picky 4.3.0 → 4.3.1

data/lib/picky/category_realtime.rb

@@ -1,6 +1,8 @@
 module Picky
 
   class Category
+    
+    class Picky::IdNotGivenException < StandardError; end
 
     # Removes an indexed object with the
     # given id.
@@ -31,11 +33,9 @@ module Picky
     # Note: Takes a hash as opposed to the add/replace method.
     #
     def replace_from hash
-      # TODO Decide on a format.
-      #
       return unless text = hash[from] || hash[from.to_s]
       
-      id = hash[:id] || hash['id']
+      raise IdNotGivenException.new unless id = hash[:id] || hash['id']
       id = id.send key_format
       
       remove id

data/lib/picky/sinatra/index_actions.rb

@@ -4,18 +4,44 @@ module Picky
     module IndexActions
       
       def self.extended base
+        # Updates the given item and returns HTTP codes:
+        #  * 200 if the index has been updated or no error case has occurred.
+        #  * 404 if the index cannot be found.
+        #  * 400 if no data or item id has been provided in the data.
+        #
+        # Note: 200 returns no data yet.
+        #
         base.post '/' do
           index_name = params['index']
-          index = Picky::Indexes[index_name.to_sym]
-          data = params['data']
-          index.replace_from Yajl::Parser.parse(data) if data
+          begin
+            index = Picky::Indexes[index_name.to_sym]
+            data = params['data']
+            return 400 unless data
+            data && index.replace_from(Yajl::Parser.parse data) && 200
+          rescue IdNotGivenException
+            400
+          rescue StandardError
+            404
+          end
         end
+        
+        # Deletes the given item and returns:
+        #  * 200 if the index has been updated or no error case has occurred.
+        #  * 404 if the index cannot be found.
+        #  * 400 if no data or item id has been provided in the data.
+        #
+        # Note: 200 returns no data yet.
+        #
         base.delete '/' do
           index_name = params['index']
-          index = Picky::Indexes[index_name.to_sym]
-          data = Yajl::Parser.parse params['data']
-          id = data['id']
-          index.remove id if id
+          begin
+            index = Picky::Indexes[index_name.to_sym]
+            data = Yajl::Parser.parse params['data']
+            id = data['id']
+            id ? index.remove(id) && 200 : 400
+          rescue StandardError
+            404
+          end
         end
       end
       

data/spec/category_realtime_spec.rb

@@ -29,5 +29,16 @@ describe Picky::Category, "Realtime API" do
   it 'offers an unshift method' do
     category.unshift Thing.new(1, 'text')
   end
+  it 'offers a replace_from method' do
+    category.replace_from id: 1, text: "some text"
+  end
+  it 'raises on no id given' do
+    expect {
+      category.replace_from text: "some text"
+    }.to raise_error
+  end
+  it 'shrugs off no data given' do
+    category.replace_from id: 1
+  end
 
