picky 4.15.0 → 4.15.1

checksums.yaml

@@ -0,0 +1,15 @@
+---
+!binary "U0hBMQ==":
+  metadata.gz: !binary |-
+    M2MxMTQxZjI1MDcwYzA1N2UzNWY2MTQ5ZGNkMTllYWFjYmI1MDk0Nw==
+  data.tar.gz: !binary |-
+    NjA3MTc5MTEwMWUzZWQ1NGZjNWU0ZjRhZDJmMTJhZjhhYjdiZWIzNA==
+!binary "U0hBNTEy":
+  metadata.gz: !binary |-
+    NjdhYmYyYTkwOGJlZmUzZjljYTA2MjIxNzg0NWEwYzZkZWE2ZmE3OTdmZGEz
+    Mjc4YjYwMWZlMTU4Zjk0NGQyNGE4NzM2MjA0ZWY2MGRkODlhM2Y5Y2ZhYWZk
+    MzFjOWM2ZTQ4MzQzYTFiYTA3ZjUyZmQyZTU1NGU0ZjExMGQyYTA=
+  data.tar.gz: !binary |-
+    MzlhODU0NmZlNGRkOGM4YzlkZDY3MTQxYmJlYTA2MWJkYWEwMGI4MjEwNzA1
+    M2VhYjExNzQ2YzdhODcyZjE0ZWVlN2QzMDRmNTU0YTI0YjVjMjFkNzc1ODk2
+    ZjQ3OTQyNTVhNmY3ZjhiN2ZhZWY4MWM0OTcyMDYxYjIzMTQ1YTY=

data/lib/picky/analyzer.rb

@@ -1,6 +1,17 @@
 # encoding: utf-8
 #
 
+class Range
+  def expand_with thing
+    return (thing..thing) unless min
+    if max < thing
+      (min..thing)
+    else
+      thing < min ? (thing..max) : self
+    end
+  end
+end
+
 # Analyzes indexes (index bundles, actually).
 #
 class Analyzer
@@ -38,46 +49,39 @@ class Analyzer
 
     self
   end
+    
   def cardinality identifier, index
-    return if index.size.zero?
-    return unless index.respond_to?(:each_pair)
-
-    key_length_average = 0
-    ids_length_average = 0
+    return unless can_calculate_cardinality? index
 
-    min_ids_length = 1.0/0 # Infinity
-    max_ids_length =     0
-    min_key_length = 1.0/0 # Infinity
-    max_key_length =     0
+    key_length_sum = 0
+    ids_length_sum = 0
+    
+    key_length = (1.0/0..0)
+    ids_length = (1.0/0..0)
 
-    key_size, ids_size = 0, 0
     index.each_pair do |key, ids|
-      key_size = key.size
-      if key_size < min_key_length
-        min_key_length = key_size
-      else
-        max_key_length = key_size if key_size > max_key_length
-      end
-      key_length_average += key_size
-
-      ids_size = ids.size
-      if ids_size < min_ids_length
-        min_ids_length = ids_size
-      else
-        max_ids_length = ids_size if ids_size > max_ids_length
-      end
-      ids_length_average += ids_size
+      key_length = key_length.expand_with key.size
+      key_length_sum += key.size
+      ids_length = ids_length.expand_with ids.size
+      ids_length_sum += ids.size
     end
-    index_size = index.size
-    key_length_average = key_length_average.to_f / index_size
-    ids_length_average = ids_length_average.to_f / index_size
-
-    analysis[identifier] ||= {}
-    analysis[identifier][:key_length]         = (min_key_length..max_key_length)
-    analysis[identifier][:ids_length]         = (min_ids_length..max_ids_length)
-    analysis[identifier][:key_length_average] = key_length_average
-    analysis[identifier][:ids_length_average] = ids_length_average
+
+    report_cardinality identifier, index, key_length, ids_length, key_length_sum, ids_length_sum
+  end
+  def report_cardinality identifier, index, key_length, ids_length, key_length_sum, ids_length_sum
+    analysis_identifier = analysis[identifier] ||= {}
+    analysis_identifier[:key_length]         = key_length
+    analysis_identifier[:ids_length]         = ids_length
+    analysis_identifier[:key_length_average] = key_length_sum.to_f / index.size
+    analysis_identifier[:ids_length_average] = ids_length_sum.to_f / index.size    
+  end
+  
+  def can_calculate_cardinality? index
+    return if index.size.zero?
+    return unless index.respond_to? :each_pair
+    true
   end
+  
   def index_analysis
     return unless analysis[:index]
 
