tsuga 0.0.1

data/spec/model/tile_spec.rb

@@ -0,0 +1,116 @@
+require 'spec_helper'
+require 'tsuga/model/point'
+require 'tsuga/model/tile'
+require 'ostruct'
+
+describe Tsuga::Model::Tile do
+
+  describe '.including' do
+    let(:depth)  { OpenStruct.new :value => 1 }
+    let(:point)  { Tsuga::Model::Point.new(lat:0, lng:0) }
+    let(:result) { described_class.including(point, :depth => depth.value) }
+
+    it 'creates a tile from a point' do
+      result.should be_a_kind_of(Tsuga::Model::Tile)
+    end
+
+    context "on equator/greenwidth" do
+      before do
+        point.lat   = 0
+        point.lng   = 0
+        depth.value = 18
+      end
+
+      it 'calculates southwest' do
+        result.southwest.should =~ point
+      end
+
+      it 'calculates northeast' do
+        result.northeast.lng.should be_within(1e-6).of(360.0 * (2 ** -18))
+        result.northeast.lat.should be_within(1e-6).of(180.0 * (2 ** -18))
+      end
+
+      it 'respects geohash ordering' do
+        result.southwest.geohash.should < result.northeast.geohash
+      end
+    end
+  end
+
+  describe '#contains?' do
+    let(:point)  { Tsuga::Model::Point.new }
+
+    subject do
+      described_class.including(
+        Tsuga::Model::Point.new(lat:0, lng:0), :depth => 2)
+    end
+
+    let(:result) { subject.contains?(point) }
+
+    it 'includes the northwest corner' do
+      point.lat = 0
+      point.lng = 0
+      result.should be_true
+    end
+
+    it 'excludes the southeast corner' do
+      point.lat = 45
+      point.lng = 90
+      result.should be_false
+    end
+
+    it 'includes point close to the southeast corner' do
+      point.lat = 45 - 1e-6
+      point.lng = 90 - 1e-6
+      result.should be_true
+    end
+
+    it 'includes the center point' do
+      point.lat = 22.5
+      point.lng = 45
+      result.should be_true
+    end
+
+    it 'excludes points north of the border' do
+      point.lat = -1
+      point.lng = 45
+      result.should be_false
+    end
+
+    it 'excludes points south of the border' do
+      point.lat = 48
+      point.lng = 45
+      result.should be_false
+    end
+
+    it 'excludes points west of the border' do
+      point.lat = 22.5
+      point.lng = -1
+      result.should be_false
+    end
+
+    it 'excludes points east of the border' do
+      point.lat = 22.5
+      point.lng = 91
+      result.should be_false
+    end
+  end
+
+  describe '#neighbour' do
+    subject { described_class.new(prefix:'300') }
+    # . . . . . . . .   13
+    # . . . . . . . .   02
+    # . . . . - . . .
+    # . . . - X - . .
+    # . . . . - . . .
+    # . . . . . . . .
+    # . . . . . . . .
+    # . . . . . . . .
+
+
+    it('works to the east' ) { subject.neighbour(lat: 0, lng: 1).prefix.should == '302' }
+    it('works to the west' ) { subject.neighbour(lat: 0, lng:-1).prefix.should == '122' }
+    it('works to the north') { subject.neighbour(lat: 1, lng: 0).prefix.should == '301' }
+    it('works to the south') { subject.neighbour(lat:-1, lng: 0).prefix.should == '211' }
+  end
+
+end

