muck-solr 0.4.0

data/lib/acts_as_solr/lazy_document.rb

@@ -0,0 +1,18 @@
+module ActsAsSolr
+  class LazyDocument
+    attr_reader :id, :clazz
+    
+    def initialize(id, clazz)
+      @id = id
+      @clazz = clazz
+    end
+  
+    def method_missing(name, *args)
+      unless @__instance
+        @__instance = @clazz.find(@id)
+      end
+      
+      @__instance.send(name, *args)
+    end
+  end
+end

data/lib/acts_as_solr/parser_methods.rb

@@ -0,0 +1,203 @@
+module ActsAsSolr #:nodoc:
+  module ParserMethods
+    protected    
+    
+    # Method used by mostly all the ClassMethods when doing a search
+    def parse_query(query=nil, options={}, models=nil)
+      valid_options = [:offset, :limit, :facets, :models, :results_format, :order, :scores, :operator, :include, :lazy, :joins, :select, :core]
+      query_options = {}
+
+      return nil if (query.nil? || query.strip == '')
+
+      raise "Invalid parameters: #{(options.keys - valid_options).join(',')}" unless (options.keys - valid_options).empty?
+      begin
+        Deprecation.validate_query(options)
+        query_options[:start] = options[:offset]
+        query_options[:rows] = options[:limit]
+        query_options[:operator] = options[:operator]
+        
+        # first steps on the facet parameter processing
+        if options[:facets]
+          query_options[:facets] = {}
+          query_options[:facets][:limit] = -1  # TODO: make this configurable
+          query_options[:facets][:sort] = :count if options[:facets][:sort]
+          query_options[:facets][:mincount] = 0
+          query_options[:facets][:mincount] = 1 if options[:facets][:zeros] == false
+          # override the :zeros (it's deprecated anyway) if :mincount exists
+          query_options[:facets][:mincount] = options[:facets][:mincount] if options[:facets][:mincount]
+          query_options[:facets][:fields] = options[:facets][:fields].collect{|k| "#{k}_facet"} if options[:facets][:fields]
+          query_options[:filter_queries] = replace_types([*options[:facets][:browse]].collect{|k| "#{k.sub!(/ *: */,"_facet:")}"}) if options[:facets][:browse]
+          query_options[:facets][:queries] = replace_types(options[:facets][:query].collect{|k| "#{k.sub!(/ *: */,"_t:")}"}) if options[:facets][:query]
+          
+          
+          if options[:facets][:dates]
+            query_options[:date_facets] = {}
+            # if options[:facets][:dates][:fields] exists then :start, :end, and :gap must be there
+            if options[:facets][:dates][:fields]
+              [:start, :end, :gap].each { |k| raise "#{k} must be present in faceted date query" unless options[:facets][:dates].include?(k) }
+              query_options[:date_facets][:fields] = []
+              options[:facets][:dates][:fields].each { |f|
+                if f.kind_of? Hash
+                  key = f.keys[0]
+                  query_options[:date_facets][:fields] << {"#{key}_d" => f[key]}
+                  validate_date_facet_other_options(f[key][:other]) if f[key][:other]
+                else
+                  query_options[:date_facets][:fields] << "#{f}_d"
+                end
+              }
+            end
+            
+            query_options[:date_facets][:start]   = options[:facets][:dates][:start] if options[:facets][:dates][:start]
+            query_options[:date_facets][:end]     = options[:facets][:dates][:end] if options[:facets][:dates][:end]
+            query_options[:date_facets][:gap]     = options[:facets][:dates][:gap] if options[:facets][:dates][:gap]
+            query_options[:date_facets][:hardend] = options[:facets][:dates][:hardend] if options[:facets][:dates][:hardend]
+            query_options[:date_facets][:filter]  = replace_types([*options[:facets][:dates][:filter]].collect{|k| "#{k.sub!(/ *:(?!\d) */,"_d:")}"}) if options[:facets][:dates][:filter]
+
+            if options[:facets][:dates][:other]
+              validate_date_facet_other_options(options[:facets][:dates][:other])
+              query_options[:date_facets][:other]   = options[:facets][:dates][:other]
+            end
+            
+          end          
+        end
+        
+        if models.nil?
+          # TODO: use a filter query for type, allowing Solr to cache it individually
+          models = "AND #{solr_type_condition}"
+          field_list = solr_configuration[:primary_key_field]
+        else
+          field_list = "id"
+        end
+        
+        query_options[:field_list] = [field_list, 'score']
+        query = "(#{query.gsub(/ *: */,"_t:")}) #{models}"
+        order = options[:order].split(/\s*,\s*/).collect{|e| e.gsub(/\s+/,'_t ').gsub(/\bscore_t\b/, 'score')  }.join(',') if options[:order] 
+        query_options[:query] = replace_types([query])[0] # TODO adjust replace_types to work with String or Array  
+
+        if options[:order]
+          # TODO: set the sort parameter instead of the old ;order. style.
+          query_options[:query] << ';' << replace_types([order], false)[0]
