postgresql_cursor 0.2.0

data/.document

@@ -0,0 +1,5 @@
+README.rdoc
+lib/**/*.rb
+bin/*
+features/**/*.feature
+LICENSE

data/.gitignore

@@ -0,0 +1,21 @@
+## MAC OS
+.DS_Store
+
+## TEXTMATE
+*.tmproj
+tmtags
+
+## EMACS
+*~
+\#*
+.\#*
+
+## VIM
+*.swp
+
+## PROJECT::GENERAL
+coverage
+rdoc
+pkg
+
+## PROJECT::SPECIFIC

data/LICENSE

@@ -0,0 +1,20 @@
+Copyright (c) 2009 Allen Fair
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

data/README.rdoc

@@ -0,0 +1,78 @@
+= PostgreSQLCursor
+
+PostgreSQL Cursor is an extension to the ActiveRecord PostgreSQLAdapter for very large result sets.
+It provides a cursor open/fetch/close interface to access data without loading all rows into memory,
+and instead loads the result rows in "chunks" (default of 10_000 rows), buffers them, and returns the
+rows one at a time.
+
+For web pages, an application would not want to process a large amount of data, usually employing a 
+Pagination scheme to present it to the users. Background processes sometimes need to generate a large 
+amount of data, and the ActiveRecord approach to load all data into memory is not the best fit here.
+
+Previous solutions employ pagination to fetch each block, 
+then re-running the query for the next "page". This gem avoids re-executing
+the query by using the PostgreSQL cursors. 
+
+Like the #find_by_sql method, #find_cursor returns each row as a hash instead of an instantiated 
+model class. The rationale for this is performance, though an option to return instances is available. 
+Julian's benchmarks showed returning instances was a factor of 4 slower than return the hash.
+
+==Installation
+ [sudo] gem install postgresql_cursor
+
+This does not require Rails to work, just ActiveRecord < 3.0.0 and the 'pg' gem.
+
+==Usage
+
+A Rails/ActiveRecord plugin for the PostgreSQL database adapter that will add
+cursors to a find_cursor() method to process very large result sets. 
+
+the *find_cursor* method uses cursors to pull in one data block (of x records) at a time, and 
+return each record as a Hash to a procedural block. When each data block is 
+exhausted, it will fetch the next one.
+
+*find_by_sql_with_cursor* takes a custom SQL statement and returns each row.
+
+==Examples
+
+ Account.find_with_cursor(:conditions=>["status = ?", 'good']).each do |row|
+   puts row.to_json
+ end
+ 
+ Account.find_by_sql_with_cursor("select ...", :buffer_size=>1000) do |row|
+   row.process
+ end
+
+ Account.transaction do 
+   cursor = Account.find_with_cursor(...) { |record| record.symbolize_keys }
+   while record = cursor.next do
+     record.process # => {:column=>value, ...}
+     cursor.close if cursor.count > 1000 # Halts loop after 1000 records
+   end
+ end
+
+ Account.find_with_cursor(...) { |record| record.symbolize_keys }.each do |row|
+   row.process
+ end
+
+==Authors
+Allen Fair, allen.fair@gmail.com, http://github.com/afair
+
+Thank you to:
+* Iulian Dogariu, http://github.com/iulianu (Fixes)
+* Julian Mehnle, http://www.mehnle.net (Suggestions)
+
+
+== Note on Patches/Pull Requests
+ 
+* Fork the project.
+* Make your feature addition or bug fix.
+* Add tests for it. This is important so I don't break it in a
+  future version unintentionally.
+* Commit, do not mess with rakefile, version, or history.
+  (if you want to have your own version, that is fine but bump version in a commit by itself I can ignore when I pull)
+* Send me a pull request. Bonus points for topic branches.
+
+== Copyright
+
+Copyright (c) 2010 Allen Fair. See LICENSE for details.