data/spec/service/aggregator_spec.rb

@@ -0,0 +1,143 @@
+require 'spec_helper'
+require 'tsuga/service/aggregator'
+require 'tsuga/adapter/memory/cluster'
+
+describe Tsuga::Service::Aggregator do
+  let(:adapter) { Class.new { include Tsuga::Adapter::Memory::Cluster } }
+
+  def new_cluster(depth, lat, lng)
+    adapter.new.tap do |cluster|
+      cluster.depth   = depth
+      cluster.lat     = lat
+      cluster.lng     = lng
+      cluster.weight  = 1
+      cluster.sum_lat = lat
+      cluster.sum_lng = lng
+      cluster.ssq_lat = lat * lat
+      cluster.ssq_lng = lng * lng
+      cluster.children_ids = []
+      cluster.persist!
+    end
+  end
+
+  subject { described_class.new(clusters:clusters, ratio:0.2) }
+
+  describe '#min_distance' do
+    let(:clusters) { [new_cluster(2,0,0)] }
+
+    it 'is about 20% of a tile diagonal' do
+      subject.min_distance.should == be_within(1e-6).of(Math.sqrt(45*45 + 90*90) / 5)
+    end
+  end
+
+  describe '#run' do
+    context 'when the list of clusters is empty' do
+      let(:clusters) { [] }
+      it('passes') { subject.run }
+    end
+
+    context 'with a single cluster' do
+      let(:clusters) { [new_cluster(2,0,0)] }
+      it('passes') { subject.run }
+    end
+
+    context 'with 4 distant clusters' do
+      let(:north) { 0 }
+      let(:south) { 45 - 1e-4 }
+      let(:west)  { 0 }
+      let(:east)  { 90 - 1e-4 }
+
+      let(:clusters) {[
+        new_cluster(2, north, west),
+        new_cluster(2, south, west),
+        new_cluster(2, south, east),
+        new_cluster(2, north, east)
+      ]}
+      
+      it('passes') { subject.run }
+
+      it('preserves clusers') do 
+        subject.run
+        clusters.length.should == 4
+      end
+    end
+
+
+    context 'with 3 close clusters' do
+      let(:lat) { 22.5 }
+      let(:lng) { 45 }
+
+      let(:clusters) {[
+        new_cluster(2, lat + 0, lng),
+        new_cluster(2, lat + 1, lng),
+        new_cluster(2, lat - 1, lng)
+      ]}
+      
+      it('passes') { subject.run }
+
+      it 'groups clusters' do 
+        subject.run
+        clusters.length.should == 1
+      end
+
+      it 'computes centroid' do
+        subject.run
+        clusters.first.tap do |cluster|
+          cluster.lat.should == lat
+          cluster.lng.should == lng
+        end
+      end
+    end
+
+
+    context 'with superimposed points' do
+      let(:lat) { 22.5 }
+      let(:lng) { 45 }
+
+      let(:clusters) {[
+        new_cluster(2, lat, lng),
+        new_cluster(2, lat, lng),
+        new_cluster(2, lat, lng)
+      ]}
+      
+      it('passes') { subject.run }
+
+      it 'groups clusters' do 
+        subject.run
+        clusters.length.should == 1
+      end
+    end
+
+    context 'with 2 close and 2 distant clusters' do
+      let(:lat) { 22.5 }
+      let(:lng) { 45 }
+
+      let(:clusters) {[
+        new_cluster(2, 22, 45),
+        new_cluster(2, 23, 46),
+        new_cluster(2,  0,  0),
+        new_cluster(2,  1,  1)
+      ]}
+      
+      it('passes') { subject.run }
+
+      it 'groups clusters' do 
+        subject.run
+        clusters.length.should == 2
+      end
+    end
+
+    context 'with 100 random clusters' do
+      let(:lat_max) { 45 - 1e-4 }
+      let(:lng_max) { 90 - 1e-4 }
+
+      let(:clusters) { 
+        (1...100).map { new_cluster(2, rand*lat_max, rand*lng_max) }
+      }
+      
+      it('passes') { subject.run }
+    end
+
+  end
+
+end

