picky 0.9.0 → 0.9.1

data/lib/picky/indexes.rb

@@ -128,8 +128,8 @@ module Indexes
   #
   def self.clear_caches
     each_bundle do |exact, partial|
-      exact.delete_all
-      partial.delete_all
+      exact.delete
+      partial.delete
     end
   end
 

data/lib/picky/loader.rb

@@ -155,14 +155,20 @@ module Loader
     load_relative 'cacher/weights_generator'
     load_relative 'cacher/similarity_generator'
     
+    # Index file handling.
+    #
+    load_relative 'index/file/basic'
+    load_relative 'index/file/text'
+    load_relative 'index/file/marshal'
+    load_relative 'index/file/json'
+    load_relative 'index/files'
+    
     # Index types.
     #
     load_relative 'index/bundle'
     load_relative 'index/category'
     load_relative 'index/type'
     
-    load_relative 'index/bundle_checker'
-    
     load_relative 'index/wrappers/exact_first'
     
     # Tokens.

data/lib/picky/sources/delicious.rb

@@ -3,10 +3,16 @@ module Sources
   class Delicious < Base
     
     def initialize username, password
-      require 'www/delicious'
+      check_gem
       @username = username
       @password = password
     end
+    def check_gem
+      require 'www/delicious'
+    rescue LoadError
+      puts "Delicious gem missing!\nTo use the delicious source, you need to:\n  1. Add the following line to Gemfile:\n     gem 'www-delicious'\n  2. Then, run:\n     bundle update\n"
+      exit(1)
+    end
     
     # Harvests the data to index.
     #

data/project_prototype/app/application.rb

@@ -17,14 +17,16 @@ class PickySearch < Application
   books_index = index :books,
                       Sources::CSV.new(:title, :author, :isbn, :year, :publisher, :subjects, :file => 'app/library.csv'),
                       # Use a database as source:
-                      # Sources::DB.new('SELECT id, title, author, isbn13 as isbn FROM books', :file => 'app/db.yml'), 
+                      # Sources::DB.new('SELECT id, title, author, isbn13 as isbn FROM books', :file => 'app/db.yml'),
+                      # Or delicious:
+                      # Sources::Delicious.new('username', 'password'), # offers title, tags, url fields.
                       field(:title,
-                            :partial => Partial::Subtoken.new(:down_to => 1), # Index partial down to character 1 (default: -3),
-                                                                              # e.g. florian -> floria, flori, flor, flo, fl, f
-                                                                              # Like this, you'll find florian even when entering just an "f".
-                            :similarity => Similarity::DoubleLevenshtone.new(3)), # Up to three similar title word indexed.
-                      field(:author, :partial => Partial::Subtoken.new(:down_to => 1)),
-                      field(:isbn,   :partial => Partial::None.new) # Partially searching on an ISBN makes not much sense, neither does similarity.
+                            :partial => Partial::Substring.new(:from => 1), # Index substrings upwards from character 1 (default: -3),
+                                                                              # e.g. picky -> p, pi, pic, pick, picky
+                                                                              # Like this, you'll find picky even when entering just a "p".
+                            :similarity => Similarity::DoubleLevenshtone.new(3)), # Up to three similar title word indexed (default: no similarity).
+                      field(:author, :partial => Partial::Substring.new(:from => 1)),
+                      field(:isbn,   :partial => Partial::None.new) # Partial substring searching on an ISBN makes not much sense, neither does similarity.
   
   # Defines the maximum tokens (words) that pass through to the engine.
   #

data/spec/lib/cacher/partial/default_spec.rb

@@ -3,13 +3,13 @@ require 'spec_helper'
 describe Cacher::Partial::Default do
   
   it "should be a subtoken" do
-    Cacher::Partial::Default.should be_kind_of(Cacher::Partial::Subtoken)
+    Cacher::Partial::Default.should be_kind_of(Cacher::Partial::Substring)
   end
   it "should be a the right down to" do
-    Cacher::Partial::Default.down_to.should == -3
+    Cacher::Partial::Default.from.should == -3
   end
   it "should be a the right starting at" do
-    Cacher::Partial::Default.starting_at.should == -1
+    Cacher::Partial::Default.to.should == -1
   end
   
 end

data/spec/lib/cacher/partial/subtoken_spec.rb

@@ -1,14 +1,14 @@
 require 'spec_helper'
 
