documentcloud-cloud-crowd 0.0.3 → 0.0.4

data/cloud-crowd.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name      = 'cloud-crowd'
-  s.version   = '0.0.3'         # Keep version in sync with cloud-cloud.rb
+  s.version   = '0.0.4'         # Keep version in sync with cloud-cloud.rb
   s.date      = '2009-08-23'
 
   s.homepage    = "http://documentcloud.org" # wiki page on github?  
@@ -49,12 +49,11 @@ lib/cloud_crowd/action.rb
 lib/cloud_crowd/app.rb
 lib/cloud_crowd/asset_store.rb
 lib/cloud_crowd/command_line.rb
-lib/cloud_crowd/core_ext.rb
 lib/cloud_crowd/daemon.rb
 lib/cloud_crowd/helpers/authorization.rb
 lib/cloud_crowd/helpers/resources.rb
-lib/cloud_crowd/helpers/urls.rb
 lib/cloud_crowd/helpers.rb
+lib/cloud_crowd/inflector.rb
 lib/cloud_crowd/models/job.rb
 lib/cloud_crowd/models/work_unit.rb
 lib/cloud_crowd/models.rb

data/lib/cloud-crowd.rb

@@ -11,9 +11,6 @@ gem 'rest-client'
 gem 'right_aws'
 gem 'sinatra'
 
-# Common CloudCrowd libs:
-require 'cloud_crowd/core_ext'
-
 # Autoloading for all the pieces which may or may not be needed:
 autoload :ActiveRecord, 'activerecord'
 autoload :Benchmark,    'benchmark'
@@ -34,6 +31,7 @@ module CloudCrowd
   autoload :Action,     'cloud_crowd/action'
   autoload :AssetStore, 'cloud_crowd/asset_store'
   autoload :Helpers,    'cloud_crowd/helpers'
+  autoload :Inflector,  'cloud_crowd/inflector'
   autoload :Job,        'cloud_crowd/models'
   autoload :WorkUnit,   'cloud_crowd/models'
   
@@ -41,7 +39,7 @@ module CloudCrowd
   ROOT        = File.expand_path(File.dirname(__FILE__) + '/..')
   
   # Keep the version in sync with the gemspec.
-  VERSION     = '0.0.3'
+  VERSION     = '0.0.4'
     
   # A Job is processing if its WorkUnits in the queue to be handled by workers.
   PROCESSING  = 1
@@ -88,6 +86,15 @@ module CloudCrowd
       configuration = YAML.load_file(config_path)
       ActiveRecord::Base.establish_connection(configuration)
     end
+    
+    # Keep an authenticated (if configured to enable authentication) resource 
+    # for the central server.
+    def central_server
+      return @central_server if @central_server
+      params = [CloudCrowd.config[:central_server]]
+      params += [CloudCrowd.config[:login], CloudCrowd.config[:password]] if CloudCrowd.config[:use_http_authentication]
+      @central_server = RestClient::Resource.new(*params)
+    end
 
     # Return the readable status name of an internal CloudCrowd status number.
     def display_status(status)
@@ -98,13 +105,13 @@ module CloudCrowd
     # so we lazy-load them. Think about a variant of this for installing and
     # loading actions into a running CloudCrowd cluster on the fly.
     def actions(name)
-      action_class = name.camelize
+      action_class = Inflector.camelize(name)
       begin
         raise NameError, "can't find the #{action_class} Action" unless Module.constants.include?(action_class)
         Module.const_get(action_class)
       rescue NameError => e
         user_action     = "#{@config_path}/actions/#{name}"
-        default_action  = "#{CloudCrowd::ROOT}/actions/#{name}"
+        default_action  = "#{ROOT}/actions/#{name}"
         require user_action and retry    if File.exists? "#{user_action}.rb"
         require default_action and retry if File.exists? "#{default_action}.rb"
         raise e

