bumbleworks 0.0.6 → 0.0.7

data/.gitignore

@@ -1,2 +1,17 @@
+*.gem
+*.rbc
+.bundle
+.config
+.yardoc
+Gemfile.lock
+InstalledFiles
+_yardoc
 coverage
-pkg/*
+doc/
+lib/bundler/man
+pkg
+rdoc
+spec/reports
+test/tmp
+test/version_tmp
+tmp

data/README.md

@@ -24,11 +24,35 @@ Or you can install it yourself using
 
 ## Configuration
 
+#### Example Initializer File
+
+```ruby
+Bumbleworks.configure do |c|
+  c.storage = Redis.new # requires bumbleworks-redis gem
+  # Or, for dev/test, if you don't have Redis set up yet:
+  # c.storage = {}
+end
+
+# this next block is optional - it's only needed
+# if you want to load custom participants
+Bumbleworks.register_participants do
+  # foo FooParticipant
+  # bar BarParticipant
+  # ...
+end
+
+# Load all process definitions in lib/process_definitions
+Bumbleworks.load_definitions!
+
+# Start a worker in the background
+Bumbleworks.start_worker!
+```
+
 ### The Process Storage
 
 Bumbleworks uses a dedicated key-value storage for process information; this is where all process instance state is stored.  We consider it a best practice to keep this storage in a separate place from your business information; see Process vs. Business Information for more discussion.
 
-Before you can load process definitions, register participants, and spin up workers in Bumbleworks, you need to configure Bumbleworks's process storage.  Right now, Bumbleworks supports two storage methods - [Redis](http://redis.io/) and [Sequel](http://sequel.rubyforge.org/) (an ORM that itself supports MySQL, Postgres, etc.).
+Before you can load process definitions, register participants, and spin up workers in Bumbleworks, you need to configure Bumbleworks's process storage.  Right now, Bumbleworks supports three storage methods - [Redis](http://redis.io/), [Sequel](http://sequel.rubyforge.org/) (an ORM that itself supports MySQL, Postgres, etc.), and a simple Hash (only for development and testing, since it won't persist )
 
 #### Redis
 
@@ -44,8 +68,8 @@ If you want to use Redis:
 
   ```ruby
   Bumbleworks.configure do |c|
-    c.storage = Redis.new(:host => '127.0.0.1', :db => 0, :thread_safe => true)
-    # ...
+      c.storage = Redis.new(:host => '127.0.0.1', :db => 0, :thread_safe => true)
+      # ...
   end
   ```
 
@@ -63,14 +87,16 @@ If you want to use Sequel:
 
   ```ruby
   Bumbleworks.configure do |c|
-    c.storage = Sequel.connect('postgres://user:password@host:port/database_name')
-    # ...
+      c.storage = Sequel.connect('postgres://user:password@host:port/database_name')
+      # ...
   end
   ```
 
-### Process Definition Loading
+### Process Definitions Directory
+
+Bumbleworks uses [ruote](http://github.com/jmettraux/ruote), which allows process definitions to be written using a [Ruby DSL](http://ruote.rubyforge.org/definitions.html#ruby).
 
-Bumbleworks uses [ruote](http://github.com/jmettraux/ruote), which allows process definitions to be written using a [Ruby DSL](http://ruote.rubyforge.org/definitions.html#ruby).  By default, your process definitions will be loaded from the `lib/process_definitions` directory at `Bumbleworks.root` (see Determining the Root Directory for more info).  This directory can have as many subdirectories as you want, and Bumbleworks will load everything recursively; note, however, that the directory hierarchy doesn't mean anything to Bumbleworks, and is only for your own organization.  The directory is configurable by setting Bumbleworks.definitions_directory:
+By default, your process definitions will be loaded from the `lib/process_definitions` directory at `Bumbleworks.root` (see Determining the Root Directory for more info).  This directory can have as many subdirectories as you want, and Bumbleworks will load everything recursively; note, however, that the directory hierarchy doesn't mean anything to Bumbleworks, and is only for your own organization.  The directory is configurable by setting Bumbleworks.definitions_directory:
 
 ```ruby
 Bumbleworks.configure do |c|
@@ -79,24 +105,11 @@ Bumbleworks.configure do |c|
 end
 ```
 
-Note that if you override the default path, you can either specify an absolute path or a relative path - just use a leading slash if you want it to be interpreted as absolute.
+Note that if you override the default path, you can either specify an absolute path or a relative path - just use a leading slash if you want it to be interpreted as absolute.  Relative paths will be relative to `Bumbleworks.root`.
 
