each row so the sequencer will fire later
if !column_names.include?(primary_key) && sequence_name && connection.prefetch_primary_key?
column_names << primary_key
array_of_attributes.each { |a| a << nil }
end
is_validating = options.delete( :validate )
# dup the passed in array so we don't modify it unintentionally
array_of_attributes = array_of_attributes.dup
# record timestamps unless disabled in ActiveRecord::Base
if record_timestamps && options.delete( :timestamps )
add_special_rails_stamps column_names, array_of_attributes, options
end
return_obj = if is_validating
import_with_validations( column_names, array_of_attributes, options )
else
num_inserts = import_without_validations_or_callbacks( column_names, array_of_attributes, options )
OpenStruct.new :failed_instances=>[], :num_inserts=>num_inserts
end
if options[:synchronize]
synchronize( options[:synchronize] )
end
return_obj.num_inserts = 0 if return_obj.num_inserts.nil?
return_obj
end
# TODO import_from_table needs to be implemented.
def import_from_table( options ) # :nodoc:
end
# Imports the passed in +column_names+ and +array_of_attributes+
# given the passed in +options+ Hash with validations. Returns an
# object with the methods +failed_instances+ and +num_inserts+.
# +failed_instances+ is an array of instances that failed validations.
# +num_inserts+ is the number of inserts it took to import the data. See
# ActiveRecord::Base.import for more information on
# +column_names+, +array_of_attributes+ and +options+.
def import_with_validations( column_names, array_of_attributes, options={} )
failed_instances = []
# create instances for each of our column/value sets
arr = validations_array_for_column_names_and_attributes( column_names, array_of_attributes )
# keep track of the instance and the position it is currently at. if this fails
# validation we'll use the index to remove it from the array_of_attributes
arr.each_with_index do |hsh,i|
instance = new( hsh )
if not instance.valid?
array_of_attributes[ i ] = nil
failed_instances << instance
end
end
array_of_attributes.compact!
num_inserts = array_of_attributes.empty? ? 0 : import_without_validations_or_callbacks( column_names, array_of_attributes, options )
OpenStruct.new :failed_instances=>failed_instances, :num_inserts => num_inserts
end
# Imports the passed in +column_names+ and +array_of_attributes+
# given the passed in +options+ Hash. This will return the number
# of insert operations it took to create these records without
# validations or callbacks. See ActiveRecord::Base.import for more
# information on +column_names+, +array_of_attributes_ and
# +options+.
def import_without_validations_or_callbacks( column_names, array_of_attributes, options={} )
escaped_column_names = quote_column_names( column_names )
columns = []
array_of_attributes.first.each_with_index { |arr,i| columns << columns_hash[ column_names[i] ] }
if not supports_import?
columns_sql = "(" + escaped_column_names.join( ',' ) + ")"
insert_statements, values = [], []
number_inserted = 0
array_of_attributes.each do |arr|
my_values = []
arr.each_with_index do |val,j|
if !sequence_name.blank? && column_names[j] == primary_key && val.nil?
my_values << connection.next_value_for_sequence(sequence_name)
else
my_values << connection.quote( val, columns[j] )
end
end
insert_statements << "INSERT INTO #{quoted_table_name} #{columns_sql} VALUES(" + my_values.join( ',' ) + ")"
connection.execute( insert_statements.last )
number_inserted += 1
end
else
# generate the sql
insert_sql = connection.multiple_value_sets_insert_sql( quoted_table_name, escaped_column_names, options )
values_sql = connection.values_sql_for_column_names_and_attributes( columns, array_of_attributes )
post_sql_statements = connection.post_sql_statements( quoted_table_name, options )
# perform the inserts
number_inserted = connection.insert_many( [ insert_sql, post_sql_statements ].flatten,
values_sql,
"#{self.class.name} Create Many Without Validations Or Callbacks" )
end
number_inserted
end
# Returns an array of quoted column names
def quote_column_names( names )
names.map{ |name| connection.quote_column_name( name ) }
end
private
def add_special_rails_stamps( column_names, array_of_attributes, options )
AREXT_RAILS_COLUMNS[:create].each_pair do |key, blk|
if self.column_names.include?(key)
value = blk.call
if index=column_names.index(key)
# replace every instance of the array of attributes with our value
array_of_attributes.each{ |arr| arr[index] = value }
else
column_names << key
array_of_attributes.each { |arr| arr << value }
end
end
end
AREXT_RAILS_COLUMNS[:update].each_pair do |key, blk|
if self.column_names.include?(key)
value = blk.call
if index=column_names.index(key)
# replace every instance of the array of attributes with our value
array_of_attributes.each{ |arr| arr[index] = value }
else
column_names << key
array_of_attributes.each { |arr| arr << value }
end
if options[:on_duplicate_key_update]
options[:on_duplicate_key_update] << key.to_sym if options[:on_duplicate_key_update].is_a?(Array)
options[:on_duplicate_key_update][key.to_sym] = key.to_sym if options[:on_duplicate_key_update].is_a?(Hash)
else
options[:on_duplicate_key_update] = [ key.to_sym ]
end
end
end
end
# Returns an Array of Hashes for the passed in +column_names+ and +array_of_attributes+.
def validations_array_for_column_names_and_attributes( column_names, array_of_attributes ) # :nodoc:
arr = []
array_of_attributes.each do |attributes|
c = 0
hsh = attributes.inject( {} ){|hsh,attr| hsh[ column_names[c] ] = attr ; c+=1 ; hsh }
arr << hsh
end
arr
end
end
end

