mdquery 0.3.0

data/.document

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

data/.rspec

@@ -0,0 +1 @@
+--color

data/.travis.yml

@@ -0,0 +1,5 @@
+script: "JRUBY_OPTS=-J-Djruby.objectspace.enabled=true rake"
+rvm:
+  - 1.8.7
+  - 1.9.3
+  - jruby

data/Gemfile

@@ -0,0 +1,13 @@
+source "http://rubygems.org"
+
+gem "activerecord", ">= 3.1.0"
+
+group :development do
+  gem "rake", "~> 0.9.2"
+  gem "rspec", "~> 2.8.0"
+  gem "rdoc", "~> 3.12"
+  gem "bundler", "~> 1.1.0"
+  gem "jeweler", "~> 1.8.3"
+#  gem "rcov", ">= 0"
+ gem 'rr', ">= 1.0.4"
+end

data/Gemfile.lock

@@ -0,0 +1,53 @@
+GEM
+  remote: http://rubygems.org/
+  specs:
+    activemodel (3.2.2)
+      activesupport (= 3.2.2)
+      builder (~> 3.0.0)
+    activerecord (3.2.2)
+      activemodel (= 3.2.2)
+      activesupport (= 3.2.2)
+      arel (~> 3.0.2)
+      tzinfo (~> 0.3.29)
+    activesupport (3.2.2)
+      i18n (~> 0.6)
+      multi_json (~> 1.0)
+    arel (3.0.2)
+    builder (3.0.0)
+    diff-lcs (1.1.3)
+    git (1.2.5)
+    i18n (0.6.0)
+    jeweler (1.8.3)
+      bundler (~> 1.0)
+      git (>= 1.2.5)
+      rake
+      rdoc
+    json (1.6.5)
+    json (1.6.5-java)
+    multi_json (1.1.0)
+    rake (0.9.2.2)
+    rdoc (3.12)
+      json (~> 1.4)
+    rr (1.0.4)
+    rspec (2.8.0)
+      rspec-core (~> 2.8.0)
+      rspec-expectations (~> 2.8.0)
+      rspec-mocks (~> 2.8.0)
+    rspec-core (2.8.0)
+    rspec-expectations (2.8.0)
+      diff-lcs (~> 1.1.2)
+    rspec-mocks (2.8.0)
+    tzinfo (0.3.32)
+
+PLATFORMS
+  java
+  ruby
+
+DEPENDENCIES
+  activerecord (>= 3.1.0)
+  bundler (~> 1.1.0)
+  jeweler (~> 1.8.3)
+  rake (~> 0.9.2)
+  rdoc (~> 3.12)
+  rr (>= 1.0.4)
+  rspec (~> 2.8.0)