data/Rakefile

@@ -0,0 +1,54 @@
+require 'rubygems'
+require 'rake'
+
+begin
+  require 'jeweler'
+  Jeweler::Tasks.new do |gem|
+    gem.name = "postgresql_cursor"
+    gem.summary = %Q{ActiveRecord PostgreSQL Adapter extension for using a cursor to return a large result set}
+    gem.description = %Q{PostgreSQL Cursor is an extension to the ActiveRecord PostgreSQLAdapter for very large result sets. It provides a cursor open/fetch/close interface to access data without loading all rows into memory, and instead loads the result rows in "chunks" (default of 10_000 rows), buffers them, and returns the rows one at a time.}
+    gem.email = "allen.fair@gmail.com"
+    gem.homepage = "http://github.com/afair/postgresql_cursor"
+    gem.authors = ["Allen Fair"]
+    gem.add_dependency 'activerecord', '<=2.3.5'
+    gem.add_dependency 'pg'
+    # 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
+
+require 'rake/testtask'
+Rake::TestTask.new(:test) do |test|
+  test.libs << 'lib' << 'test'
+  test.pattern = 'test/**/test_*.rb'
+  test.verbose = true
+end
+
+begin
+  require 'rcov/rcovtask'
+  Rcov::RcovTask.new do |test|
+    test.libs << 'test'
+    test.pattern = 'test/**/test_*.rb'
+    test.verbose = true
+  end
+rescue LoadError
+  task :rcov do
+    abort "RCov is not available. In order to run rcov, you must: sudo gem install spicycode-rcov"
+  end
+end
+
+task :test => :check_dependencies
+
+task :default => :test
+
+require 'rake/rdoctask'
+Rake::RDocTask.new do |rdoc|
+  version = File.exist?('VERSION') ? File.read('VERSION') : ""
+
+  rdoc.rdoc_dir = 'rdoc'
+  rdoc.title = "postgresql_cursor #{version}"
+  rdoc.rdoc_files.include('README*')
+  rdoc.rdoc_files.include('lib/**/*.rb')
+end

data/VERSION

@@ -0,0 +1 @@
+0.2.0

data/lib/postgresql_cursor.rb

@@ -0,0 +1,121 @@
+gem 'activerecord', '<=2.3.5'
+require 'active_record'
+
+# Class to operate a PostgreSQL cursor to buffer a set of rows, and return single rows for processing.
+# Use this class when processing a very large number of records, which would otherwise all be instantiated
+# in memory by find(). This also adds helpers to ActiveRecord::Base for *find_with_cursor()* and 
+# *find_by_sql_with_cursor()* to return instances of the cursor ready to fetch. 
+#
+# Use each() with a block to accept an instance of the Model (or whatever you define with a block on 
+# initialize()). It will open, buffer, yield each row, then close the cursor.
+#
+# PostgreSQL requires that a cursor is executed within a transaction block, which you must provide unless
+# you use each() to run through the result set.
+class PostgreSQLCursor
+  attr_reader :count, :buffer_reads
+  @@counter=0
+  
+  # Define a new cursor, with a SQL statement, as a string with parameters already replaced, and options for 
+  # the cursor
+  # * :buffer_size=>number of records to buffer, default 10000.
+  # Pass a optional block which takes a Hash of "column"=>"value", and returns an object to be yielded for each row.
+  def initialize(sql,*args, &block)
+    @options = args.last.is_a?(Hash) ? args.pop : {}
+    @@counter += 1
+    @instantiator = block || lambda {|r| r }
+    @sql = sql
+    @name = "pgcursor_#{@@counter}"
+    @connection = ActiveRecord::Base.connection
+    @buffer_size = @options[:buffer_size] || 10_000 
+    @count = 0
+    @state = :ready
+  end
+  
+  # Iterates through the rows, yields them to the block. It wraps the processing in a transaction 
+  # (required by PostgreSQL), opens the cursor, buffers the results, returns each row, and closes the cursor.
+  def each
+    @connection.transaction do
+      @result = open 
+      while (row = fetch ) do
+        yield row
+      end
+      close 
+      @count
+    end
+  end
+  
+  # Starts buffered result set processing for a given SQL statement. The DB
+  def open
+    raise "Open Cursor state not ready" unless @state == :ready
+    @result = @connection.execute("declare #{@name} cursor for #{@sql}")
+    @state = :empty
+    @buffer_reads = 0
+    @buffer = nil
+  end
+
+  # Returns a string of the current status
+  def status #:nodoc:
+    "row=#{@count} buffer=#{@buffer.size} state=#{@state} buffer_size=#{@buffer_size} reads=#{@buffer_reads}"
+  end
+
+  # Fetches the next block of rows into memory
+  def fetch_buffer #:nodoc:
+    return unless @state == :empty
+    @result = @connection.execute("fetch #{@buffer_size} from #{@name}")
+    @buffer = @result.collect {|row| row }
+    @state  = @buffer.size > 0 ? :buffered : :eof
+    @buffer_reads += 1
+    @buffer
+  end
+
+  # Returns the next row from the cursor, or nil when end of data.
+  # The row returned is a hash[:colname]
+  def fetch
+    open         if @state == :ready
+    fetch_buffer if @state == :empty
+    return nil   if @state == :eof || @state == :closed
+    @state = :empty if @buffer.size <= 1
+    @count+= 1
+    row = @buffer.shift
+    @instantiator.call(row)
+  end
+  
+  alias_method :next, :fetch 
+  
+  # Closes the cursor to clean up resources. Call this method during process of each() to 
+  # exit the loop
+  def close 
+    pg_result = @connection.execute("close #{@name}")
+    @state = :closed
+  end
+  
+end
+
+class ActiveRecord::Base
+  class <<self
+  
+    # Returns a PostgreSQLCursor instance to access the results, on which you are able to call
+    # each (though the cursor is not Enumerable and no other methods are available).
+    # No :all argument is needed, and other find() options can be specified.
+    # Specify the :cursor=>{...} option to override options for the cursor such has :buffer_size=>n.
+    # Pass an optional block that takes a Hash of the record and returns what you want to return.
+    # For example, return the Hash back to process a Hash instead of a table instance for better speed.
+    def find_with_cursor(*args, &block)
+      find_options = args.last.is_a?(Hash) ? args.pop : {}
+      options = find_options.delete(:cursor) || {}
+      validate_find_options(find_options)
+      set_readonly_option!(find_options)
+      sql = construct_finder_sql(find_options)
+      PostgreSQLCursor.new(sql, options) { |r| block_given? ? yield(r) : instantiate(r) }
+    end
+  
+    # Returns a PostgreSQLCursor instance to access the results of the sql
+    # Specify the :cursor=>{...} option to override options for the cursor such has :buffer_size=>n.
+    # Pass an optional block that takes a Hash of the record and returns what you want to return.
+    # For example, return the Hash back to process a Hash instead of a table instance for better speed.
+    def find_by_sql_with_cursor(sql, options={})
+      PostgreSQLCursor.new(sql, options) { |r| block_given? ? yield(r) : instantiate(r) }
+    end
+
+  end
+end

