picky 2.6.0 → 2.7.0

data/lib/tasks/try.rake

@@ -1,34 +1,19 @@
 # Tasks for manually testing your engine configuration.
 #
-namespace :try do
+desc "Try the given text in the indexer/query (index and category optional)."
+task :try, [:text, :index, :category] => :application do |_, options|
+  text, index, category = options.text, options.index, options.category
 
-  # desc "Try how a given word would be tokenized when indexing (type:category optional)."
-  task :index, [:text, :index, :category] => :application do |_, options|
-    text, index, category = options.text, options.index, options.category
+  puts
+  fail "\x1b[31mrake try needs a text to try indexing and query preparation\x1b[m, e.g. rake 'try[yourtext]'." unless text
 
-    tokenizer = category ? Indexes.find(index, category).tokenizer : Tokenizers::Index.default
+  specific = Indexes
+  specific = specific[index]    if index
+  specific = specific[category] if category
 
-    puts "\"#{text}\" is saved in the index as              #{tokenizer.tokenize(text.dup).to_a}"
-  end
-
-  # desc "Try how a given word would be tokenized when querying."
-  task :query, [:text] => :application do |_, options|
-    text = options.text
-
-    puts "\"#{text}\" as a search will be preprocessed into #{Tokenizers::Query.default.tokenize(text.dup).to_a.map(&:to_s).map(&:to_sym)}"
-    puts
-    puts "(category qualifiers, e.g. title: are removed if they do not exist as a qualifier, so 'toitle:bla' -> 'bla')"
-  end
-
-  # desc "Try the given text with both the index and the query (type:category optional)."
-  task :both, [:text, :index, :category] => :application do |_, options|
-    text, index, category = options.text, options.index, options.category
-
-    puts
-    fail "\x1b[31mrake try needs a text to try indexing and query preparation\x1b[m, e.g. rake 'try[yourtext]'." unless text
-
-    Rake::Task[:"try:index"].invoke text, index, category
-    Rake::Task[:"try:query"].invoke text
-  end
+  puts "\"#{text}\" is saved in the #{specific.identifier} index as #{specific.tokenizer.tokenize(text.dup).to_a}"
 
+  puts "\"#{text}\" as a search will be tokenized into #{Tokenizers::Query.default.tokenize(text.dup).to_a.map(&:to_s).map(&:to_sym)}"
+  puts
+  puts "(category qualifiers, e.g. title: are removed if they do not exist as a qualifier, so 'toitle:bla' -> 'bla')"
 end

data/spec/lib/application_spec.rb

@@ -37,7 +37,7 @@ describe Application do
                     normalizes_words:   [[/Deoxyribonucleic Acid/i, 'DNA']],
                     
                     substitutes_characters_with: CharacterSubstituters::WestEuropean.new,
-                    maximum_tokens: 5 # TODO maximum_words?
+                    maximum_tokens: 5
           
           books_index = Index::Memory.new :books,
                                           source: Sources::DB.new('SELECT id, title, author, isbn13 as isbn FROM books', :file => 'app/db.yml')

data/spec/lib/backend/file/basic_spec.rb

@@ -10,16 +10,16 @@ describe Backend::File::Basic do
     end
   end
   
-  describe 'to_s' do
-    it 'returns the cache path with the default file extension' do
-      file.to_s.should == 'some/cache/path/to/file.index'
-    end
-  end
-  
   describe 'backup_directory' do
     it "returns the cache path's backup path" do
       file.backup_directory.should == 'some/cache/path/to/backup'
     end
   end
   
+  describe 'to_s' do
+    it 'returns the cache path with the default file extension' do
+      file.to_s.should == 'Backend::File::Basic(some/cache/path/to/file.index)'
+    end
+  end
+  
 end

data/spec/lib/backend/file/json_spec.rb

@@ -2,25 +2,30 @@ require 'spec_helper'
 
 describe Backend::File::JSON do
   
-  before(:each) do
-    @file = described_class.new "some_cache_path"
-  end
+  let(:file) { described_class.new 'some/cache/path/to/file' }
   
   describe "dump" do
     it "delegates to the given hash" do
       hash = stub :hash
       
-      hash.should_receive(:dump_json).once.with "some_cache_path.json"
+      hash.should_receive(:dump_json).once.with "some/cache/path/to/file.json"
       
-      @file.dump hash
+      file.dump hash
     end
   end
+  
   describe "retrieve" do
     it "raises" do
       lambda do
-        @file.retrieve
+        file.retrieve
       end.should raise_error("Can't retrieve from JSON file. Use text file.")
     end
   end
   
