lucid_works 0.3.9 → 0.4.9

data/lib/lucid_works/collection/settings.rb

@@ -1,3 +1,5 @@
+require 'lucid_works/field'
+
 module LucidWorks
   class Collection
     
@@ -5,18 +7,36 @@ module LucidWorks
       self.singleton = true
       belongs_to :collection
 
-      DEDUP_OPTIONS = %w{ off overwrite tag }
+      DEDUP_OPTIONS     = %w{ off overwrite tag }
+      QUERY_PARSERS     = %w{ lucid dismax edismax lucene }
+      DEFAULT_SORTS     = %w{ relevance date random }
+      FEEDBACK_EMPHASIS = %w{ relevancy recall }
 
       schema do
-        attributes :unsupervised_feedback_emphasis, :unknown_type_handling,
-                   :click_boost_field, :click_boost_data, :query_parser, :default_sort,
-                   :type => :string
+        # Indexing Settings
+        attribute  :unknown_type_handling, :string, :values => LucidWorks::Field::TYPES
         attribute  :de_duplication, :string, :values => DEDUP_OPTIONS
-        attributes :spellcheck, :display_facets, :ssl, :unsupervised_feedback, :query_time_stopwords,
-                   :auto_complete, :boost_recent, :click_enabled, :show_similar, :query_time_synonyms,
-                   :index_time_stopwords, :type => :boolean
-        attributes :search_server_list, :update_server_list, :stopword_list, :boosts, :synonym_list # Arrays
+        attribute  :index_time_stopwords, :boolean
+
+        # Querying Settings
+        attribute :unsupervised_feedback_emphasis, :string, :values => FEEDBACK_EMPHASIS
+        attribute :default_sort, :string, :values => DEFAULT_SORTS
+        attribute :query_parser, :string, :values => QUERY_PARSERS
+        attributes :spellcheck, :display_facets, :unsupervised_feedback, :query_time_stopwords,
+                   :auto_complete, :boost_recent, :show_similar, :query_time_synonyms,
+                   :type => :boolean
+        attributes :stopword_list, :boosts, :synonym_list # Arrays
+
+        # Click Settings
+        attributes :click_boost_field, :click_boost_data, :type => :string
+        attribute :click_enabled, :boolean
+
+        # Other Settings
+        attribute :ssl, :boolean
         attribute :elevations # Hash
+
+        # Distrubuted Search Settings
+        attributes :search_server_list, :update_server_list # Arrays
       end
     end
   end

data/lib/lucid_works/crawler.rb

@@ -1,8 +1,8 @@
 module LucidWorks
 
   class Crawler < Base
-
-    self.primary_key = :name
-
+    schema do
+      primary_key :name
+    end
   end
 end

data/lib/lucid_works/datasource.rb

@@ -4,12 +4,12 @@ module LucidWorks
     belongs_to :collection
     has_many   :histories, :class_name => :history
     has_one :status, :schedule, :crawldata
-    has_one :index, :has_content => false
+    has_one :index, :job, :has_content => false
 
     schema do
       # common
       attributes :name, :type, :crawler
-      attributes :crawl_depth, :max_bytes, :type => :integer
+      attributes :crawl_depth, :max_bytes, :max_docs, :type => :integer
       attribute :include_paths
       attribute :exclude_paths
       attribute :mapping # Hash
@@ -20,6 +20,9 @@ module LucidWorks
       # file
       attribute :path
       attribute :follow_links, :boolean
+
+      # new un-evaluated
+      # attributes :commitWithin
     end
 
     validates_presence_of :type, :crawler, :name, :crawl_depth
@@ -28,7 +31,7 @@ module LucidWorks
 
     before_save :remove_blank_max_bytes
 
-    TYPES = %w{ file web solrxml jdbc sharepoint }
+    TYPES = %w{ external file lwelogs web solrxml jdbc sharepoint }
     BOUNDS = %w{ tree none }
     CRAWLERS = {
       # Later we may change these to be arrays if we decide to support more than one choice
@@ -45,6 +48,29 @@ module LucidWorks
       index.destroy
     end
 
