classiccms 0.2.4.pre → 0.3.0

data/vendor/bundle/gems/dragonfly-0.9.12/spec/dragonfly/job_spec.rb

@@ -0,0 +1,1059 @@
+require 'spec_helper'
+
+# Matchers
+RSpec::Matchers.define :match_steps do |steps|
+  match do |given|
+    given.map{|step| step.class } == steps
+  end
+end
+
+describe Dragonfly::Job do
+
+  describe "Step types" do
+
+    {
+      Dragonfly::Job::Fetch => :fetch,
+      Dragonfly::Job::Process => :process,
+      Dragonfly::Job::Encode => :encode,
+      Dragonfly::Job::Generate => :generate,
+      Dragonfly::Job::FetchFile => :fetch_file,
+      Dragonfly::Job::FetchUrl => :fetch_url
+    }.each do |klass, step_name|
+      it "should return the correct step name for #{klass}" do
+        klass.step_name.should == step_name
+      end
+    end
+
+    {
+      Dragonfly::Job::Fetch => :f,
+      Dragonfly::Job::Process => :p,
+      Dragonfly::Job::Encode => :e,
+      Dragonfly::Job::Generate => :g,
+      Dragonfly::Job::FetchFile => :ff,
+      Dragonfly::Job::FetchUrl => :fu
+    }.each do |klass, abbreviation|
+      it "should return the correct abbreviation for #{klass}" do
+        klass.abbreviation.should == abbreviation
+      end
+    end
+
+    describe "step_names" do
+      it "should return the available step names" do
+        Dragonfly::Job.step_names.should == [:fetch, :process, :encode, :generate, :fetch_file, :fetch_url]
+      end
+    end
+
+  end
+
+  it "should allow initializing with content" do
+    job = Dragonfly::Job.new(@app, 'eggheads')
+    job.data.should == 'eggheads'
+  end
+
+  describe "without content" do
+
+    before(:each) do
+      @app = test_app
+      @job = Dragonfly::Job.new(@app)
+    end
+
+    describe "fetch" do
+      before(:each) do
+        @job.fetch!('some_uid')
+      end
+
+      it { @job.steps.should match_steps([Dragonfly::Job::Fetch]) }
+
+      it "should retrieve from the app's datastore when applied" do
+        @app.datastore.should_receive(:retrieve).with('some_uid').and_return('HELLO')
+        @job.data.should == 'HELLO'
+      end
+
+      it "should set extra data if returned from the datastore" do
+        @app.datastore.should_receive(:retrieve).with('some_uid').and_return(['HELLO', {:name => 'test.txt'}])
+        @job.data.should == 'HELLO'
+        @job.meta.should == {:name => 'test.txt'}
+      end
+      
+      it "shouldn't set any url_attrs" do
+        @job.url_attrs.should == {}
+      end
+    end
+
+    describe "process" do
+      it "should raise an error when applying" do
+        @job.process!(:resize, '20x30')
+        lambda{
+          @job.apply
+        }.should raise_error(Dragonfly::Job::NothingToProcess)
+      end
+    end
+
+    describe "encode" do
+      it "should raise an error when applying" do
+        @job.encode!(:gif)
+        lambda{
+          @job.apply
+        }.should raise_error(Dragonfly::Job::NothingToEncode)
+      end
+    end
+
+    describe "analyse" do
+      it "should raise a NoContent error" do
+        job = @app.new_job # This will define #analyser on it
+        lambda{
+          job.analyse(:width)
+        }.should raise_error(Dragonfly::Job::NoContent)
+      end
+    end
+
+    describe "data" do
+      it "should raise a NoContent error" do
+        lambda{
+          @job.data
+        }.should raise_error(Dragonfly::Job::NoContent)
+      end
+    end
+
+    describe "generate" do
+      before(:each) do
+        @job.generate!(:plasma, 20, 30)
+      end
+
+      it { @job.steps.should match_steps([Dragonfly::Job::Generate]) }
+
+      it "should use the generator when applied" do
+        @app.generator.should_receive(:generate).with(:plasma, 20, 30).and_return('hi')
+        @job.data.should == 'hi'
+      end
+
+      it "should save extra data if the generator returns it" do
+        @app.generator.should_receive(:generate).with(:plasma, 20, 30).and_return(['hi', {:name => 'plasma.png'}])
+        @job.data.should == 'hi'
+        @job.meta.should == {:name => 'plasma.png'}
+      end
+      
+      it "shouldn't set any url_attrs" do
+        @job.url_attrs.should == {}
+      end
+    end
+
+    describe "fetch_file" do
+      before(:each) do
+        @job.fetch_file!(File.dirname(__FILE__) + '/../../samples/egg.png')
+      end
+
+      it { @job.steps.should match_steps([Dragonfly::Job::FetchFile]) }
+
+      it "should fetch the specified file when applied" do
+        @job.size.should == 62664
+      end
+      
+      it "should set the url_attrs" do
+        @job.url_attrs.should == {:name => 'egg.png'}
+      end
+      
+      it "should set the name" do
+        @job.meta[:name].should == 'egg.png'
+      end
+    end
+
+    describe "fetch_url" do
+      before(:each) do
+        stub_request(:get, %r{http://some\.place\.com/.*}).to_return(:body => 'result!')