data/lib/cloud_crowd/action.rb

@@ -18,14 +18,14 @@ module CloudCrowd
     
     # Configuring a new Action sets up all of the read-only variables that
     # form the bulk of the API for action subclasses. (Paths to read from and
-    # write to).
+    # write to). It creates the work_directory and moves into it.
     def configure(status, input, options, store)
       @input, @options, @store = input, options, store
       @job_id, @work_unit_id = options['job_id'], options['work_unit_id']
       @work_directory = File.expand_path(File.join(@store.temp_storage_path, storage_prefix))
       FileUtils.mkdir_p(@work_directory) unless File.exists?(@work_directory)
       Dir.chdir @work_directory
-      unless status == CloudCrowd::MERGING
+      unless status == MERGING
         @input_path = File.join(@work_directory, File.basename(@input))
         @file_name = File.basename(@input_path, File.extname(@input_path))
         download(@input, @input_path)
@@ -43,15 +43,16 @@ module CloudCrowd
       path
     end
     
-    # Takes a local filesystem path, and returns the public url on S3 where the 
-    # file was saved. 
+    # Takes a local filesystem path, and returns the public (or authenticated) 
+    # url on S3 where the file was saved. 
     def save(file_path)
       save_path = File.join(s3_storage_path, File.basename(file_path))
       @store.save(file_path, save_path)
       return @store.url(save_path)
     end
     
-    # After the Action has finished, we remove the work directory.
+    # After the Action has finished, we remove the work directory and return
+    # to the root directory (where daemons run by default).
     def cleanup_work_directory
       Dir.chdir '/'
       FileUtils.rm_r(@work_directory)
@@ -64,7 +65,7 @@ module CloudCrowd
     # [action_name]/job_[job_id]/unit_[work_unit_it]
     def storage_prefix
       path_parts = []
-      path_parts << underscore(self.class.to_s)
+      path_parts << Inflector.underscore(self.class)
       path_parts << "job_#{@job_id}"
       path_parts << "unit_#{@work_unit_id}" if @work_unit_id
       @storage_prefix ||= File.join(path_parts)
@@ -74,15 +75,6 @@ module CloudCrowd
       @s3_storage_path ||= storage_prefix
     end
     
-    # Pilfered from the ActiveSupport::Inflector.
-    def underscore(word)
-      word.to_s.gsub(/::/, '/').
-        gsub(/([A-Z]+)([A-Z][a-z])/,'\1_\2').
-        gsub(/([a-z\d])([A-Z])/,'\1_\2').
-        tr("-", "_").
-        downcase
-    end
-    
   end
   
 end

data/lib/cloud_crowd/app.rb

@@ -1,15 +1,16 @@
 module CloudCrowd
   
   class App < Sinatra::Default
-        
-    # static serves files from /public, methodoverride allows the _method param.
-    enable :static, :methodoverride
     
-    set :root, CloudCrowd::ROOT
+    set :root, ROOT
     set :authorization_realm, "CloudCrowd"
     
-    helpers CloudCrowd::Helpers
+    helpers Helpers
+    
+    # static serves files from /public, methodoverride allows the _method param.
+    enable :static, :methodoverride
     
+    # Enabling HTTP Authentication turns it on for all requests.
     before do
       login_required if CloudCrowd.config[:use_http_authentication]
     end
@@ -51,7 +52,7 @@ module CloudCrowd
           current_work_unit.fail(params[:output], params[:time])
           dequeue_work_unit(1)
         else             
-          return error(500, "Completing a work unit must specify status.")
+          error(500, "Completing a work unit must specify status.")
         end
       end
     end

data/lib/cloud_crowd/command_line.rb

@@ -6,7 +6,7 @@ module CloudCrowd
     # Configuration files required for the `crowd` command to function.
     CONFIG_FILES = ['config.yml', 'config.ru', 'database.yml']
     
-    # Reference the absolute path to the root, because we're about to chdir.
+    # Reference the absolute path to the root.
     CC_ROOT = File.expand_path(File.dirname(__FILE__) + '/../..')
     
