codez-tarantula 0.5.0

data/spec/relevance/tarantula/log_grabber_spec.rb

@@ -0,0 +1,26 @@
+require "spec_helper"
+
+describe 'Relevance::Tarantula::LogGrabber' do
+  before do
+    @grabber = Relevance::Tarantula::LogGrabber.new(log_file)
+    FileUtils.mkdir_p(test_output_dir)
+  end
+
+  def log_file
+    File.join(File.join(test_output_dir, "example.log"))
+  end
+
+  it "can clear the log file" do
+    File.open(log_file, "w") {|f| f.print "sample log"}
+    File.size(log_file).should == 10
+    @grabber.clear!
+    File.size(log_file).should == 0
+  end
+
+  it "can grab the log file" do
+    File.open(log_file, "w") {|f| f.print "sample log"}
+    @grabber.grab!.should == "sample log"
+    File.size(log_file).should == 0
+  end
+
+end

data/spec/relevance/tarantula/rails_integration_proxy_spec.rb

@@ -0,0 +1,100 @@
+require "spec_helper"
+
+describe "Relevance::Tarantula::RailsIntegrationProxy rails_integration_test" do
+  before {
+    Relevance::Tarantula::Crawler.any_instance.stubs(:crawl)
+    Relevance::Tarantula::Crawler.any_instance.stubs(:rails_root).returns("STUB_RAILS_ROOT")
+    Relevance::Tarantula::RailsIntegrationProxy.stubs(:rails_root).returns("STUB_RAILS_ROOT")
+    Relevance::Tarantula::RailsIntegrationProxy.stubs(:new).returns(stub(:integration_test => stub(:method_name => @test_name)))
+    @test_name = "test_user_pages"
+  }
+
+  it "strips leading hostname from link urls" do    
+    crawler = Relevance::Tarantula::RailsIntegrationProxy.rails_integration_test(stub(:host => "foo.com"))
+    crawler.transform_url("http://foo.com/path").should == "/path"
+    crawler.transform_url("http://bar.com/path").should == "http://bar.com/path"
+  end
+  
+  it "allows override of max_url_length" do
+    crawler = Relevance::Tarantula::RailsIntegrationProxy.rails_integration_test(stub(:host => "foo.com"), 
+                                             :max_url_length => 16)
+    crawler.max_url_length.should == 16
+  end
+
+  it "has some useful defaults" do
+    crawler = Relevance::Tarantula::RailsIntegrationProxy.rails_integration_test(stub(:host => "foo.com")) 
+    crawler.log_grabber.should_not be_nil
+  end
+end
+
+
+describe "Relevance::Tarantula::RailsIntegrationProxy" do
+  %w{get post}.each do |http_method|
+    it "can #{http_method}" do
+      @rip = Relevance::Tarantula::RailsIntegrationProxy.new(stub)
+      @response = stub({:code => :foo})
+      @rip.integration_test = stub_everything(:response => @response)
+      @rip.send(http_method, "/url").should == @response
+    end
+  end
+  
+  it "adds a response accessor to its delegate rails integration test" do
+    o = Object.new
+    Relevance::Tarantula::RailsIntegrationProxy.new(o)
+    o.methods(false).map(&:to_s).sort.should == %w{response response=}
+  end
+end
+
+describe "Relevance::Tarantula::RailsIntegrationProxy errors" do
+  it "alters response to contain the error message" do
+    @rip = Relevance::Tarantula::RailsIntegrationProxy.new(stub)
+    @response = Struct.new(:body, :code).new(nil, nil)
+    @rip.integration_test = stub_everything(:response => @response) do |stub|
+      stub.stubs(:get).raises(Exception, 'Internal Server Error')
+    end
+    response = @rip.get("/url")
+    response.code.should == '500'
+    response.body.should =~ /\AInternal Server Error\n\n/
+  end
+end
+
+describe "Relevance::Tarantula::RailsIntegrationProxy patching" do
+  before do
+    @rip = Relevance::Tarantula::RailsIntegrationProxy.new(stub)
+    @rip.stubs(:rails_root).returns("faux_rails_root")
+    @response = stub_everything({:code => "404", :headers => {}})
+    File.stubs(:exist?).returns(true)
+  end
+  
+  it "patches in Relevance::CoreExtensions::Response" do
+    @rip = Relevance::Tarantula::RailsIntegrationProxy.new(stub)
+    @rip.stubs(:rails_root).returns("faux_rails_root")
+    @response = stub_everything({:code => "404", :headers => {}, :content_type => Mime::Type.new("text/html")})
+    @response.meta.ancestors.should_not include(Relevance::CoreExtensions::Response)
+    @rip.patch_response("/url", @response)
+    @response.meta.ancestors.should include(Relevance::CoreExtensions::Response)
+    @response.should be_html
+  end
+    
+  it "replaces 404s with 200s, pulling content from public, for known text types" do
+    File.expects(:extension).returns("html")
+    @rip.expects(:static_content_file).with("/url").returns("File body")
+    @rip.patch_response("/url", @response)
+    @response.headers.should == {"type" => "text/html"}
+  end
+  
+  it "logs and skips types we haven't dealt with yet" do
+    File.expects(:extension).returns("whizzy")
+    @rip.expects(:log).with("Skipping unknown type /url")
+    @rip.patch_response("/url", @response)
+  end
+  
+  it "can find static content relative to rails root" do
+    @rip.static_content_path("foo").should == File.expand_path("faux_rails_root/public/foo")
+  end
+  
+  it "can read static content relative to rails root" do
+    File.expects(:read).with(@rip.static_content_path("foo"))
+    @rip.static_content_file("foo")
+  end
+end