+  describe 'to_s' do
+    it 'returns the cache path with the default file extension' do
+      file.to_s.should == 'Backend::File::JSON(some/cache/path/to/file.json)'
+    end
+  end
+  
 end

data/spec/lib/backend/file/marshal_spec.rb

@@ -2,25 +2,30 @@ require 'spec_helper'
 
 describe Backend::File::Marshal do
   
-  before(:each) do
-    @file = described_class.new "some_cache_path"
-  end
+  let(:file) { described_class.new 'some/cache/path/to/file' }
   
   describe "dump" do
     it "delegates to the given hash" do
       hash = stub :hash
       
-      hash.should_receive(:dump_marshalled).once.with "some_cache_path.dump"
+      hash.should_receive(:dump_marshalled).once.with "some/cache/path/to/file.dump"
       
-      @file.dump hash
+      file.dump hash
     end
   end
+  
   describe "retrieve" do
     it "raises" do
       lambda do
-        @file.retrieve
+        file.retrieve
       end.should raise_error("Can't retrieve from marshalled file. Use text file.")
     end
   end
   
+  describe 'to_s' do
+    it 'returns the cache path with the default file extension' do
+      file.to_s.should == 'Backend::File::Marshal(some/cache/path/to/file.dump)'
+    end
+  end
+  
 end

data/spec/lib/backend/files_spec.rb

@@ -10,7 +10,7 @@ describe Backend::Files do
     
     @prepared      = @files.prepared
     
-    @index         = @files.index
+    @index         = @files.inverted
     @weights       = @files.weights
     @similarity    = @files.similarity
     @configuration = @files.configuration
@@ -25,11 +25,14 @@ describe Backend::Files do
   end
   
   describe "dump indexes" do
+    before(:each) do
+      @files.stub! :timed_exclaim
+    end
     describe "dump_index" do
       it "uses the right file" do
         @index.should_receive(:dump).once.with :some_hash
         
-        @files.dump_index :some_hash
+        @files.dump_inverted :some_hash
       end
     end
     describe "dump_weights" do
@@ -63,9 +66,9 @@ describe Backend::Files do
       it "uses the right file" do
         Yajl::Parser.stub! :parse
         
-        File.should_receive(:open).once.with 'spec/test_directory/index/test/some_index/some_category_some_name_index.json', 'r'
+        File.should_receive(:open).once.with 'spec/test_directory/index/test/some_index/some_category_some_name_inverted.json', 'r'
         
-        @files.load_index
+        @files.load_inverted
       end
     end
     describe "load_weights" do
@@ -102,7 +105,7 @@ describe Backend::Files do
       it 'uses the right method' do
         @index.should_receive(:cache_ok?).once.with
         
-        @files.index_cache_ok?
+        @files.inverted_cache_ok?
       end
     end
     describe "weights_cache_ok?" do
@@ -122,11 +125,11 @@ describe Backend::Files do
   end
   
   describe 'dump indexes' do
-    describe 'index_cache_small?' do
+    describe 'inverted_cache_small?' do
       it 'uses the right method' do
         @index.should_receive(:cache_small?).once.with
         
-        @files.index_cache_small?
+        @files.inverted_cache_small?
       end
     end
     describe 'weights_cache_small?' do
@@ -181,5 +184,16 @@ describe Backend::Files do
       @files.bundle_name.should == :some_name
     end
   end
+  
+  describe 'to_s' do
+    it 'returns the right value' do
+      category = stub :category,
+                      :identifier => "category_identifier",
+                      :prepared_index_path => "prepared/index/path",
+                      :index_path => 'index/path'
+      
+      described_class.new("some_bundle", category).to_s.should == "Backend::Files(Backend::File::Text(prepared/index/path.txt), Backend::File::JSON(index/path.json), Backend::File::JSON(index/path.json), Backend::File::Marshal(index/path.dump), Backend::File::JSON(index/path.json))"
+    end
+  end
 
 end

data/spec/lib/backend/redis/basic_spec.rb

@@ -64,4 +64,10 @@ describe Backend::Redis::Basic do
     end
   end
   
+  describe 'to_s' do
+    it 'returns the cache path with the default file extension' do
+      redis.to_s.should == 'Backend::Redis::Basic(some_namespace:*)'
+    end
+  end
+  
 end

data/spec/lib/backend/redis/list_hash_spec.rb

@@ -8,7 +8,7 @@ describe Backend::Redis::ListHash do
     it 'raises an error' do
       expect {
         index.member :some_sym
-      }.to raise_error("Can't retrieve a single value from a Redis ListHash. Use Index::Redis::StringHash.")
+      }.to raise_error("Can't retrieve single value :some_sym from a Redis ListHash. Use Index::Redis::StringHash.")
     end
   end
   