+        stub_request(:get, 'https://some.place.com').to_return(:body => 'secure result!')
+      end
+
+      it {
+        @job.fetch_url!('some.url')
+        @job.steps.should match_steps([Dragonfly::Job::FetchUrl])
+      }
+
+      it "should fetch the specified url when applied" do
+        @job.fetch_url!('http://some.place.com')
+        @job.data.should == "result!"
+      end
+
+      it "should default to http" do
+        @job.fetch_url!('some.place.com')
+        @job.data.should == "result!"
+      end
+
+      it "should also work with https" do
+        @job.fetch_url!('https://some.place.com')
+        @job.data.should == "secure result!"
+      end
+
+      it "should set the name if there is one" do
+        @job.fetch_url!('some.place.com/dung.beetle')
+        @job.meta[:name].should == 'dung.beetle'
+      end
+      
+      it "should set the name url_attr if there is one" do
+        @job.fetch_url!('some.place.com/dung.beetle')
+        @job.url_attrs.should == {:name =>'dung.beetle'}
+      end
+
+      ["some.place.com", "some.place.com/", "some.place.com/eggs/"].each do |url|
+        it "should not set the name if there isn't one, e.g. #{url}" do
+          @job.fetch_url!(url)
+          @job.meta[:name].should be_nil
+        end
+
+        it "should not set the name url_attr if there isn't one, e.g. #{url}" do
+          @job.fetch_url!(url)
+          @job.url_attrs[:name].should be_nil
+        end
+      end
+    end
+
+  end
+
+  describe "with content already there" do
+
+    before(:each) do
+      @app = mock_app
+      @job = Dragonfly::Job.new(@app, 'HELLO', :name => 'hello.txt', :a => :b)
+      @temp_object = @job.temp_object
+    end
+
+    describe "apply" do
+      it "should return itself" do
+        @job.apply.should == @job
+      end
+    end
+
+    describe "process" do
+      before(:each) do
+        @job.url_attrs = {:some => 'thing'}
+        @job.process!(:resize, '20x30')
+      end
+
+      it { @job.steps.should match_steps([Dragonfly::Job::Process]) }
+
+      it "should use the processor when applied" do
+        @app.processor.should_receive(:process).with(@temp_object, :resize, '20x30').and_return('hi')
+        @job.data.should == 'hi'
+      end
+
+      it "should maintain the meta attributes" do
+        @app.processor.should_receive(:process).with(@temp_object, :resize, '20x30').and_return('hi')
+        @job.meta.should == {:name => 'hello.txt', :a => :b}
+      end
+
+      it "should maintain the url_attrs" do
+        @job.url_attrs.should == {:some => 'thing'}
+      end
+
+      it "should allow returning an array with extra attributes from the processor" do
+        @app.processor.should_receive(:process).with(@temp_object, :resize, '20x30').and_return(['hi', {:name => 'hello_20x30.txt', :eggs => 'asdf'}])
+        @job.data.should == 'hi'
+        @job.meta.should == {:name => 'hello_20x30.txt', :a => :b, :eggs => 'asdf'}
+      end
+    end
+
+    describe "encode" do
+      before(:each) do
+        @job.url_attrs = {:name => 'boid'}
+        @job.encode!(:gif, :bitrate => 'mumma')
+      end
+
+      it { @job.steps.should match_steps([Dragonfly::Job::Encode]) }
+
+      it "should use the encoder when applied" do
+        @app.encoder.should_receive(:encode).with(@temp_object, :gif, :bitrate => 'mumma').and_return('alo')
+        @job.data.should == 'alo'
+      end
+
+      it "should maintain the meta and update the format" do
+        @app.encoder.should_receive(:encode).with(@temp_object, :gif, :bitrate => 'mumma').and_return('alo')
+        @job.meta.should == {:name => 'hello.txt', :a => :b, :format => :gif}
+      end
+
+      it "should update the format on the url_attrs" do
+        @app.encoder.should_not_receive(:encode)
+        @job.url_attrs[:format].should == :gif
+      end
+
+      it "should allow returning an array with extra attributes form the encoder" do
+        @app.encoder.should_receive(:encode).with(@temp_object, :gif, :bitrate => 'mumma').and_return(['alo', {:name => 'doobie', :eggs => 'fish'}])
