picky 2.1.2 → 2.2.0

data/{lib → aux}/picky/cli.rb

@@ -19,7 +19,7 @@ module Picky
 
     class Base
       def usage name, params
-        puts "Usage\n  picky #{name} #{params_to_s(params)}"
+        puts "Usage:\n  picky #{name} #{params_to_s(params)}"
       end
       # String params are optional, Symbol params aren't.
       #
@@ -27,25 +27,21 @@ module Picky
         params.map { |param| param.respond_to?(:to_str) ? "[#{param}]" : param }.join(' ') if params
       end
     end
-    class Statistics < Base
+    class Generate < Base
       def execute name, args, params
-        relative_log_file = args.shift
-        port              = args.shift
-
-        usage(name, params) || exit(1) unless relative_log_file
-
-        ENV['PICKY_LOG_FILE']        = File.expand_path relative_log_file
-        ENV['PICKY_STATISTICS_PORT'] = port
-
-        begin
-          require 'picky-statistics'
-        rescue LoadError => e
-          require 'picky/extensions/object'
-          warn_gem_missing 'picky-statistics', 'the Picky statistics'
-          exit 1
-        end
+        Kernel.system "picky-generate #{args.join(' ')}"
+      end
+    end
+    class Help < Base
+      # Displays usage information.
+      #
+      def execute name, args, params
+        commands = Picky::CLI.mapping.map do |command, object_and_params|
+          _, *params = object_and_params
+          "  picky #{command} #{params_to_s(params)}"
+        end.join(?\n)
 
-        require 'picky-statistics/application/app'
+        Kernel.puts "Possible commands:\n#{commands}\n"
       end
     end
     class Live < Base
@@ -58,32 +54,47 @@ module Picky
         ENV['PICKY_LIVE_URL']  = url
         ENV['PICKY_LIVE_PORT'] = port
 
-        begin
-          require 'picky-live'
-        rescue LoadError => e
-          require 'picky/extensions/object'
-          warn_gem_missing 'picky-live', 'the Picky Live Interface'
-          exit 1
-        end
-
+        require 'picky-live'
         require 'picky-live/application/app'
+      rescue LoadError => e
+        require 'picky/extensions/object'
+        warn_gem_missing 'picky-live', 'the Picky Live Interface'
+        exit 1
       end
     end
-    class Generate < Base
+    class Search < Base
       def execute name, args, params
-        Kernel.system "picky-generate #{args.join(' ')}"
+        url_or_path = args.shift
+        ids         = args.shift
+
+        usage(name, params) || exit(1) unless url_or_path
+
+        require 'picky-client'
+        require 'picky-client/aux/terminal'
+        terminal = Terminal.new url_or_path, ids
+        terminal.run
+      rescue LoadError => e
+        require 'picky/extensions/object'
+        warn_gem_missing 'picky-client', 'the Picky client'
+        exit 1
       end
     end
-    class Help < Base
-      # Displays usage information.
-      #
+    class Statistics < Base
       def execute name, args, params
-        commands = Picky::CLI.mapping.map do |command, object_and_params|
-          _, *params = object_and_params
-          "  picky #{command} #{params_to_s(params)}"
-        end.join(?\n)
+        relative_log_file = args.shift
+        port              = args.shift
 
-        Kernel.puts "Possible commands:\n#{commands}\n"
+        usage(name, params) || exit(1) unless relative_log_file
+
+        ENV['PICKY_LOG_FILE']        = File.expand_path relative_log_file
+        ENV['PICKY_STATISTICS_PORT'] = port
+
+        require 'picky-statistics'
+        require 'picky-statistics/application/app'
+      rescue LoadError => e
+        require 'picky/extensions/object'
+        warn_gem_missing 'picky-statistics', 'the Picky statistics'
+        exit 1
       end
     end
 
