picky 0.2.4 → 0.3.0

data/lib/picky/cacher/partial_generator.rb

@@ -4,7 +4,7 @@ module Cacher
   #
   class PartialGenerator < Generator
     
-    # Generate a similarity index based on the given index.
+    # Generate a partial index based on the given index.
     #
     def generate strategy = Partial::Subtoken.new(:down_to => 1)
       strategy.generate_from self.index

data/lib/picky/extensions/hash.rb

@@ -2,16 +2,22 @@
 #
 class Hash
   
-  # Dumps binary self to the path given.
+  # Dumps jsonized self to the path given. Minus extension.
   #
-  # TODO Still used? If yes, spec!
-  #
-  def dump_to path
-    File.open(path, 'w:binary') do |out_file|
+  def dump_to_json path
+    File.open("#{path}.json", 'w') do |out_file|
       Yajl::Encoder.encode self, out_file
     end
   end
   
+  # Dumps binary self to the path given. Minus extension.
+  #
+  def dump_to_marshalled path
+    File.open("#{path}.dump", 'w:binary') do |out_file|
+      Marshal.dump self, out_file
+    end
+  end
+  
   # Use yajl's encoding.
   #
   def to_json options = {}

data/lib/picky/generator.rb

@@ -40,7 +40,7 @@ module Picky
     def generator_for identifier, *args
       generator_class = types[identifier.to_sym]
       raise NoGeneratorException unless generator_class
-      generator_for_class generator_class, *args
+      generator_for_class generator_class, identifier, *args
     end
     
     #
@@ -56,11 +56,11 @@ module Picky
     #
     class Project
       
-      attr_reader :name, :prototype_project_basedir
+      attr_reader :name, :project_prototype_basedir
       
-      def initialize name, *args
+      def initialize identifier, name, *args
         @name = name
-        @prototype_project_basedir = File.expand_path '../../../prototype_project', __FILE__
+        @project_prototype_basedir = File.expand_path '../../../project_prototype', __FILE__
       end
       
       #
@@ -70,12 +70,14 @@ module Picky
         create_target_directory
         copy_all_files
         exclaim "\"#{name}\" is a great project name! Have fun :)\n"
-        exclaim "Next steps:"
         exclaim ""
-        exclaim "cd #{name}"
-        exclaim "cat Gemfile # <- Do you need the mysql gem, for example?"
-        exclaim "bundle install"
-        exclaim "rake        # <- shows you where Picky needs input from you."
+        exclaim "Next steps:"
+        exclaim "1. cd #{name}"
+        exclaim "2. bundle install"
+        exclaim "3. rake index"
+        exclaim "4. rake start"
+        exclaim "5. rake           # (optional) shows you where Picky needs input from you"
+        exclaim "                  #            if you want to define your own search."
       end
       
       #
@@ -101,7 +103,7 @@ module Picky
       #
       #
       def target_filename_for filename
-        filename.gsub(%r{#{prototype_project_basedir}}, target_directory)
+        filename.gsub(%r{#{project_prototype_basedir}}, target_directory)
       end
       #
       #
@@ -145,7 +147,7 @@ module Picky
       #
       #
       def all_prototype_files
-        Dir[File.join(prototype_project_basedir, '**', '*')]
+        Dir[File.join(project_prototype_basedir, '**', '*')]
       end
       
       #

data/lib/picky/index/bundle.rb

@@ -152,7 +152,7 @@ module Index
     # Generates a cache path.
     #
     def cache_path text
-      File.join cache_directory, "#{name}_#{text}.json"
+      File.join cache_directory, "#{name}_#{text}"
     end
     def index_cache_path
       cache_path "#{category.name}_index"
@@ -171,20 +171,23 @@ module Index
       load_similarity
       load_weights
     end
-    def load_the index_method_name, path
-      self.send "#{index_method_name}=", Yajl::Parser.parse(File.open(path, 'r'), :symbolize_keys => true)
+    def load_the_json path
+      Yajl::Parser.parse File.open("#{path}.json", 'r'), :symbolize_keys => true
+    end
+    def load_the_marshalled path
+      Marshal.load File.open("#{path}.dump", 'r:binary')
     end
     def load_index
       timed_exclaim "Loading the index for #{identifier} from the cache."
-      load_the :index, index_cache_path
+      self.index = load_the_json index_cache_path
     end
     def load_similarity
       timed_exclaim "Loading the similarity for #{identifier} from the cache."
-      load_the :similarity, similarity_cache_path
+      self.similarity = load_the_marshalled similarity_cache_path
     end
     def load_weights
       timed_exclaim "Loading the weights for #{identifier} from the cache."
-      load_the :weights, weights_cache_path
+      self.weights = load_the_json weights_cache_path
     end
     
