activerecord-import 0.3.1 → 0.4.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: b643c184e9c5ecb05a667d0845794f7a5282417d
+  data.tar.gz: f02c166d88df6237b5ca09350d85a6fbd41927d5
+SHA512:
+  metadata.gz: 22c96aa8eed0dc9594b7e7bc0592266506c7a4c93bbe2d5cbb0b08550728a00a91e95deb9adf087c5a74008f1fecc93588a603402d91ce70b44fd6f2495b5fa6
+  data.tar.gz: caf4b4f41c6e34f55d340f43d1fa82e5b33dd1dfb52d27970ad61d29a478b2988d310521a0aac580d76a26ef269f967f06c7baf828b89da4948dd028d9c5e950

data/.gitignore

@@ -0,0 +1,31 @@
+## MAC OS
+.DS_Store
+
+## TEXTMATE
+*.tmproj
+tmtags
+
+## EMACS
+*~
+\#*
+.\#*
+
+## VIM
+*.swp
+
+## PROJECT::GENERAL
+coverage
+rdoc
+pkg
+*.gem
+*.lock
+
+## PROJECT::SPECIFIC
+log/*.log
+test.db
+test/database.yml
+
+.bundle/
+.redcar/
+.rvmrc
+docsite/

data/Brewfile

@@ -0,0 +1,3 @@
+mysql
+postgresql
+sqlite

data/Gemfile

@@ -0,0 +1,39 @@
+source 'https://rubygems.org'
+
+gemspec
+
+# Database Adapters
+platforms :ruby do
+  gem "em-synchrony",           "~> 1.0.3"
+  gem "mysql2",                 "~> 0.3.0"
+  gem "pg",                     "~> 0.9"
+  gem "sqlite3-ruby",           "~> 1.3.1"
+  gem "seamless_database_pool", "~> 1.0.13"
+end
+
+platforms :jruby do
+  gem "jdbc-mysql"
+  gem "activerecord-jdbcmysql-adapter"
+end
+
+# Support libs
+gem "factory_girl", "~> 4.2.0"
+gem "delorean",     "~> 0.2.0"
+
+# Debugging
+platforms :mri_18 do
+  gem "ruby-debug", "= 0.10.4"
+end
+
+platforms :jruby do
+  gem "ruby-debug-base", "= 0.10.4"
+  gem "ruby-debug",      "= 0.10.4"
+end
+
+platforms :mri_19 do
+  gem "debugger"
+end
+
+version = ENV['RAILS_VERSION'] || "3.2"
+
+eval_gemfile File.expand_path("../gemfiles/#{version}.gemfile", __FILE__)

data/Rakefile

@@ -4,29 +4,6 @@ Bundler.setup
 require 'rake'
 require 'rake/testtask'
 
-begin
-  require 'jeweler'
-  Jeweler::Tasks.new do |gem|
-    gem.name = "activerecord-import"
-    gem.summary = %Q{Bulk-loading extension for ActiveRecord}
-    gem.description = %Q{Extraction of the ActiveRecord::Base#import functionality from ar-extensions for Rails 3 and beyond}
-    gem.email = "zach.dennis@gmail.com"
-    gem.homepage = "http://github.com/zdennis/activerecord-import"
-    gem.authors = ["Zach Dennis"]
-    gem.files = FileList["VERSION", "Rakefile", "README*", "lib/**/*"]
-
-    bundler = Bundler.load
-    bundler.dependencies_for(:default).each do |dependency|
-      gem.add_dependency dependency.name, *dependency.requirements_list
-    end
-
-    # gem is a Gem::Specification... see http://www.rubygems.org/read/chapter/20 for additional settings
-  end
-  Jeweler::GemcutterTasks.new
-rescue LoadError
-  puts "Jeweler (or a dependency) not available. Install it with: gem install jeweler"
-end
-
 namespace :display do
   task :notice do
     puts
@@ -36,7 +13,7 @@ namespace :display do
 end
 task :default => ["display:notice"]
 
-ADAPTERS = %w(mysql mysql2 jdbcmysql postgresql sqlite3 seamless_database_pool mysqlspatial mysql2spatial spatialite postgis)
+ADAPTERS = %w(mysql mysql2 em_mysql2 jdbcmysql postgresql sqlite3 seamless_database_pool mysqlspatial mysql2spatial spatialite postgis)
 ADAPTERS.each do |adapter|
   namespace :test do
     desc "Runs #{adapter} database tests."