@@ -91,6 +95,7 @@ class Analyzer
       comments << "\033[33mIndex matches single characters.\033[m"
     end
   end
+  
   def weights index
     return if !index.respond_to?(:size) || index.size.zero?
     return unless index.respond_to?(:each_pair)
@@ -115,6 +120,7 @@ class Analyzer
     analysis[:weights][:weight_range]   = (min_weight..max_weight)
     analysis[:weights][:weight_average] = weight_average
   end
+  
   def weights_analysis
     return unless analysis[:weights]
 
@@ -131,25 +137,36 @@ class Analyzer
   def to_s
     [*comments, index_to_s, weights_to_s, similarity_to_s, configuration_to_s].compact.join "\n"
   end
+  
   def index_to_s
     return if analysis[:__keys].zero?
-    ary = ["index key cardinality:                #{"%10d" % analysis[:__keys]}"]
+    ary = ["index key cardinality:                #{"%9d" % analysis[:__keys]}"]
     return ary.join "\n" unless analysis[:index]
-    ary << formatted('',        :key_length)
+    ary << formatted(nil,       :key_length)
     ary << formatted('ids per', :ids_length)
     ary.join "\n"
   end
+  
   def formatted description, key, index = :index
-    "index #{description} key length range (avg): #{"%10s" % analysis[index][key]} (#{analysis[index][:"#{key}_average"].round(2)})"
+    what    = "%-40s" % ["index", description, "key length range (avg):"].compact.join(' ')
+    range   = "%7s" % analysis[index][key]
+    average = "%8s" % "(#{analysis[index][:"#{key}_average"].round(2)})"
+    what + range + average
   end
+  
   def weights_to_s
     return unless analysis[:weights]
-    %Q{weights range (avg):                  #{"%10s" % analysis[:weights][:weight_range]} (#{analysis[:weights][:weight_average].round(2)})}
+    what    = "%-30s" % "weights range (avg):"
+    range   = "%17s" % analysis[:weights][:weight_range]
+    average = "%8s" % "(#{analysis[:weights][:weight_average].round(2)})"
+    what + range + average
   end
+  
   def similarity_to_s
     return unless analysis[:similarity]
     formatted('similarity', :key_length, :similarity)
   end
+  
   def configuration_to_s
     # analysis[:configuration]
   end

data/lib/picky/category_realtime.rb

@@ -17,7 +17,11 @@ module Picky
     # given object.
     #
     def add object, where = :unshift
-      add_text object.id, object.send(from), where
+      if from.respond_to? :call
+        add_text object.id, from.call(object), where
+      else
+        add_text object.id, object.send(from), where
+      end
     end
 
     # Removes the object's id, and then
@@ -65,8 +69,13 @@ module Picky
       else
         tokens = text_or_tokens
       end
+      
+      # TODO Have an "as is" key_format?
+      #
       tokens.each { |text| add_tokenized_token id.send(key_format), text, where, false }
     rescue NoMethodError
+      # TODO This also is raised on a wrong key_format.
+      # TODO Improve error message by pointing out what exactly goes wrong: thing xy does not have an #each method.
       raise %Q{You probably set tokenize: false on category "#{name}". It will need an Enumerator of previously tokenized tokens.}
     end
 
