splashy 0.1.0 → 0.1.1

data/README.markdown

@@ -41,20 +41,24 @@ buckets.select
 #     :b => ["20", "21"],
 #     :c => ["30", "31", "32", "33", "34", "35", "36", "37"]
 #   }
+
+# Or even grab randomly:
+
+buckets.select( :random => true )
 ```
 
 Changelog
 =========
 
-* 0.1.0 - Several bug fixes, add "neediest_buckets" method to Buckets to allow
-you to choose which buckets to add to first if an element can be put in
+* *0.1.1* - Add support for random selection from buckets.
+* *0.1.0* - Several bug fixes, add `#neediest_buckets` method to `Buckets` to
+allow you to choose which buckets to add to first if an element can be put in
 multiple buckets, final distributions can now have empty buckets if it means
-we meet the wanted distribution better (i.e. a 99% % 1% distribution with 5
-and 1 elements, respectively, which will now select 4 and 0 elements if your
-wanted count is 4).
-* 0.0.2 - Raise `ArgumentError` when trying to add to a bucket that doesn't 
+we meet the wanted distribution better (i.e. a 99% to 1% wanted distribution 
+with 5 and 1 elements, respectively, which will now select 4 and 0 elements if your wanted count is 4).
+* *0.0.2* - Raise `ArgumentError` when trying to add to a bucket that doesn't 
 exist, don't consider an empty bucket "satisfied".
-* 0.0.1 - Initial release.
+* *0.0.1* - Initial release.
 
 Contributing
 ============

data/VERSION

@@ -1 +1 @@
-0.1.0
+0.1.1

data/lib/splashy/bucket.rb

@@ -1,7 +1,7 @@
 module Splashy
   private
   
-  # Private: Collector of elements.
+  # Private: Collects elements and maintains a count.
   class Bucket
     attr_reader :name
     
@@ -22,6 +22,14 @@ module Splashy
       end
     end
     
+    def random_elements( count = nil )
+      if @elements.respond_to?( :sample )
+        @elements.sample( count )
+      else
+        count ? @elements.sort_by{ rand }[0, count] : @elements.sort_by{ rand }
+      end
+    end
+    
     def empty?
       self.count == 0
     end
@@ -30,4 +38,4 @@ module Splashy
       @elements.count
     end
   end
-end
+end

data/lib/splashy/buckets.rb

@@ -69,9 +69,12 @@ module Splashy
     # 
     # Returns a Hash of elements matching the desired distribution, keyed by
     # the bucket names.
-    def select
+    def select( opts = {} )
       self.assert_satisfied!
-      selected = self._select_wanted
+      opts = { :random => false }.merge( opts )
+      
+      selected = self._select_wanted( opts[:random] )
+      
       # Sometimes we need to fudge by a few to meet the `@wanted_count`
       selected = self.trim( selected, @wanted_count ) if @wanted_count
       selected
@@ -101,13 +104,17 @@ module Splashy
     # 
     # Returns Hash of bucket elements, matching the wanted distribution as
     # closely as possible.
-    def _select_wanted
+    def _select_wanted( randomly = false )
       final_count = self.estimated_final_count
       
       @buckets.values.inject({}) do |memo, bucket|
         count = ( final_count * @wanted_distribution[bucket.name] ).round
         count = [1, count].max # Ensure every bucket has at least one element
-        memo[bucket.name] = bucket.elements( count )
+        if randomly
+          memo[bucket.name] = bucket.random_elements( count )
+        else
+          memo[bucket.name] = bucket.elements( count )
+        end
         memo
       end
     end

data/splashy.gemspec

@@ -4,14 +4,14 @@
 # -*- encoding: utf-8 -*-
 
 Gem::Specification.new do |s|
-  s.name = %q{splashy}
-  s.version = "0.1.0"
+  s.name = "splashy"
+  s.version = "0.1.1"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Tyson Tate"]
-  s.date = %q{2011-12-09}
-  s.description = %q{Simple distribution-based sampling of arbitrary objects from any number of buckets. Splashy. Buckets. Get it!?}
-  s.email = %q{tyson@tysontate.com}
+  s.date = "2011-12-13"
+  s.description = "Simple distribution-based sampling of arbitrary objects from any number of buckets. Splashy. Buckets. Get it!?"