@@ -92,8 +103,9 @@ module Picky
     @@mapping = {
       :generate => [Generate, :'{sinatra_client,unicorn_server,empty_unicorn_server}', :'app_directory_name'],
       :help     => [Help],
-      :stats    => [Statistics, :'logfile (e.g. log/search.log)', 'port (default: 4567)'],
-      :live     => [Live, 'host:port/path (default: localhost:8080/admin)', 'port (default: 4568)']
+      :live     => [Live, 'host:port/path (default: localhost:8080/admin)', 'port (default: 4568)'],
+      :search   => [Search, :url_or_path, 'amount of ids (default 20)'],
+      :stats    => [Statistics, :'logfile (e.g. log/search.log)', 'port (default: 4567)']
     }
     def self.mapping
       @@mapping

data/bin/picky

@@ -5,7 +5,7 @@ begin
   require 'picky/cli'
 rescue LoadError => e
   require 'rubygems'
-  picky_path = File.expand_path '../../lib', __FILE__
+  picky_path = File.expand_path '../../aux', __FILE__
   $:.unshift(picky_path) if File.directory?(picky_path) && !$:.include?(picky_path)
   require 'picky/cli'
 end

data/lib/picky/application.rb

@@ -152,16 +152,19 @@ class Application
     # Returns a configured tokenizer that
     # is used for indexing by default.
     #
-    def default_indexing options = {}
+    def indexing options = {}
       Internals::Tokenizers::Index.default = Internals::Tokenizers::Index.new(options)
     end
+    alias default_indexing indexing
 
     # Returns a configured tokenizer that
     # is used for querying by default.
     #
-    def default_querying options = {}
+    def searching options = {}
       Internals::Tokenizers::Query.default = Internals::Tokenizers::Query.new(options)
     end
+    alias default_querying searching
+    alias querying searching
 
     # Routes.
     #

data/lib/picky/index/base.rb

@@ -9,29 +9,36 @@ module Index
   #
   class Base
 
-    attr_reader :name, :indexing, :indexed
+    attr_reader :name
 
     # Create a new index with a given source.
     #
     # === Parameters
     # * name: A name that will be used for the index directory and in the Picky front end.
-    # * source: Where the data comes from, e.g. Sources::CSV.new(...)
+    # * source: Where the data comes from, e.g. Sources::CSV.new(...). Optional, can be defined in the block using #source.
     #
     # === Options
     # * result_identifier: Use if you'd like a different identifier/name in the results than the name of the index.
     # * after_indexing: As of this writing only used in the db source. Executes the given after_indexing as SQL after the indexing process.
+    # * tokenizer: The tokenizer to use for this index.
     #
-    # Example:
+    # Examples:
     #   my_index = Index::Memory.new(:my_index, some_source) do
-    #     define_category :bla
+    #     category :bla
+    #   end
+    #
+    #   my_index = Index::Memory.new(:my_index) do
+    #     source   Sources::CSV.new(file: 'data/index.csv')
+    #     category :bla
     #   end
     #
     #
-    def initialize name, source, options = {}
-      check name, source
+    def initialize name, options = {}
+      check_name name
+      @name = name.to_sym
 
-      @name     = name.to_sym
-      @indexing = Internals::Indexing::Index.new name, source, options
+      check_options options
+      @indexing = Internals::Indexing::Index.new name, options
       @indexed  = Internals::Indexed::Index.new  name, options
 