-## Participant Class Registration
+### Participant Class Directory
 
-Registering participants with Bumbleworks is done using Bumbleworks.register_participants, which takes a block, and follows [Ruote's #register syntax](http://ruote.rubyforge.org/participants.html#registering).  For example:
-
-```ruby
-Bumbleworks.register_participants do
-  update_status StatusChangerParticipant
-  acquire_lock LockerParticipant, 'action' => 'acquire'
-  release_lock LockerParticipant, 'action' => 'release'
-  notify_applicant ApplicantNotifierParticipant
-end
-```
-
-By default, Bumbleworks will register a "catchall" participant at the end of your participant list, which will catch any workitems not picked up by a participant higher in the list.  Those workitems then fall into ruote's StorageParticipant, from where Bumbleworks will assemble its task queue.
-
-If your app has a `participants` or `app/participants` directory at the root (see Determining the Root Directory), Bumbleworks will require all files in that directory by default before running the `register_participants` block.  You can customize this directory by setting Bumbleworks.participants_directory:
+If your app has a `participants` or `app/participants` directory at the root (see Determining the Root Directory), Bumbleworks will require all files in that directory by default before running your `register_participants` block (see below).  You can customize this directory by setting Bumbleworks.participants_directory:
 
 ```ruby
 Bumbleworks.configure do |c|
@@ -107,30 +120,53 @@ end
 
 ### Determining the Root Directory
 
-By default, Bumbleworks will attempt in several ways to find your root directory.  In the most common cases (Rails, Sinatra, running via Rake), it usually won't have trouble guessing the directory.
+By default, Bumbleworks will attempt in several ways to find your root directory.  In the most common cases (Rails, Sinatra, or Rory), it usually won't have trouble guessing the directory.
+
+If you're not using Rails, Sinatra, or Rory, Bumbleworks will complain when you call `bootstrap!` or `update!`, **unless** both your definitions directory and your participants directory (see above) are specified as absolute paths.
 
 ## Usage
 
-### Starting Work
+### Loading Definitions and Participants
+
+#### Process Definitions
 
-Without running a "worker," Bumbleworks won't do anything behind the scenes - no workitems will proceed through their workflow, no schedules will be checked, etc.  To run a worker, you can either set the `autostart_worker` option in configuration, before starting Bumbleworks:
+Process definitions are just ruby files with blocks following Ruote's [Ruby DSL syntax](http://ruote.rubyforge.org/definitions.html#ruby).  The only difference is that instead of `Ruote.define`, the method to call is `Bumbleworks.define_process`.  No arguments are necessary - the name will be taken from the base name of the ruby file (e.g. 'foo.rb' defines a process named 'foo').  All process definition files should be in your app's `definitions_directory` (see above); they can be in subdirectories for your own organization, but all filenames must be unique within the `definition_directory`'s tree.
+
+To actually load your process definitions from the directory:
 
 ```ruby
-Bumbleworks.configure do |c|
-  # ...
-  # NOTE: NOT RECOMMENDED IN PRODUCTION!
-  c.autostart_worker = true
-end
+Bumbleworks.load_definitions!
+```
+
+Keep in mind that any changed process definitions will overwrite previously loaded ones - in other words, after running this command successfully, all process definitions loaded into Bumbleworks will be in sync with the files in your definitions directory.
+
+Process definitions will be named (and `launch`able) using the name of the file itself.  Therefore, Bumbleworks will throw an exception if any files in your definitions directory (and subdirectories) have the same name.
 
-Bumbleworks.start! # this will now start a worker automatically
+#### Participants
+
+Registering participants with Bumbleworks is done using `Bumbleworks.register_participants`, which takes a block, and follows [Ruote's #register syntax](http://ruote.rubyforge.org/participants.html#registering).  For example:
+
+```ruby
+Bumbleworks.register_participants do
+  update_status StatusChangerParticipant
+  acquire_lock LockerParticipant, 'action' => 'acquire'
+  release_lock LockerParticipant, 'action' => 'release'
+  notify_applicant ApplicantNotifierParticipant
+end
 ```
 
-... but, while this is handy in development and testing, it's not a good practice to follow in production.  In an actual production environment, you will likely have multiple workers running in their own threads, or even on separate servers.  So the **preferred way** is to do the following (most likely in a Rake task that has your environment loaded):
+Unless you add it yourself, Bumbleworks will register a "catchall" participant at the end of your participant list, which will catch any workitems not picked up by a participant higher in the list.  Those workitems then fall into ruote's StorageParticipant, from where Bumbleworks will assemble its task queue.
+
+### Starting Work
+
+Without running a "worker," Bumbleworks won't do anything behind the scenes - no workitems will proceed through their workflow, no schedules will be checked, etc.  Running a worker is done using the following command:
 