     # Path to the Daemons gem script which launches workers.
@@ -58,7 +58,7 @@ OPTIONS:
     def run_server
       ensure_config
       require 'rubygems'
-      rackup_path = File.expand_path('config.ru')
+      rackup_path = File.expand_path("#{@options[:config_path]}/config.ru")
       if Gem.available? 'thin'
         exec "thin -e production -p #{@options[:port]} -R #{rackup_path} start"
       else
@@ -106,13 +106,13 @@ OPTIONS:
       load_code
       num_workers = @options[:num_workers] || CloudCrowd.config[:num_workers]
       num_workers.times do
-        `CLOUD_CROWD_CONFIG='#{File.expand_path('config.yml')}' ruby #{WORKER_RUNNER} start`
+        `CLOUD_CROWD_CONFIG='#{File.expand_path(@options[:config_path] + "/config.yml")}' ruby #{WORKER_RUNNER} start`
       end
     end
     
     # For debugging, run a single worker in the current process, showing output.
     def run_worker
-      exec "CLOUD_CROWD_CONFIG='#{File.expand_path('config.yml')}' ruby #{WORKER_RUNNER} run"
+      exec "CLOUD_CROWD_CONFIG='#{File.expand_path(@options[:config_path] + "/config.yml")}' ruby #{WORKER_RUNNER} run"
     end
     
     # Stop all active workers.
@@ -137,25 +137,21 @@ OPTIONS:
     # the CLOUD_CROWD_CONFIG environment variable. Exit if they're not found.
     def ensure_config
       return if @config_found
-      Dir.chdir @options[:config_path]
-      CONFIG_FILES.all? {|f| File.exists? f } ? @config_dir = true : config_not_found
+      found = CONFIG_FILES.all? {|f| File.exists? "#{@options[:config_path]}/#{f}" }
+      found ? @config_dir = true : config_not_found
     end
     
     # Parse all options for all actions.
     # TODO: Think about parsing options per sub-command separately.
     def parse_options
       @options = {
-        :db_config    => 'database.yml',
         :port         => 9173,
-        :config_path  => ENV['CLOUD_CROWD_CONFIG'] || '.',
+        :config_path  => ENV['CLOUD_CROWD_CONFIG'] || '.'
       }
       @option_parser = OptionParser.new do |opts|
         opts.on('-c', '--config PATH', 'path to configuration directory') do |conf_path|
           @options[:config_path] = conf_path
         end
-        opts.on('-d', '--database-config PATH', 'path to database.yml') do |conf_path|
-          @options[:db_config] = conf_path
-        end
         opts.on('-n', '--num-workers NUM', OptionParser::DecimalInteger, 'number of worker processes') do |num|
           @options[:num_workers] = num
         end
@@ -164,7 +160,7 @@ OPTIONS:
         end
         opts.on_tail('-v', '--version', 'show version') do
           load_code
-          puts "CloudCrowd version #{CloudCrowd::VERSION}"
+          puts "CloudCrowd version #{VERSION}"
           exit
         end
       end
@@ -178,14 +174,14 @@ OPTIONS:
       ensure_config
       require 'rubygems'
       require "#{CC_ROOT}/lib/cloud-crowd"
-      CloudCrowd.configure('config.yml')
+      CloudCrowd.configure("#{@options[:config_path]}/config.yml")
     end
     
     # Establish a connection to the central server's database. Not all commands
     # require this.
     def connect_to_database
       require 'cloud_crowd/models'
-      CloudCrowd.configure_database(@options[:db_config])
+      CloudCrowd.configure_database("#{@options[:config_path]}/database.yml")
     end
     
     # Exit with an explanation if the configuration files couldn't be found.

data/lib/cloud_crowd/daemon.rb

@@ -16,7 +16,7 @@ module CloudCrowd
     
     def initialize
       @wait_time = MIN_WAIT
-      @worker = CloudCrowd::Worker.new
+      @worker = Worker.new
       Signal.trap('INT',  'EXIT')
       Signal.trap('KILL', 'EXIT')
       Signal.trap('TERM', 'EXIT')

data/lib/cloud_crowd/helpers.rb

@@ -1,9 +1,8 @@
 require 'cloud_crowd/helpers/authorization'
 require 'cloud_crowd/helpers/resources'
-require 'cloud_crowd/helpers/urls'
 
 module CloudCrowd
   module Helpers
-    include Authorization, Resources, Urls #, Rack::Utils
+    include Authorization, Resources #, Rack::Utils
   end
 end

data/lib/cloud_crowd/helpers/authorization.rb

@@ -4,6 +4,7 @@ module CloudCrowd
   module Helpers
     module Authorization
       
+      # Ensure that the request includes the correct credentials.
       def login_required
         return if authorized?
         unauthorized! unless auth.provided?
@@ -12,14 +13,13 @@ module CloudCrowd
         request.env['REMOTE_USER'] = auth.username
       end
       