data/LICENSE.txt

@@ -0,0 +1,20 @@
+Copyright (c) 2012 mccraigmccraig
+
+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 @@
+= mdquery
+
+{<img src="https://secure.travis-ci.org/mccraigmccraig/mdquery.png" />}[http://travis-ci.org/mccraigmccraig/mdquery]
+
+Defines a DSL for specifying and executing segmented multi-dimensional queries on your ActiveRecord-3 models
+
+== Installation
+
+  gem install mdquery
+
+== Usage
+
+A Dataset consists of some Measures over some Dimensions of a Source
+
+* Measures are SQL aggregate functions to be computed, e.g. "count(*)" or "avg(age)"
+* Each Dimension consists of 1 or more Segments, and each Segment either fixes a value for the Dimension or extracts values from the Source. Each Segment may narrow the Source scope according to the Dimension value or value-range for the Segment
+* A Source is an ActiveRecord-3 Scope
+
+Given the definition of the Dataset, it will query the Sources to extract data points, each of which will have a value for each Dimension and a value for each Measure
+
+  require 'mdquery'
+
+  Q = MDQuery.dataset do
+    source Foo
+
+    measure :count, "count(*)", :int
+
+    dimension :time do
+      label "Time"
+
+      segment(:all) do
+        fix_dimension :all
+      end
+
+      segment(:five_years) do
+        narrow{|scope| scope.where("foos.created_at > now() - interval '5 years'")}
+        extract_dimension "extract(year from foos.created_at)"
+        values{|scope| (Date.today.year-4..Date.today.year).to_a.map(&:to_s)}
+        label{|value| "Year: #{value}"}
+      end
+    end
+
+    dimension :users do
+      label "Users"
+
+      segment(:all) do
+        fix_dimension :all
+      end
+
+      segment(:by_type) do
+        extract_dimension "user_types.name"
+        narrow{|scope| scope.joins(:user_types)}
+      end
+    end
+  end
+
+  # run queries, collect data. returns an MDQuery::Dataset::Dataset
+  dataset = Q.collect
+
+  # retrieve dimension values, segment values and labels
+  dataset.dimensions[:time][:five_years].values # => ["2008", "2009", "2010", "2011", "2012"]
+  dataset.dimensions[:time].label_for("2008") #=> "Year: 2008"
+  dataset.datapoint({:time=>'2008', :users=>'all'}, :count) # => 100 # 100 users in 2008
+
+== Contributing to mdquery
+
+* Check out the latest master to make sure the feature hasn't been implemented or the bug hasn't been fixed yet.
+* Check out the issue tracker to make sure someone already hasn't requested it and/or contributed it.
+* Fork the project.
+* Start a feature/bugfix branch.
+* Commit and push until you are happy with your contribution.
+* Make sure to add tests for it. This is important so I don't break it in a future version unintentionally.
+* Please try not to mess with the Rakefile, version, or history. If you want to have your own version, or is otherwise necessary, that is fine, but please isolate to its own commit so I can cherry-pick around it.
+
+== Copyright
+
+Copyright (c) 2012 mccraigmccraig. See LICENSE.txt for
+further details.

data/Rakefile

@@ -0,0 +1,49 @@
+# encoding: utf-8
+
+require 'rubygems'
+require 'bundler'
+begin
+  Bundler.setup(:default, :development)
+rescue Bundler::BundlerError => e
+  $stderr.puts e.message
+  $stderr.puts "Run `bundle install` to install missing gems"
+  exit e.status_code
+end
+require 'rake'
+
+require 'jeweler'
+Jeweler::Tasks.new do |gem|
+  # gem is a Gem::Specification... see http://docs.rubygems.org/read/chapter/20 for more options
+  gem.name = "mdquery"
+  gem.homepage = "http://github.com/mccraigmccraig/mdquery"
+  gem.license = "MIT"
+  gem.summary = %Q{simple multi-dimensional queries on top of active-record-3}
+  gem.description = %Q{provides a DSL for simply specifying and executing segmented multi-dimensional queries on your active-record-3 models}
+  gem.email = "mccraigmccraig@gmail.com"
+  gem.authors = ["mccraigmccraig"]
+  # dependencies defined in Gemfile
+end
+Jeweler::RubygemsDotOrgTasks.new
+
+require 'rspec/core'
+require 'rspec/core/rake_task'
+RSpec::Core::RakeTask.new(:spec) do |spec|
+  spec.pattern = FileList['spec/**/*_spec.rb']
+end
+
+# RSpec::Core::RakeTask.new(:rcov) do |spec|
+#   spec.pattern = 'spec/**/*_spec.rb'
+#   spec.rcov = true
+# end
+
+task :default => :spec
+
+require 'rdoc/task'
+Rake::RDocTask.new do |rdoc|
+  version = File.exist?('VERSION') ? File.read('VERSION') : ""
+
+  rdoc.rdoc_dir = 'rdoc'
+  rdoc.title = "mdquery #{version}"
+  rdoc.rdoc_files.include('README*')
+  rdoc.rdoc_files.include('lib/**/*.rb')
+end

data/VERSION

@@ -0,0 +1 @@
+0.3.0

data/lib/mdquery.rb

@@ -0,0 +1,13 @@
+$: << File.expand_path('..', __FILE__) if !$:.include?(File.expand_path('..', __FILE__))
+require 'mdquery/dsl'
+
+# a DSL for specifying analytic queries
+module MDQuery
+
+  # define a Dataset with the DSL
+  # * +proc+ the DatasetDSL Proc
+  def self.dataset(&proc)
+    MDQuery::DSL::DatasetDSL.new(&proc).send(:build)
+  end
+
+end