+    def editable?
+      # lwelogs is not editable
+      type != 'lwelogs'
+    end
+
+    def destroyable?
+      # Don't let user destroy 'lwelogs'
+      type != 'lwelogs'
+    end
+
+    def crawlable?
+      # Don't let user schedule crawl of 'lwelogs'
+      !%w{ external lwelogs }.include?(type)
+    end
+
+    def start_crawl!
+      job.save
+    end
+
+    def stop_crawl!
+      job.destroy
+    end
+    
     def t_type
       I18n.t(type, :scope => 'activemodel.models.lucid_works.datasource.type')
     end

data/lib/lucid_works/datasource/job.rb

@@ -0,0 +1,9 @@
+module LucidWorks
+  class Datasource
+
+    class Job < Base
+      self.singleton = true
+      belongs_to :datasource
+    end
+  end
+end

data/lib/lucid_works/datasource/status.rb

@@ -5,15 +5,16 @@ module LucidWorks
       self.singleton = true
       belongs_to :datasource
 
-      schema do
-        attributes :crawlStarted, :crawlState, :crawlStopped, :jobId
-        attributes :numUnchanged, :numUpdated, :numNew, :numFailed, :numDeleted, :type => :integer
-      end
-
       STOPPED_STATES =         %w{ IDLE STOPPED ABORTED EXCEPTION FINISHED }
       POST_PROCESSING_STATES = %w{ STOPPING ABORTING }
       CRAWLSTATES =            STOPPED_STATES + [ 'RUNNING' ] + POST_PROCESSING_STATES
 
+      schema do
+        attribute :crawlState, :string, :values => CRAWLSTATES
+        attributes :crawlStarted, :crawlStopped, :jobId
+        attributes :numUnchanged, :numUpdated, :numNew, :numFailed, :numDeleted, :numTotal, :type => :integer
+      end
+
       # Create predicate methods for all the crawl states
       CRAWLSTATES.each do |state|
         method_name = state.downcase + "?"
@@ -36,12 +37,6 @@ module LucidWorks
         numUpdated + numNew + numUnchanged
       end
 
-      def t_crawl_state
-        I18n.translate(crawlState.downcase,
-                       :scope => 'activemodel.models.lucid_works.datasource.status.crawl_state',
-                       :default => crawlState)
-      end
-
       def crawl_started
         Time.iso8601 crawlStarted
       end

data/lib/lucid_works/patch_time.rb

@@ -0,0 +1,13 @@
+require 'time'
+
+class Time
+  class << self
+    # Ruby's 'Time.is08601 can't handle timezone offsets expressed as +nnnn.
+    # It prefers +nn:nn.  Morph the former to the latter.
+    def iso8601_with_support_for_timezones_without_colons(datetime_string)
+        iso8601_without_support_for_timezones_without_colons(datetime_string.gsub /([+-])(\d\d)(\d\d)$/, '\1\2:\3')
+    end
+
+    alias_method_chain :iso8601, :support_for_timezones_without_colons
+  end
+end

data/lib/lucid_works/schema.rb

@@ -1,10 +1,30 @@
 module LucidWorks
 
+  # Specify an attributes for a model.
   class Schema < ActiveSupport::HashWithIndifferentAccess
 
-    # Specify an attribute for this model.
+    def initialize
+      @primary_key = :id
+      @dynamic_attributes = true
+    end
+
+    def primary_key(key=nil)
+      @primary_key = key.to_sym if key
+      @primary_key
+    end
+
+    # Specify whether attributes unrecognized during retrieval should raise an error (false)
+    # or cause a new attribute to be added to the schema (true) [default].
+    #
+    def dynamic_attributes(on=nil)
+      @dynamic_attributes = !!on unless on.nil?
+      @dynamic_attributes
+    end
+    
+    alias :dynamic_attributes? :dynamic_attributes
     
     def attribute(name, type=:string, options={})
+      primary_key name if options.delete(:primary_key)
       self[name] = options.merge(:type => type)
     end
 
@@ -13,7 +33,7 @@ module LucidWorks
     #   schema do
     #     attributes :first_name, :last_name, :type => :string
     #   end
-
+    #
     def attributes(*attributes_and_options)
       options = attributes_and_options.last.is_a?(Hash) ? attributes_and_options.pop : {}
       attributes = attributes_and_options
@@ -26,6 +46,10 @@ module LucidWorks
       has_key? sanitize_identifier(name)
     end
 