+      # Has the request been authenticated?
       def authorized?
         !!request.env['REMOTE_USER']
       end
       
-      def current_user
-        request.env['REMOTE_USER']
-      end
-      
+      # A request is authorized if its login and password match those stored
+      # in config.yml, or if authentication is disabled.
       def authorize(login, password)
         return true unless CloudCrowd.config[:use_http_authentication]
         return CloudCrowd.config[:login] == login &&
@@ -33,7 +33,7 @@ module CloudCrowd
         @auth ||= Rack::Auth::Basic::Request.new(request.env)
       end
       
-      def unauthorized!(realm = CloudCrowd::App.authorization_realm)
+      def unauthorized!(realm = App.authorization_realm)
         response['WWW-Authenticate'] = "Basic realm=\"#{realm}\""
         halt 401, 'Authorization Required'
       end

data/lib/cloud_crowd/inflector.rb

@@ -0,0 +1,19 @@
+module CloudCrowd
+  
+  # Pilfered in parts from the ActiveSupport::Inflector.
+  module Inflector
+  
+    def self.camelize(word)
+      word.to_s.gsub(/\/(.?)/) { "::#{$1.upcase}" }.gsub(/(?:^|_)(.)/) { $1.upcase }
+    end
+    
+    def self.underscore(word)
+      word.to_s.gsub(/::/, '/').
+        gsub(/([A-Z]+)([A-Z][a-z])/,'\1_\2').
+        gsub(/([a-z\d])([A-Z])/,'\1_\2').
+        tr("-", "_").
+        downcase
+    end
+  
+  end
+end

data/lib/cloud_crowd/models.rb

@@ -1,28 +1,31 @@
 module CloudCrowd
+  
+  # Adds named scopes and query methods for every CloudCrowd status to
+  # both Jobs and WorkUnits.
   module ModelStatus
     
     def self.included(klass)
       
       klass.class_eval do
         # Note that COMPLETE and INCOMPLETE are unions of other states.
-        named_scope 'processing', :conditions => {:status => CloudCrowd::PROCESSING}
-        named_scope 'succeeded',  :conditions => {:status => CloudCrowd::SUCCEEDED}
-        named_scope 'failed',     :conditions => {:status => CloudCrowd::FAILED}
-        named_scope 'splitting',  :conditions => {:status => CloudCrowd::SPLITTING}
-        named_scope 'merging',    :conditions => {:status => CloudCrowd::MERGING}
-        named_scope 'complete',   :conditions => {:status => CloudCrowd::COMPLETE}
-        named_scope 'incomplete', :conditions => {:status => CloudCrowd::INCOMPLETE}
+        named_scope 'processing', :conditions => {:status => PROCESSING}
+        named_scope 'succeeded',  :conditions => {:status => SUCCEEDED}
+        named_scope 'failed',     :conditions => {:status => FAILED}
+        named_scope 'splitting',  :conditions => {:status => SPLITTING}
+        named_scope 'merging',    :conditions => {:status => MERGING}
+        named_scope 'complete',   :conditions => {:status => COMPLETE}
+        named_scope 'incomplete', :conditions => {:status => INCOMPLETE}
       end
       
     end
     
-    def processing?;  self.status == CloudCrowd::PROCESSING;          end
-    def succeeded?;   self.status == CloudCrowd::SUCCEEDED;           end
-    def failed?;      self.status == CloudCrowd::FAILED;              end
-    def splitting?;   self.status == CloudCrowd::SPLITTING;           end
-    def merging?;     self.status == CloudCrowd::MERGING;             end
-    def complete?;    CloudCrowd::COMPLETE.include?(self.status);     end
-    def incomplete?;  CloudCrowd::INCOMPLETE.include?(self.status);   end
+    def processing?;  self.status == PROCESSING;          end
+    def succeeded?;   self.status == SUCCEEDED;           end
+    def failed?;      self.status == FAILED;              end
+    def splitting?;   self.status == SPLITTING;           end
+    def merging?;     self.status == MERGING;             end
+    def complete?;    COMPLETE.include?(self.status);     end
+    def incomplete?;  INCOMPLETE.include?(self.status);   end
     