data/lib/mdquery/dataset.rb

@@ -0,0 +1,283 @@
+require 'mdquery/util'
+
+module MDQuery
+  module Dataset
+
+    # describes a value on a segment of a dimension
+    class DimensionValue
+      # DimensionSegment this value belongs to
+      attr_reader :dimension_segment
+
+      # the value
+      attr_reader :value
+
+      # Optional label for the value
+      attr_reader :label
+
+      def initialize(dimension_segment, value, label)
+        @dimension_segment = dimension_segment
+        @value = value
+        @label = label
+        validate
+      end
+
+      def validate
+        raise "no dimension_segment!" if !dimension_segment
+        raise "no value!" if !value
+      end
+
+      def inspect
+        "#<DimensionValue: value=#{value.inspect}, label=#{label.inspect}>"
+      end
+    end
+
+    # describes a segment of a dimension, a segment being some part of
+    # the dimension value line. Dimension values should not be present
+    # in more than one segment of a dimension, or results will be
+    # unexpected, though it's fine for aggregate values to be present
+    # which cover the same range as other values. e.g. having values of
+    # "jan", "feb", "march"... in one segment and "q1","q2","q3","q4"
+    # in another segment is fine
+    class DimensionSegment
+      # Dimension this Segment belongs to
+      attr_reader :dimension
+
+      # key of segment, unique within Dimension
+      attr_reader :key
+
+      # ordered list of DimensionValues in segment
+      attr_reader :dimension_values
+
+      # ordered list of all values in segment
+      attr_reader :values
+
+      def initialize(model, dimension)
+        @dimension = dimension
+        @key = model.key
+
+        @values = model.get_values(dimension.dataset.model.source)
+
+        label_index = model.labels(@values)
+        @dimension_values = @values.map{|v| DimensionValue.new(self, v, label_index[v]) }
+        @dimension_value_index = @dimension_values.reduce({}){|dvi,dv| dvi[dv.value] = dv ; dvi}
+
+        validate
+      end
+
+      def validate
+        raise "no dimension!" if !dimension
+        raise "no key!" if !key
+        raise "no values!" if !values
+      end
+
+      def inspect
+        "#<DimensionSegment: key=#{key.inspect}, dimension_values=#{dimension_values.inspect}>"
+      end
+
+      # retrieve a DimensionValue describing the given +value+
+      def dimension_value_for(value)
+        @dimension_value_index[value]
+      end
+
+      # retrieve a DimensionValue describing the given +value+
+      def [](value)
+        dimension_value_for(value)
+      end
+
+      # retrieve a label describing the given +value+
+      def label_for(value)
+        (dv = dimension_value_for(value)) && dv.label
+      end
+    end
+
+    # describes a Dimension consisting of one or more segments
+    class Dimension
+      # Dataset this Dimension belongs to
+      attr_reader :dataset
+
+      # key for this Dimension
+      attr_reader :key
+
+      # Optional label of the Dimension
+      attr_reader :label
+
+      # ordered list of one or more DimensionSegments
+      attr_reader :segments
+
+      # an ordered list of values for the dimension. May be static or
+      # extracted from the data source, depending on DimensionSegment
+      # definitions. It is the concatentation of the +values+ from each
+      # DimensionSegment in the Dimension
+      attr_reader :values
+
+      def initialize(model, dataset)
+        @dataset = dataset
+        @key = model.key
+        @label = model.label
+
+        @segments = model.segment_models.map{|sm| DimensionSegment.new(sm, self) }
+        @segment_index = @segments.reduce({}){|si, s| si[s.key] = s ; si}
+
+        @values = segments.map(&:values).reduce(&:+)
+        @dimension_value_index = segments.map(&:dimension_values).reduce(&:+).reduce({}){|dvi,dv| dvi[dv.value] = dv ; dvi}
+
+        validate
+      end
+
+      def validate
+        raise "no dataset!" if !dataset
+        raise "no key!" if !key
+        raise "no segments!" if !segments || segments.empty?
+      end
+
+      def inspect
+        "#<Dimension: key=#{key.inspect}, label=#{label.inspect}, segments=#{segments.inspect}>"
+      end
+
+      # lookup a segment by +key+
+      def segment(key)
+        @segment_index[key]
+      end
+
+      # lookup a segment by +key+
+      def [](key)