       # Centralized registry.
@@ -41,37 +48,97 @@ module Index
       #
       #
       instance_eval(&Proc.new) if block_given?
+
+      check_source internal_indexing.source
+    end
+    def internal_indexing
+      @indexing
+    end
+    def internal_indexed
+      @indexed
     end
     #
     # Since this is an API, we fail hard quickly.
     #
-    def check name, source
+    def check_name name
       raise ArgumentError.new(<<-NAME
+
+
 The index identifier (you gave "#{name}") for Index::Memory/Index::Redis should be a String/Symbol,
 Examples:
   Index::Memory.new(:my_cool_index, ...) # Recommended
   Index::Redis.new("a-redis-index", ...)
 NAME
+
+
 ) unless name.respond_to?(:to_sym)
+    end
+    def check_options options
+      raise ArgumentError.new(<<-OPTIONS
+
+
+Sources are not passed in as second parameter for #{self.class.name} anymore, but either
+* as :source option:
+  #{self.class.name}.new(#{name.inspect}, source: #{options})
+or
+* given to the #source method inside the config block:
+  #{self.class.name}.new(#{name.inspect}) do
+    source #{options}
+  end
+
+Sorry about that breaking change (in 2.2.0), didn't want to go to 3.0.0 yet!
+
+All the best
+  -- Picky
+
+
+OPTIONS
+) unless options.respond_to?(:[])
+    end
+    def check_source source
       raise ArgumentError.new(<<-SOURCE
-The index "#{name}" should use a data source that responds to the method #harvest, which yields(id, text).
+
+
+The index "#{name}" should use a data source that responds to either the method #each, or the method #harvest, which yields(id, text).
 Or it could use one of the built-in sources:
   Sources::#{(Sources.constants - [:Base, :Wrappers, :NoCSVFileGiven, :NoCouchDBGiven]).join(',
   Sources::')}
+
+
 SOURCE
-) unless source.respond_to?(:harvest)
+) unless source.respond_to?(:each) || source.respond_to?(:harvest)
     end
 
     def to_stats
       stats = <<-INDEX
 #{name} (#{self.class}):
-  #{"source:            #{indexing.source}".indented_to_s}
-  #{"categories:        #{indexing.categories.categories.map(&:name).join(', ')}".indented_to_s}
+  #{"source:            #{internal_indexing.source}".indented_to_s}
+  #{"categories:        #{internal_indexing.categories.map(&:name).join(', ')}".indented_to_s}
 INDEX
-      stats << "  result identifier: \"#{indexed.result_identifier}\"".indented_to_s unless indexed.result_identifier.to_s == indexed.name.to_s
+      stats << "  result identifier: \"#{internal_indexed.result_identifier}\"".indented_to_s unless internal_indexed.result_identifier.to_s == internal_indexed.name.to_s
       stats
     end
 
+    # Define an index tokenizer on the index.
+    #
+    # Parameters are the exact same as for the default_indexing.
+    #
+    def define_indexing options = {}
+      internal_indexing.define_indexing options
+    end
+    alias indexing define_indexing
+
+    # Define a source on the index.
+    #
+    # Parameter is a source, either one of the standard sources or
+    # anything responding to #each and returning objects that
+    # respond to id and the category names (or the category from option).
+    #
+    def define_source source
+      internal_indexing.define_source source
+    end
+    alias source define_source
+
     # Defines a searchable category on the index.
     #
     # === Parameters
@@ -88,8 +155,8 @@ INDEX
     def define_category category_name, options = {}
       category_name = category_name.to_sym
 
-      indexing_category = indexing.define_category category_name, options
-      indexed_category  = indexed.define_category  category_name, options
+      indexing_category = internal_indexing.define_category category_name, options
+      indexed_category  = internal_indexed.define_category  category_name, options
 
       yield indexing_category, indexed_category if block_given?
 
@@ -157,22 +224,18 @@ INDEX
     # * ... all options of #define_category.
     #
     def define_ranged_category category_name, range, options = {}
-      precision = options[:precision]
+      precision = options[:precision] || 1
 
       options = { partial: Partial::None.new }.merge options
 
-      define_category category_name, options do |indexing, indexed|
-        indexing.source    = Sources::Wrappers::Location.new indexing, grid: range, precision: precision
-        indexing.tokenizer = Internals::Tokenizers::Index.new
-
-        exact_bundle    = Indexed::Wrappers::Bundle::Location.new indexed.exact, grid: range, precision: precision
-        indexed.exact   = exact_bundle
-        indexed.partial = exact_bundle # A partial token also uses the exact index.
+      define_category category_name, options do |indexing_category, indexed_category|
+        Internals::Indexing::Wrappers::Category::Location.install_on indexing_category, range, precision
+        Internals::Indexed::Wrappers::Category::Location.install_on indexed_category, range, precision
       end
     end
     alias ranged_category define_ranged_category
 