@@ -17,7 +17,7 @@ describe Backend::Redis::ListHash do
       backend = stub :backend
       index.should_receive(:backend).and_return backend
       
-      backend.stub! :lrange => :some_lrange_result
+      backend.stub! :zrange => :some_lrange_result
       
       index.collection(:anything).should == :some_lrange_result
     end
@@ -25,10 +25,16 @@ describe Backend::Redis::ListHash do
       backend = stub :backend
       index.should_receive(:backend).and_return backend
       
-      backend.should_receive(:lrange).once.with "some_namespace:some_sym", 0, -1
+      backend.should_receive(:zrange).once.with "some_namespace:some_sym", 0, -1
       
       index.collection :some_sym
     end
   end
   
+  describe 'to_s' do
+    it 'returns the cache path with the default file extension' do
+      index.to_s.should == 'Backend::Redis::ListHash(some_namespace:*)'
+    end
+  end
+  
 end

data/spec/lib/backend/redis/string_hash_spec.rb

@@ -17,7 +17,7 @@ describe Backend::Redis::StringHash do
   
   describe 'collection' do
     it 'raises' do
-      expect { index.collection :anything }.to raise_error("Can't retrieve a collection from a StringHash. Use Index::Redis::ListHash.")
+      expect { index.collection :anything }.to raise_error("Can't retrieve collection for :anything from a StringHash. Use Index::Redis::ListHash.")
     end
   end
   
@@ -38,4 +38,10 @@ describe Backend::Redis::StringHash do
     end
   end
   
+  describe 'to_s' do
+    it 'returns the cache path with the default file extension' do
+      index.to_s.should == 'Backend::Redis::StringHash(some_namespace:*)'
+    end
+  end
+  
 end

data/spec/lib/backend/redis_spec.rb

@@ -58,18 +58,18 @@ describe Backend::Redis do
     end
     describe 'ids' do
       it 'delegates to the index' do
-        index = stub :index
-        redis.stub! :index => index
+        inverted = stub :inverted
+        redis.stub! :inverted => inverted
 
-        index.should_receive(:collection).once.with :some_sym
+        inverted.should_receive(:collection).once.with :some_sym
 
         redis.ids :some_sym
       end
       it 'returns the value' do
-        index = stub :index
-        redis.stub! :index => index
+        inverted = stub :inverted
+        redis.stub! :inverted => inverted
 
-        index.should_receive(:collection).any_number_of_times.and_return [1,2,3]
+        inverted.should_receive(:collection).any_number_of_times.and_return [1,2,3]
 
         redis.ids(:any).should == [1,2,3]
       end
@@ -132,22 +132,32 @@ describe Backend::Redis do
     end
     describe 'ids' do
       it 'delegates to the index' do
-        index = stub :index
-        redis.stub! :index => index
+        inverted = stub :inverted
+        redis.stub! :inverted => inverted
 
-        index.should_receive(:collection).once.with :some_sym
+        inverted.should_receive(:collection).once.with :some_sym
 
         redis.ids :some_sym
       end
       it 'returns the value' do
-        index = stub :index
-        redis.stub! :index => index
+        inverted = stub :inverted
+        redis.stub! :inverted => inverted
 
-        index.should_receive(:collection).any_number_of_times.and_return [1,2,3]
+        inverted.should_receive(:collection).any_number_of_times.and_return [1,2,3]
 
         redis.ids(:any).should == [1,2,3]
       end
     end
   end
+  
+  describe 'to_s' do
+    it 'returns the right value' do
+      category = stub :category,
+                      :identifier => "category_identifier",
+                      :prepared_index_path => "prepared/index/path"
+      
+      described_class.new("some_bundle", category).to_s.should == "Backend::Redis(Backend::File::Text(prepared/index/path.txt), some_bundle, category_identifier)"
+    end
+  end
 
 end

data/spec/lib/categories_indexed_spec.rb

@@ -39,7 +39,7 @@ describe Categories do
     end
   end
   
-  describe 'clear' do
+  describe 'clear_categories' do
     before(:each) do
       @categories = described_class.new
     end
@@ -56,7 +56,7 @@ describe Categories do
     it "is clear again after clearing" do
       @categories << stub(:category, :name => :some_name)
       