+  s.email = "tyson@tysontate.com"
   s.extra_rdoc_files = [
     "LICENSE.txt",
     "README.markdown"
@@ -23,7 +23,6 @@ Gem::Specification.new do |s|
     "README.markdown",
     "Rakefile",
     "VERSION",
-    "buckets.gemspec",
     "lib/splashy.rb",
     "lib/splashy/bucket.rb",
     "lib/splashy/buckets.rb",
@@ -31,11 +30,11 @@ Gem::Specification.new do |s|
     "test/helper.rb",
     "test/test_splashy_buckets.rb"
   ]
-  s.homepage = %q{http://github.com/tysontate/splashy}
+  s.homepage = "http://github.com/tysontate/splashy"
   s.licenses = ["MIT"]
   s.require_paths = ["lib"]
-  s.rubygems_version = %q{1.4.2}
-  s.summary = %q{Simple distribution-based sampling of arbitrary objects from buckets.}
+  s.rubygems_version = "1.8.11"
+  s.summary = "Simple distribution-based sampling of arbitrary objects from buckets."
 
   if s.respond_to? :specification_version then
     s.specification_version = 3

data/test/helper.rb

@@ -8,7 +8,6 @@ rescue Bundler::BundlerError => e
   exit e.status_code
 end
 require 'minitest/autorun'
-# require 'minitest/unit'
 require 'minitest/benchmark'
 
 $LOAD_PATH.unshift(File.dirname(__FILE__))

data/test/test_splashy_buckets.rb

@@ -213,7 +213,7 @@ describe Splashy::Buckets do
     end
     
     it "selects from a pool with an uneven distribution" do
-      @buckets = Splashy::Buckets.new( {:a => 0.33, :b => 0.33, :c => 0.34}, 5 )
+      @buckets = Splashy::Buckets.new( {:a => 0.34, :b => 0.32, :c => 0.34}, 5 )
       fill_with_counts( 10, 2, 40 )
       assert @buckets.satisfied?
       assert_equal(
@@ -241,6 +241,18 @@ describe Splashy::Buckets do
         @buckets.select
       )
     end
+    
+    it "selects randomly" do
+      @buckets = Splashy::Buckets.new( {:a => 0.5, :b => 0.20, :c => 0.30}, 10 )
+      fill_with_counts( 10, 10, 10 )
+      assert @buckets.satisfied?
+      randomized = @buckets.select( :random => true )
+      assert_equal( 5, randomized[:a].size )
+      assert_equal( 2, randomized[:b].size )
+      assert_equal( 3, randomized[:c].size )
+      # Non-deterministic, obv, but whatever.
+      assert_equal( false, randomized[:a].sort == randomized[:a] )
+    end
   end
   
   describe "variances" do
@@ -263,7 +275,7 @@ describe Splashy::Buckets do
     end
     
     it "reports on a pool with a wacky distribution" do
-      @buckets = Splashy::Buckets.new( {:a => 0.01, :b => 0.01, :c => 0.98} )
+      @buckets = Splashy::Buckets.new( {:a => 0.02, :b => 0.01, :c => 0.97} )
       fill_with_counts( 3, 3, 3 )
       assert_equal( [:c, :a, :b], @buckets.neediest_buckets )
     end
@@ -283,4 +295,3 @@ describe Splashy::Buckets do
     end
   end
 end
-

metadata

@@ -1,125 +1,96 @@
---- !ruby/object:Gem::Specification 
+--- !ruby/object:Gem::Specification
 name: splashy
-version: !ruby/object:Gem::Version 
-  hash: 27
+version: !ruby/object:Gem::Version
+  version: 0.1.1
   prerelease: 
-  segments: 
-  - 0
-  - 1
-  - 0
-  version: 0.1.0
 platform: ruby
-authors: 
+authors:
 - Tyson Tate
 autorequire: 
 bindir: bin
 cert_chain: []
-
-date: 2011-12-09 00:00:00 -08:00
-default_executable: 
-dependencies: 
-- !ruby/object:Gem::Dependency 
-  prerelease: false
+date: 2011-12-13 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
   name: minitest
-  type: :development
-  version_requirements: &id001 !ruby/object:Gem::Requirement 
+  requirement: &70277083322720 !ruby/object:Gem::Requirement
     none: false
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        hash: 3
-        segments: 
-        - 0
-        version: "0"
-  requirement: *id001
-- !ruby/object:Gem::Dependency 
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
   prerelease: false
+  version_requirements: *70277083322720
+- !ruby/object:Gem::Dependency
   name: bundler
-  type: :development
-  version_requirements: &id002 !ruby/object:Gem::Requirement 
+  requirement: &70277083321600 !ruby/object:Gem::Requirement
     none: false
-    requirements: 
+    requirements:
     - - ~>
-      - !ruby/object:Gem::Version 
-        hash: 23
-        segments: 
-        - 1
-        - 0
-        - 0
+      - !ruby/object:Gem::Version
         version: 1.0.0
-  requirement: *id002
-- !ruby/object:Gem::Dependency 
+  type: :development
   prerelease: false
+  version_requirements: *70277083321600
+- !ruby/object:Gem::Dependency
   name: jeweler
-  type: :development
-  version_requirements: &id003 !ruby/object:Gem::Requirement 
+  requirement: &70277083320380 !ruby/object:Gem::Requirement
     none: false
-    requirements: 
+    requirements:
     - - ~>
-      - !ruby/object:Gem::Version 
-        hash: 7
-        segments: 
-        - 1
-        - 6
-        - 4
+      - !ruby/object:Gem::Version
         version: 1.6.4
-  requirement: *id003
-description: Simple distribution-based sampling of arbitrary objects from any number of buckets. Splashy. Buckets. Get it!?
