group_by_hash 0.1.0

data/Rakefile

@@ -0,0 +1,32 @@
+task :default => :spec
+task :test    => :spec
+  
+desc "Run specs"
+task :spec do
+  exec "spec spec/"
+end
+
+desc "Build a gem"
+task :gem => [ :gemspec, :build ]
+ 
+desc "Run specs"
+task :spec do
+  exec "spec spec/"
+end
+
+begin
+  require 'jeweler'
+  Jeweler::Tasks.new do |gemspec|
+    gemspec.name = "group_by_hash"
+    gemspec.summary = "ruby hash with a sql-esque group by method"
+    gemspec.description = <<END
+A ruby hash with a sql-esque group by method for when you can't do a group by in SQL because it is too complicated
+END
+    gemspec.email = "paulboone@mindbucket.com"
+    gemspec.homepage = "http://github.com/paulboone/group_by_hash"
+    gemspec.authors = ["Paul Boone"]
+  end
+rescue LoadError
+  warn "Jeweler not available. Install it with:"
+  warn "gem install jeweler"
+end

data/VERSION

@@ -0,0 +1 @@
+0.1.0

data/lib/group_by_hash.rb

@@ -0,0 +1,41 @@
+=begin
+
+=end
+class GroupByHash < Hash
+  
+  #
+  # group_by_fields is a list of symbols that MUST be available on datasets passed in via <<
+  # blk is a block that returns a hash where each field will be += together 
+  def initialize(group_by_fields, &blk)
+    @group_by_fields = group_by_fields
+    @blk = blk
+  end
+  
+  def flatten
+    a = []
+    each_pair do |k,v|
+      a << k.merge(v)
+    end
+    a
+  end
+  
+  def <<(array_or_hash)
+    h = array_or_hash.clone
+    key = h.reject{|k,v| ! @group_by_fields.include?(k)}
+    val = h.reject{|k,v|   @group_by_fields.include?(k)}
+    if self[key]
+      self[key] = add_hash_fields(self[key],@blk.call(val))
+    else
+      self[key] = @blk.call(val)
+    end
+    self
+  end
+  
+  def add_hash_fields(h1,h2)
+    h = {}
+    h1.each_pair do |k,v|
+      h[k] = v + h2[k]
+    end
+    h
+  end
+end

data/spec/group_by_hash_spec.rb

@@ -0,0 +1,92 @@
+require 'spec_helper'
+
+describe GroupByHash do
+  it "should be defineable" do
+    g = GroupByHash.new([:col1,:col2,:col3]) {|h| {:count => 1, :random_sum => rand(500)}}    
+  end
+  
+  describe "add_hash_fields" do
+    it "should handle simple addition" do
+      GroupByHash.new(nil).add_hash_fields({:count => 5},{:count => 25}).should == {:count => 30}      
+    end
+  end
+  
+  describe "when flattening" do
+    before(:each) do      
+      @g = GroupByHash.new([:col1]) {|h| {:count => 1}}        
+    end
+    
+    it "flattening and empty hash should == []" do
+      @g.flatten.should == []
+    end
+    
+    it "adding one row" do
+      @g << {:col1 => 'disc1'}
+      fg = @g.flatten
+      fg[0].has_key?(:col1).should == true
+      fg[0].has_key?(:count).should == true
+      fg[0][:col1].should == 'disc1'
+      fg[0][:count].should == 1
+    end
+    
+    it "adding three rows across two keys" do
+      @g << {:col1 => 'disc1'}
+      @g << {:col1 => 'disc2'}
+      @g << {:col1 => 'disc1'}
+      fg = @g.flatten
+      fg.size.should == 2
+      fg.each do |r|
+        if r[:col1] == 'disc1'
+          r[:count].should == 2
+        elsif r[:col1] == 'disc2'
+          r[:count].should == 1
+        else
+          fail
+        end
+      end      
+    end
+  end
+  
+  it "can use extra columns to calculate values" do
+    @g = GroupByHash.new([:col1]) {|h| {:units => h[:quantity]}}
+    @g << {:col1 => 1, :quantity => 4}
+    @g << {:col1 => 1, :quantity => 2}
+    @g << {:col1 => 1, :quantity => 5}
+    @g[{:col1 => 1}][:units].should == 11
+  end
+  
+  describe "with col1/col2/col3 count" do
+    before(:each) do
+      @g = GroupByHash.new([:col1,:col2,:col3]) {|h| {:count => 1}}
+    end
+  
+    describe "<<" do
+      it "should add hash key with only defined cols" do
+        @g << {:col1 => 1,:col2 => 2,:col3 => 3, :extra_col1 => 123}
+        @g.size.should == 1
+        @g.has_key?({:col1 => 1,:col2 => 2,:col3 => 3}).should == true
+      end
+      
+      it "should add value with only block-defined cols" do
+        @g << {:col1 => 1,:col2 => 2,:col3 => 3, :extra_col1 => 123}
+        @g.size.should == 1
+        @g[{:col1 => 1,:col2 => 2,:col3 => 3}].should == {:count => 1}
+      end
+    end
+  
+    
+    it "groups right" do
+      @g << {:col1 => 1,:col2 => 2,:col3 => 3, :extra_col1 => 123}
+      @g << {:col1 => 1,:col2 => 2,:col3 => 3, :extra_col1 => 123}
+      @g << {:col1 => 1,:col2 => 2,:col3 => 3, :extra_col1 => 123}
+      @g << {:col1 => 1,:col2 => 2,:col3 => 5, :extra_col1 => 123}
+      @g << {:col1 => 1,:col2 => 4,:col3 => 5, :extra_col1 => 123}
+      @g.has_key?({:col1 => 1,:col2 => 2,:col3 => 3})
+      @g[{:col1 => 1,:col2 => 2,:col3 => 3}].should == {:count => 3}
+      @g.has_key?({:col1 => 1,:col2 => 2,:col3 => 5})
+      @g[{:col1 => 1,:col2 => 2,:col3 => 5}].should == {:count => 1}
+      @g.has_key?({:col1 => 1,:col2 => 4,:col3 => 5})
+      @g[{:col1 => 1,:col2 => 4,:col3 => 5}].should == {:count => 1}      
+    end
+  end
+end

data/spec/spec_helper.rb

@@ -0,0 +1,4 @@
+require 'rubygems'
+
+$:.unshift File.join(File.dirname(__FILE__), '..', 'lib')
+require 'group_by_hash'

metadata

@@ -0,0 +1,62 @@
+--- !ruby/object:Gem::Specification 
+name: group_by_hash
+version: !ruby/object:Gem::Version 
+  version: 0.1.0
+platform: ruby
+authors: 
+- Paul Boone
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2010-07-22 00:00:00 -07:00
+default_executable: 
+dependencies: []
+
+description: |
+  A ruby hash with a sql-esque group by method for when you can't do a group by in SQL because it is too complicated
+
+email: paulboone@mindbucket.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: []
+
+files: 
+- Rakefile
+- VERSION
+- lib/group_by_hash.rb
+- spec/group_by_hash_spec.rb
+- spec/spec_helper.rb
+has_rdoc: true
+homepage: http://github.com/paulboone/group_by_hash
+licenses: []
+
+post_install_message: 
+rdoc_options: 
+- --charset=UTF-8
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.5
+signing_key: 
+specification_version: 3
+summary: ruby hash with a sql-esque group by method
+test_files: 
+- spec/group_by_hash_spec.rb
+- spec/spec_helper.rb