View file

@ -0,0 +1,53 @@
require "active_record/connection_adapters/mysql_adapter"
require File.join File.dirname(__FILE__), "base"
module ActiveRecord::Extensions::ConnectionAdapters::MysqlAdapter # :nodoc:
include ActiveRecord::Extensions::Import::ImportSupport
include ActiveRecord::Extensions::Import::OnDuplicateKeyUpdateSupport
# Returns a generated ON DUPLICATE KEY UPDATE statement given the passed
# in +args+.
def sql_for_on_duplicate_key_update( table_name, *args ) # :nodoc:
sql = ' ON DUPLICATE KEY UPDATE '
arg = args.first
if arg.is_a?( Array )
sql << sql_for_on_duplicate_key_update_as_array( table_name, arg )
elsif arg.is_a?( Hash )
sql << sql_for_on_duplicate_key_update_as_hash( table_name, arg )
elsif arg.is_a?( String )
sql << arg
else
raise ArgumentError.new( "Expected Array or Hash" )
end
sql
end
def sql_for_on_duplicate_key_update_as_array( table_name, arr ) # :nodoc:
results = arr.map do |column|
qc = quote_column_name( column )
"#{table_name}.#{qc}=VALUES(#{qc})"
end
results.join( ',' )
end
def sql_for_on_duplicate_key_update_as_hash( table_name, hsh ) # :nodoc:
sql = ' ON DUPLICATE KEY UPDATE '
results = hsh.map do |column1, column2|
qc1 = quote_column_name( column1 )
qc2 = quote_column_name( column2 )
"#{table_name}.#{qc1}=VALUES( #{qc2} )"
end
results.join( ',')
end
#return true if the statement is a duplicate key record error
def duplicate_key_update_error?(exception)# :nodoc:
exception.is_a?(ActiveRecord::StatementInvalid) && exception.to_s.include?('Duplicate entry')
end
end
ActiveRecord::ConnectionAdapters::MysqlAdapter.class_eval do
include ActiveRecord::Extensions::ConnectionAdapters::MysqlAdapter
end

View file

@ -0,0 +1,2 @@
require File.join File.dirname(__FILE__), "base"
ActiveRecord::Extensions.require_adapter "postgresql"

View file

@ -0,0 +1,2 @@
require "active_record/connection_adapters/sqlite3_adapter"
require File.join File.dirname(__FILE__), "base"