+        @job.data.should == 'alo'
+        @job.meta.should == {:name => 'doobie', :a => :b, :eggs => 'fish', :format => :gif}
+      end
+
+      it "not allow overriding the format" do
+        @app.encoder.should_receive(:encode).with(@temp_object, :gif, :bitrate => 'mumma').and_return(['alo', {:format => :png}])
+        @job.apply.meta[:format].should == :gif
+      end
+
+      it "should maintain the url_attrs" do
+        @job.url_attrs[:name].should == 'boid'
+      end
+      
+    end
+  end
+
+  describe "analysis" do
+    before(:each) do
+      @app = test_app
+      @job = @app.new_job('HELLO')
+      @app.analyser.add(:num_letters){|temp_object, letter| temp_object.data.count(letter) }
+    end
+    it "should return correctly when calling analyse" do
+      @job.analyse(:num_letters, 'L').should == 2
+    end
+    it "should have mixed in the analyser method" do
+      @job.num_letters('L').should == 2
+    end
+    it "should return nil from analyse if calling any old method" do
+      @job.analyse(:robin_van_persie).should be_nil
+    end
+    it "should not allow calling any old method" do
+      lambda{
+        @job.robin_van_persie
+      }.should raise_error(NoMethodError)
+    end
+    it "should work correctly with chained jobs, applying before analysing" do
+      @app.processor.add(:double){|temp_object| temp_object.data * 2 }
+      @job.process(:double).num_letters('L').should == 4
+    end
+  end
+
+  describe "defining extra steps after applying" do
+    before(:each) do
+      @app = mock_app
+      @job = Dragonfly::Job.new(@app)
+      @job.temp_object = Dragonfly::TempObject.new("hello")
+      @job.process! :resize
+      @job.apply
+      @job.encode! :micky
+    end
+    it "should not call apply on already applied steps" do
+      @job.steps[0].should_not_receive(:apply)
+      @job.apply
+    end
+    it "should call apply on not yet applied steps" do
+      @job.steps[1].should_receive(:apply)
+      @job.apply
+    end
+    it "should return all steps" do
+      @job.steps.should match_steps([
+        Dragonfly::Job::Process,
+        Dragonfly::Job::Encode
+      ])
+    end
+    it "should return applied steps" do
+      @job.applied_steps.should match_steps([
+        Dragonfly::Job::Process
+      ])
+    end
+    it "should return the pending steps" do
+      @job.pending_steps.should match_steps([
+        Dragonfly::Job::Encode
+      ])
+    end
+    it "should not call apply on any steps when already applied" do
+      @job.apply
+      @job.steps[0].should_not_receive(:apply)
+      @job.steps[1].should_not_receive(:apply)
+      @job.apply
+    end
+  end
+
+  describe "chaining" do
+
+    before(:each) do
+      @app = mock_app
+      @job = Dragonfly::Job.new(@app)
+    end
+
+    it "should return itself if bang is used" do
+      @job.fetch!('some_uid').should == @job
+    end
+
+    it "should return a new job if bang is not used" do
+      @job.fetch('some_uid').should_not == @job
+    end
+
+    describe "when a chained job is defined" do
+      before(:each) do
+        @job.fetch!('some_uid')
+        @job2 = @job.process(:resize, '30x30')
+      end
+
+      it "should return the correct steps for the original job" do
+        @job.applied_steps.should match_steps([
+        ])
+        @job.pending_steps.should match_steps([
+          Dragonfly::Job::Fetch
+        ])
+      end
+
+      it "should return the correct data for the original job" do
+        @job.data.should == 'SOME_DATA'
+      end
+
+      it "should return the correct steps for the new job" do
+        @job2.applied_steps.should match_steps([
+        ])
+        @job2.pending_steps.should match_steps([
+          Dragonfly::Job::Fetch,
+          Dragonfly::Job::Process
+        ])
+      end
+
+      it "should return the correct data for the new job" do
+        @job2.data.should == 'SOME_PROCESSED_DATA'
+      end
+