 end

data/spec/functional/tokenizer_spec.rb

@@ -0,0 +1,72 @@
+# encoding: utf-8
+#
+require 'spec_helper'
+
+describe Picky::Tokenizer do
+  describe 'examples' do
+    it 'works correctly' do
+      tokenizer = described_class.new(split_words_on: /\&/, normalizes_words: [[/\&/, 'and']])
+      
+      # TODO Is this really correct? Shouldn't we split after normalizing? 
+      #
+      tokenizer.tokenize('M & M').should == [['m', 'and', 'm'], ['m', 'and', 'm']]
+    end
+    it 'works correctly' do
+      tokenizer = described_class.new(stopwords: /\b(and)\b/, normalizes_words: [[/\&/, 'and']])
+      
+      # TODO Is this really correct? Shouldn't we stop words after normalizing? 
+      #
+      tokenizer.tokenize('M & M').should == [['m', 'and', 'm'], ['m', 'and', 'm']]
+    end
+    it 'removes all stopwords if they do not occur alone' do
+      tokenizer = described_class.new(stopwords: /\b(and|then)\b/)
+      tokenizer.tokenize('and then').should == [[], []]
+    end
+    it 'does not remove a stopword if it occurs alone' do
+      tokenizer = described_class.new(stopwords: /\b(and|then)\b/)
+      tokenizer.tokenize('and').should == [['and'], ['and']]
+    end
+    it 'does not remove a stopword if it occurs alone (even with qualifier)' do
+      tokenizer = described_class.new(stopwords: /\b(and|then)\b/)
+      tokenizer.tokenize('title:and').should == [['title:and'], ['title:and']]
+    end
+    it 'does not remove a stopword if it occurs alone (even with qualifier)' do
+      tokenizer = described_class.new(stopwords: /\b(and|then)\b/)
+      tokenizer.tokenize('title:and').should == [['title:and'], ['title:and']]
+    end
+    it 'removes stopwords, then only lets through max words words' do
+      tokenizer = described_class.new(stopwords: /\b(and|then|to|the)\b/, max_words: 2)
+      tokenizer.tokenize('and then they went to the house').should == [['they', 'went'], ['they', 'went']]
+    end
+    it 'can take freaky splits_text_on options' do
+      tokenizer = described_class.new(splits_text_on: /([A-Z]?[a-z]+)/, case_sensitive: false) # Explicit case, is false by default.
+      tokenizer.tokenize('TOTALCamelCaseExample').should == [
+        ["total", "camel", "case", "example"],
+        ["total", "camel", "case", "example"]
+      ]
+    end
+    it 'substitutes and removes in the right order' do
+      tokenizer = described_class.new(
+        substitutes_characters_with: Picky::CharacterSubstituters::WestEuropean.new,
+        removes_characters: /e/
+      )
+      
+      # Ä -> Ae -> A
+      #
+      tokenizer.tokenize('Ä ä').should == [['a', 'a'], ['a', 'a']]
+    end
+    it 'removes characters, then only lets through an ok sized token' do
+      tokenizer = described_class.new(rejects_token_if: ->(token){ token.size >= 5 }, removes_characters: /e/)
+      tokenizer.tokenize('hullo').should == [[], []]
+      tokenizer.tokenize('hello').should == [['hllo'], ['hllo']]
+    end
+    it 'is case sensitive' do
+      tokenizer = described_class.new(case_sensitive: true)
+      tokenizer.tokenize('Kaspar codes').should == [['Kaspar', 'codes'], ['Kaspar', 'codes']]
+    end
+    it 'is case sensitive, also for removing characters' do
+      tokenizer = described_class.new(case_sensitive: true, removes_characters: /K/)
+      tokenizer.tokenize('Kaspar codes').should == [['aspar', 'codes'], ['aspar', 'codes']]
+    end
+  end
+end

data/spec/integration/sinatra_index_actions_spec.rb

@@ -1,61 +1,196 @@
 require 'yajl'
 require 'sinatra'
-require 'picky-client'
+require_relative '../../../client/lib/picky-client'
+require_relative '../../../client/lib/picky-client/spec'
 require 'spec_helper'
 
 describe 'Sinatra Index Actions' do
   
-  # This is the application that is tested.
-  #
-  class MyPickyServer < Sinatra::Base
-    extend Picky::Sinatra::IndexActions
+  before(:all) do
     
-    data = Picky::Index.new :index do
-      category :name
-      category :surname
-    end
+    # This is the application that is tested.
+    #
+    class MyIndexActionsPickyServer < Sinatra::Base
+      extend Picky::Sinatra::IndexActions
+    
+      data = Picky::Index.new :some_index do
+        category :name
+        category :surname
+      end
     
-    people = Picky::Search.new data
+      people = Picky::Search.new data
     
-    get '/people' do
-      results = people.search params[:query], params[:ids] || 20, params[:offset] || 0
-      results.to_json
+      get '/people' do
+        results = people.search params[:query], params[:ids] || 20, params[:offset] || 0
+        results.to_json
+      end
     end
+    
   end
   