 ```ruby
 Bumbleworks.start_worker!
 ```
 
+You can add this to the end of your initializer, but, while this is handy in development and testing, it's not a good practice to follow in production.  In an actual production environment, you will likely have multiple workers running in their own threads, or even on separate servers.  So the **preferred way** is to call the `Bumbleworks.start_worker!` method outside of the initializer, most likely in a Rake task that has your environment loaded.
+
 > Strictly speaking, the entire environment doesn't need to be loaded; only Bumbleworks.storage needs to be set before starting a worker.  However, it's best practice to configure Bumbleworks in one place, to ensure you don't get your storage configurations out of sync.
 
 You can run as many workers as you want in parallel, and as long as they're accessing the same storage, no concurrency issues should arise.
@@ -157,4 +193,36 @@ Bumbleworks::Task.for_roles(['trombonist', 'admin']) # returns both tasks
 
 Call Bumbleworks::Task#complete to finish a task and proceed to the next expression.
 
-See the Bumbleworks::Task class for more details.
+See the Bumbleworks::Task class for more details.
+
+## Rake Tasks
+
+*TODO:  Actually implement these, so I'm not a liar.*
+
+If you...
+
+```ruby
+require 'bumbleworks/rake_tasks'
+```
+
+... in your Rakefile, Bumbleworks will give you a couple of rake tasks.  Both of these rake tasks expect an `environment` Rake task to be specified that loads your application's environment (Rails, for example, gives you this automatically).
+
+The tasks are:
+
+1. `rake bumbleworks:start_worker`
+
+  This task starts a Bumbleworks worker, and does not return.  It will expect Bumbleworks to be required and for Bumbleworks' storage to be configured.
+
+2. `rake bumbleworks:reload_definitions`
+
+  All process definitions will be reloaded from the configured `definitions_directory`.
+
+## Contributing
+
+1. Check out the latest master to make sure the feature hasn't been implemented or the bug hasn't been fixed yet.
+1. Check out the issue tracker to make sure someone already hasn't requested it and/or contributed it.
+1. Fork the project.
+1. Start a feature/bugfix branch.
+1. Commit and push until you are happy with your contribution.
+1. Make sure to add tests for it. This is important so I don't break it in a future version unintentionally.
+1. Please try not to mess with the Rakefile, version, or history. If you want to have your own version, or is otherwise necessary, that is fine, but please isolate to its own commit so I can cherry-pick around it.

data/doc/GUIDE.md

@@ -66,10 +66,10 @@ Put the following in a `config/initializers/bumbleworks.rb` file:
 
 Bumbleworks.configure do |c|
   c.storage = {}
-  c.autostart_worker = true
 end
 
-Bumbleworks.start!
+Bumbleworks.load_definitions!
+Bumbleworks.start_worker!
 ```
 
 #### The Bumbleworks Data Store
@@ -82,13 +82,19 @@ Bumbleworks needs its own data store, used only for process state.  We'll explor
 
 Out of the box, Bumbleworks supports three storage types - [Redis](http://redis.io), [Sequel](http://sequel.rubyforge.org), and a simple Ruby Hash.  We're using the latter for now, since it requires no setup.  You would never use the Hash storage type for production - it's in-memory and in-process, so it won't survive a restart and you can't run multiple workers.  But for testing, it's ideal.
 
-#### Auto-starting Workers
+#### Loading Process Definitions
 
-*Okay, intelligent pants.  What about `c.autostart_worker = true`?*
+*Why are you yelling at me about loading definitions?*
 
-In a production environment, your "workers" - Ruby processes that actually run the workflow, stepping through your process definitions and generating task queues, etc. - should be instantiated in separate daemon processes.
+The `Bumbleworks.load_definitions!` call loads all process definitions in the configured directory.  See "Writing our First Process Definition" below for more on this.
 