-      @categories.clear
+      @categories.clear_categories
       
       @categories.categories.should be_empty
       @categories.category_hash.should be_empty      

data/spec/lib/category_indexing_spec.rb

@@ -12,36 +12,36 @@ describe Category do
     let(:exact) { category.indexing_exact }
     let(:partial) { category.indexing_partial }
     
-    describe 'backup_caches' do
+    describe 'backup' do
       it 'delegates to both bundles' do
         exact.should_receive(:backup).once.with()
         partial.should_receive(:backup).once.with()
         
-        category.backup_caches
+        category.backup
       end
     end
-    describe 'restore_caches' do
+    describe 'restore' do
       it 'delegates to both bundles' do
         exact.should_receive(:restore).once.with()
         partial.should_receive(:restore).once.with()
         
-        category.restore_caches
+        category.restore
       end
     end
-    describe 'check_caches' do
+    describe 'check' do
       it 'delegates to both bundles' do
         exact.should_receive(:raise_unless_cache_exists).once.with()
         partial.should_receive(:raise_unless_cache_exists).once.with()
         
-        category.check_caches
+        category.check
       end
     end
-    describe 'clear_caches' do
+    describe 'clear' do
       it 'delegates to both bundles' do
         exact.should_receive(:delete).once.with()
         partial.should_receive(:delete).once.with()
         
-        category.clear_caches
+        category.clear
       end
     end
     
@@ -79,7 +79,7 @@ describe Category do
       end
       it 'should use the exact index to generate the partial index' do
         exact_index = stub :exact_index
-        exact.stub! :index => exact_index
+        exact.stub! :inverted => exact_index
         partial.should_receive(:generate_partial_from).once.with(exact_index)
 
         category.generate_partial
@@ -94,7 +94,7 @@ describe Category do
       end
     end
 
-    describe 'generate_caches' do
+    describe 'cache' do
       it 'should call multiple methods in order' do
         category.should_receive(:configure).once.with().ordered
         category.should_receive(:generate_caches_from_source).once.with().ordered
@@ -103,7 +103,7 @@ describe Category do
         category.should_receive(:dump_caches).once.with().ordered
         category.should_receive(:timed_exclaim).once.ordered
         
-        category.generate_caches
+        category.cache
       end
     end
     
@@ -159,34 +159,13 @@ describe Category do
       end
     end
     
-    describe "cache" do
-      before(:each) do
-        category.stub! :generate_caches
-        category.stub! :configure
-      end
-      it "prepares the cache directory" do
-        category.should_receive(:prepare_index_directory).once.with
-        
-        category.cache
-      end
-      it "tells the indexer to index" do
-        category.should_receive(:generate_caches).once.with
-        
-        category.cache
-      end
-    end
     describe "index" do
       before(:each) do
         @indexer = stub :indexer, :index => nil
         category.stub! :indexer => @indexer
       end
-      it "prepares the cache directory" do
-        category.should_receive(:prepare_index_directory).once.with
-        
-        category.prepare
-      end
       it "tells the indexer to index" do
-        @indexer.should_receive(:index).once.with
+        @indexer.should_receive(:index).once.with [category]
         
         category.prepare
       end

data/spec/lib/category_spec.rb

@@ -0,0 +1,22 @@
+require 'spec_helper'
+
+describe Category do
+  
+  let(:index) { Index::Memory.new :some_index, source: [] }
+  let(:category) { described_class.new :some_category, index }
+  
+  it 'should set defaults correctly' do
+    category.indexing_exact.weights_strategy.should == Generators::Weights::Default
+    category.indexing_exact.partial_strategy.should be_kind_of(Generators::Partial::None)
+    category.indexing_exact.similarity_strategy.should == Generators::Similarity::Default
+    
+    category.indexing_partial.weights_strategy.should be_kind_of(Generators::Weights::Logarithmic)
+    category.indexing_partial.partial_strategy.should == Generators::Partial::Default
+    category.indexing_partial.similarity_strategy.should be_kind_of(Generators::Similarity::None)
+    
+    category.indexed_exact.similarity_strategy.should == Generators::Similarity::Default
+    
+    category.indexed_partial.similarity_strategy.should be_kind_of(Generators::Similarity::None)
+  end
+  
+end

data/spec/lib/index/base_indexing_spec.rb

@@ -2,6 +2,36 @@ require 'spec_helper'
 