@@ -98,4 +107,4 @@ module Picky
 
   end
 
-end
+end

data/lib/picky/index_indexing.rb

@@ -5,7 +5,9 @@ module Picky
   class Index
     include Helpers::Indexing
 
-    forward :cache, :clear, :to => :categories
+    forward :cache,
+            :clear,
+            :to => :categories
 
     # Define an index tokenizer on the index.
     #
@@ -21,6 +23,8 @@ module Picky
     # Decides whether to use a parallel indexer or whether to
     # forward to each category to prepare themselves.
     #
+    # TODO Do a critical reading of this on the blog.
+    #
     def prepare scheduler = Scheduler.new
       if source.respond_to?(:each)
         check_source_empty
@@ -79,7 +83,7 @@ module Picky
       some_source ? (@source = Source.from(some_source, false, name)) : unblock_source
     end
     # Get the actual source if it is wrapped in a time
-    # capsule, i.e. a block/lambda.
+    # capsule, ie. a block/lambda.
     #
     def unblock_source
       @source.respond_to?(:call) ? @source.call : @source
@@ -87,7 +91,9 @@ module Picky
 
     # Define a key_format on the index.
     #
-    # Parameter is a method name to use on the key (e.g. :to_i, :to_s, :strip).
+    # Parameter is a method name to use on the key (e.g. :to_i, :to_s, :strip, :split).
+    #
+    # TODO Rename id_format?
     #
     def key_format key_format = nil
       key_format ? (@key_format = key_format) : @key_format

data/spec/functional/arrays_as_ids_spec.rb

@@ -0,0 +1,57 @@
+# encoding: utf-8
+#
+require 'spec_helper'
+
+describe "Array IDs" do
+
+  # This tests the weights option.
+  #
+  it 'can use Arrays as IDs' do
+    index = Picky::Index.new :arrays do
+      key_format :to_a
+      
+      category :text1
+    end
+
+    require 'ostruct'
+
+    thing = OpenStruct.new id: ['id1', 'thing1'], text1: "ohai"
+    other = OpenStruct.new id: ['id2', 'thing2'], text1: "ohai kthxbye"
+
+    index.add thing
+    index.add other
+
+    try = Picky::Search.new index
+
+    try.search("text1:ohai").ids.should == [
+      ["id2", "thing2"],
+      ["id1", "thing1"]
+    ] # WAT
+  end
+
+  # This tests the weights option.
+  #
+  it 'can use split as key_format' do
+    index = Picky::Index.new :arrays do
+      key_format :split
+      
+      category :text1
+    end
+
+    require 'ostruct'
+    
+    thing = OpenStruct.new id: "id1 thing1", text1: "ohai"
+    other = OpenStruct.new id: "id2 thing2", text1: "ohai kthxbye"
+
+    index.add thing
+    index.add other
+
+    try = Picky::Search.new index
+
+    try.search("text1:ohai").ids.should == [
+      ["id2", "thing2"],
+      ["id1", "thing1"]
+    ] # WAT
+  end
+
+end

data/spec/functional/from_spec.rb

@@ -0,0 +1,25 @@
+# encoding: utf-8
+#
+require 'spec_helper'
+
+describe "From option" do
+
+  it 'can be given a lambda' do
+    index = Picky::Index.new :lambda do
+      category :text, from: ->(thing){ thing.some_text * 2 } # Anything, really.
+    end
+
+    require 'ostruct'
+
+    thing = OpenStruct.new id: 1, some_text: "ohai"
+    other = OpenStruct.new id: 2, some_text: "ohai kthxbye"
+
+    index.add thing
+    index.add other
+
+    try = Picky::Search.new index
+
+    try.search("text:ohaiohai").ids.should == [1]
+  end
+
+end