+    def attrs_to_omit_during_update
+      select { |k,v| v['omit_during_update'] == true }.keys
+    end
+
     # Used for classes that have no predefined schema.
     # When the class is retrieved.
     def add_attribute(klass, name, type=:string) # :nodoc:
@@ -47,10 +71,6 @@ module LucidWorks
         def #{attribute}                            # def foo
           @attributes[:#{attribute}]                #   @attributes[:foo]
         end                                         # end
-
-        def #{attribute}=(new_value)                # def foo=(new_value)
-          @attributes[:#{attribute}] = new_value    #   @attributes[:foo] = new_value
-        end                                         # end
       EOF
       
       if self[attribute][:type] == :boolean
@@ -58,12 +78,20 @@ module LucidWorks
           def #{attribute}?                           # def foo?
             @attributes[:#{attribute}]                #   @attributes[:foo]
           end                                         # end
+
+          def #{attribute}=(new_value)                         # def foo=(new_value)
+            @attributes[:#{attribute}] = to_bool(new_value)    #   @attributes[:foo] = to_bool(new_value)
+          end                                                  # end
+        EOF
+      else
+        klass.class_eval <<-EOF, __FILE__, __LINE__+1
+          def #{attribute}=(new_value)                # def foo=(new_value)
+            @attributes[:#{attribute}] = new_value    #   @attributes[:foo] = new_value
+          end                                         # end
         EOF
       end
     end
 
-    private
-
      # Change any characters illegal for an identifier to _
     def sanitize_identifier(identifier) # :nodoc:
       identifier.to_s.gsub(/[^\w]/, '_').to_sym

data/lib/lucid_works/utils.rb

@@ -0,0 +1,22 @@
+module LucidWorks
+  module Utils
+    module BoolConverter
+      TRUE_VALUES  = %w(t true  1 yes)
+      FALSE_VALUES = %w(f false 0 no)
+
+      def to_bool(thing)
+        if thing.kind_of?(TrueClass) || thing.kind_of?(FalseClass)
+          thing
+        elsif thing.kind_of?(String)
+          raise "Sorry, I can't to_bool \"#{thing}\"" unless (TRUE_VALUES + FALSE_VALUES).include?(thing.downcase)
+          TRUE_VALUES.include?(thing.downcase)
+        elsif thing.kind_of?(Integer)
+          raise "Sorry, I can't to_bool #{thing}" unless [0, 1].include?(thing)
+          thing == 1
+        else
+          raise "Sorry, I can't to_bool things of class #{thing.class.name}"
+        end
+      end
+    end
+  end
+end

data/lib/lucid_works/version.rb

@@ -1,3 +1,3 @@
 module LucidWorks
-  VERSION = "0.3.9"
+  VERSION = "0.4.9"
 end

data/lucid_works.gemspec

@@ -23,4 +23,6 @@ Gem::Specification.new do |s|
   s.add_runtime_dependency 'activemodel',   '>= 3'
   s.add_runtime_dependency 'rest-client',   '>= 1.6.1'
   s.add_runtime_dependency 'json'
+  s.add_runtime_dependency 'rsolr'
+  s.add_runtime_dependency 'nokogiri'
 end

data/spec/lib/lucid_works/associations_spec.rb

@@ -30,9 +30,20 @@ describe LucidWorks::Associations do
   describe ".has_one" do
     context "without content" do
       describe "#<child>" do
+        it "should call child! the first time then return the cached value thereafter" do
+          mock_launch_party = double('launch_party')
+          LaunchParty.should_receive(:new).once.and_return(mock_launch_party)
+
+          @blog.launch_party.should == mock_launch_party
+          @blog.launch_party.should == mock_launch_party
+        end
+      end
+
+      describe "#<child!>" do
         it "should build a new model, and not call REST API to retrieve" do
           LaunchParty.should_not_receive(:find)
-          launch_party = @blog.launch_party
+          
+          launch_party = @blog.launch_party!
 
           launch_party.should be_a(LaunchParty)
           launch_party.should be_persisted # All singletons are always persisted

data/spec/lib/lucid_works/base_spec.rb

@@ -38,10 +38,6 @@ describe LucidWorks::Base do
 
     describe ".schema" do
       context "for a class with a schema" do
-        it "should set has_schema" do
-          WidgetWithSchema.has_schema.should be_true
-        end
-
         it "should have accessors for the attributes defined in the schema" do
           [:name, :name=, :frooble, :frooble=].each do |method|
             WidgetWithSchema.new(:parent => @server).should respond_to(method)
@@ -66,10 +62,7 @@ describe LucidWorks::Base do
         end
       end
 