+      it "should not affect the other one when one is applied" do
+        @job.apply
+        @job.applied_steps.should match_steps([
+          Dragonfly::Job::Fetch
+        ])
+        @job.pending_steps.should match_steps([
+        ])
+        @job.temp_object.data.should == 'SOME_DATA'
+        @job2.applied_steps.should match_steps([
+        ])
+        @job2.pending_steps.should match_steps([
+          Dragonfly::Job::Fetch,
+          Dragonfly::Job::Process
+        ])
+        @job2.temp_object.should be_nil
+      end
+    end
+
+  end
+
+  describe "applied?" do
+    before(:each) do
+      @app = test_app
+    end
+    it "should return true when empty" do
+      @app.new_job.should be_applied
+    end
+    it "should return false when not applied" do
+      @app.fetch('eggs').should_not be_applied
+    end
+    it "should return true when applied" do
+      @app.datastore.should_receive(:retrieve).with('eggs').and_return("cracked")
+      job = @app.fetch('eggs').apply
+      job.should be_applied
+    end
+  end
+
+  describe "to_a" do
+    before(:each) do
+      @app = mock_app
+    end
+    it "should represent all the steps in array form" do
+      job = Dragonfly::Job.new(@app)
+      job.fetch! 'some_uid'
+      job.generate! :plasma # you wouldn't really call this after fetch but still works
+      job.process! :resize, '30x40'
+      job.encode! :gif, :bitrate => 20
+      job.to_a.should == [
+        [:f, 'some_uid'],
+        [:g, :plasma],
+        [:p, :resize, '30x40'],
+        [:e, :gif, {:bitrate => 20}]
+      ]
+    end
+  end
+
+  describe "from_a" do
+
+    before(:each) do
+      @app = test_app
+    end
+
+    describe "a well-defined array" do
+      before(:each) do
+        @job = Dragonfly::Job.from_a([
+          [:f, 'some_uid'],
+          [:g, :plasma],
+          [:p, :resize, '30x40'],
+          [:e, :gif, {:bitrate => 20}]
+        ], @app)
+      end
+      it "should have the correct step types" do
+        @job.steps.should match_steps([
+          Dragonfly::Job::Fetch,
+          Dragonfly::Job::Generate,
+          Dragonfly::Job::Process,
+          Dragonfly::Job::Encode
+        ])
+      end
+      it "should have the correct args" do
+        @job.steps[0].args.should == ['some_uid']
+        @job.steps[1].args.should == [:plasma]
+        @job.steps[2].args.should == [:resize, '30x40']
+        @job.steps[3].args.should == [:gif, {:bitrate => 20}]
+      end
+      it "should have no applied steps" do
+        @job.applied_steps.should be_empty
+      end
+      it "should have all steps pending" do
+        @job.steps.should == @job.pending_steps
+      end
+    end
+
+    [
+      :f,
+      [:f],
+      [[]],
+      [[:egg]]
+    ].each do |object|
+      it "should raise an error if the object passed in is #{object.inspect}" do
+        lambda {
+          Dragonfly::Job.from_a(object, @app)
+        }.should raise_error(Dragonfly::Job::InvalidArray)
+      end
+    end
+
+    it "should initialize an empty job if the array is empty" do
+      job = Dragonfly::Job.from_a([], @app)
+      job.steps.should be_empty
+    end
+  end
+
+  describe "serialization" do
+    before(:each) do
+      @app = test_app
+      @job = Dragonfly::Job.new(@app).fetch('uid').process(:resize_and_crop, :width => 270, :height => 92, :gravity => 'n')
+    end
+    it "should serialize itself" do
+      @job.serialize.should =~ /^\w{1,}$/
+    end
+    it "should deserialize to the same as the original" do
+      new_job = Dragonfly::Job.deserialize(@job.serialize, @app)
+      new_job.to_a.should == @job.to_a
+    end
+    it "should correctly deserialize a string serialized with ruby 1.8.7" do
+      job = Dragonfly::Job.deserialize('BAhbB1sHOgZmSSIIdWlkBjoGRVRbCDoGcDoUcmVzaXplX2FuZF9jcm9wewg6CndpZHRoaQIOAToLaGVpZ2h0aWE6DGdyYXZpdHlJIgZuBjsGVA', @app)
+      job.to_a.should == @job.to_a
+    end
+    it "should correctly deserialize a string serialized with ruby 1.9.2" do