-    #  HIGHLY EXPERIMENTAL Not correctly working yet. Try it if you feel "beta".
+    # HIGHLY EXPERIMENTAL Not correctly working yet. Try it if you feel "beta".
     #
     # Also a range search see #define_ranged_category, but on the earth's surface.
     #

data/lib/picky/index/memory.rb

@@ -1,9 +1,9 @@
 module Index
-  
+
   # An index that is persisted in files, loaded at startup and kept in memory at runtime.
   #
   class Memory < Base
-    
+
     # Create a new memory index for indexing and for querying.
     #
     # Parameters:
@@ -14,15 +14,15 @@ module Index
     # * source: The source the data comes from. See Sources::Base.
     #
     # Options:
-    # * result_identifier: Use if you'd like a different identifier/name in the results JSON than the name of the index. 
+    # * result_identifier: Use if you'd like a different identifier/name in the results JSON than the name of the index.
     #
-    def initialize name, source, options = {}
+    def initialize name, options = {}
+      super name, options
+
       options[:indexing_bundle_class] ||= Internals::Indexing::Bundle::Memory
       options[:indexed_bundle_class]  ||= Internals::Indexed::Bundle::Memory
-      
-      super name, source, options
     end
-    
+
   end
-  
+
 end

data/lib/picky/index/redis.rb

@@ -1,9 +1,9 @@
 module Index
-  
+
   # An index that is persisted in Redis.
   #
   class Redis < Base
-    
+
     # Create a new Redis index for indexing and for querying.
     #
     # Parameters:
@@ -14,15 +14,15 @@ module Index
     # * source: The source the data comes from. See Sources::Base.
     #
     # Options:
-    # * result_identifier: Use if you'd like a different identifier/name in the results JSON than the name of the index. 
+    # * result_identifier: Use if you'd like a different identifier/name in the results JSON than the name of the index.
     #
-    def initialize name, source, options = {}
+    def initialize name, options = {}
+      super name, options
+
       options[:indexing_bundle_class] ||= Internals::Indexing::Bundle::Redis
       options[:indexed_bundle_class]  ||= Internals::Indexed::Bundle::Redis
-      
-      super name, source, options
     end
-    
+
   end
-  
+
 end

data/lib/picky/index_bundle.rb

@@ -37,8 +37,8 @@ class IndexBundle # :nodoc:all
     self.indexes << index
     self.index_mapping[index.name] = index
 
-    indexing.register index.indexing
-    indexed.register  index.indexed
+    indexing.register index.internal_indexing
+    indexed.register  index.internal_indexed
   end
 
   def [] name

data/lib/picky/indexing/indexes.rb

@@ -47,8 +47,8 @@ module Indexing # :nodoc:all
       # Run indexing/caching forked.
       #
       Cores.forked self.indexes, { randomly: randomly } do |an_index|
-        an_index.index
-        an_index.cache
+        an_index.index!
+        an_index.cache!
       end
 
       timed_exclaim "Indexing finished."
@@ -60,8 +60,8 @@ module Indexing # :nodoc:all
       take_snapshot
 
       self.indexes.each do |an_index|
-        an_index.index
-        an_index.cache
+        an_index.index!
+        an_index.cache!
       end
     end
 
@@ -69,7 +69,7 @@ module Indexing # :nodoc:all
     #
     def generate_index_only index_name, category_name = nil
       found = find index_name, category_name
-      found.index if found
+      found.index! if found
     end
     # Generate only the cache for the given index:category pair.
     #
@@ -88,7 +88,7 @@ module Indexing # :nodoc:all
 
         return index unless category_name
 
-        found = index.categories.find category_name
+        found = index.find category_name
         return found if found
       end
 