+        end
+        
+        ActsAsSolr::Post.execute(Solr::Request::Standard.new(query_options), options[:core])
+      rescue
+        raise "There was a problem executing your search\n#{query_options.inspect}\n: #{$!} in #{$!.backtrace.first}"
+      end            
+    end
+    
+    def solr_type_condition
+      subclasses.inject("(#{solr_configuration[:type_field]}:#{self.name}") do |condition, subclass|
+        condition << " OR #{solr_configuration[:type_field]}:#{subclass.name}"
+      end << ')'
+    end
+    
+    # Parses the data returned from Solr
+    def parse_results(solr_data, options = {})
+      results = {
+        :docs => [],
+        :total => 0
+      }
+      
+      configuration = {
+        :format => :objects
+      }
+      results.update(:facets => {'facet_fields' => []}) if options[:facets]
+      return SearchResults.new(results) if (solr_data.nil? || solr_data.total_hits == 0)
+      
+      configuration.update(options) if options.is_a?(Hash)
+
+      ids = solr_data.hits.collect {|doc| doc["#{solr_configuration[:primary_key_field]}"]}.flatten
+      
+      result = find_objects(ids, options, configuration)
+      
+      add_scores(result, solr_data) if configuration[:format] == :objects && options[:scores]
+      
+      results.update(:facets => solr_data.data['facet_counts']) if options[:facets]
+      results.update({:docs => result, :total => solr_data.total_hits, :max_score => solr_data.max_score, :query_time => solr_data.data['responseHeader']['QTime']})
+      SearchResults.new(results)
+    end
+    
+    
+    def find_objects(ids, options, configuration)
+      result = if configuration[:lazy] && configuration[:format] != :ids
+        ids.collect {|id| ActsAsSolr::LazyDocument.new(id, self)}
+      elsif configuration[:format] == :objects
+        conditions = [ "#{self.table_name}.#{primary_key} in (?)", ids ]
+        find_options = {:conditions => conditions}
+        find_options[:include] = options[:include] if options[:include]
+        find_options[:select] = options[:select] if options[:select]
+        find_options[:joins] = options[:joins] if options[:joins]
+        result = reorder(self.find(:all, find_options), ids)
+      else
+        ids
+      end
+        
+      result
+    end
+    
+    # Reorders the instances keeping the order returned from Solr
+    def reorder(things, ids)
+      ordered_things = Array.new(things.size)
+      raise "Out of sync! Found #{ids.size} items in index, but only #{things.size} were found in database!" unless things.size == ids.size
+      things.each do |thing|
+        position = ids.index(thing.id)
+        ordered_things[position] = thing
+      end
+      ordered_things
+    end
+
+    # Replaces the field types based on the types (if any) specified
+    # on the acts_as_solr call
+    def replace_types(strings, include_colon=true)
+      suffix = include_colon ? ":" : ""
+      if configuration[:solr_fields]
+        configuration[:solr_fields].each do |name, options|
+          solr_name = options[:as] || name.to_s
+          solr_type = get_solr_field_type(options[:type])
+          field = "#{solr_name}_#{solr_type}#{suffix}"
+          strings.each_with_index {|s,i| strings[i] = s.gsub(/#{solr_name.to_s}_t#{suffix}/,field) }
+        end
+      end
+      if configuration[:solr_includes]
+        configuration[:solr_includes].each do |association, options|
+          solr_name = options[:as] || association.to_s.singularize
+          solr_type = get_solr_field_type(options[:type])
+          field = "#{solr_name}_#{solr_type}#{suffix}"
+          strings.each_with_index {|s,i| strings[i] = s.gsub(/#{solr_name.to_s}_t#{suffix}/,field) }
+        end
+      end
+      strings
+    end
+    
+    # Adds the score to each one of the instances found
+    def add_scores(results, solr_data)
+      with_score = []
+      solr_data.hits.each do |doc|
+        with_score.push([doc["score"], 
+          results.find {|record| scorable_record?(record, doc) }])
+      end
+      with_score.each do |score, object| 
+        class << object; attr_accessor :solr_score; end
+        object.solr_score = score
+      end
+    end
+    
+    def scorable_record?(record, doc)
+      doc_id = doc["#{solr_configuration[:primary_key_field]}"]
+      if doc_id.nil?
+        doc_id = doc["id"]
+        "#{record.class.name}:#{record_id(record)}" == doc_id.first.to_s
+      else
+        record_id(record).to_s == doc_id.to_s
+      end
+    end
+    
+    def validate_date_facet_other_options(options)
+      valid_other_options = [:after, :all, :before, :between, :none]
+      options = [options] unless options.kind_of? Array
+      bad_options = options.map {|x| x.to_sym} - valid_other_options
+      raise "Invalid option#{'s' if bad_options.size > 1} for faceted date's other param: #{bad_options.join(', ')}. May only be one of :after, :all, :before, :between, :none" if bad_options.size > 0
+    end
+    
+  end
+end