+      job = Dragonfly::Job.deserialize('BAhbB1sHOgZmIgh1aWRbCDoGcDoUcmVzaXplX2FuZF9jcm9wewg6CndpZHRoaQIOAToLaGVpZ2h0aWE6DGdyYXZpdHkiBm4', @app)
+      job.to_a.should == @job.to_a
+    end
+  end
+
+  describe "to_app" do
+    before(:each) do
+      @app = test_app
+      @job = Dragonfly::Job.new(@app)
+    end
+    it "should return an endpoint" do
+      endpoint = @job.to_app
+      endpoint.should be_a(Dragonfly::JobEndpoint)
+      endpoint.job.should == @job
+    end
+  end
+
+  describe "url" do
+    before(:each) do
+      @app = test_app
+      @job = Dragonfly::Job.new(@app)
+    end
+
+    it "should return nil if there are no steps" do
+      @job.url.should be_nil
+    end
+    
+    describe "using url_attrs in the url" do
+      before(:each) do
+        @app.server.url_format = '/media/:job/:zoo'
+        @job.generate!(:fish)
+      end
+      it "should act as per usual if no params given" do
+        @job.url.should == "/media/#{@job.serialize}"
+      end
+      it "should add given params" do
+        @job.url(:zoo => 'jokes', :on => 'me').should == "/media/#{@job.serialize}/jokes?on=me"
+      end
+      it "should use the url_attr if it exists" do
+        @job.url_attrs[:zoo] = 'hair'
+        @job.url.should == "/media/#{@job.serialize}/hair"
+      end
+      it "should not add any url_attrs that aren't needed" do
+        @job.url_attrs[:gump] = 'flub'
+        @job.url.should == "/media/#{@job.serialize}"
+      end
+      it "should override if a param is passed in" do
+        @job.url_attrs[:zoo] = 'hair'
+        @job.url(:zoo => 'dare').should == "/media/#{@job.serialize}/dare"
+      end
+      
+      describe "basename" do
+        before(:each) do
+          @app.server.url_format = '/:job/:basename'
+        end
+        it "should use the name" do
+          @job.url_attrs = {:name => 'hello.egg'}
+          @job.url.should == "/#{@job.serialize}/hello"
+        end
+        it "should not set if neither exist" do
+          @job.url_attrs = {}
+          @job.url.should == "/#{@job.serialize}"
+        end
+      end
+
+      describe "ext" do
+        before(:each) do
+          @app.server.url_format = '/:job.:ext'
+          @job.url_attrs = {:name => 'hello.egg', :ext => 'hi'}
+        end
+        it "should use the name" do
+          @job.url_attrs = {:name => 'hello.egg'}
+          @job.url.should == "/#{@job.serialize}.egg"
+        end
+        it "should not set if neither exist" do
+          @job.url_attrs = {}
+          @job.url.should == "/#{@job.serialize}"
+        end
+      end
+
+      describe "format" do
+        before(:each) do
+          @app.server.url_format = '/:job.:format'
+        end
+        it "should use the url_attr if it exists" do
+          @job.url_attrs = {:ext => 'hi', :format => :txt}
+          @job.url.should == "/#{@job.serialize}.txt"
+        end
+        it "should use the ext if format url_attr doesn't exist" do
+          @job.url_attrs = {:name => 'hup.hi'}
+          @job.url.should == "/#{@job.serialize}.hi"
+        end
+        it "should not use the ext if format meta doesn't exist and trust_file_extensions is switched off" do
+          @job.url_attrs = {:name => 'hup.hi'}
+          @app.trust_file_extensions = false
+          @job.url.should == "/#{@job.serialize}"
+        end
+        it "should not set if neither exist" do
+          @job.url_attrs = {}
+          @job.url.should == "/#{@job.serialize}"
+        end
+      end
+    end
+  end
+
+  describe "to_fetched_job" do
+    before(:each) do
+      @app = test_app
+      @job = @app.create("HELLO")
+    end
+    it "should maintain the same temp_object and be already applied" do
+      new_job = @job.to_fetched_job('some_uid')
+      new_job.data.should == 'HELLO'
+      new_job.to_a.should == [
+        [:f, 'some_uid']
+      ]
+      new_job.pending_steps.should be_empty
+    end
+    it "should maintain the meta" do
+      @job.meta = {:right => 'said fred'}
+      new_job = @job.to_fetched_job('some_uid')