-To support this best practice, we don't run a worker by default when you call `Bumbleworks.start!`.  This means launching processes won't actually do anything, and you'll slowly succumb to a suffocating despair.  For ease of development and testing, you can set `autostart_worker` to true, which will (surprisingly) automatically start a worker when you call `Bumbleworks.start!`.
+#### Starting a Worker
+
+*Okay, intelligent pants.  What about `Bumbleworks.start_worker!`?*
+
+In a production environment, your "workers" - Ruby processes that actually run the workflow, stepping through your process definitions and generating task queues, etc. - should be instantiated in separate daemon processes.  To support this best practice, Bumbleworks doesn't ever automatically start a worker for you - this is something you'd most likely do through the provided rake task (`rake bumbleworks:start_worker`, if you're one of those nerds who likes to know stuff).
+
+When we're developing locally or running tests, though, running a separate worker would be overkill.  However, not running a worker means launching processes won't actually do anything, and you'll slowly succumb to a suffocating despair.  For ease of development and testing, for now we're going to spin up a worker in the initializer with `Bumbleworks.start_worker!`, which (surprisingly) starts a worker.
 
 ## Writing our First Process Definition
 

data/lib/bumbleworks.rb

@@ -70,8 +70,7 @@ module Bumbleworks
     end
 
     # @public
-    # Accepts a block for registering participants which
-    # is envoked when start! is called. Note that a
+    # Accepts a block for registering participants.  Note that a
     # 'catchall Ruote::StorageParticipant' is always added to
     # the end of the list (unless it is defined in the block).
     #
@@ -82,18 +81,16 @@ module Bumbleworks
     #     plumber PlumberClass
     #   end
     def register_participants(&block)
-      @participant_block = block
+      Bumbleworks::ParticipantRegistration.autoload_all
+      Bumbleworks::Ruote.register_participants(&block)
     end
 
     # @public
-    # Registers participants and process_definitions with the Ruote engine.
-    # Also calls Bumbleworks::Ruote.dashboard for the first time, which
-    # implicitly instantiates the storage, and might start a worker (if
-    # configuration.autostart_worker is true)
+    # Registers all process_definitions in the configured definitions_directory
+    # with the Ruote engine.
     #
-    def start!
-      autoload_and_register_participants
-      load_process_definitions
+    def load_definitions!
+      Bumbleworks::ProcessDefinition.create_all_from_directory!(definitions_directory)
     end
 
     # @public
@@ -113,16 +110,5 @@ module Bumbleworks
     def launch!(process_definition_name, options = {})
       Bumbleworks::Ruote.launch(process_definition_name, options)
     end
-
-  private
-
-    def autoload_and_register_participants
-      Bumbleworks::ParticipantRegistration.autoload_all
-      Bumbleworks::Ruote.register_participants(&@participant_block)
-    end
-
-    def load_process_definitions
-      Bumbleworks::ProcessDefinition.create_all_from_directory!(definitions_directory)
-    end
   end
 end

data/lib/bumbleworks/configuration.rb

@@ -65,15 +65,6 @@ module Bumbleworks
     #
     define_setting :storage
 
-    # By default, a worker will NOT be started when the storage is initialized;
-    # this is the recommended practice since workers should be instantiated in
-    # their own threads, and multiple workers (even on different hosts) can run
-    # simultaneously.  However, if you want a worker to start up immediately
-    # (useful for testing or development), set autostart_worker to true.
-    #
-    # default: false
-    define_setting :autostart_worker
-
     def initialize
       @storage_adapters = []
     end
@@ -116,13 +107,6 @@ module Bumbleworks
       end
     end
 
-    # Whether or not we should start a worker when initializing the dashboard
-    # and storage.  Only returns true if set explicitly to true.
-    #
-    def autostart_worker
-      @autostart_worker == true
-    end
-
     # Add a storage adapter to the set of possible adapters.  Takes an object
     # that responds to `driver`, `use?`, `storage_class`, and `display_name`.
     #

data/lib/bumbleworks/participant_registration.rb

@@ -10,7 +10,7 @@ module Bumbleworks
       #
       def autoload_all(options = {})
         options[:directory] ||= Bumbleworks.participants_directory
-        Bumbleworks::Support.all_files(options[:directory], :camelize => true) do |path, name|
+        Bumbleworks::Support.all_files(options[:directory], :camelize => true).each do |path, name|
           Object.autoload name.to_sym, path
         end
       end

data/lib/bumbleworks/process_definition.rb

@@ -5,6 +5,7 @@ module Bumbleworks
     class NotFound < StandardError; end
     class FileNotFound < StandardError; end
     class Invalid < StandardError; end
+    class DuplicatesInDirectory < StandardError; end
 