-describe Cacher::Partial::Subtoken do
+describe Cacher::Partial::Substring do
   
-  context 'default down_to' do
+  context 'default from' do
     before(:each) do
-      @cacher = Cacher::Partial::Subtoken.new
+      @cacher = Cacher::Partial::Substring.new
     end
-    describe 'down_to' do
+    describe 'from' do
       it 'should return the right value' do
-        @cacher.down_to.should == 1
+        @cacher.from.should == 1
       end
     end
     describe 'generate_from' do
@@ -40,10 +40,10 @@ describe Cacher::Partial::Subtoken do
       end
     end
   end
-  context 'down_to set' do
-    describe 'negative down_to' do
+  context 'from set' do
+    describe 'negative from' do
       before(:each) do
-        @cacher = Cacher::Partial::Subtoken.new :down_to => -2
+        @cacher = Cacher::Partial::Substring.new :from => -2
       end
       it 'should generate the right index' do
         @cacher.generate_from( :florian => [1], :flavia => [2] ).should == {
@@ -54,9 +54,9 @@ describe Cacher::Partial::Subtoken do
         }
       end
     end
-    context "large down_to" do
+    context "large from" do
       before(:each) do
-        @cacher = Cacher::Partial::Subtoken.new :down_to => 10
+        @cacher = Cacher::Partial::Substring.new :from => 10
       end
       describe 'generate_from' do
         it 'should generate the right index' do
@@ -68,18 +68,18 @@ describe Cacher::Partial::Subtoken do
         end
       end
     end
-    context 'default starting_at' do
+    context 'default to' do
       before(:each) do
-        @cacher = Cacher::Partial::Subtoken.new :down_to => 4
+        @cacher = Cacher::Partial::Substring.new :from => 4
       end
-      describe 'starting_at' do
+      describe 'to' do
         it 'should return the right value' do
-          @cacher.starting_at.should == -1
+          @cacher.to.should == -1
         end
       end
-      describe 'down_to' do
+      describe 'from' do
         it 'should return the right value' do
-          @cacher.down_to.should == 4
+          @cacher.from.should == 4
         end
       end
       describe 'generate_from' do
@@ -122,18 +122,18 @@ describe Cacher::Partial::Subtoken do
         end
       end
     end
-    context 'starting_at set' do
+    context 'to set' do
       before(:each) do
-        @cacher = Cacher::Partial::Subtoken.new :down_to => 4, :starting_at => -2
+        @cacher = Cacher::Partial::Substring.new :from => 4, :to => -2
       end
-      describe 'starting_at' do
+      describe 'to' do
         it 'should return the right value' do
-          @cacher.starting_at.should == -2
+          @cacher.to.should == -2
         end
       end
-      describe 'down_to' do
+      describe 'from' do
         it 'should return the right value' do
-          @cacher.down_to.should == 4
+          @cacher.from.should == 4
         end
       end
       describe 'generate_from' do
@@ -148,18 +148,18 @@ describe Cacher::Partial::Subtoken do
         end
       end
     end
-    context 'starting_at set' do
+    context 'to set' do
       before(:each) do
-        @cacher = Cacher::Partial::Subtoken.new :down_to => 4, :starting_at => 0
+        @cacher = Cacher::Partial::Substring.new :from => 4, :to => 0
       end
-      describe 'starting_at' do
+      describe 'to' do
         it 'should return the right value' do
-          @cacher.starting_at.should == 0
+          @cacher.to.should == 0
         end
       end
-      describe 'down_to' do
+      describe 'from' do
         it 'should return the right value' do
-          @cacher.down_to.should == 4
+          @cacher.from.should == 4
         end
       end
       describe 'generate_from' do

data/spec/lib/cacher/partial_generator_spec.rb

@@ -6,7 +6,7 @@ describe Cacher::PartialGenerator do
     it 'should generate the correct values with a given strategy' do
       generator = Cacher::PartialGenerator.new :meier => [1], :maier => [2]
       
-      generator.generate(Cacher::Partial::Subtoken.new).should == {
+      generator.generate(Cacher::Partial::Substring.new).should == {
         :meier => [1],
         :meie => [1],
         :mei => [1],
@@ -21,7 +21,7 @@ describe Cacher::PartialGenerator do
     it 'should generate the correct values with a given specific strategy' do
       generator = Cacher::PartialGenerator.new :meier => [1], :maier => [2]
       
