picky 4.31.0 → 4.31.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 1399f4e15ee1e94d7ec6c91ab88b9721702524f6
-  data.tar.gz: 24e20ab22278fddf14d13a3ce5f22319aac4fe7b
+  metadata.gz: 2b3c54b9a8b3db32d2cfb2ddbc7acd92889ae4bf
+  data.tar.gz: dc1355c15b92450a9b4d5ac2d05cdd1754938e94
 SHA512:
-  metadata.gz: f0948fcb8da05a7174713c5dae9e408eb69808173de4d55dacfba1bdfd024016b63d030293a929973b3c805b29c319afc59bb85959f10609b438c4058fa81897
-  data.tar.gz: 930e22468a92b53f87fa06c51c96b94e3142d6d33457bafa936bebd65f5e14ab5a089d90f4cafb295005c65e157230b3d4e1e2694cb0a0d801c1798e8cf9837b
+  metadata.gz: 5c8267d3a6d40cb9170d9f068d296dca257c83651db5848c2e8d43c1e8547bb07a57a5a58bf20b8e9d3c3e1e0c5b64564daefa5a0e7cc434df0b2a7fbded6154
+  data.tar.gz: 4a5e283823b8558e5e030cb583b491f5be88669ca6e6b70904a78a5d92e1aeae10183b96e27deea2c816af677bd4038a8fa7251982c684ea0606fe1b9d55d68d

data/lib/picky/category.rb

@@ -62,6 +62,10 @@ module Picky
 
       @symbol_keys = options[:symbol_keys] || @index.symbol_keys # SYMBOLS.
     end
+    
+    def symbol_keys?
+      @symbol_keys
+    end
 
     # TODO I do a lot of helper method calls here. Refactor?
     #

data/lib/picky/splitters/automatic.rb

@@ -26,6 +26,7 @@ module Picky
     class Automatic
       
       def initialize category, options = {}
+        @category     = category
         @exact        = category.exact
         @partial      = category.partial
         @with_partial = options[:partial]
@@ -58,13 +59,16 @@ module Picky
       
       def segment text, use_partial = false
         segments, score = segment_recursively text, use_partial
+        segments.collect!(&:to_s) if @category.symbol_keys?
         [segments, score && score-text.size+segments.size]
       end
       
       # Segments the given text recursively.
       #
       def segment_recursively text, use_partial = false
+        text = text.to_sym if @category.symbol_keys?
         (use_partial ? @partial_memo : @exact_memo)[text] ||= splits(text).inject([[], nil]) do |(current, heaviest), (head, tail)|
+          tail = tail.to_sym if @category.symbol_keys?
           tail_weight = use_partial ? @partial.weight(tail) : @exact.weight(tail)
           tail_weight && tail_weight += (tail.size-1)
           

data/spec/functional/automatic_segmentation_spec.rb

@@ -4,117 +4,123 @@ require 'spec_helper'
 
 describe "automatic splitting" do
   
-  let(:index) do
-    index = Picky::Index.new :automatic_text_splitting do
-      indexing removes_characters: /[^a-z\s]/i,
-               stopwords: /\b(in|a)\b/
-      category :text
-    end
+  [false, true].each do |sym_keys|
+    context "symbol index? #{sym_keys}" do
+      let(:index) do
+        index = Picky::Index.new :automatic_text_splitting do
+          symbol_keys sym_keys
+        
+          indexing removes_characters: /[^a-z\s]/i,
+                   stopwords: /\b(in|a)\b/
+          category :text
+        end
 
-    require 'ostruct'
-    index.add OpenStruct.new(id: 1, text: 'It does rain in Spain. Purple is a new color. Bow to the king.')
-    index.add OpenStruct.new(id: 2, text: 'Rainbow rainbow.')
-    index.add OpenStruct.new(id: 3, text: 'Bow and arrow in Papua New Guinea.')
-    index.add OpenStruct.new(id: 4, text: 'The color purple.')
-    index.add OpenStruct.new(id: 5, text: 'Sun and rain.')
-    index.add OpenStruct.new(id: 6, text: 'The king is in Spain.')
+        require 'ostruct'
+        index.add OpenStruct.new(id: 1, text: 'It does rain in Spain. Purple is a new color. Bow to the king.')
+        index.add OpenStruct.new(id: 2, text: 'Rainbow rainbow.')
+        index.add OpenStruct.new(id: 3, text: 'Bow and arrow in Papua New Guinea.')
+        index.add OpenStruct.new(id: 4, text: 'The color purple.')
+        index.add OpenStruct.new(id: 5, text: 'Sun and rain.')
+        index.add OpenStruct.new(id: 6, text: 'The king is in Spain.')
     