   describe 'updating' do
     before(:each) do
       Picky::Indexes.clear
     end
-    let(:request) { ::Rack::MockRequest.new MyPickyServer }
-    it 'should update the index correctly' do
-      request.post('/', params: { index: 'index', data: %Q{{ "id":"1", "name":"Florian", "surname":"Hanke" }} })
+    let(:request) { ::Rack::MockRequest.new MyIndexActionsPickyServer }
+    context 'return values' do
+      describe 'update' do
+        it 'returns a correct code after updating without problems' do
+          result = request.post('/', params: {
+            index: 'some_index',
+            data: %Q{{ "id":"1", "name":"Florian", "surname":"Hanke" }}
+          })
+          result.status.should == 200
+        end
+        it 'returns a correct code after updating with just the id' do
+          result = request.post('/', params: {
+            index: 'some_index',
+            data: %Q{{ "id":"1" }}
+          })
+          result.status.should == 200
+        end
+        it 'returns a correct code after updating without id' do
+          result = request.post('/', params: {
+            index: 'some_index',
+            data: %Q{{ "name":"Florian", "surname":"Hanke" }}
+          })
+          result.status.should == 400
+        end
+        it 'returns a correct code after updating with the wrong index' do
+          result = request.post('/', params: {
+            index: 'some_wrong_index',
+            data: %Q{{ "id":"1", "name":"Florian", "surname":"Hanke" }}
+          })
+          result.status.should == 404
+        end
+      end
+      describe 'delete' do
+        before(:each) do
+          request.post('/', params: {
+            index: 'some_index',
+            data: %Q{{ "id":"1", "name":"Florian", "surname":"Hanke" }}
+          })
+        end
+        it 'returns a correct code after deleting without problems' do
+          result = request.delete('/', params: {
+            index: 'some_index',
+            data: %Q{{ "id":"1" }}
+          })
+          result.status.should == 200
+        end
+        it 'returns a correct code after deleting twice' do
+          result = request.delete('/', params: {
+            index: 'some_index',
+            data: %Q{{ "id":"1" }}
+          })
+          result = request.delete('/', params: {
+            index: 'some_index',
+            data: %Q{{ "id":"1" }}
+          })
+          result.status.should == 200
+        end
+        it 'returns a correct code after deleting without id' do
+          result = request.delete('/', params: {
+            index: 'some_index',
+            data: %Q{{}}
+          })
+          result.status.should == 400
+        end
+        it 'returns a correct code after deleting with the wrong index' do
+          result = request.delete('/', params: {
+            index: 'some_wrong_index',
+            data: %Q{{ "id":"1" }}
+          })
+          result.status.should == 404
+        end        
+      end
+    end
+    context '' do
+      it 'updates the index correctly' do
+        request.post('/', params: {
+          index: 'some_index',
+          data: %Q{{ "id":"1", "name":"Florian", "surname":"Hanke" }}
+        })
       
-      results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
-      results['total'].should == 1
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
+        results['total'].should == 1
       
-      request.post('/', params: { index: 'index', data: %Q{{ "id":"2", "name":"Florian", "surname":"Meier" }} })
+        request.post('/', params: {
+          index: 'some_index',
+          data: %Q{{ "id":"2", "name":"Florian", "surname":"Meier" }}
+        })
       
-      results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
-      results['total'].should == 2
-    end
-    it 'should delete entries from the index correctly' do
-      request.post('/', params: { index: 'index', data: %Q{{ "id":"1", "name":"Florian", "surname":"Hanke" }} })
-      request.post('/', params: { index: 'index', data: %Q{{ "id":"2", "name":"Florian", "surname":"Meier" }} })
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
+        results['total'].should == 2
+      end
+      it 'updates the index correctly' do
+        request.post('/', params: {
+          index: 'some_index',
+          data: %Q{{ "id":"1", "name":"Flarian", "surname":"Hanke" }}
+        })
       
-      results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
-      results['total'].should == 2
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'hanke' }).body
+        results['total'].should == 1
       
-      request.delete('/', params: { index: 'index', data: %Q{{ "id":"1" }} })
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
+        results['total'].should == 0
       
-      results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
-      results['total'].should == 1
-    end
-    it 'should have no problem with a superfluous delete' do
-      request.delete('/', params: { index: 'index', data: %Q{{ "id":"1" }} })
+        # Whoops, typo. Let's fix it.
+        #
+        request.post('/', params: {
+          index: 'some_index',
+          data: %Q{{ "id":"1", "name":"Florian", "surname":"Hanke" }}
+        })
+      
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'hanke' }).body
+        results['total'].should == 1
+      
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'flarian' }).body
+        results['total'].should == 0
+      
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
+        results['total'].should == 1
+      end
+      it 'deletes entries from the index correctly' do
+        request.post('/', params: {
+          index: 'some_index',
+          data: %Q{{ "id":"1", "name":"Florian", "surname":"Hanke" }}
+        })
+        request.post('/', params: {
+          index: 'some_index',
+          data: %Q{{ "id":"2", "name":"Florian", "surname":"Meier" }}
+        })
+      
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
+        results['total'].should == 2
+      
+        request.delete('/', params: {
+          index: 'some_index',
+          data: %Q{{ "id":"1" }}
+        })
+      
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
+        results['total'].should == 1
+      end
+      it 'has no problem with a superfluous delete' do
+        request.delete('/', params: {
+          index: 'some_index',
+          data: %Q{{ "id":"1" }}
+        })
+      
+        results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
+        results['total'].should == 0
+      end
+      it 'works with the (test) client' do
+        client = Picky::TestClient.new MyIndexActionsPickyServer, :path => '/people'
+      
+        request.post('/', params: {
+          index: 'some_index',
+          data: %Q{{ "id":"1", "name":"Florian", "surname":"Hanke" }}
+        })
       