data/lib/acts_as_solr/search_results.rb

@@ -0,0 +1,68 @@
+module ActsAsSolr #:nodoc:
+  
+  # TODO: Possibly looking into hooking it up with Solr::Response::Standard
+  # 
+  # Class that returns the search results with four methods.
+  # 
+  #   books = Book.find_by_solr 'ruby'
+  # 
+  # the above will return a SearchResults class with 4 methods:
+  # 
+  # docs|results|records: will return an array of records found
+  # 
+  #   books.records.empty?
+  #   => false
+  # 
+  # total|num_found|total_hits: will return the total number of records found
+  # 
+  #   books.total
+  #   => 2
+  # 
+  # facets: will return the facets when doing a faceted search
+  # 
+  # max_score|highest_score: returns the highest score found
+  # 
+  #   books.max_score
+  #   => 1.3213213
+  # 
+  # 
+  class SearchResults
+    def initialize(solr_data={})
+      @solr_data = solr_data
+    end
+    
+    # Returns an array with the instances. This method
+    # is also aliased as docs and records
+    def results
+      @solr_data[:docs]
+    end
+    
+    # Returns the total records found. This method is
+    # also aliased as num_found and total_hits
+    def total
+      @solr_data[:total]
+    end
+    
+    # Returns the facets when doing a faceted search
+    def facets
+      @solr_data[:facets]
+    end
+    
+    # Returns the highest score found. This method is
+    # also aliased as highest_score
+    def max_score
+      @solr_data[:max_score]
+    end
+    
+    def query_time
+      @solr_data[:query_time]
+    end
+    
+    alias docs results
+    alias records results
+    alias num_found total
+    alias total_hits total
+    alias highest_score max_score
+  end
+  
+end