-      generator.generate(Cacher::Partial::Subtoken.new(:down_to => 3)).should == {
+      generator.generate(Cacher::Partial::Substring.new(:from => 3)).should == {
         :meier => [1],
         :meie => [1],
         :mei => [1],

data/spec/lib/extensions/hash_spec.rb

@@ -6,7 +6,7 @@ describe Hash do
     it 'uses the right file' do
       File.should_receive(:open).once.with('some/file/path.json', 'w')
       
-      {}.dump_to_json 'some/file/path'
+      {}.dump_json 'some/file/path.json'
     end
     it "uses the right encoder" do
       file = stub :file
@@ -14,7 +14,7 @@ describe Hash do
       
       Yajl::Encoder.should_receive(:encode).once.with({ :some => :hash }, file)
       
-      { :some => :hash }.dump_to_json 'unimportant'
+      { :some => :hash }.dump_json 'unimportant'
     end
   end
   
@@ -22,7 +22,7 @@ describe Hash do
     it 'uses the right file' do
       File.should_receive(:open).once.with('some/file/path.dump', 'w:binary')
       
-      {}.dump_to_marshalled 'some/file/path'
+      {}.dump_marshalled 'some/file/path.dump'
     end
     it "uses the right encoder" do
       file = stub :file
@@ -30,7 +30,7 @@ describe Hash do
       
       Marshal.should_receive(:dump).once.with({ :some => :hash }, file)
       
-      { :some => :hash }.dump_to_marshalled 'unimportant'
+      { :some => :hash }.dump_marshalled 'unimportant'
     end
   end
   

data/spec/lib/index/bundle_partial_generation_speed_spec.rb

@@ -5,7 +5,7 @@ describe Index::Bundle do
   before(:each) do
     @category         = stub :category, :name => :some_category
     @type             = stub :type, :name => :some_type
-    @partial_strategy = Cacher::Partial::Subtoken.new :down_to => 1
+    @partial_strategy = Cacher::Partial::Substring.new :from => 1
     @exact            = Index::Bundle.new :some_name, @category, @type, @partial_strategy, nil, nil
   end
 

data/spec/lib/index/bundle_spec.rb

@@ -17,21 +17,7 @@ describe Index::Bundle do
       @index.identifier.should == 'some_name: some_type some_category'
     end
   end
-
-  describe 'create_directory' do
-    it 'should use makedirs to create the necessary directory structure' do
-      FileUtils.should_receive(:mkdir_p).once.with 'some/search/root/index/test/some_type'
-
-      @index.create_directory
-    end
-  end
-
-  describe 'cache_directory' do
-    it 'should be correct' do
-      @index.cache_directory.should == 'some/search/root/index/test/some_type'
-    end
-  end
-
+  
   describe 'initialize_index_for' do
     context 'token not yet assigned' do
       before(:each) do
@@ -127,17 +113,7 @@ describe Index::Bundle do
       @index.weight(:existing).should == :specific
     end
   end
-
-  describe 'delete_all' do
-    it 'should call delete with all paths' do
-      @index.should_receive(:delete).once.with @index.index_cache_path
-      @index.should_receive(:delete).once.with @index.similarity_cache_path
-      @index.should_receive(:delete).once.with @index.weights_cache_path
-
-      @index.delete_all
-    end
-  end
-
+  
   describe 'load' do
     it 'should trigger loads' do
       @index.should_receive(:load_index).once.with
@@ -190,25 +166,56 @@ describe Index::Bundle do
     end
   end
   