+      new_job.meta.should == {:right => 'said fred'}
+    end
+    it "should maintain the url_attrs" do
+      @job.url_attrs = {:dang => 'that dawg'}
+      new_job = @job.to_fetched_job('some_uid')
+      new_job.url_attrs.should == {:dang => 'that dawg'}
+    end
+  end
+
+  describe "to_unique_s" do
+    it "should use the arrays of args to create the string" do
+      job = test_app.fetch('uid').process(:gug, 4, :some => 'arg', :and => 'more')
+      job.to_unique_s.should == 'fuidpgug4andmoresomearg'
+    end
+  end
+
+  describe "sha" do
+    before(:each) do
+      @app = test_app
+      @job = @app.fetch('eggs')
+    end
+    
+    it "should be of the correct format" do
+      @job.sha.should =~ /^\w{8}$/
+    end
+    
+    it "should be the same for the same job steps" do
+      @app.fetch('eggs').sha.should == @job.sha
+    end
+    
+    it "should be different for different jobs" do
+      @app.fetch('figs').sha.should_not == @job.sha
+    end
+  end
+
+  describe "validate_sha!" do
+    before(:each) do
+      @app = test_app
+      @job = @app.fetch('eggs')
+    end
+    it "should raise an error if nothing is given" do
+      lambda{
+        @job.validate_sha!(nil)
+      }.should raise_error(Dragonfly::Job::NoSHAGiven)
+    end
+    it "should raise an error if the wrong SHA is given" do
+      lambda{
+        @job.validate_sha!('asdf')
+      }.should raise_error(Dragonfly::Job::IncorrectSHA)
+    end
+    it "should return self if ok" do
+      @job.validate_sha!(@job.sha).should == @job
+    end
+  end
+
+  describe "setting the name" do
+    before(:each) do
+      @app = test_app
+      @job = @app.new_job("HELLO", :name => 'not.me')
+    end
+    it "should allow setting the name" do
+      @job.name = 'wassup.doc'
+      @job.name.should == 'wassup.doc'
+    end
+  end
+
+  describe "setting the meta" do
+    before(:each) do
+      @app = test_app
+      @job = @app.new_job("HiThere", :five => 'beans')
+    end
+    it "should allow setting the meta" do
+      @job.meta = {:doogie => 'ladders'}
+      @job.meta.should == {:doogie => 'ladders'}
+    end
+    it "should allow updating the meta" do
+      @job.meta[:doogie] = 'ladders'
+      @job.meta.should == {:five => 'beans', :doogie => 'ladders'}
+    end
+  end
+
+  describe "b64_data" do
+    before(:each) do
+      @app = test_app
+    end
+    it "should return a string using the data:URI schema" do
+      job = @app.new_job("HELLO", :name => 'text.txt')
+      job.b64_data.should == "data:text/plain;base64,SEVMTE8=\n"
+    end
+  end
+
+  describe "querying stuff without applying steps" do
+    before(:each) do
+      @app = test_app
+    end
+    
+    describe "fetch_step" do
+      it "should return nil if it doesn't exist" do
+        @app.generate(:ponies).process(:jam).fetch_step.should be_nil
+      end
+      it "should return the fetch step otherwise" do
+        step = @app.fetch('hello').process(:cheese).fetch_step
+        step.should be_a(Dragonfly::Job::Fetch)
+        step.uid.should == 'hello'
+      end
+    end
+    describe "uid" do
+      describe "when there's no fetch step" do
+        before(:each) do
+          @job = @app.new_job("AGG")
+        end
+        it "should return nil for uid" do
+          @job.uid.should be_nil
+        end
+      end
+      describe "when there is a fetch step" do
+        before(:each) do
+          @job = @app.fetch('gungedin/innit.blud')
+        end
+        it "should return the uid" do
+          @job.uid.should == 'gungedin/innit.blud'
+        end
+      end
+    end
+
+    describe "fetch_file_step" do
+      it "should return nil if it doesn't exist" do
+        @app.generate(:ponies).process(:jam).fetch_file_step.should be_nil
+      end
+      it "should return the fetch_file step otherwise" do
+        step = @app.fetch_file('/my/file.png').process(:cheese).fetch_file_step
+        step.should be_a(Dragonfly::Job::FetchFile)
+        if Dragonfly.running_on_windows?