data/lib/acts_as_solr/solr_fixtures.rb

@@ -0,0 +1,13 @@
+module ActsAsSolr
+
+  class SolrFixtures
+    def self.load(table_names)
+      [table_names].flatten.map { |n| n.to_s }.each do |table_name|
+        klass = instance_eval(File.split(table_name.to_s).last.to_s.gsub('_',' ').split(" ").collect{|w| w.capitalize}.to_s.singularize)
+        klass.rebuild_solr_index if klass.respond_to?(:rebuild_solr_index)
+      end
+      ActsAsSolr::Post.execute(Solr::Request::Commit.new)
+    end
+  end
+  
+end

data/lib/acts_as_solr/tasks.rb

@@ -0,0 +1,10 @@
+dir = File.dirname(__FILE__)
+require 'rubygems'
+require 'rake'
+require 'net/http'
+require 'active_record'
+require File.expand_path("#{dir}/solr_fixtures")
+
+load File.expand_path("#{dir}/tasks/database.rake")
+load File.expand_path("#{dir}/tasks/solr.rake")
+load File.expand_path("#{dir}/tasks/test.rake")

data/lib/acts_as_solr/tasks/database.rake

@@ -0,0 +1,16 @@
+namespace :db do
+  namespace :fixtures do
+    desc "Load fixtures into the current environment's database. Load specific fixtures using FIXTURES=x,y"
+    task :load => :environment do
+      begin
+        ActsAsSolr::Post.execute(Solr::Request::Delete.new(:query => "*:*"))
+        ActsAsSolr::Post.execute(Solr::Request::Commit.new)
+        (ENV['FIXTURES'] ? ENV['FIXTURES'].split(/,/) : Dir.glob(File.join(RAILS_ROOT, 'test', 'fixtures', '*.{yml,csv}'))).each do |fixture_file|    
+          ActsAsSolr::SolrFixtures.load(File.basename(fixture_file, '.*'))
+        end 
+        puts "The fixtures loaded have been added to Solr"       
+      rescue
+      end
+    end
+  end
+end