 describe Index::Base do
   
+  context 'in general' do
+    context 'with #each source' do
+      let(:index) { described_class.new :some_name, source: [] }
+
+      it 'does things in order' do
+        index.should_receive(:check_source_empty).once.with.ordered
+        index.should_receive(:index_in_parallel).once.with.ordered
+
+        index.index
+      end
+    end
+    context 'with non#each source' do
+      let(:source) { stub :source, :harvest => nil }
+      let(:index) { described_class.new :some_name, source: source }
+
+      it 'does things in order' do
+        category = stub :category        
+        index.stub! :categories => [
+          category,
+          category,
+          category
+        ]
+
+        category.should_receive(:index).exactly(3).times
+
+        index.index
+      end
+    end
+  end
+  
   context "with categories" do
     before(:each) do
       @source = []

data/spec/lib/indexed/bundle/memory_spec.rb

@@ -12,28 +12,21 @@ describe Indexed::Bundle::Memory do
   
   describe 'to_s' do
     it 'does something' do
-      @bundle.to_s.should == <<-TO_S
-Memory
-  Files:
-    Index:      spec/test_directory/index/test/some_index/some_category_some_name_index.json
-    Weights:    spec/test_directory/index/test/some_index/some_category_some_name_weights.json
-    Similarity: spec/test_directory/index/test/some_index/some_category_some_name_similarity.dump
-    Config:     spec/test_directory/index/test/some_index/some_category_some_name_configuration.json
-      TO_S
+      @bundle.to_s.should == "Indexed::Bundle::Memory(test:some_index:some_category:some_name, Backend::Files(Backend::File::Text(spec/test_directory/index/test/some_index/prepared_some_category_index.txt), Backend::File::JSON(spec/test_directory/index/test/some_index/some_category_some_name_inverted.json), Backend::File::JSON(spec/test_directory/index/test/some_index/some_category_some_name_weights.json), Backend::File::Marshal(spec/test_directory/index/test/some_index/some_category_some_name_similarity.dump), Backend::File::JSON(spec/test_directory/index/test/some_index/some_category_some_name_configuration.json)))"
     end
   end
   
   describe 'clear_index' do
     before(:each) do
-      @bundle.instance_variable_set(:@index, :not_empty)
+      @bundle.instance_variable_set(:@inverted, :not_empty)
     end
-    it 'has not cleared the index' do
-      @bundle.index.should == :not_empty
+    it 'has not cleared the inverted index' do
+      @bundle.inverted.should == :not_empty
     end
-    it 'clears the index' do
-      @bundle.clear_index
+    it 'clears the inverted index' do
+      @bundle.clear_inverted
       
-      @bundle.index.should be_empty
+      @bundle.inverted.should be_empty
     end
   end
   describe 'clear_weights' do
@@ -78,13 +71,13 @@ Memory
 
   describe 'identifier' do
     it 'should return a specific identifier' do
-      @bundle.identifier.should == 'some_index:some_category:some_name'
+      @bundle.identifier.should == 'test:some_index:some_category:some_name'
     end
   end
 
   describe 'ids' do
     before(:each) do
-      @bundle.instance_variable_set :@index, { :existing => :some_ids }
+      @bundle.instance_variable_set :@inverted, { :existing => :some_ids }
     end
     it 'should return an empty array if not found' do
       @bundle.ids(:non_existing).should == []
@@ -108,7 +101,7 @@ Memory
   
   describe 'load' do
     it 'should trigger loads' do
-      @bundle.should_receive(:load_index).once.with
+      @bundle.should_receive(:load_inverted).once.with
       @bundle.should_receive(:load_weights).once.with
       @bundle.should_receive(:load_similarity).once.with
       @bundle.should_receive(:load_configuration).once.with
@@ -124,9 +117,9 @@ Memory
       it "uses the right file" do
         Yajl::Parser.stub! :parse
         
-        File.should_receive(:open).once.with 'spec/test_directory/index/test/some_index/some_category_some_name_index.json', 'r'
+        File.should_receive(:open).once.with 'spec/test_directory/index/test/some_index/some_category_some_name_inverted.json', 'r'
         
-        @bundle.load_index
+        @bundle.load_inverted
       end
     end
     describe "load_weights" do
@@ -166,7 +159,7 @@ Memory
       @bundle = described_class.new :some_name, @category, :similarity
     end
     it 'should initialize the index correctly' do
-      @bundle.index.should == {}
+      @bundle.inverted.should == {}
     end
     it 'should initialize the weights index correctly' do
       @bundle.weights.should == {}