-    index
-  end
+        index
+      end
 
-  context 'splitting the text automatically' do
-    let(:automatic_splitter) { Picky::Splitters::Automatic.new index[:text] }
+      context 'splitting the text automatically' do
+        let(:automatic_splitter) { Picky::Splitters::Automatic.new index[:text] }
     
-    # It splits the text correctly.
-    #
-    it do
-      automatic_splitter.segment('purplerainbow').should == [
-        ['purple', 'rain', 'bow'],
-        2.078999999999999
-      ]
-    end
-  end
+        # It splits the text correctly.
+        #
+        it do
+          automatic_splitter.segment('purplerainbow').should == [
+            ['purple', 'rain', 'bow'],
+            2.078999999999999
+          ]
+        end
+      end
   
-  context 'splitting the text automatically' do
-    let(:automatic_splitter) { Picky::Splitters::Automatic.new index[:text] }
+      context 'splitting the text automatically' do
+        let(:automatic_splitter) { Picky::Splitters::Automatic.new index[:text] }
     
-    # It splits the text correctly.
-    #
-    it { automatic_splitter.split('purplerainbow').should == ['purple', 'rain', 'bow'] }
-    it { automatic_splitter.split('purplerain').should == ['purple', 'rain'] }
-    it { automatic_splitter.split('purple').should == ['purple'] }
+        # It splits the text correctly.
+        #
+        it { automatic_splitter.split('purplerainbow').should == ['purple', 'rain', 'bow'] }
+        it { automatic_splitter.split('purplerain').should == ['purple', 'rain'] }
+        it { automatic_splitter.split('purple').should == ['purple'] }
     
-    # When it can't, it splits it using the partial index (correctly).
-    #
-    it { automatic_splitter.split('purplerainbo').should == ['purple', 'rain'] }
-    it { automatic_splitter.split('purplerainb').should  == ['purple', 'rain'] }
-    #
-    it { automatic_splitter.split('purplerai').should == ['purple'] }
-    it { automatic_splitter.split('purplera').should  == ['purple'] }
-    it { automatic_splitter.split('purpler').should   == ['purple'] }
-    #
-    it { automatic_splitter.split('purpl').should == [] }
-    it { automatic_splitter.split('purp').should  == [] }
-    it { automatic_splitter.split('pur').should   == [] }
-    it { automatic_splitter.split('pu').should    == [] }
-    it { automatic_splitter.split('p').should     == [] }
-  end
+        # When it can't, it splits it using the partial index (correctly).
+        #
+        it { automatic_splitter.split('purplerainbo').should == ['purple', 'rain'] }
+        it { automatic_splitter.split('purplerainb').should  == ['purple', 'rain'] }
+        #
+        it { automatic_splitter.split('purplerai').should == ['purple'] }
+        it { automatic_splitter.split('purplera').should  == ['purple'] }
+        it { automatic_splitter.split('purpler').should   == ['purple'] }
+        #
+        it { automatic_splitter.split('purpl').should == [] }
+        it { automatic_splitter.split('purp').should  == [] }
+        it { automatic_splitter.split('pur').should   == [] }
+        it { automatic_splitter.split('pu').should    == [] }
+        it { automatic_splitter.split('p').should     == [] }
+      end
   
-  context 'splitting text automatically (with partial)' do
-    let(:automatic_splitter) { Picky::Splitters::Automatic.new index[:text], partial: true }
+      context 'splitting text automatically (with partial)' do
+        let(:automatic_splitter) { Picky::Splitters::Automatic.new index[:text], partial: true }
     
-    # It splits the text correctly.
-    #
-    it { automatic_splitter.split('purplerainbow').should == ['purple', 'rain', 'bow'] }
-    it { automatic_splitter.split('purplerain').should == ['purple', 'rain'] }
-    it { automatic_splitter.split('purple').should == ['purple'] }
+        # It splits the text correctly.
+        #
+        it { automatic_splitter.split('purplerainbow').should == ['purple', 'rain', 'bow'] }
+        it { automatic_splitter.split('purplerain').should == ['purple', 'rain'] }
+        it { automatic_splitter.split('purple').should == ['purple'] }
     