-  describe 'weights_cache_path' do
-    it 'should return the correct file name' do
-      @index.weights_cache_path.should == 'some/search/root/index/test/some_type/some_name_some_category_weights'
+  describe 'raise_unless_cache_exists' do
+    before(:each) do
+      @files = stub :files
+      @files.stub! :index_cache_ok? => true
+      @files.stub! :similarity_cache_ok? => true
+      @files.stub! :weights_cache_ok? => true
+      @files.stub! :index_cache_small? => false
+      @files.stub! :similarity_cache_small? => false
+      @files.stub! :weights_cache_small? => false
+      
+      @index.stub! :files => @files
     end
-  end
-  describe 'similarity_cache_path' do
-    it 'should return the correct file name' do
-      @index.similarity_cache_path.should == 'some/search/root/index/test/some_type/some_name_some_category_similarity'
+    context 'weights cache missing' do
+      before(:each) do
+        @files.stub! :weights_cache_ok? => false
+      end
+      it 'should raise' do
+        lambda do
+          @index.raise_unless_cache_exists
+        end.should raise_error("weights cache for some_name: some_type some_category missing.")
+      end
     end
-  end
-  describe 'index_cache_path' do
-    it 'should return the correct file name' do
-      @index.index_cache_path.should == 'some/search/root/index/test/some_type/some_name_some_category_index'
+    context 'similarity cache missing' do
+      before(:each) do
+        @files.stub! :similarity_cache_ok? => false
+      end
+      it 'should raise' do
+        lambda do
+          @index.raise_unless_cache_exists
+        end.should raise_error("similarity cache for some_name: some_type some_category missing.")
+      end
+    end
+    context 'index cache missing' do
+      before(:each) do
+        @files.stub! :index_cache_ok? => false
+      end
+      it 'should raise' do
+        lambda do
+          @index.raise_unless_cache_exists
+        end.should raise_error("index cache for some_name: some_type some_category missing.")
+      end
     end
   end
-
+  
   describe 'initialization' do
     before(:each) do
-      @index = @index_class.new :some_name, :some_category, :some_type, :partial, :weights, :similarity
+      @category = stub :category, :name => :some_category
+      @type     = stub :type, :name => :some_type
+      
+      @index = @index_class.new :some_name, @category, @type, :partial, :weights, :similarity
     end
     it 'should initialize the index correctly' do
       @index.index.should == {}
@@ -220,13 +227,7 @@ describe Index::Bundle do
       @index.similarity.should == {}
     end
     it 'should initialize the name correctly' do
-      @index.name.should == :some_name
-    end
-    it 'should initialize the name correctly' do
-      @index.category.should == :some_category
-    end
-    it 'should initialize the name correctly' do
-      @index.type.should == :some_type
+      @index.category.should == @category
     end
     it 'should initialize the partial strategy correctly' do
       @index.partial_strategy.should == :partial

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

@@ -0,0 +1,15 @@
+require 'spec_helper'
+
+describe Index::File::Basic do
+  
+  before(:each) do
+    @file = Index::File::Basic.new "some/cache/path/to/file"
+  end
+  
+  describe "backup_file_path_of" do
+    it "returns a backup path relative to the path" do
+      @file.backup_file_path_of('some/path/to/some.index').should == 'some/path/to/backup/some.index'
+    end
+  end
+  
+end

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

@@ -0,0 +1,26 @@
+require 'spec_helper'
+
+describe Index::File::JSON do
+  
+  before(:each) do
+    @file = Index::File::JSON.new "some_cache_path"
+  end
+  
+  describe "dump" do
+    it "delegates to the given hash" do
+      hash = stub :hash
+      
+      hash.should_receive(:dump_json).once.with "some_cache_path.json"
+      
+      @file.dump hash
+    end
+  end
+  describe "retrieve" do
+    it "raises" do
+      lambda do
+        @file.retrieve
+      end.should raise_error("Can't retrieve from marshalled file. Use text file.")
+    end
+  end
+  
+end

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

@@ -0,0 +1,26 @@
+require 'spec_helper'
+
+describe Index::File::Marshal do
+  
+  before(:each) do
+    @file = Index::File::Marshal.new "some_cache_path"
+  end
+  
+  describe "dump" do
+    it "delegates to the given hash" do
+      hash = stub :hash
+      
+      hash.should_receive(:dump_marshalled).once.with "some_cache_path.dump"
+      
+      @file.dump hash
+    end
+  end
+  describe "retrieve" do
+    it "raises" do
+      lambda do
+        @file.retrieve
+      end.should raise_error("Can't retrieve from marshalled file. Use text file.")
+    end
+  end
+  
+end

data/spec/lib/index/file/text_spec.rb

@@ -0,0 +1,27 @@
+require 'spec_helper'
+
+describe Index::File::Text do
+  
+  before(:each) do
+    @file = Index::File::Text.new "some_cache_path"
+  end
+  
+  describe "load" do
+    it "raises" do
+      lambda do
+        @file.load
+      end.should raise_error("Can't load from text file. Use JSON or Marshal.")
+    end
+  end
+  describe "dump" do
+    it "raises" do
+      lambda do
+        @file.dump :anything
+      end.should raise_error("Can't dump to text file. Use JSON or Marshal.")
+    end
+  end
+  describe "retrieve" do
+    it
+  end
+  
+end