data/activerecord-import.gemspec

@@ -0,0 +1,22 @@
+# -*- encoding: utf-8 -*-
+require File.expand_path('../lib/activerecord-import/version', __FILE__)
+
+Gem::Specification.new do |gem|
+  gem.authors       = ["Zach Dennis"]
+  gem.email         = ["zach.dennis@gmail.com"]
+  gem.summary       = "Bulk-loading extension for ActiveRecord"
+  gem.description   = "Extraction of the ActiveRecord::Base#import functionality from ar-extensions for Rails 3 and beyond"
+  gem.homepage      = "http://github.com/zdennis/activerecord-import"
+
+  gem.files         = `git ls-files`.split($\)
+  gem.executables   = gem.files.grep(%r{^bin/}).map{ |f| File.basename(f) }
+  gem.test_files    = gem.files.grep(%r{^(test|spec|features)/})
+  gem.name          = "activerecord-import"
+  gem.require_paths = ["lib"]
+  gem.version       = ActiveRecord::Import::VERSION
+
+  gem.required_ruby_version = ">= 1.9.2"
+
+  gem.add_runtime_dependency "activerecord", ">= 3.0"
+  gem.add_development_dependency "rake"
+end

data/benchmarks/README

@@ -0,0 +1,32 @@
+To run the benchmarks, from within the benchmarks run:
+   ruby benchmark.rb [options]
+
+The following options are supported:
+  --adapter [String]   The database adapter to use. IE: mysql, postgresql, oracle
+             
+  --do-not-delete       By default all records in the benchmark tables will be deleted at the end of the benchmark. This flag indicates not to delete the benchmark data.
+  --num [Integer]      The number of objects to benchmark. (Required!)
+  --table-type [String]  The table type to test. This can be used multiple times. By default it is all table types.
+  --to-csv [String]      Print results in a CSV file format
+  --to-html [String]     Print results in HTML format (String filename must be supplied)
+
+See "ruby benchmark.rb -h" for the complete listing of options.
+
+EXAMPLES
+--------
+To output to html format:
+  ruby benchmark.rb --adapter=mysql --to-html=results.html
+
+To output to csv format:
+  ruby benchmark.rb --adapter=mysql --to-csv=results.csv
+
+LIMITATIONS
+-----------
+Currently MySQL is the only supported adapter to benchmark.
+
+AUTHOR
+------
+Zach Dennis
+zach.dennis@gmail.com
+http://www.continuousthinking.com
+

data/benchmarks/benchmark.rb

@@ -0,0 +1,64 @@
+require "pathname"
+this_dir = Pathname.new File.dirname(__FILE__)
+require this_dir.join('boot')
+
+# Parse the options passed in via the command line
+options = BenchmarkOptionParser.parse( ARGV )
+
+# The support directory where we use to load our connections and models for the 
+# benchmarks.
+SUPPORT_DIR = this_dir.join('../test')
+
+# Load the database adapter
+adapter = options.adapter
+
+# load the library
+LIB_DIR = this_dir.join("../lib")
+require LIB_DIR.join("activerecord-import/#{adapter}")
+
+ActiveRecord::Base.logger = Logger.new("log/test.log")
+ActiveRecord::Base.logger.level = Logger::DEBUG
+ActiveRecord::Base.configurations["test"] = YAML.load(SUPPORT_DIR.join("database.yml").open)[adapter]
+ActiveRecord::Base.establish_connection "test"
+
+ActiveSupport::Notifications.subscribe(/active_record.sql/) do |event, _, _, _, hsh|
+  ActiveRecord::Base.logger.info hsh[:sql]
+end
+
+adapter_schema = SUPPORT_DIR.join("schema/#{adapter}_schema.rb")
+require adapter_schema if File.exists?(adapter_schema)
+Dir[this_dir.join("models/*.rb")].each{ |file| require file }
+
+# Load databse specific benchmarks
+require File.join( File.dirname( __FILE__ ), 'lib', "#{adapter}_benchmark" )
+
+# TODO implement method/table-type selection
+table_types = nil
+if options.benchmark_all_types
+  table_types = [ "all" ]
+else
+  table_types = options.table_types.keys
+end
+puts
+
+letter = options.adapter[0].chr
+clazz_str = letter.upcase + options.adapter[1..-1].downcase
+clazz = Object.const_get( clazz_str + "Benchmark" )
+
+benchmarks = []
+options.number_of_objects.each do |num|
+  benchmarks << (benchmark = clazz.new)
+  benchmark.send( "benchmark", table_types, num )
+end
+
+options.outputs.each do |output|
+  format = output.format.downcase
+  output_module = Object.const_get( "OutputTo#{format.upcase}" )
+  benchmarks.each do |benchmark|
+    output_module.output_results( output.filename, benchmark.results )
+  end
+end
+
+puts
+puts "Done with benchmark!"
+