metadata

@@ -1,93 +1,113 @@
 --- !ruby/object:Gem::Specification
 name: picky
 version: !ruby/object:Gem::Version
-  version: 4.15.0
-  prerelease: 
+  version: 4.15.1
 platform: ruby
 authors:
 - Florian Hanke
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-05-02 00:00:00.000000000 Z
+date: 2013-06-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec
-  requirement: &70116179207700 !ruby/object:Gem::Requirement
-    none: false
+  requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *70116179207700
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rake-compiler
-  requirement: &70116179229800 !ruby/object:Gem::Requirement
-    none: false
+  requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *70116179229800
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: picky-client
-  requirement: &70116179229120 !ruby/object:Gem::Requirement
-    none: false
+  requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: 4.15.0
+        version: 4.15.1
   type: :development
   prerelease: false
-  version_requirements: *70116179229120
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 4.15.1
 - !ruby/object:Gem::Dependency
   name: text
-  requirement: &70116179228440 !ruby/object:Gem::Requirement
-    none: false
+  requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *70116179228440
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: multi_json
-  requirement: &70116179227980 !ruby/object:Gem::Requirement
-    none: false
+  requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *70116179227980
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: activesupport
-  requirement: &70116179227480 !ruby/object:Gem::Requirement
-    none: false
+  requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '3.0'
   type: :runtime
   prerelease: false
-  version_requirements: *70116179227480
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '3.0'
 - !ruby/object:Gem::Dependency
   name: rack_fast_escape
-  requirement: &70116179227040 !ruby/object:Gem::Requirement
-    none: false
+  requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *70116179227040
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 description: Fast Ruby semantic text search engine with comfortable single field interface.
 email: florian.hanke+picky@gmail.com
 executables:
@@ -244,6 +264,7 @@ files:
 - ext/picky/picky.c
 - spec/aux/picky/cli_spec.rb
 - spec/functional/allocations_uniq_by_definition_spec.rb
+- spec/functional/arrays_as_ids_spec.rb
 - spec/functional/automatic_segmentation_spec.rb
 - spec/functional/backends/file_spec.rb
 - spec/functional/backends/memory_bundle_realtime_spec.rb
@@ -259,6 +280,7 @@ files:
 - spec/functional/error_messages_spec.rb
 - spec/functional/exact_first_spec.rb
 - spec/functional/facets_spec.rb
+- spec/functional/from_spec.rb
 - spec/functional/ignore_allocations_spec.rb
 - spec/functional/ignore_spec.rb
 - spec/functional/max_allocations_spec.rb
@@ -385,35 +407,36 @@ files:
 - spec/lib/tasks/try_spec.rb
 - spec/lib/tokenizer_spec.rb
 - spec/performant_spec.rb
-- bin/picky
+- !binary |-
+  YmluL3BpY2t5
 - ext/picky/extconf.rb
 homepage: http://florianhanke.com/picky
 licenses: []
+metadata: {}
 post_install_message: 
 rdoc_options: []
 require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
-  none: false
   requirements:
   - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
-  none: false
   requirements:
   - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: http://rubyforge.org/projects/picky
-rubygems_version: 1.8.10
+rubygems_version: 2.0.3
 signing_key: 
-specification_version: 3
+specification_version: 4
 summary: ! 'Picky: Semantic Search Engine. Clever Interface. Good Tools.'
 test_files:
 - spec/aux/picky/cli_spec.rb
 - spec/functional/allocations_uniq_by_definition_spec.rb
+- spec/functional/arrays_as_ids_spec.rb
 - spec/functional/automatic_segmentation_spec.rb
 - spec/functional/backends/file_spec.rb
 - spec/functional/backends/memory_bundle_realtime_spec.rb
@@ -429,6 +452,7 @@ test_files:
 - spec/functional/error_messages_spec.rb
 - spec/functional/exact_first_spec.rb
 - spec/functional/facets_spec.rb
+- spec/functional/from_spec.rb
 - spec/functional/ignore_allocations_spec.rb
 - spec/functional/ignore_spec.rb
 - spec/functional/max_allocations_spec.rb