     attr_reader :name, :definition, :tree
 
@@ -30,7 +31,6 @@ module Bumbleworks
       errors = []
       errors << "Name must be specified" unless @name
       errors << "Definition or tree must be specified" unless @definition || @tree
-      errors << "Name is not unique" if Bumbleworks.dashboard.variables[@name]
       begin
         build_tree!
       rescue Invalid
@@ -123,7 +123,11 @@ module Bumbleworks
       #
       def create_all_from_directory!(directory, opts = {})
         added_names = []
-        Bumbleworks::Support.all_files(directory) do |path, basename|
+        definition_files = Bumbleworks::Support.all_files(directory)
+        if definition_files.values.uniq.count != definition_files.count
+          raise DuplicatesInDirectory, "Definitions directory contains duplicate filenames"
+        end
+        definition_files.each do |path, basename|
           puts "Registering process definition #{basename} from file #{path}" if opts[:verbose] == true
           begin
             create!(:name => basename, :definition => File.read(path))

data/lib/bumbleworks/rake_tasks.rb

@@ -0,0 +1,3 @@
+Dir[File.join(File.dirname(__FILE__), '..', 'tasks', '*.rake')].each do |path|
+  load path
+end

data/lib/bumbleworks/ruote.rb

@@ -5,7 +5,7 @@ module Bumbleworks
     class << self
       def dashboard(options = {})
         @dashboard ||= begin
-          context = if Bumbleworks.autostart_worker || options[:start_worker] == true
+          context = if options[:start_worker] == true
             ::Ruote::Worker.new(storage)
           else
             storage
@@ -21,8 +21,8 @@ module Bumbleworks
         dashboard.worker
       end
 
-      def launch(name, options)
-        dashboard.launch(dashboard.variables[name], options)
+      def launch(name, *args)
+        dashboard.launch(dashboard.variables[name], *args)
       end
 
       def register_participants(&block)

data/lib/bumbleworks/support.rb

@@ -10,10 +10,11 @@ module Bumbleworks
     end
 
     def all_files(directory, options = {})
-      Dir["#{directory}/**/*.rb"].each do |path|
+      Dir["#{directory}/**/*.rb"].inject({}) do |memo, path|
         name = File.basename(path, '.rb')
         name = camelize(name) if options[:camelize] == true
-        yield path, name
+        memo[path] = name
+        memo
       end
     end
   end

data/lib/bumbleworks/version.rb

@@ -1,3 +1,3 @@
 module Bumbleworks
-  VERSION = "0.0.6"
+  VERSION = "0.0.7"
 end

data/lib/tasks/bumbleworks.rake

@@ -0,0 +1,21 @@
+namespace :bumbleworks do
+  desc 'Start a Bumbleworks worker'
+  task :start_worker => :environment do
+    puts "Starting Bumbleworks worker..." if verbose == true
+    Bumbleworks.start_worker!(:join => true)
+  end
+
+  desc 'Reload all process definitions from directory'
+  task :reload_definitions => :environment do
+    puts "Reloading all Bumbleworks process definitions..." if verbose == true
+    Bumbleworks.load_definitions!
+  end
+
+  desc 'Launch a given Bumbleworks process'
+  task :launch, [:process] => :environment do |task, args|
+    process = args[:process]
+    raise ArgumentError, "Process name required" unless process
+    puts "Launching process '#{process}'..." if verbose == true
+    Bumbleworks.launch!(process)
+  end
+end

data/spec/fixtures/apps/with_default_directories/full_initializer.rb

@@ -1,7 +1,6 @@
 Bumbleworks.configure! do |c|
   c.root = File.dirname(__FILE__)
   c.storage = {}
-  c.autostart_worker = true
 end
 
 Bumbleworks.register_participants do
@@ -9,4 +8,5 @@ Bumbleworks.register_participants do
   molasses_maker MolassesParticipant
 end
 
-Bumbleworks.start!
+Bumbleworks.load_definitions!
+Bumbleworks.start_worker!