   end
 end

data/lib/cloud_crowd/models/job.rb

@@ -5,7 +5,7 @@ module CloudCrowd
   # of inputs (usually public urls to files), an action (the name of a script that 
   # CloudCrowd knows how to run), and, eventually a corresponding list of output.
   class Job < ActiveRecord::Base
-    include CloudCrowd::ModelStatus
+    include ModelStatus
     
     has_many :work_units, :dependent => :destroy
     
@@ -23,16 +23,18 @@ module CloudCrowd
       )
     end
     
+    # Creating a job creates its corresponding work units, adding them 
+    # to the queue.
     def after_create
       self.queue_for_workers(JSON.parse(self.inputs))
     end
     
     def before_validation_on_create
-      self.status = self.splittable? ? CloudCrowd::SPLITTING : CloudCrowd::PROCESSING
+      self.status = self.splittable? ? SPLITTING : PROCESSING
     end
     
     # After work units are marked successful, we check to see if all of them have
-    # finished, if so, this job is complete.
+    # finished, if so, continue on to the next phase of the job. 
     def check_for_completion
       return unless all_work_units_complete?
       transition_to_next_phase               
@@ -45,19 +47,19 @@ module CloudCrowd
       self.save
       
       case self.status
-      when CloudCrowd::PROCESSING then queue_for_workers(output_list.map {|o| JSON.parse(o) }.flatten)
-      when CloudCrowd::MERGING    then queue_for_workers(output_list.to_json)
-      else                             fire_callback
+      when PROCESSING then queue_for_workers(output_list.map {|o| JSON.parse(o) }.flatten)
+      when MERGING    then queue_for_workers(output_list.to_json)
+      else                 fire_callback
       end
       self
     end
     
-    # Transition from the current phase to the next one.
+    # Transition this Job's status to the following one.
     def transition_to_next_phase
-      self.status = any_work_units_failed? ? CloudCrowd::FAILED     :
-                    self.splitting?        ? CloudCrowd::PROCESSING :
-                    self.should_merge?     ? CloudCrowd::MERGING    :
-                                             CloudCrowd::SUCCEEDED
+      self.status = any_work_units_failed? ? FAILED     :
+                    self.splitting?        ? PROCESSING :
+                    self.mergeable?        ? MERGING    :
+                                             SUCCEEDED
     end
     
     # If a callback_url is defined, post the Job's JSON to it upon completion.
@@ -71,7 +73,7 @@ module CloudCrowd
     
     # Cleaning up after a job will remove all of its files from S3.
     def cleanup
-      CloudCrowd::AssetStore.new.cleanup_job(self)
+      AssetStore.new.cleanup_job(self)
     end
     
     # Have all of the WorkUnits finished? We could trade reads for writes here
@@ -85,18 +87,23 @@ module CloudCrowd
       self.work_units.failed.count > 0
     end
     
+    # This job is splittable if its Action has a +split+ method.
     def splittable?
-      self.action_class.new.respond_to? :split
+      self.action_class.public_instance_methods.include? 'split'
     end
     
-    def should_merge?
-      self.processing? && self.action_class.new.respond_to?(:merge)
+    # This job is mergeable if its Action has a +merge+ method.
+    def mergeable?
+      self.processing? && self.action_class.public_instance_methods.include?('merge')
     end
     
+    # Retrieve the class for this Job's Action, loading it if necessary.
     def action_class
       CloudCrowd.actions(self.action)
     end
     
+    # When the WorkUnits are all finished, gather all their outputs together
+    # before removing them from the database entirely.
     def gather_outputs_from_work_units
       outs = self.work_units.complete.map {|wu| wu.output }
       self.work_units.complete.destroy_all
@@ -107,14 +114,18 @@ module CloudCrowd
       CloudCrowd.display_status(self.status)
     end
     
-    def work_units_remaining
-      self.work_units.incomplete.count
+    # How complete is this Job?
+    def percent_complete
+      return 0   if splitting?
+      return 100 if complete?
+      return 99  if merging?
+      (work_units.complete.count / work_units.count.to_f * 100).round
     end
     