data/benchmarks/boot.rb

@@ -0,0 +1,18 @@
+begin ; require 'rubygems' ; rescue LoadError ; end
+require 'active_record' # ActiveRecord loads the Benchmark library automatically
+require 'active_record/version'
+require 'fastercsv'
+require 'fileutils'
+require 'logger'
+
+# Files are loaded alphabetically. If this is a problem then manually specify the files
+# that need to be loaded here.
+Dir[ File.join( File.dirname( __FILE__ ), 'lib', '*.rb' ) ].sort.each{ |f| require f }
+
+ActiveRecord::Base.logger = Logger.new STDOUT
+
+
+
+
+
+

data/benchmarks/lib/base.rb

@@ -0,0 +1,137 @@
+class BenchmarkBase
+
+  attr_reader :results
+
+  # The main benchmark method dispatcher. This dispatches the benchmarks
+  # to actual benchmark_xxxx methods.
+  # 
+  # == PARAMETERS
+  #  * table_types - an array of table types to benchmark
+  #  * num - the number of record insertions to test
+  def benchmark( table_types, num )
+    array_of_cols_and_vals = build_array_of_cols_and_vals( num )
+    table_types.each do |table_type|
+      self.send( "benchmark_#{table_type}", array_of_cols_and_vals )
+    end
+  end
+
+  # Returns an OpenStruct which contains two attritues, +description+ and +tms+ after performing an
+  # actual benchmark.
+  #
+  # == PARAMETERS
+  #  * description - the description of the block that is getting benchmarked
+  #  * blk - the block of code to benchmark
+  #
+  # == RETURNS
+  # An OpenStruct object with the following attributes:
+  #   * description - the description of the benchmark ran
+  #   * tms - a Benchmark::Tms containing the results of the benchmark
+  def bm( description, &blk )
+    tms  = nil
+    puts "Benchmarking #{description}"
+    
+    Benchmark.bm { |x| tms = x.report { blk.call } }
+    delete_all
+    failed = false
+
+    OpenStruct.new :description=>description, :tms=>tms, :failed=>failed
+  end
+  
+  # Given a model class (ie: Topic), and an array of columns and value sets
+  # this will perform all of the benchmarks necessary for this library.
+  # 
+  # == PARAMETERS
+  #  * model_clazz - the model class to benchmark (ie: Topic)
+  #  * array_of_cols_and_vals - an array of column identifiers and value sets
+  #
+  # == RETURNS
+  #  returns true
+  def bm_model( model_clazz, array_of_cols_and_vals )
+    puts
+    puts "------ Benchmarking #{model_clazz.name} -------"
+
+    cols,vals = array_of_cols_and_vals
+    num_inserts = vals.size
+
+    # add a new result group for this particular benchmark
+    group = []
+    @results << group
+
+    description = "#{model_clazz.name}.create (#{num_inserts} records)"
+    group << bm( description ) { 
+      vals.each do |values|
+        model_clazz.create create_hash_for_cols_and_vals( cols, values )
+      end  }
+
+    description = "#{model_clazz.name}.import(column, values) for #{num_inserts} records with validations"
+    group << bm( description ) { model_clazz.import cols, vals, :validate=>true }
+
+    description = "#{model_clazz.name}.import(columns, values) for #{num_inserts} records without validations"
+    group << bm( description ) { model_clazz.import cols, vals, :validate=>false }
+
+    models = []
+    array_of_attrs = []
+    
+    vals.each do |arr|
+      array_of_attrs << (attrs={})
+      arr.each_with_index { |value, i| attrs[cols[i]] = value }
+    end
+    array_of_attrs.each{ |attrs| models << model_clazz.new(attrs) }
+
+    description = "#{model_clazz.name}.import(models) for #{num_inserts} records with validations"
+    group << bm( description ) { model_clazz.import models, :validate=>true }
+
+    description = "#{model_clazz.name}.import(models) for #{num_inserts} records without validations"
+    group << bm( description ) { model_clazz.import models, :validate=>false }
+
+    true
+  end
+
+  # Returns a two element array composing of an array of columns and an array of
+  # value sets given the passed +num+.
+  #
+  # === What is a value set?
+  # A value set is an array of arrays. Each child array represents an array of value sets
+  # for a given row of data. 
+  #
+  # For example, say we wanted to represent an insertion of two records:
+  #   column_names = [ 'id', 'name', 'description' ]
+  #   record1 = [ 1, 'John Doe', 'A plumber' ]
+  #   record2 = [ 2, 'John Smith', 'A painter' ]
+  #   value_set [ record1, record2 ]
+  #
+  # == PARAMETER
+  #  * num - the number of records to create
+  def build_array_of_cols_and_vals( num )
+    cols = [ :my_name, :description ]
+    value_sets = []
+    num.times { |i| value_sets << [ "My Name #{i}", "My Description #{i}" ] }
+    [ cols, value_sets ]
+  end
+
+  # Returns a hash of column identifier to value mappings giving the passed in
+  # value array. 
+  #
+  # Example:
+  #   cols = [ 'id', 'name', 'description' ]
+  #   values = [ 1, 'John Doe', 'A plumber' ]
+  #   hsh = create_hash_for_cols_and_vals( cols, values )   
+  #   # hsh => { 'id'=>1, 'name'=>'John Doe', 'description'=>'A plumber' }
+  def create_hash_for_cols_and_vals( cols, vals )
+    h = {}
+    cols.zip( vals ){ |col,val| h[col] = val }
+    h
+  end
+
+  # Deletes all records from all ActiveRecord subclasses
+  def delete_all
+    ActiveRecord::Base.send( :subclasses ).each do |subclass|
+      subclass.delete_all if subclass.respond_to? :delete_all
+    end
+  end
+
+  def initialize   # :nodoc:
+    @results = []
+  end
+	
+end