+  type: :development
+  prerelease: false
+  version_requirements: *70277083320380
+description: Simple distribution-based sampling of arbitrary objects from any number
+  of buckets. Splashy. Buckets. Get it!?
 email: tyson@tysontate.com
 executables: []
-
 extensions: []
-
-extra_rdoc_files: 
+extra_rdoc_files:
 - LICENSE.txt
 - README.markdown
-files: 
+files:
 - Gemfile
 - Gemfile.lock
 - LICENSE.txt
 - README.markdown
 - Rakefile
 - VERSION
-- buckets.gemspec
 - lib/splashy.rb
 - lib/splashy/bucket.rb
 - lib/splashy/buckets.rb
 - splashy.gemspec
 - test/helper.rb
 - test/test_splashy_buckets.rb
-has_rdoc: true
 homepage: http://github.com/tysontate/splashy
-licenses: 
+licenses:
 - MIT
 post_install_message: 
 rdoc_options: []
-
-require_paths: 
+require_paths:
 - lib
-required_ruby_version: !ruby/object:Gem::Requirement 
+required_ruby_version: !ruby/object:Gem::Requirement
   none: false
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      hash: 3
-      segments: 
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+      segments:
       - 0
-      version: "0"
-required_rubygems_version: !ruby/object:Gem::Requirement 
+      hash: 4086114583339023412
+required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      hash: 3
-      segments: 
-      - 0
-      version: "0"
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
 requirements: []
-
 rubyforge_project: 
-rubygems_version: 1.4.2
+rubygems_version: 1.8.11
 signing_key: 
 specification_version: 3
 summary: Simple distribution-based sampling of arbitrary objects from buckets.
 test_files: []
-

data/buckets.gemspec

@@ -1,57 +0,0 @@
-# Generated by jeweler
-# DO NOT EDIT THIS FILE DIRECTLY
-# Instead, edit Jeweler::Tasks in Rakefile, and run 'rake gemspec'
-# -*- encoding: utf-8 -*-
-
-Gem::Specification.new do |s|
-  s.name = %q{buckets}
-  s.version = "0.0.1"
-
-  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
-  s.authors = ["Tyson Tate"]
-  s.date = %q{2011-12-08}
-  s.description = %q{Simple distribution-based sampling of arbitrary objects via the use of, well, buckets.}
-  s.email = %q{tyson@tysontate.com}
-  s.extra_rdoc_files = [
-    "LICENSE.txt",
-    "README.markdown"
-  ]
-  s.files = [
-    "Gemfile",
-    "Gemfile.lock",
-    "LICENSE.txt",
-    "README.markdown",
-    "Rakefile",
-    "VERSION",
-    "buckets.gemspec",
-    "lib/splashy.rb",
-    "lib/splashy/bucket.rb",
-    "lib/splashy/buckets.rb",
-    "test/helper.rb",
-    "test/test_splashy_buckets.rb"
-  ]
-  s.homepage = %q{http://github.com/tysontate/buckets}
-  s.licenses = ["MIT"]
-  s.require_paths = ["lib"]
-  s.rubygems_version = %q{1.4.2}
-  s.summary = %q{Simple distribution-based sampling of arbitrary objects.}
-
-  if s.respond_to? :specification_version then
-    s.specification_version = 3
-
-    if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
-      s.add_development_dependency(%q<minitest>, [">= 0"])
-      s.add_development_dependency(%q<bundler>, ["~> 1.0.0"])
-      s.add_development_dependency(%q<jeweler>, ["~> 1.6.4"])
-    else
-      s.add_dependency(%q<minitest>, [">= 0"])
-      s.add_dependency(%q<bundler>, ["~> 1.0.0"])
-      s.add_dependency(%q<jeweler>, ["~> 1.6.4"])
-    end
-  else
-    s.add_dependency(%q<minitest>, [">= 0"])
-    s.add_dependency(%q<bundler>, ["~> 1.0.0"])
-    s.add_dependency(%q<jeweler>, ["~> 1.6.4"])
-  end
-end
-