     # A JSON representation of this job includes the statuses of its component
     # WorkUnits, as well as any completed outputs.
     def to_json(opts={})
-      atts = {'id' => self.id, 'status' => self.display_status, 'work_units_remaining' => self.work_units_remaining}
+      atts = {'id' => self.id, 'status' => self.display_status, 'percent_complete' => self.percent_complete}
       atts.merge!({'outputs' => JSON.parse(self.outputs)}) if self.outputs
       atts.merge!({'time' => self.time}) if self.time
       atts.to_json

data/lib/cloud_crowd/models/work_unit.rb

@@ -3,7 +3,7 @@ module CloudCrowd
   # A WorkUnit is an atomic chunk of work from a job, processing a single input
   # through a single action. All WorkUnits receive the same options.
   class WorkUnit < ActiveRecord::Base
-    include CloudCrowd::ModelStatus
+    include ModelStatus
     
     belongs_to :job
     
@@ -14,7 +14,7 @@ module CloudCrowd
     # Find the Nth available WorkUnit in the queue, and take it out.
     def self.dequeue(offset=0)
       unit = self.first(
-        :conditions => {:status => CloudCrowd::INCOMPLETE, :taken => false}, 
+        :conditions => {:status => INCOMPLETE, :taken => false}, 
         :order      => "created_at asc",
         :offset     => offset
       )
@@ -29,7 +29,7 @@ module CloudCrowd
     # Mark this unit as having finished successfully.
     def finish(output, time_taken)
       update_attributes({
-        :status   => CloudCrowd::SUCCEEDED,
+        :status   => SUCCEEDED,
         :taken    => false,
         :attempts => self.attempts + 1,
         :output   => output,
@@ -42,7 +42,7 @@ module CloudCrowd
       tries = self.attempts + 1
       return try_again if tries < CloudCrowd.config[:work_unit_retries]
       update_attributes({
-        :status   => CloudCrowd::FAILED,
+        :status   => FAILED,
         :taken    => false,
         :attempts => tries,
         :output   => output,

data/lib/cloud_crowd/runner.rb

@@ -1,7 +1,6 @@
 # This is the script that kicks off a single CloudCrowd::Daemon. Rely on 
 # cloud-crowd.rb for autoloading of all the code we need.
 
-# Daemon/Worker Dependencies.
 require "#{File.dirname(__FILE__)}/../cloud-crowd"
 
 FileUtils.mkdir('log') unless File.exists?('log')

data/lib/cloud_crowd/worker.rb

@@ -1,10 +1,7 @@
 module CloudCrowd
   
   class Worker
-        
-    CENTRAL_URL = CloudCrowd.config[:central_server]
-    RETRY_WAIT = CloudCrowd.config[:worker_retry_wait]
-    
+            
     attr_reader :action
     
     # Spinning up a worker will create a new AssetStore with a persistent
@@ -13,8 +10,8 @@ module CloudCrowd
     def initialize
       @id       = $$
       @hostname = Socket.gethostname
-      @store    = CloudCrowd::AssetStore.new
-      @server   = central_server_resource
+      @store    = AssetStore.new
+      @server   = CloudCrowd.central_server
       log 'started'
     end
     
@@ -48,14 +45,18 @@ module CloudCrowd
       end
     end
     
+    # We expect and require internal communication between the central server
+    # and the workers to succeed. If it fails for any reason, log it, and then 
+    # keep trying the same request.
     def keep_trying_to(title)
       begin
         yield
       rescue Exception => e
-        log "failed to #{title} -- retry in #{RETRY_WAIT} seconds"
+        wait_time = CloudCrowd.config[:worker_retry_wait] 
+        log "failed to #{title} -- retry in #{wait_time} seconds"
         log e.message
         log e.backtrace
-        sleep RETRY_WAIT
+        sleep wait_time
         retry
       end
     end
@@ -71,9 +72,9 @@ module CloudCrowd
         @action = CloudCrowd.actions(@action_name).new
         @action.configure(@status, @input, @options, @store)
         result = case @status
-        when CloudCrowd::PROCESSING then @action.process
-        when CloudCrowd::SPLITTING  then @action.split
-        when CloudCrowd::MERGING    then @action.merge
+        when PROCESSING then @action.process
+        when SPLITTING  then @action.split
+        when MERGING    then @action.merge
         else raise "Work units must specify their status."
         end
         complete_work_unit(result)
@@ -92,14 +93,6 @@ module CloudCrowd
     