data/lib/picky/internals/calculations/location.rb

@@ -1,47 +1,59 @@
-module Calculations # :nodoc:all
-  
-  # A location calculation recalculates a 1-d location
-  # to the Picky internal 1-d "grid".
-  #
-  # For example, if you have a location x == 12.3456,
-  # it will be recalculated into 3, if the minimum is 9
-  # and the gridlength is 1.
-  #
-  class Location
-    
-    attr_reader :minimum
-    
-    def initialize user_grid, precision = nil
-      @user_grid = user_grid
-      @precision = precision || 1
-      @grid      = @user_grid / (@precision + 0.5)
-    end
-    
-    def minimum= minimum
-      # Add a margin of 1 user grid.
-      #
-      minimum -= @user_grid
-      
-      # Add plus 1 grid so that the index key never falls on 0.
-      # Why? to_i maps by default to 0.
-      #
-      minimum -= @grid
-      
-      @minimum = minimum
-    end
-    
+module Internals
+  module Calculations # :nodoc:all
+
+    # A location calculation recalculates a 1-d location
+    # to the Picky internal 1-d "grid".
     #
+    # For example, if you have a location x == 12.3456,
+    # it will be recalculated into 3, if the minimum is 9
+    # and the gridlength is 1.
     #
-    def add_margin length
-      @minimum -= length
-    end
-    
-    #
-    #
-    def recalculate location
-      ((location - @minimum) / @grid).floor
+    class Location
+
+      attr_reader :minimum, :precision, :grid
+
+      def initialize user_grid, precision = nil
+        @user_grid = user_grid
+        @precision = precision || 1
+        @grid      = @user_grid / (@precision + 0.5)
+      end
+
+      def minimum= minimum
+        # Add a margin of 1 user grid.
+        #
+        minimum -= @user_grid
+
+        # Add plus 1 grid so that the index key never falls on 0.
+        # Why? to_i maps by default to 0.
+        #
+        minimum -= @grid
+
+        @minimum = minimum
+      end
+
+      #
+      #
+      def add_margin length
+        @minimum -= length
+      end
+
+      #
+      #
+      def recalculated_range location
+        range recalculate(location)
+      end
+      #
+      #
+      def range around_location
+        (around_location - @precision)..(around_location + @precision)
+      end
+      #
+      #
+      def recalculate location
+        ((location - @minimum) / @grid).floor
+      end
+
     end
-    
+
   end
-  
 end

data/lib/picky/internals/index/backend.rb

@@ -1,29 +1,29 @@
 module Internals
-  
+
   module Index
-  
+
     class Backend
-    
+
       attr_reader :bundle_name
       attr_reader :prepared, :index, :weights, :similarity, :configuration
-    
-      delegate :index_name, :category_name, :to => :@config
-    
-      def initialize bundle_name, config
+
+      delegate :index_name, :category_name, :to => :@category
+
+      def initialize bundle_name, category
         @bundle_name = bundle_name
-        @config      = config
-        @prepared    = File::Text.new config.prepared_index_path
+        @category    = category
+        @prepared    = File::Text.new category.prepared_index_path
       end
-    
+
       # Delegators.
       #
-    
+
       # Retrieving data.
       #
       def retrieve &block
         prepared.retrieve &block
       end
-    
+
       # Dumping.
       #
       def dump_index index_hash
@@ -38,7 +38,7 @@ module Internals
       def dump_configuration configuration_hash
         configuration.dump configuration_hash
       end
-    
+
       # Loading.
       #
       def load_index
@@ -53,7 +53,7 @@ module Internals
       def load_configuration
         configuration.load
       end
-    
+
       # Cache ok?
       #
       def index_cache_ok?
@@ -65,7 +65,7 @@ module Internals
       def weights_cache_ok?
         weights.cache_ok?
       end
-    
+
       # Cache small?
       #
       def index_cache_small?
@@ -77,7 +77,7 @@ module Internals
       def weights_cache_small?
         weights.cache_small?
       end
-    
+
       # Copies the indexes to the "backup" directory.
       #
       def backup
@@ -86,7 +86,7 @@ module Internals
         similarity.backup
         configuration.backup
       end
-    
+
       # Restores the indexes from the "backup" directory.
       #
       def restore
@@ -95,7 +95,7 @@ module Internals
         similarity.restore
         configuration.restore
       end
-      
+
       # Delete all index files.
       #
       def delete
@@ -104,9 +104,9 @@ module Internals
         similarity.delete
         configuration.delete
       end
-    
+
     end
-  
+
   end
-  
+
 end