-    # Creates the right queries (see below).
-    #
-    it { automatic_splitter.split('colorpurple').should == ['color', 'purple'] }
-    it { automatic_splitter.split('bownew').should == ['bow', 'new'] }
-    it { automatic_splitter.split('spainisking').should == ['spain', 'is', 'king'] }
+        # Creates the right queries (see below).
+        #
+        it { automatic_splitter.split('colorpurple').should == ['color', 'purple'] }
+        it { automatic_splitter.split('bownew').should == ['bow', 'new'] }
+        it { automatic_splitter.split('spainisking').should == ['spain', 'is', 'king'] }
     
-    # When it can't, it splits it using the partial index (correctly).
-    #
-    it { automatic_splitter.split('purplerainbo').should == ['purple', 'rain', 'bo'] }
-    it { automatic_splitter.split('purplerainb').should == ['purple', 'rain', 'b'] }
-    #
-    it { automatic_splitter.split('purplerai').should == ['purple', 'rai'] }
-    it { automatic_splitter.split('purplera').should == ['purple', 'ra'] }
-    it { automatic_splitter.split('purpler').should == ['purple'] } # No 'r' in partial index.
-    #
-    it { automatic_splitter.split('purpl').should == ['purpl'] }
-    it { automatic_splitter.split('purp').should == ['purp'] }
-    it { automatic_splitter.split('pur').should == [] } # No 'pur' in partial index etc.
-    it { automatic_splitter.split('pu').should == [] }
-    it { automatic_splitter.split('p').should == [] }
+        # When it can't, it splits it using the partial index (correctly).
+        #
+        it { automatic_splitter.split('purplerainbo').should == ['purple', 'rain', 'bo'] }
+        it { automatic_splitter.split('purplerainb').should == ['purple', 'rain', 'b'] }
+        #
+        it { automatic_splitter.split('purplerai').should == ['purple', 'rai'] }
+        it { automatic_splitter.split('purplera').should == ['purple', 'ra'] }
+        it { automatic_splitter.split('purpler').should == ['purple'] } # No 'r' in partial index.
+        #
+        it { automatic_splitter.split('purpl').should == ['purpl'] }
+        it { automatic_splitter.split('purp').should == ['purp'] }
+        it { automatic_splitter.split('pur').should == [] } # No 'pur' in partial index etc.
+        it { automatic_splitter.split('pu').should == [] }
+        it { automatic_splitter.split('p').should == [] }
     
-    let(:try) do
-      splitter = automatic_splitter
-      Picky::Search.new index do
-        searching splits_text_on: splitter
-      end
-    end
+        let(:try) do
+          splitter = automatic_splitter
+          Picky::Search.new index do
+            searching splits_text_on: splitter
+          end
+        end
     
-    # Should find the one with all parts.
-    #
-    it { try.search('purplerainbow').ids.should == [1] }
-    it { try.search('sunandrain').ids.should == [5] }
+        # Should find the one with all parts.
+        #
+        it { try.search('purplerainbow').ids.should == [1] }
+        it { try.search('sunandrain').ids.should == [5] }
     
-    # Common parts are found in multiple examples.
-    #
-    it { try.search('colorpurple').ids.should == [4,1] }
-    it { try.search('bownew').ids.should      == [3,1] }
-    it { try.search('spainisking').ids.should == [6,1] }
-  end
+        # Common parts are found in multiple examples.
+        #
+        it { try.search('colorpurple').ids.should == [4,1] }
+        it { try.search('bownew').ids.should      == [3,1] }
+        it { try.search('spainisking').ids.should == [6,1] }
+      end
   
-  it 'is fast enough' do
-    automatic_splitter = Picky::Splitters::Automatic.new index[:text]
+      it 'is fast enough' do
+        automatic_splitter = Picky::Splitters::Automatic.new index[:text]
     
-    performance_of do
-      automatic_splitter.split('purplerainbow')
-    end.should < 0.0002
+        performance_of do
+          automatic_splitter.split('purplerainbow')
+        end.should < 0.0002
+      end
+    end
   end
 
 end

data/spec/functional/optimize_memory_spec.rb

@@ -75,7 +75,7 @@ describe "Memory optimization" do
     GC.start
     
     # Optimize saves some memory.
-    (memsize_with_optimized_memory + 2952).should == memsize_with_readded_thing
+    (memsize_with_optimized_memory + 2000).should <= memsize_with_readded_thing
   end
 
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: picky
 version: !ruby/object:Gem::Version
-  version: 4.31.0
+  version: 4.31.1
 platform: ruby
 authors:
 - Florian Hanke