data/benchmarks/lib/cli_parser.rb

@@ -0,0 +1,103 @@
+require 'optparse'
+require 'ostruct'
+
+# 
+# == PARAMETERS
+# * a - database adapter. ie: mysql, postgresql, oracle, etc.
+# * n - number of objects to test with. ie: 1, 100, 1000, etc.
+# * t - the table types to test. ie: myisam, innodb, memory, temporary, etc.
+#
+module BenchmarkOptionParser
+  BANNER = "Usage: ruby #{$0} [options]\nSee ruby #{$0} -h for more options."
+
+  def self.print_banner
+    puts BANNER
+  end
+  
+  def self.print_banner!
+    print_banner
+    exit
+  end
+  
+  def self.print_options( options )
+    puts "Benchmarking the following options:"
+    puts "  Database adapter: #{options.adapter}"
+    puts "  Number of objects: #{options.number_of_objects}"
+    puts "  Table types:"
+    print_valid_table_types( options, :prefix=>"    " )
+  end
+  
+  # TODO IMPLEMENT THIS
+  def self.print_valid_table_types( options, hsh={:prefix=>''} )
+    if options.table_types.keys.size > 0
+      options.table_types.keys.sort.each{ |type| puts hsh[:prefix].to_s + type.to_s }
+    else
+      puts 'No table types defined.'
+    end
+  end
+  
+  def self.parse( args )
+    options = OpenStruct.new( 
+      :table_types => {}, 
+      :delete_on_finish => true,
+      :number_of_objects => [],
+      :outputs => [] )
+    
+    opts = OptionParser.new do |opts|
+      opts.banner = BANNER
+      
+      # parse the database adapter
+      opts.on( "a", "--adapter [String]", 
+        "The database adapter to use. IE: mysql, postgresql, oracle" ) do |arg|
+        options.adapter = arg
+      end
+      
+      # parse do_not_delete flag
+      opts.on( "d", "--do-not-delete", 
+        "By default all records in the benchmark tables will be deleted at the end of the benchmark. " +
+        "This flag indicates not to delete the benchmark data." ) do |arg|
+        options.delete_on_finish = false
+      end
+      
+      # parse the number of row objects to test
+      opts.on( "n", "--num [Integer]",
+        "The number of objects to benchmark." ) do |arg|
+        options.number_of_objects << arg.to_i
+      end
+      
+      # parse the table types to test
+      opts.on( "t", "--table-type [String]",
+        "The table type to test. This can be used multiple times." ) do |arg|
+        if arg =~ /^all$/
+          options.table_types['all'] = options.benchmark_all_types = true
+        else
+          options.table_types[arg] = true
+        end        
+      end
+
+      # print results in CSV format
+      opts.on( "--to-csv [String]", "Print results in a CSV file format" ) do |filename|
+        options.outputs << OpenStruct.new( :format=>'csv', :filename=>filename)
+      end
+
+      # print results in HTML format
+      opts.on( "--to-html [String]", "Print results in HTML format" ) do |filename|
+        options.outputs << OpenStruct.new( :format=>'html', :filename=>filename )
+      end
+    end #end opt.parse!
+
+    begin 
+      opts.parse!( args )
+      if options.table_types.size == 0
+        options.table_types['all'] = options.benchmark_all_types = true
+      end
+    rescue Exception => ex
+      print_banner!
+    end
+    
+    print_options( options )
+    
+    options
+  end
+
+end