     private
     
-    # Keep an authenticated (if configured to enable authentication) resource 
-    # for the central server.
-    def central_server_resource
-      params = [CENTRAL_URL]
-      params += [CloudCrowd.config[:login], CloudCrowd.config[:password]] if CloudCrowd.config[:use_http_authentication]
-      RestClient::Resource.new(*params)
-    end
-    
     # Common parameters to send back to central, regardless of success or failure.
     def completion_params
       {:id => @options['work_unit_id'], :time => Time.now - @start_time}

data/test/acceptance/test_failing_work_units.rb

@@ -17,7 +17,7 @@ class FailingWorkUnitsTest < Test::Unit::TestCase
     (CloudCrowd.config[:work_unit_retries] - 1).times do
       job.work_units.each {|unit| unit.fail('failed', 10) }
     end
-    assert job.reload.work_units_remaining == 3
+    assert job.reload.percent_complete == 0
     job.work_units.reload.each_with_index do |unit, i|
       assert unit.processing?
       assert unit.attempts == CloudCrowd.config[:work_unit_retries] - 1

data/test/unit/test_job.rb

@@ -17,7 +17,7 @@ class JobTest < Test::Unit::TestCase
     
     should "create all of its work units as soon as the job is created" do
       assert @job.work_units.count >= 1
-      assert @job.work_units_remaining == 1
+      assert @job.percent_complete == 0
       assert @job.processing?
       assert @unit.processing?
       assert !@job.all_work_units_complete?
@@ -27,7 +27,7 @@ class JobTest < Test::Unit::TestCase
       assert !@job.all_work_units_complete?
       @unit.update_attributes(:status => CloudCrowd::SUCCEEDED, :output => 'hello')
       assert @job.reload.all_work_units_complete?
-      assert @job.work_units_remaining == 0
+      assert @job.percent_complete == 100
       assert @job.outputs == "[\"hello\"]"
     end
     

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: documentcloud-cloud-crowd
 version: !ruby/object:Gem::Version 
-  version: 0.0.3
+  version: 0.0.4
 platform: ruby
 authors: 
 - Jeremy Ashkenas
@@ -141,12 +141,11 @@ files:
 - lib/cloud_crowd/app.rb
 - lib/cloud_crowd/asset_store.rb
 - lib/cloud_crowd/command_line.rb
-- lib/cloud_crowd/core_ext.rb
 - lib/cloud_crowd/daemon.rb
 - lib/cloud_crowd/helpers/authorization.rb
 - lib/cloud_crowd/helpers/resources.rb
-- lib/cloud_crowd/helpers/urls.rb
 - lib/cloud_crowd/helpers.rb
+- lib/cloud_crowd/inflector.rb
 - lib/cloud_crowd/models/job.rb
 - lib/cloud_crowd/models/work_unit.rb
 - lib/cloud_crowd/models.rb
@@ -163,6 +162,7 @@ files:
 - test/unit/test_work_unit.rb
 has_rdoc: true
 homepage: http://documentcloud.org
+licenses: 
 post_install_message: 
 rdoc_options: []
 
@@ -183,7 +183,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: cloud-crowd
-rubygems_version: 1.2.0
+rubygems_version: 1.3.5
 signing_key: 
 specification_version: 2
 summary: Better living through Map --> Ruby --> Reduce

data/lib/cloud_crowd/core_ext.rb

@@ -1,10 +0,0 @@
-# Extensions to core Ruby.
-
-class String
-  
-  # Stolen-ish in parts from ActiveSupport::Inflector.
-  def camelize
-    self.gsub(/\/(.?)/) { "::#{$1.upcase}" }.gsub(/(?:^|_)(.)/) { $1.upcase }
-  end
-  
-end

data/lib/cloud_crowd/helpers/urls.rb

@@ -1,7 +0,0 @@
-module CloudCrowd
-  module Helpers
-    module Urls
-      
-    end
-  end
-end