data/postgresql_cursor.gemspec

@@ -0,0 +1,57 @@
+# Generated by jeweler
+# DO NOT EDIT THIS FILE DIRECTLY
+# Instead, edit Jeweler::Tasks in Rakefile, and run the gemspec command
+# -*- encoding: utf-8 -*-
+
+Gem::Specification.new do |s|
+  s.name = %q{postgresql_cursor}
+  s.version = "0.2.0"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
+  s.authors = ["Allen Fair"]
+  s.date = %q{2010-05-17}
+  s.description = %q{PostgreSQL Cursor is an extension to the ActiveRecord PostgreSQLAdapter for very large result sets. It provides a cursor open/fetch/close interface to access data without loading all rows into memory, and instead loads the result rows in "chunks" (default of 10_000 rows), buffers them, and returns the rows one at a time.}
+  s.email = %q{allen.fair@gmail.com}
+  s.extra_rdoc_files = [
+    "LICENSE",
+     "README.rdoc"
+  ]
+  s.files = [
+    ".document",
+     ".gitignore",
+     "LICENSE",
+     "README.rdoc",
+     "Rakefile",
+     "VERSION",
+     "lib/postgresql_cursor.rb",
+     "postgresql_cursor.gemspec",
+     "test/helper.rb",
+     "test/test_postgresql_cursor.rb"
+  ]
+  s.homepage = %q{http://github.com/afair/postgresql_cursor}
+  s.rdoc_options = ["--charset=UTF-8"]
+  s.require_paths = ["lib"]
+  s.rubygems_version = %q{1.3.6}
+  s.summary = %q{ActiveRecord PostgreSQL Adapter extension for using a cursor to return a large result set}
+  s.test_files = [
+    "test/helper.rb",
+     "test/test_postgresql_cursor.rb"
+  ]
+
+  if s.respond_to? :specification_version then
+    current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
+    s.specification_version = 3
+
+    if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
+      s.add_runtime_dependency(%q<activerecord>, ["<= 2.3.5"])
+      s.add_runtime_dependency(%q<pg>, [">= 0"])
+    else
+      s.add_dependency(%q<activerecord>, ["<= 2.3.5"])
+      s.add_dependency(%q<pg>, [">= 0"])
+    end
+  else
+    s.add_dependency(%q<activerecord>, ["<= 2.3.5"])
+    s.add_dependency(%q<pg>, [">= 0"])
+  end
+end
+