+          step.path.should =~ %r(:/my/file\.png$)
+        else
+          step.path.should == '/my/file.png'
+        end
+      end
+    end
+
+    describe "fetch_url_step" do
+      it "should return nil if it doesn't exist" do
+        @app.generate(:ponies).fetch_url_step.should be_nil
+      end
+      it "should return the fetch_url step otherwise" do
+        step = @app.fetch_url('egg.heads').process(:cheese).fetch_url_step
+        step.should be_a(Dragonfly::Job::FetchUrl)
+        step.url.should == 'http://egg.heads'
+      end
+    end
+
+    describe "generate_step" do
+      it "should return nil if it doesn't exist" do
+        @app.fetch('many/ponies').process(:jam).generate_step.should be_nil
+      end
+      it "should return the generate step otherwise" do
+        step = @app.generate(:ponies).process(:cheese).generate_step
+        step.should be_a(Dragonfly::Job::Generate)
+        step.args.should == [:ponies]
+      end
+    end
+    
+    describe "process_steps" do
+      it "should return the processing steps" do
+        job = @app.fetch('many/ponies').process(:jam).process(:eggs).encode(:gif)
+        job.process_steps.should match_steps([
+          Dragonfly::Job::Process,
+          Dragonfly::Job::Process
+        ])
+      end
+    end
+
+    describe "encode_step" do
+      it "should return nil if it doesn't exist" do
+        @app.generate(:ponies).encode_step.should be_nil
+      end
+      it "should return the last encode step otherwise" do
+        step = @app.fetch('hello').encode(:smells).encode(:cheese).encode_step
+        step.should be_a(Dragonfly::Job::Encode)
+        step.format.should == :cheese
+      end
+    end
+    
+    describe "step_types" do
+      it "should return the step types" do
+        job = @app.fetch('eggs').process(:beat, 'strongly').encode(:meringue)
+        job.step_types.should == [:fetch, :process, :encode]
+      end
+    end
+  end
+
+  describe "meta" do
+    before(:each) do
+      @app = test_app
+      @app.generator.add(:gollum){|t| "OK"}
+      @job = @app.new_job("Goo")
+    end
+    it "should default meta to an empty hash" do
+      @job.meta.should == {}
+    end
+    it "should allow setting" do
+      @job.meta = {:a => :b}
+      @job.meta.should == {:a => :b}
+    end
+    it "should apply the job" do
+      @job.should_receive :apply
+      @job.meta
+    end
+    it "should apply the job before setting (for consistency)" do
+      @job.should_receive :apply
+      @job.meta = {}
+    end
+    it "should allow setting on initialize" do
+      job = @app.new_job('asdf', :b => :c)
+      job.meta.should == {:b => :c}
+    end
+    it "should keep them when chained" do
+      @job.meta[:darn] = 'it'
+      @job.generate(:gollum).meta.should == {:darn => 'it'}
+    end
+  end
+
+  describe "sanity check for name, basename, ext" do
+    before(:each) do
+      @app = test_app
+      @job = @app.new_job('asdf')
+    end
+
+    it "should default to nil" do
+      @job.name.should be_nil
+    end
+
+    it "reflect the meta" do
+      @job.meta[:name] = 'monkey.egg'
+      @job.name.should == 'monkey.egg'
+      @job.basename.should == 'monkey'
+      @job.ext.should == 'egg'
+    end
+  end
+
+  describe "format" do
+    before(:each) do
+      @app = test_app
+    end
+    it "should default to nil" do
+      job = @app.new_job("HELLO")
+      job.format.should be_nil
+    end
+    it "should use the meta format if it exists" do
+      job = @app.new_job("HELLO")
+      job.meta[:format] = :txt
+      job.format.should == :txt
+    end
+    it "should use the analyser format if it exists" do
+      @app.analyser.add :format do |temp_object|
+        :egg
+      end
+      job = @app.new_job("HELLO")
+      job.format.should == :egg
+    end
+    it "should use the file extension if it has no format" do
+      job = @app.new_job("HIMATE", :name => 'test.pdf')
+      job.format.should == :pdf
+    end
+    it "should not use the file extension if it's been switched off" do
+      @app.trust_file_extensions = false
+      job = @app.new_job("HIMATE", :name => 'test.pdf')
+      job.format.should be_nil
+    end
+    it "should prefer the set format over the file extension" do
+      job = @app.new_job("HELLO", :name => 'test.pdf', :format => :txt)
+      job.format.should == :txt
+    end