+        segment(key)
+      end
+
+      # return an ordered list of values for 0 or more segments.
+      # * +segment_keys+ a list of segment keys. if empty, methods returns +values+,
+      # otherwise returns the concatentation of +values+ for each identified segment
+      def values_for_segments(segment_keys)
+        if segment_keys && !segment_keys.empty?
+          segment_keys.map{|sk| segment(sk)}.map(&:values).reduce(&:+)
+        else
+          values
+        end
+      end
+
+      # return an ordered list of DimensionValues for 0 or more segments.
+      # * +segment_keys+ a list of segment keys. if empty, methods return all DimensionValues
+      # for all segments, otherwise returns the concatenation of DimensionValues for
+      # each identified segment
+      def dimension_values_for_segments(segment_keys)
+        if segment_keys && !segment_keys.empty?
+          segment_keys.map{|sk| segment(sk)}.map(&:dimension_values).reduce(&:+)
+        else
+          dimension_values
+        end
+      end
+
+      def dimension_values
+        segments.map(&:dimension_values).reduce(&:+)
+      end
+
+      # the DimensionValue describing +value+ or nil
+      def dimension_value_for(value)
+        @dimension_value_index[value]
+      end
+
+      # the label for the +value+ or nil
+      def label_for(value)
+        (dv = dimension_value_for(value)) && dv.label
+      end
+    end
+
+    # describes a Measure computed from the source data over the Dimensions
+    class Measure
+      # the +dataset+ this Measure belongs to
+      attr_reader :dataset
+
+      # the +key+ identifying this Measure
+      attr_reader :key
+
+      # the SQL fragment definition of the Measure
+      attr_reader :definition
+
+      def initialize(model, dataset)
+        @dataset = dataset
+        @key = model.key
+        @definition = model.definition
+        validate
+      end
+
+      def validate
+        raise "no dataset" if !dataset
+        raise "no key!" if !key
+        raise "no definition!" if !definition || definition=~/^\s*$/
+      end
+
+      def inspect
+        "#<Measure: key=#{key.inspect}, definition=#{definition.inspect}>"
+      end
+    end
+
+    # a Dataset is defined over a number of Dimensions with a number of Measures.
+    #
+    class Dataset
+      # the +Model+ describing how the +Dataset+ is to be assembled
+      attr_reader :model
+
+      # a list of points. each point is a Hash with a value for each +Dimension+ and a value for each +Measure+.
+      # keys are as given in the +Dimension+ and +Measure+ objects
+      attr_reader :data
+
+      # a Hash of +Dimensions+ keyed by their +keys+
+      attr_reader :dimensions
+
+      # a Hash of +Measures+ keyed by their +keys+
+      attr_reader :measures
+
+      # index of points from +data+, where key is a Hash of all Dimension {key=>value} pairs, and value is all Measure {key=>value} pairs
+      attr_reader :indexed_data
+
+      def initialize(model, data)
+        @model = model
+        @data = data
+
+        @measures = model.measure_models.map{|mm| Measure.new(mm, self) }.reduce({}){|mi,m| mi[m.key] = m ; mi}
+        @dimensions = model.dimension_models.map{|dm| Dimension.new(dm, self) }.reduce({}){|di,d| di[d.key] = d ; di}
+
+        validate
+        index
+      end
+
+      def validate
+        raise "no model!" if !model
+        raise "no data!" if !data
+        raise "no dimensions!" if !dimensions || dimensions.empty?
+        raise "no measures!" if !measures || measures.empty?
+      end
+
+      def inspect
+        "#<Dataset: dimensions=#{dimensions.inspect}, measures=#{measures.inspect}, data=#{data.inspect}>"
+      end
+
+      # retrieve a datapoint given a hash of {dimension_key=>dimension_values}
+      def datapoint(dimension_values, measure)
+        d = @indexed_data[dimension_values]
+        d[measure] if d
+      end
+
+      private
+
+      def index_key(point)
+        Hash[dimensions.keys.map{|k| [k, point[k]]}]
+      end
+
+      def index_data(point)
+        pc = point.clone
+        dks = dimensions.keys
+        pc.delete_if{|k,v| dks.include?(k)}
+        pc
+      end
+
+      def index
+        @indexed_data = {}
+        data.each do |point|
+          @indexed_data[index_key(point)] = index_data(point)
+        end
+        @indexed_data
+      end
+    end
+  end
+end