-      context "for a class without a schema" do
-        it "should return false for has_schema" do
-          WidgetWithoutSchema.has_schema.should be_false
-        end
+      context "for a class with default schema (i.e. dynamic_attributes == true)" do
 
         it "should create accessors upon retrieval" do
           widget = WidgetWithoutSchema.new(:parent => @server)
@@ -354,7 +347,7 @@ describe LucidWorks::Base do
         it "should create an array of arrays suitable for a form select" do
           # Use Collection::Settings because we know de_duplication as translations
           LucidWorks::Collection::Settings.to_select(:de_duplication).should ==
-            [['off', 'Off'], ['overwrite', 'Overwrite'], ['tag', 'Tag']]
+            [['Off', 'off'], ['Overwrite', 'overwrite'], ['Tag', 'tag']]
         end
       end
     end
@@ -472,6 +465,27 @@ describe LucidWorks::Base do
           @widget.save
         end
       end
+
+      context "for a persisted instance of a model with an attribute marked :omit_during_update" do
+        before do
+          class WidgetWithOmit < LucidWorks::Base
+            schema do
+              attribute :foo, :string
+              attribute :bar, :string, :omit_during_update => true
+            end
+          end
+          @widget_attrs = { :foo => 'frooble', :bar => 'barfalicious' }
+          @widget_with_omit = WidgetWithOmit.new(@widget_attrs.merge :id => 1234, :parent => @server, :persisted => true)
+        end
+
+        it "should omit said attribute when performing the save" do
+          expected_json = @widget_attrs.reject { |k,v| k.to_sym == :bar }.to_json
+          correct_url = "#{@fake_server_uri}/widget_with_omits/1234"
+          RestClient.should_receive(:put).with(correct_url, expected_json, :content_type => :json)
+
+          @widget_with_omit.save
+        end
+      end
     end
 
     describe "#update_attributes" do
@@ -507,7 +521,9 @@ describe LucidWorks::Base do
   describe "for a model with primary key other than 'id'" do
     before :all do
       class NamedWidget < LucidWorks::Base
-        self.primary_key = :name
+        schema do
+          primary_key :name
+        end
       end
       NAMED_WIDGET1_JSON = '{"name":"widget1","size":"small"}'
       NAMED_WIDGET2_JSON = '{"name":"widget2","size":"medium"}'

data/spec/lib/lucid_works/collection/activity/history_spec.rb

@@ -0,0 +1,33 @@
+require 'spec_helper'
+require 'active_support/core_ext'
+
+describe LucidWorks::Collection::Activity::History do
+  before :all do
+    @server = connect_to_live_server
+    @server.reset_collections!
+    @collection = @server.collections!.first
+    @activity = @collection.build_activity(:type => 'optimize', :start_time => 0, :active => true)
+  end
+
+  context "after running" do
+    before :all do
+      @activity.save.should be_true
+      sleep 1
+    end
+
+    it "should have a history array with one element" do
+      histories = @activity.histories
+      histories.count.should == 1
+    end
+
+    it "should subtract crawlStarted form crawlStopped and return the difference in seconds" do
+      include ActiveSupport::Duration
+
+      history = LucidWorks::Collection::Activity::History.new(:parent => @activity)
+      now = Time.now
+      history.stub(:activity_started) { now.advance(:seconds => -15) }
+      history.stub(:activity_finished) { now.advance(:seconds => -10) }
+      history.duration.should == 5.seconds
+    end
+  end
+end

data/spec/lib/lucid_works/collection/activity/status_spec.rb

@@ -0,0 +1,20 @@
+require 'spec_helper'
+
+describe LucidWorks::Collection::Activity::Status do
+  before :all do
+    @server = connect_to_live_server
+    @server.reset_collections!
+    @collection = @server.collections!.first
+    @activity = @collection.create_activity(:type => 'optimize', :start_time => 8600)
+  end
+
+  describe "find" do
+    it "should retrieve status" do
+      status = @activity.status
+      status.should be_a LucidWorks::Collection::Activity::Status
+
+      status.running.should be_false
+      status.type.should == 'optimize' 
+    end
+  end
+end