-      results = Yajl::Parser.parse request.get('/people', params: { query: 'florian' }).body
-      results['total'].should == 0
+        client.search('florian').total.should == 1
+      end
     end
   end
   

data/spec/lib/tokenizer_spec.rb

@@ -96,7 +96,9 @@ ERROR
         before(:each) do
           tokenizer.normalizes_words([
             [/st\./, 'sankt'],
-            [/stras?s?e?/, 'str']
+            [/stras?s?e?/, 'str'],
+            [/\+/, 'plus'],
+            [/\&/, 'and']
           ])
         end
         it "has normalize_with_patterns" do
@@ -105,6 +107,12 @@ ERROR
         it "normalizes, but just the first one" do
           tokenizer.normalize_with_patterns('st. wegstrasse').should == 'sankt wegstrasse'
         end
+        it "works correctly" do
+          tokenizer.normalize_with_patterns('camera +').should == 'camera plus'
+        end
+        it "works correctly" do
+          tokenizer.normalize_with_patterns('alice & bob').should == 'alice and bob'
+        end
       end
     end
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: picky
 version: !ruby/object:Gem::Version
-  version: 4.3.0
+  version: 4.3.1
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-03-13 00:00:00.000000000 Z
+date: 2012-03-24 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec
-  requirement: &70126521728720 !ruby/object:Gem::Requirement
+  requirement: &70193563400740 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,21 +21,21 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *70126521728720
+  version_requirements: *70193563400740
 - !ruby/object:Gem::Dependency
   name: picky-client
-  requirement: &70126521728220 !ruby/object:Gem::Requirement
+  requirement: &70193563400240 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: 4.3.0
+        version: 4.3.1
   type: :development
   prerelease: false
-  version_requirements: *70126521728220
+  version_requirements: *70193563400240
 - !ruby/object:Gem::Dependency
   name: text
-  requirement: &70126521727800 !ruby/object:Gem::Requirement
+  requirement: &70193563399820 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -43,10 +43,10 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *70126521727800
+  version_requirements: *70193563399820
 - !ruby/object:Gem::Dependency
   name: yajl-ruby
-  requirement: &70126521727340 !ruby/object:Gem::Requirement
+  requirement: &70193563399360 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -54,10 +54,10 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *70126521727340
+  version_requirements: *70193563399360
 - !ruby/object:Gem::Dependency
   name: activesupport
-  requirement: &70126521726800 !ruby/object:Gem::Requirement
+  requirement: &70193563415200 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -65,10 +65,10 @@ dependencies:
         version: '3.0'
   type: :runtime
   prerelease: false
-  version_requirements: *70126521726800
+  version_requirements: *70193563415200
 - !ruby/object:Gem::Dependency
   name: procrastinate
-  requirement: &70126521726280 !ruby/object:Gem::Requirement
+  requirement: &70193563414580 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -76,10 +76,10 @@ dependencies:
         version: '0.4'
   type: :runtime
   prerelease: false
-  version_requirements: *70126521726280
+  version_requirements: *70193563414580
 - !ruby/object:Gem::Dependency
   name: rack_fast_escape
-  requirement: &70126521725880 !ruby/object:Gem::Requirement
+  requirement: &70193563414120 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -87,7 +87,7 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *70126521725880
+  version_requirements: *70193563414120
 description: Fast Ruby semantic text search engine with comfortable single field interface.
 email: florian.hanke+picky@gmail.com
 executables:
@@ -259,6 +259,7 @@ files:
 - spec/functional/regression_spec.rb
 - spec/functional/speed_spec.rb
 - spec/functional/terminate_early_spec.rb
+- spec/functional/tokenizer_spec.rb
 - spec/integration/sinatra_index_actions_spec.rb
 - spec/lib/analytics_spec.rb
 - spec/lib/analyzer_spec.rb
@@ -408,6 +409,7 @@ test_files:
 - spec/functional/regression_spec.rb
 - spec/functional/speed_spec.rb
 - spec/functional/terminate_early_spec.rb
+- spec/functional/tokenizer_spec.rb
 - spec/integration/sinatra_index_actions_spec.rb
 - spec/lib/analytics_spec.rb
 - spec/lib/analyzer_spec.rb