+    it "should prefer the file extension over the analysed format" do
+      @app.analyser.add :format do |temp_object|
+        :egg
+      end
+      job = @app.new_job("HELLO", :name => 'test.pdf')
+      job.format.should == :pdf
+    end
+    it "should apply the job" do
+      @app.generator.add(:test){ ["skid marks", {:name => 'terry.burton'}] }
+      job = @app.generate(:test)
+      job.format.should == :burton
+      job.should be_applied
+    end
+  end
+
+  describe "mime_type" do
+    before(:each) do
+      @app = test_app
+    end
+    it "should return the correct mime_type if the format is given" do
+      @job = @app.new_job("HIMATE")
+      @job.should_receive(:format).and_return(:tiff)
+      @job.mime_type.should == 'image/tiff'
+    end
+    it "should fall back to the mime_type analyser if the format is nil" do
+      @app.analyser.add :mime_type do |temp_object|
+        'image/jpeg'
+      end
+      @job = @app.new_job("HIMATE")
+      @job.mime_type.should == 'image/jpeg'
+    end
+    it "should fall back to the fallback mime_type if neither format or analyser exist" do
+      @app.new_job("HIMATE").mime_type.should == 'application/octet-stream'
+    end
+  end
+
+  describe "store" do
+    before(:each) do
+      @app = test_app
+      @app.generator.add(:test){ ["Toes", {:name => 'doogie.txt'}] }
+      @job = @app.generate(:test)
+    end
+    it "should store its data along with the meta and mime_type" do
+      @job.meta[:eggs] = 'doolally'
+      @app.datastore.should_receive(:store).with do |temp_object, opts|
+        temp_object.data.should == "Toes"
+        temp_object.name.should == 'doogie.txt'
+        temp_object.meta[:eggs].should == 'doolally'
+        opts[:mime_type].should == 'text/plain'
+      end
+      @job.store
+    end
+    it "should add extra opts" do
+      @app.datastore.should_receive(:store).with(anything, hash_including(:path => 'blah', :mime_type => 'text/plain'))
+      @job.store(:path => 'blah')
+    end
+    it "should pass in its meta, for deprecated datastores" do
+      @job.meta[:beans] = 5
+      @app.datastore.should_receive(:store).with do |temp_object, opts|
+        opts[:meta].should == {:beans => 5, :name => 'doogie.txt'}
+      end
+      @job.store
+    end
+  end
+
+  describe "dealing with original_filename" do
+    before(:each) do
+      @string = "terry"
+      @string.stub!(:original_filename).and_return("gum.tree")
+      @app = test_app
+      @app.generator.add(:test){ @string }
+    end
+    it "should set it as the name" do
+      @app.create(@string).name.should == 'gum.tree'
+    end
+    it "should prefer the initialized name over the original_filename" do
+      @app.create(@string, :name => 'doo.berry').name.should == 'doo.berry'
+    end
+    it "should work with e.g. generators" do
+      @app.generate(:test).apply.name.should == 'gum.tree'
+    end
+    it "should favour an e.g. generator returned name" do
+      @app.generator.add(:test2){ [@string, {:name => 'gen.ome'}] }
+      @app.generate(:test2).apply.name.should == 'gen.ome'
+    end
+    it "should not overwrite a set name" do
+      job = @app.generate(:test)
+      job.name = 'egg.mumma'
+      job.apply.name.should == 'egg.mumma'
+    end
+  end
+
+  describe "deprecated meta format" do
+    before(:each) do
+      @app = test_app
+    end
+    it "should still work if the datastore/whatever returns meta nested in :meta key" do
+      @app.datastore.should_receive(:retrieve).with('some_uid').and_return(['HELLO', {:name => 'test.txt', :meta => {:some => 'meta'}}])
+      job = @app.fetch('some_uid').apply
+      job.meta.should == {:name => 'test.txt', :some => 'meta'}
+    end
+  end
+
+  describe "close" do
+    before(:each) do
+      @app = test_app
+      @app.generator.add(:toast){ "toast" }
+      @app.processor.add(:upcase){|t| t.data.upcase }
+      @job = @app.generate(:toast)
+      @path1 = @job.tempfile.path
+      @job.process!(:upcase)
+      @path2 = @job.tempfile.path
+    end
+
+    it "should clean up tempfiles for the last temp_object" do
+      File.exist?(@path2).should be_true
+      @job.close
+      File.exist?(@path2).should be_false
+    end
+
+    it "should clean up tempfiles for previous temp_objects" do
+      File.exist?(@path1).should be_true
+      @job.close
+      File.exist?(@path1).should be_false
+    end
+  end
+
+end