data/spec/lib/bumbleworks/configuration_spec.rb

@@ -127,21 +127,6 @@ describe Bumbleworks::Configuration do
     end
   end
 
-  describe '#autostart_worker' do
-    it 'returns false by default' do
-      configuration.autostart_worker.should be_false
-    end
-
-    it 'only returns true if set explicitly to true' do
-      configuration.autostart_worker = 'yes'
-      configuration.autostart_worker.should be_false
-      configuration.autostart_worker = 1
-      configuration.autostart_worker.should be_false
-      configuration.autostart_worker = true
-      configuration.autostart_worker.should be_true
-    end
-  end
-
   describe '#clear!' do
     it 'resets #root' do
       configuration.root = '/Root'

data/spec/lib/bumbleworks/process_definition_spec.rb

@@ -72,13 +72,6 @@ describe Bumbleworks::ProcessDefinition do
         pdef.validate!
       }.to raise_error(described_class::Invalid, /Definition is not a valid process definition/)
     end
-
-    it "raises error if name already taken" do
-      Bumbleworks.dashboard.variables['tasty_beans'] = 'A miraculous ingredient'
-      expect {
-        described_class.new(:name => 'tasty_beans').validate!
-      }.to raise_error(described_class::Invalid, /Name is not unique/)
-    end
   end
 
   describe "#save!" do
@@ -100,6 +93,14 @@ describe Bumbleworks::ProcessDefinition do
       Bumbleworks.dashboard.variables['monkeys'].should ==
         pdef.build_tree!
     end
+
+    it "overwrites existing variable with new definition" do
+      Bumbleworks.dashboard.variables['monkeys'] = 'A miraculous ingredient'
+      pdef = described_class.new(:name => 'monkeys', :definition => valid_definition)
+      pdef.save!
+      Bumbleworks.dashboard.variables['monkeys'].should ==
+        pdef.build_tree!
+    end
   end
 
   describe ".find_by_name" do
@@ -136,11 +137,25 @@ describe Bumbleworks::ProcessDefinition do
       }.to raise_error(described_class::Invalid)
     end
 
+    it 'raises error if duplicate filenames are encountered' do
+      Bumbleworks::Support.stub(:all_files).and_return({
+        'Rhubarb Derailleur' => 'popular_side_dish',
+        'Cantonese Phrasebook' => 'popular_side_dish',
+        'Beans' => 'unpopular_side_dish'
+      })
+      expect {
+        described_class.create_all_from_directory!(definitions_path)
+      }.to raise_error(described_class::DuplicatesInDirectory)
+    end
+
     it 'rolls back any processes defined within current transaction if error' do
       Bumbleworks.dashboard.variables['keep_me'] = 'top_secret_cat_pics'
-      Bumbleworks::Support.stub(:all_files).
-        and_yield(definition_path('test_process'), 'test_process').
-        and_yield(definition_path('a_list_of_jams'), 'a_list_of_jams')
+      # stubbing here so we can explicitly set an order which will
+      # ensure we're testing rollback
+      Bumbleworks::Support.stub(:all_files).and_return({
+        definition_path('test_process') => 'test_process',
+        definition_path('a_list_of_jams') => 'a_list_of_jams'
+      })
       expect {
         described_class.create_all_from_directory!(definitions_path)
       }.to raise_error

data/spec/lib/bumbleworks/ruote_spec.rb

@@ -14,17 +14,11 @@ describe Bumbleworks::Ruote do
       described_class.dashboard.should be_an_instance_of(Ruote::Dashboard)
     end
 
-    it 'does not start a worker if autostart is false' do
+    it 'does not start a worker by default' do
       Bumbleworks.storage = {}
       described_class.dashboard.worker.should be_nil
     end
 
-    it 'starts a worker if autostart is true' do
-      Bumbleworks.storage = {}
-      Bumbleworks.autostart_worker = true
-      described_class.dashboard.worker.should_not be_nil
-    end
-
     it 'starts a worker if :start_worker option is true' do
       Bumbleworks.storage = {}
       described_class.dashboard(:start_worker => true).worker.should_not be_nil

data/spec/lib/bumbleworks/support_spec.rb

@@ -12,29 +12,23 @@ describe Bumbleworks::Support do
   describe ".all_files" do
     let(:test_directory) { File.join(fixtures_path, 'definitions').to_s }
 
-    it "for given directory, yields given block with path and name params" do
-      assembled_hash = {}
-      described_class.all_files(test_directory) do |path, name|
-        assembled_hash[name] = path
-      end
+    it "for given directory, creates hash of basename => path pairs" do
+      assembled_hash = described_class.all_files(test_directory)
 
-      assembled_hash['test_process'].should ==
-        File.join(fixtures_path, 'definitions', 'test_process.rb').to_s
-      assembled_hash['test_nested_process'].should ==
-        File.join(fixtures_path, 'definitions', 'nested_folder', 'test_nested_process.rb').to_s
+      assembled_hash[File.join(fixtures_path, 'definitions', 'test_process.rb').to_s].should ==
+        'test_process'
+      assembled_hash[File.join(fixtures_path, 'definitions', 'nested_folder', 'test_nested_process.rb').to_s].should ==
+        'test_nested_process'
     end
 
     it "camelizes names if :camelize option is true " do
       path = File.join(fixtures_path, 'definitions')
-      assembled_hash = {}
-      described_class.all_files(test_directory, :camelize => true) do |path, name|
-        assembled_hash[name] = path
-      end
+      assembled_hash = described_class.all_files(test_directory, :camelize => true)
 