data/spec/service/clusterer_spec.rb

@@ -0,0 +1,84 @@
+require 'spec_helper'
+require 'tsuga/service/clusterer'
+require 'tsuga/adapter/memory/cluster'
+require 'ostruct'
+
+describe Tsuga::Service::Clusterer do
+  subject { described_class.new(source: records, adapter: adapter) }
+  let(:adapter) { Class.new { include Tsuga::Adapter::Memory::Cluster } }
+  let(:records) { ArrayWithFindEach.new }
+
+  def make_record(lat, lng)
+    records << OpenStruct.new(lat: lat, lng: lng)
+  end
+
+  before { adapter.delete_all }
+
+  describe '#run' do
+    context 'with no records' do
+      it('passes') { subject.run }
+    end
+
+    context 'with a single record' do
+      before do
+        make_record(0,0)
+      end
+
+      it('passes') { subject.run }
+
+      it 'creates 1 cluster per depth' do
+        subject.run
+        adapter.collect_ids.length.should == 17
+      end
+
+      it 'creates clusters with the same position' do
+        subject.run
+        hashes = Set.new
+        adapter.find_each { |c| hashes << c.geohash }
+        hashes.size.should == 1
+      end
+    end
+
+    context 'with two distant records' do
+      before do
+        make_record(0,0)
+        make_record(45,0)
+      end
+
+      it('passes') { subject.run }
+
+      it 'creates 2 clusters per depth' do
+        subject.run
+        adapter.collect_ids.length.should == 34
+      end
+    end
+
+
+    context 'with random records' do
+      before do
+        10.times { make_record(rand, rand) }
+        subject.run
+      end
+
+      let :toplevel_cluster do
+        id = adapter.at_depth(3).collect_ids.first
+        adapter.find_by_id(id)
+      end
+
+      let :barycenter do
+        sum_lat = 0
+        sum_lng = 0
+        records.find_each { |r| sum_lat += r.lat ; sum_lng += r.lng }
+        Tsuga::Model::Point.new(lat: 0.1 * sum_lat, lng: 0.1 * sum_lng)
+      end
+
+      it 'toplevel cluster has correct weight' do
+        toplevel_cluster.weight.should == 10
+      end
+
+      it 'toplevel cluster is centered' do
+        toplevel_cluster.should =~ barycenter
+      end
+    end
+  end
+end

data/spec/spec_helper.rb

@@ -0,0 +1,26 @@
+require 'bundler/setup'
+
+# This file was generated by the `rspec --init` command. Conventionally, all
+# specs live under a `spec` directory, which RSpec adds to the `$LOAD_PATH`.
+# Require this file using `require "spec_helper"` to ensure that it is only
+# loaded once.
+#
+# See http://rubydoc.info/gems/rspec-core/RSpec/Core/Configuration
+RSpec.configure do |config|
+  config.treat_symbols_as_metadata_keys_with_true_values = true
+  config.run_all_when_everything_filtered = true
+  config.filter_run :focus
+
+  # Run specs in random order to surface order dependencies. If you find an
+  # order dependency and want to debug it, you can fix the order by providing
+  # the seed, which is printed after each run.
+  #     --seed 1234
+  config.order = 'random'
+end
+
+
+class ArrayWithFindEach < Array
+  def find_each(&block)
+    each(&block)
+  end
+end

data/spec/support/mongoid.yml

@@ -0,0 +1,17 @@
+development:
+  sessions:
+    default:
+      database: tsuga_dev
+      hosts:
+        - localhost:27017
+      options:
+        safe: true
+test:
+  sessions:
+    default:
+      database: tsuga_test
+      hosts:
+        - localhost:27017
+      options:
+        safe: true
+

data/tsuga.gemspec

@@ -0,0 +1,29 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'tsuga/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = "tsuga"
+  spec.version       = Tsuga::VERSION
+  spec.authors       = ["Julien Letessier"]
+  spec.email         = ["julien.letessier@gmail.com"]
+  spec.description   = %q{Hierarchical Geo Clusterer tuned for Google Maps usage}
+  spec.summary       = %q{Hierarchical Geo Clusterer tuned for Google Maps usage}
+  spec.homepage      = "http://github.com/mezis/tsuga"
+  spec.license       = "MIT"
+
+  spec.files         = `git ls-files`.split($/)
+  spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
+  spec.require_paths = ["lib"]
+
+  spec.add_runtime_dependency "ruby-progressbar"
+
+  spec.add_development_dependency "bundler", "~> 1.3"
+  spec.add_development_dependency "rake"
+  spec.add_development_dependency "rspec"
+  spec.add_development_dependency "guard"
+  spec.add_development_dependency "guard-rspec"
+  spec.add_development_dependency "pry"
+  spec.add_development_dependency "pry-nav"
+end