     # Generation
@@ -195,16 +198,10 @@ module Index
     # * generates derived indexes
     # * dumps all the indexes into files
     #
-    # TODO Rename to Source!!!
-    #
-    def generate_caches_from_db
-      cache_from_db_generation_message
+    def generate_caches_from_source
       load_from_index_file
       generate_caches_from_memory
     end
-    def cache_from_db_generation_message
-      timed_exclaim "CACHE FROM SOURCE #{identifier}."
-    end
     # Generates derived indexes from the index and dumps.
     #
     # Note: assumes that there is something in the index
@@ -227,9 +224,13 @@ module Index
     # Load the data from the db.
     #
     def load_from_index_file
+      load_from_index_generation_message
       clear
       retrieve
     end
+    def load_from_index_generation_message
+      timed_exclaim "LOAD INDEX #{identifier}."
+    end
     # Retrieves the data into the index.
     #
     # TODO Beautify.
@@ -272,6 +273,7 @@ module Index
       self.index = generator.generate self.partial_strategy
     end
     def generate_partial_from exact_index
+      timed_exclaim "PARTIAL GENERATE #{identifier}."
       self.index = exact_index
       self.generate_partial
       self
@@ -299,13 +301,20 @@ module Index
       dump_weights
     end
     def dump_index
-      index.dump_to index_cache_path
+      timed_exclaim "DUMP INDEX #{identifier}."
+      index.dump_to_json index_cache_path
     end
+    # Note: We marshal the similarity, as the
+    #       Yajl json lib cannot load symbolized
+    #       values, just keys.
+    #
     def dump_similarity
-      similarity.dump_to similarity_cache_path
+      timed_exclaim "DUMP SIMILARITY #{identifier}."
+      similarity.dump_to_marshalled similarity_cache_path
     end
     def dump_weights
-      weights.dump_to weights_cache_path
+      timed_exclaim "DUMP WEIGHTS #{identifier}."
+      weights.dump_to_json weights_cache_path
     end
 
   end

data/lib/picky/index/category.rb

@@ -40,17 +40,14 @@ module Index
     # Generates all caches for this category.
     #
     def generate_caches
-      timed_exclaim "LOAD #{identifier}."
-      generate_caches_from_db
-      timed_exclaim "PARTIAL #{identifier}."
+      generate_caches_from_source
       generate_partial
-      timed_exclaim "CACHE #{identifier}."
       generate_caches_from_memory
-      timed_exclaim "DUMP #{identifier}."
       dump_caches
+      timed_exclaim "CACHE FINISHED #{identifier}."
     end
-    def generate_caches_from_db
-      exact.generate_caches_from_db
+    def generate_caches_from_source
+      exact.generate_caches_from_source
     end
     def generate_partial
       partial.generate_partial_from exact.index

data/lib/picky/indexers/base.rb

@@ -61,7 +61,7 @@ module Indexers
     end
     
     def indexing_message
-      timed_exclaim "INDEX #{@type.name}, #{@field.name}" #:#{@field.indexed_name}." # TODO field.identifier
+      timed_exclaim "INDEX #{@type.name} #{@field.name}" #:#{@field.indexed_name}." # TODO field.identifier
     end
     
   end

data/lib/picky/indexes.rb

@@ -16,7 +16,7 @@ module Indexes
 
     # Run in parallel.
     #
-    timed_exclaim "Indexing using #{Cores.max_processors} processors."
+    timed_exclaim "INDEXING USING #{Cores.max_processors} PROCESSORS."
     Cores.forked self.fields, :randomly => true do |field|
       # Reestablish DB connection.
       #
@@ -24,6 +24,7 @@ module Indexes
       field.index
       field.cache
     end
+    timed_exclaim "INDEXING FINISHED."
   end
   def self.index_solr
     configuration.index_solr

data/lib/picky/signals.rb

@@ -1,4 +1,4 @@
-# TODO Cleanup and move to prototype_project.
+# TODO Cleanup and move to project_prototype.
 #
 # Signal.trap 'USR1' do
 #   Indexes.reload

data/{prototype_project → project_prototype}/Gemfile

@@ -2,7 +2,7 @@ source :gemcutter
 
 # Gems required by Picky.
 #
-gem 'picky',            '~> 0.2.0'
+gem 'picky',            '~> 0.3.0'
 gem 'bundler',          '>= 0.9.26'
 gem 'rack',             '~> 1.2.1'
 gem 'rack-mount',       '~> 0.6.9'
@@ -26,4 +26,4 @@ gem 'unicorn'
 
 # Optional. Use your preferred database adapter.
 #
-gem 'mysql'
+# gem 'mysql'

data/{prototype_project → project_prototype}/app/application.rb

@@ -7,7 +7,7 @@ class PickySearch < Application
   #
   # Check the Wiki http://github.com/floere/picky/wiki for more options.
   #
-  # Ask me if you have questions or specific requests.
+  # Ask me or the google group if you have questions or specific requests.
   #
   
   indexing.removes_characters(/[^a-zA-Z0-9\s\/\-\"\&\.]/)
@@ -15,7 +15,9 @@ class PickySearch < Application
   indexing.splits_text_on(/[\s\/\-\"\&\.]/)
       
   books_index = index :books,
-                      Sources::DB.new('SELECT id, title, author, isbn13 as isbn FROM books', :file => 'app/db.yml'),
+                      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'), 
                       field(:title, :similarity => Similarity::DoubleLevenshtone.new(3)), # Up to three similar title word indexed.
                       field(:author),
                       field(:isbn,  :partial => Partial::None.new) # Partially searching on an ISBN makes not much sense.

data/{prototype_project → project_prototype}/app/db.yml

@@ -9,5 +9,5 @@ adapter: mysql
 host: localhost
 username: root
 password:
-database: books_database # Needs to contain the DB source in app/application.rb.
+database: your_database # Load this configuration and use a DB source in app/application.rb.
 encoding: utf8