-      assembled_hash['TestProcess'].should ==
-        File.join(fixtures_path, 'definitions', 'test_process.rb').to_s
-      assembled_hash['TestNestedProcess'].should ==
-        File.join(fixtures_path, 'definitions', 'nested_folder', 'test_nested_process.rb').to_s
+      assembled_hash[File.join(fixtures_path, 'definitions', 'test_process.rb').to_s].should ==
+        'TestProcess'
+      assembled_hash[File.join(fixtures_path, 'definitions', 'nested_folder', 'test_nested_process.rb').to_s].should ==
+        'TestNestedProcess'
     end
   end
 end

data/spec/lib/bumbleworks/task_spec.rb

@@ -3,9 +3,9 @@ describe Bumbleworks::Task do
 
   before :each do
     Bumbleworks.reset!
-    Bumbleworks.autostart_worker = true
     Bumbleworks.storage = {}
     Bumbleworks::Ruote.register_participants
+    Bumbleworks.start_worker!
   end
 
   describe '.new' do

data/spec/lib/bumbleworks_spec.rb

@@ -10,15 +10,14 @@ describe Bumbleworks do
     it 'allows multiple cumulative configuration blocks' do
       described_class.configure do |c|
         c.root = 'pickles'
-        c.autostart_worker = false
       end
 
       described_class.configure do |c|
-        c.autostart_worker = true
+        c.storage = 'nerfy'
       end
 
       described_class.configuration.root.should == 'pickles'
-      described_class.configuration.autostart_worker.should == true
+      described_class.configuration.storage.should == 'nerfy'
     end
 
     it 'requires a block' do
@@ -53,44 +52,20 @@ describe Bumbleworks do
   end
 
   describe '.register_participants' do
-    it 'stores the block' do
-      participant_block = lambda { bees_honey 'BeesHoney' }
-      described_class.register_participants &participant_block
-      described_class.instance_variable_get('@participant_block').should == participant_block
-    end
-  end
-
-  describe '.start!' do
-    before :each do
-      described_class.reset!
-      Bumbleworks::ParticipantRegistration.stub(:autoload_all)
-      described_class.stub(:load_process_definitions)
-    end
-
-    it 'registers pre-configured participants with ruote' do
-      the_block = lambda {}
-      described_class.register_participants &the_block
+    it 'autoloads and registers participants' do
+      the_block = lambda {  }
+      Bumbleworks::ParticipantRegistration.should_receive(:autoload_all)
       Bumbleworks::Ruote.should_receive(:register_participants).with(&the_block)
-      described_class.start!
-    end
-
-    it 'registers process definitions with dashboard' do
-      described_class.storage = {}
-      described_class.should_receive(:load_process_definitions)
-      described_class.start!
-    end
-
-    it 'does not automatically start a worker by default' do
-      described_class.storage = {}
-      described_class.start!
-      Bumbleworks::Ruote.dashboard.worker.should be_nil
+      described_class.register_participants &the_block
     end
+  end
 