metadata

@@ -0,0 +1,226 @@
+--- !ruby/object:Gem::Specification
+name: tsuga
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- Julien Letessier
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2013-11-03 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: ruby-progressbar
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.3'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.3'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: guard
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: guard-rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: pry
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: pry-nav
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+description: Hierarchical Geo Clusterer tuned for Google Maps usage
+email:
+- julien.letessier@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- .gitignore
+- .rspec
+- .ruby-version
+- .travis.yml
+- Gemfile
+- Gemfile.lock
+- Guardfile
+- LICENSE.txt
+- README.md
+- Rakefile
+- lib/tsuga.rb
+- lib/tsuga/adapter.rb
+- lib/tsuga/adapter/active_record/base.rb
+- lib/tsuga/adapter/active_record/cluster.rb
+- lib/tsuga/adapter/active_record/migration.rb
+- lib/tsuga/adapter/active_record/record.rb
+- lib/tsuga/adapter/active_record/test.rb
+- lib/tsuga/adapter/memory/base.rb
+- lib/tsuga/adapter/memory/cluster.rb
+- lib/tsuga/adapter/memory/test.rb
+- lib/tsuga/adapter/mongoid/base.rb
+- lib/tsuga/adapter/mongoid/cluster.rb
+- lib/tsuga/adapter/mongoid/record.rb
+- lib/tsuga/adapter/mongoid/test.rb
+- lib/tsuga/adapter/sequel/base.rb
+- lib/tsuga/adapter/sequel/cluster.rb
+- lib/tsuga/adapter/sequel/record.rb
+- lib/tsuga/adapter/sequel/test.rb
+- lib/tsuga/adapter/shared.rb
+- lib/tsuga/adapter/shared/cluster.rb
+- lib/tsuga/errors.rb
+- lib/tsuga/model/cluster.rb
+- lib/tsuga/model/point.rb
+- lib/tsuga/model/record.rb
+- lib/tsuga/model/tile.rb
+- lib/tsuga/service/aggregator.rb
+- lib/tsuga/service/clusterer.rb
+- lib/tsuga/service/labeler.rb
+- lib/tsuga/version.rb
+- script/benchmark-aggregator.rb
+- script/benchmark-clusterer.rb
+- spec/adapter/memory/base_spec.rb
+- spec/adapter/memory/cluster_spec.rb
+- spec/adapter/shared/cluster_spec.rb
+- spec/integration/active_record_spec.rb
+- spec/integration/memory_spec.rb
+- spec/integration/mongoid_spec.rb
+- spec/integration/sequel_spec.rb
+- spec/integration/shared.rb
+- spec/model/point_spec.rb
+- spec/model/tile_spec.rb
+- spec/service/aggregator_spec.rb
+- spec/service/clusterer_spec.rb
+- spec/spec_helper.rb
+- spec/support/mongoid.yml
+- tsuga.gemspec
+homepage: http://github.com/mezis/tsuga
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.0.3
+signing_key: 
+specification_version: 4
+summary: Hierarchical Geo Clusterer tuned for Google Maps usage
+test_files:
+- spec/adapter/memory/base_spec.rb
+- spec/adapter/memory/cluster_spec.rb
+- spec/adapter/shared/cluster_spec.rb
+- spec/integration/active_record_spec.rb
+- spec/integration/memory_spec.rb
+- spec/integration/mongoid_spec.rb
+- spec/integration/sequel_spec.rb
+- spec/integration/shared.rb
+- spec/model/point_spec.rb
+- spec/model/tile_spec.rb
+- spec/service/aggregator_spec.rb
+- spec/service/clusterer_spec.rb
+- spec/spec_helper.rb
+- spec/support/mongoid.yml