data/test/helper.rb

@@ -0,0 +1,23 @@
+require 'rubygems'
+require 'test/unit'
+
+$LOAD_PATH.unshift(File.dirname(__FILE__))
+$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+require 'postgresql_cursor'
+
+ActiveRecord::Base.establish_connection :database=>'allen_test', :adapter=>'postgresql', :username=>'allen'
+class Model < ActiveRecord::Base
+  set_table_name "records"
+
+  # create table records (id serial primary key);
+  def self.generate(max=1_000_000)
+    max.times do
+      connection.execute("insert into records values (nextval('records_id_seq'::regclass))")
+    end
+  end
+end
+
+Model.generate(1000) if Model.count == 0
+
+class Test::Unit::TestCase
+end

data/test/test_postgresql_cursor.rb

@@ -0,0 +1,44 @@
+require 'helper'
+
+class TestPostgresqlCursor < Test::Unit::TestCase
+  
+  def test_cursor
+    c = Model.find_with_cursor(:conditions=>["id>?",0], :cursor=>{:buffer_size=>10})
+    mycount=0
+    count = c.each { |r| mycount += 1 } 
+    assert_equal mycount, count
+  end
+
+  def test_empty_set
+    c = Model.find_with_cursor(:conditions=>["id<?",0])
+    count = c.each { |r| puts r.class }
+    assert_equal count, 0
+  end
+
+  def test_block
+    Model.transaction do
+      c = Model.find_with_cursor(:conditions=>["id<?",10]) { |r| r }
+      r = c.next
+      assert_equal r.class, Hash
+    end
+  end
+
+  def test_sql
+    c = Model.find_by_sql_with_cursor("select * from #{Model.table_name}")
+    mycount=0
+    count = c.each { |r| mycount += 1 } 
+    assert_equal mycount, count
+  end
+
+  def test_loop
+   Model.transaction do 
+     cursor = Model.find_with_cursor() { |record| record.symbolize_keys }
+     while record = cursor.next do
+       assert record[:id].class, Fixnum
+       cursor.close if cursor.count >= 10 
+     end
+     assert_equal cursor.count, 10
+   end
+  end
+
+end

metadata

@@ -0,0 +1,98 @@
+--- !ruby/object:Gem::Specification 
+name: postgresql_cursor
+version: !ruby/object:Gem::Version 
+  prerelease: false
+  segments: 
+  - 0
+  - 2
+  - 0
+  version: 0.2.0
+platform: ruby
+authors: 
+- Allen Fair
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2010-05-17 00:00:00 -04:00
+default_executable: 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: activerecord
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
+    requirements: 
+    - - <=
+      - !ruby/object:Gem::Version 
+        segments: 
+        - 2
+        - 3
+        - 5
+        version: 2.3.5
+  type: :runtime
+  version_requirements: *id001
+- !ruby/object:Gem::Dependency 
+  name: pg
+  prerelease: false
+  requirement: &id002 !ruby/object:Gem::Requirement 
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        segments: 
+        - 0
+        version: "0"
+  type: :runtime
+  version_requirements: *id002
+description: PostgreSQL Cursor is an extension to the ActiveRecord PostgreSQLAdapter for very large result sets. It provides a cursor open/fetch/close interface to access data without loading all rows into memory, and instead loads the result rows in "chunks" (default of 10_000 rows), buffers them, and returns the rows one at a time.
+email: allen.fair@gmail.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: 
+- LICENSE
+- README.rdoc
+files: 
+- .document
+- .gitignore
+- LICENSE
+- README.rdoc
+- Rakefile
+- VERSION
+- lib/postgresql_cursor.rb
+- postgresql_cursor.gemspec
+- test/helper.rb
+- test/test_postgresql_cursor.rb
+has_rdoc: true
+homepage: http://github.com/afair/postgresql_cursor
+licenses: []
+
+post_install_message: 
+rdoc_options: 
+- --charset=UTF-8
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      segments: 
+      - 0
+      version: "0"
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      segments: 
+      - 0
+      version: "0"
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.6
+signing_key: 
+specification_version: 3
+summary: ActiveRecord PostgreSQL Adapter extension for using a cursor to return a large result set
+test_files: 
+- test/helper.rb
+- test/test_postgresql_cursor.rb