data/lib/acts_as_solr/tasks/solr.rake

@@ -0,0 +1,135 @@
+namespace :solr do
+
+  desc 'Starts Solr. Options accepted: RAILS_ENV=your_env, PORT=XX. Defaults to development if none.'
+  task :start do
+    require File.expand_path("#{File.dirname(__FILE__)}/../../../config/solr_environment")
+    FileUtils.mkdir_p(SOLR_LOGS_PATH)
+    FileUtils.mkdir_p(SOLR_DATA_PATH)
+    FileUtils.mkdir_p(SOLR_PIDS_PATH)
+    begin
+      n = Net::HTTP.new('127.0.0.1', SOLR_PORT)
+      n.request_head('/').value 
+
+    rescue Net::HTTPServerException #responding
+      puts "Port #{SOLR_PORT} in use" and return
+
+    rescue Errno::ECONNREFUSED #not responding
+      Dir.chdir(SOLR_PATH) do
+        pid = fork do
+          #STDERR.close
+          exec "java #{SOLR_JVM_OPTIONS} -Dsolr.data.dir=#{SOLR_DATA_PATH} -Djetty.logs=#{SOLR_LOGS_PATH} -Djetty.port=#{SOLR_PORT} -jar start.jar"
+        end
+        sleep(5)
+        File.open("#{SOLR_PIDS_PATH}/#{ENV['RAILS_ENV']}_pid", "w"){ |f| f << pid}
+        puts "#{ENV['RAILS_ENV']} Solr started successfully on #{SOLR_PORT}, pid: #{pid}."
+      end
+    end
+  end
+  
+  desc 'Stops Solr. Specify the environment by using: RAILS_ENV=your_env. Defaults to development if none.'
+  task :stop do
+    require File.expand_path("#{File.dirname(__FILE__)}/../../../config/solr_environment")
+    fork do
+      file_path = "#{SOLR_PIDS_PATH}/#{ENV['RAILS_ENV']}_pid"
+      if File.exists?(file_path)
+        File.open(file_path, "r") do |f| 
+          pid = f.readline
+          Process.kill('TERM', pid.to_i)
+        end
+        File.unlink(file_path)
+        Rake::Task["solr:destroy_index"].invoke if ENV['RAILS_ENV'] == 'test'
+        puts "Solr shutdown successfully."
+      else
+        puts "PID file not found at #{file_path}. Either Solr is not running or no PID file was written."
+      end
+    end
+  end
+  
+  desc 'Remove Solr index'
+  task :destroy_index do
+    require File.expand_path("#{File.dirname(__FILE__)}/../../../config/solr_environment")
+    raise "In production mode.  I'm not going to delete the index, sorry." if ENV['RAILS_ENV'] == "production"
+    if File.exists?("#{SOLR_DATA_PATH}")
+      Dir["#{SOLR_DATA_PATH}/index/*"].each{|f| File.unlink(f) if File.exists?(f)}
+      Dir.rmdir("#{SOLR_DATA_PATH}/index")
+      puts "Index files removed under " + ENV['RAILS_ENV'] + " environment"
+    end
+  end
+  
+  # this task is by Henrik Nyh
+  # http://henrik.nyh.se/2007/06/rake-task-to-reindex-models-for-acts_as_solr
+  desc %{Reindexes data for all acts_as_solr models. Clears index first to get rid of orphaned records and optimizes index afterwards. RAILS_ENV=your_env to set environment. ONLY=book,person,magazine to only reindex those models; EXCEPT=book,magazine to exclude those models. START_SERVER=true to solr:start before and solr:stop after. BATCH=123 to post/commit in batches of that size: default is 300. CLEAR=false to not clear the index first; OPTIMIZE=false to not optimize the index afterwards.}
+  task :reindex => :environment do
+    require File.expand_path("#{File.dirname(__FILE__)}/../../../config/solr_environment")
+
+    includes = env_array_to_constants('ONLY')
+    if includes.empty?
+      includes = Dir.glob("#{RAILS_ROOT}/app/models/*.rb").map { |path| File.basename(path, ".rb").camelize.constantize }
+    end
+    excludes = env_array_to_constants('EXCEPT')
+    includes -= excludes
+    
+    optimize            = env_to_bool('OPTIMIZE',     true)
+    start_server        = env_to_bool('START_SERVER', false)
+    clear_first         = env_to_bool('CLEAR',       true)
+    batch_size          = ENV['BATCH'].to_i.nonzero? || 300
+    debug_output        = env_to_bool("DEBUG", false)
+
+    RAILS_DEFAULT_LOGGER.level = ActiveSupport::BufferedLogger::INFO unless debug_output
+
+    if start_server
+      puts "Starting Solr server..."
+      Rake::Task["solr:start"].invoke 
+    end
+    
+    # Disable solr_optimize
+    module ActsAsSolr::CommonMethods
+      def blank() end
+      alias_method :deferred_solr_optimize, :solr_optimize
+      alias_method :solr_optimize, :blank
+    end
+    
+    models = includes.select { |m| m.respond_to?(:rebuild_solr_index) }    
+    models.each do |model|
+  
+      if clear_first
+        puts "Clearing index for #{model}..."
+        ActsAsSolr::Post.execute(Solr::Request::Delete.new(:query => "#{model.solr_configuration[:type_field]}:#{model}")) 
+        ActsAsSolr::Post.execute(Solr::Request::Commit.new)
+      end
+      
+      puts "Rebuilding index for #{model}..."
+      model.rebuild_solr_index(batch_size)
+
+    end 
+
+    if models.empty?
+      puts "There were no models to reindex."
+    elsif optimize
+      puts "Optimizing..."
+      models.last.deferred_solr_optimize
+    end
+
+    if start_server
+      puts "Shutting down Solr server..."
+      Rake::Task["solr:stop"].invoke 
+    end
+    
+  end
+  
+  def env_array_to_constants(env)
+    env = ENV[env] || ''
+    env.split(/\s*,\s*/).map { |m| m.singularize.camelize.constantize }.uniq
+  end
+  
+  def env_to_bool(env, default)
+    env = ENV[env] || ''
+    case env
+      when /^true$/i then true
+      when /^false$/i then false
+      else default
+    end
+  end
+
+end
+