-    it 'starts a worker if autostart_worker config setting is true' do
+  describe '.load_definitions!' do
+    it 'creates all definitions from directory' do
+      described_class.stub(:definitions_directory).and_return(:defs_dir)
       described_class.storage = {}
-      described_class.autostart_worker = true
-      described_class.start!
-      Bumbleworks::Ruote.dashboard.worker.should_not be_nil
+      Bumbleworks::ProcessDefinition.should_receive(:create_all_from_directory!).with(:defs_dir)
+      described_class.load_definitions!
     end
   end
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: bumbleworks
 version: !ruby/object:Gem::Version
-  version: 0.0.6
+  version: 0.0.7
   prerelease: 
 platform: ruby
 authors:
@@ -12,7 +12,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-05-14 00:00:00.000000000 Z
+date: 2013-05-21 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ruote
@@ -158,7 +158,6 @@ files:
 - .ruby-version
 - .watchr
 - Gemfile
-- Gemfile.lock
 - LICENSE.txt
 - README.md
 - Rakefile
@@ -170,12 +169,14 @@ files:
 - lib/bumbleworks/hash_storage.rb
 - lib/bumbleworks/participant_registration.rb
 - lib/bumbleworks/process_definition.rb
+- lib/bumbleworks/rake_tasks.rb
 - lib/bumbleworks/ruote.rb
 - lib/bumbleworks/storage_adapter.rb
 - lib/bumbleworks/support.rb
 - lib/bumbleworks/task.rb
 - lib/bumbleworks/tree_builder.rb
 - lib/bumbleworks/version.rb
+- lib/tasks/bumbleworks.rake
 - spec/fixtures/apps/with_default_directories/app/participants/honey_participant.rb
 - spec/fixtures/apps/with_default_directories/app/participants/molasses_participant.rb
 - spec/fixtures/apps/with_default_directories/config_initializer.rb
@@ -215,18 +216,12 @@ required_ruby_version: !ruby/object:Gem::Requirement
   - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
-      segments:
-      - 0
-      hash: -3888661442558670334
 required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
   requirements:
   - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
-      segments:
-      - 0
-      hash: -3888661442558670334
 requirements: []
 rubyforge_project: 
 rubygems_version: 1.8.23
@@ -260,3 +255,4 @@ test_files:
 - spec/lib/bumbleworks_spec.rb
 - spec/spec_helper.rb
 - spec/support/path_helpers.rb
+has_rdoc: 

data/Gemfile.lock

@@ -1,84 +0,0 @@
-PATH
-  remote: .
-  specs:
-    bumbleworks (0.0.6)
-      ruote
-
-GEM
-  remote: https://rubygems.org/
-  specs:
-    blankslate (2.1.2.4)
-    columnize (0.3.6)
-    debugger (1.5.0)
-      columnize (>= 0.3.1)
-      debugger-linecache (~> 1.2.0)
-      debugger-ruby_core_source (~> 1.2.0)
-    debugger-linecache (1.2.0)
-    debugger-ruby_core_source (1.2.0)
-    diff-lcs (1.2.4)
-    file-tail (1.0.12)
-      tins (~> 0.5)
-    multi_json (1.7.3)
-    parslet (1.4.0)
-      blankslate (~> 2.0)
-    rake (10.0.4)
-    rspec (2.13.0)
-      rspec-core (~> 2.13.0)
-      rspec-expectations (~> 2.13.0)
-      rspec-mocks (~> 2.13.0)
-    rspec-core (2.13.1)
-    rspec-expectations (2.13.0)
-      diff-lcs (>= 1.1.3, < 2.0)
-    rspec-mocks (2.13.1)
-    ruby2ruby (1.3.1)
-      ruby_parser (~> 2.0)
-      sexp_processor (~> 3.0)
-    ruby_parser (2.3.1)
-      sexp_processor (~> 3.0)
-    rufus-cloche (1.0.4)
-      rufus-json (>= 1.0.3)
-    rufus-dollar (1.0.4)
-    rufus-json (1.0.4)
-    rufus-mnemo (1.2.3)
-    rufus-scheduler (2.0.19)
-      tzinfo (>= 0.3.23)
-    rufus-treechecker (1.0.8)
-      ruby_parser (>= 2.0.5)
-    ruote (2.3.0.2)
-      blankslate (= 2.1.2.4)
-      parslet (= 1.4.0)
-      ruby_parser (~> 2.3)
-      rufus-cloche (>= 1.0.2)
-      rufus-dollar (>= 1.0.4)
-      rufus-json (>= 1.0.1)
-      rufus-mnemo (>= 1.2.2)
-      rufus-scheduler (>= 2.0.16)
-      rufus-treechecker (>= 1.0.8)
-      sourcify (= 0.5.0)
-    sexp_processor (3.2.0)
-    simplecov (0.7.1)
-      multi_json (~> 1.0)
-      simplecov-html (~> 0.7.1)
-    simplecov-html (0.7.1)
-    sourcify (0.5.0)
-      file-tail (>= 1.0.5)
-      ruby2ruby (>= 1.2.5)
-      ruby_parser (>= 2.0.5)
-      sexp_processor (>= 3.0.5)
-    sqlite3 (1.3.7)
-    tins (0.8.0)
-    tzinfo (0.3.37)
-    watchr (0.7)
-
-PLATFORMS
-  ruby
-
-DEPENDENCIES
-  bumbleworks!
-  bundler (~> 1.3)
-  debugger
-  rake
-  rspec
-  simplecov
-  sqlite3
-  watchr