data/spec/relevance/tarantula/result_spec.rb

@@ -0,0 +1,85 @@
+require "spec_helper"
+
+describe "Relevance::Tarantula::Result" do
+  before do
+    @result = Relevance::Tarantula::Result.new(
+        :success => true, 
+        :method => "get", 
+        :url => "/some/url?arg1=foo&arg2=bar"
+    )
+  end
+  
+  it "has a short description" do
+    @result.short_description.should == "get /some/url?arg1=foo&arg2=bar"
+  end
+  
+  it "has a sequence number" do
+    @result.class.next_number = 0
+    @result.sequence_number.should == 1
+    @result.class.next_number.should == 1
+  end
+  
+  it "has link to the url at localhost" do
+    @result.full_url.should == "http://localhost:3000/some/url?arg1=foo&arg2=bar"
+  end
+  
+end
+
+describe "Relevance::Tarantula::Result class methods" do
+  before do
+    @rh = Relevance::Tarantula::Result
+  end
+  
+  it "defines HTTP responses that are considered 'successful' when spidering" do
+    %w{200 201 302 401}.each do |code|
+      @rh.successful?(stub(:code => code)).should == true
+    end
+  end
+  
+  it "adds successful responses to success collection" do
+    stub = stub_everything(:code => "200")
+    @rh.handle(Relevance::Tarantula::Result.new(:response => stub)).success.should == true
+  end
+
+  it "adds failed responses to failure collection" do
+    stub = stub_everything(:code => "500")
+    result = @rh.handle(Relevance::Tarantula::Result.new(:response => stub))
+    result.success.should == false
+    result.description.should == "Bad HTTP Response"
+  end
+  
+end
+
+describe "Relevance::Tarantula::Result allowed errors" do
+  before do
+    Relevance::Tarantula::Result.allow_errors_for = {}
+  end
+  
+  it "defaults to *not* skip errors" do
+    Relevance::Tarantula::Result.can_skip_error?(stub(:code => "404")).should == false
+  end
+
+  it "can skip errors matching code and url" do
+    Relevance::Tarantula::Result.allow_errors_for = {"404" => [/some_url/]}
+    Relevance::Tarantula::Result.can_skip_error?(stub(:code => "404", :url => "this_is_some_url")).should == true
+  end
+
+  it "does not skip errors matching code only" do
+    Relevance::Tarantula::Result.allow_errors_for = {"404" => [/some_other_url/]}
+    Relevance::Tarantula::Result.can_skip_error?(stub(:code => "404", :url => "this_is_some_url")).should == false
+  end
+  
+  it "users allow_nnn_for syntax to specify allowed errors" do
+    Relevance::Tarantula::Result.allow_404_for(/this_url/)
+    Relevance::Tarantula::Result.allow_errors_for.should == {"404" => [/this_url/]}
+    Relevance::Tarantula::Result.allow_404_for(/another_url/)
+    Relevance::Tarantula::Result.allow_errors_for.should == {"404" => [/this_url/, /another_url/]}
+  end
+  
+  it "chains to super method missing" do
+    lambda{Relevance::Tarantula::Result.allow_xxx_for}.should raise_error(NoMethodError)
+  end
+  
+end
+
+

data/spec/relevance/tarantula/tidy_handler_spec.rb

@@ -0,0 +1,58 @@
+require "spec_helper"
+
+if defined?(Tidy) && ENV['TIDY_PATH']
+  describe "Relevance::Tarantula::TidyHandler default" do
+    before do
+      @handler = Relevance::Tarantula::TidyHandler.new
+    end
+  
+    it "likes a good document" do
+      response = stub(:html? => true, :body => <<-BODY, :code => 200)
+<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 3.2//EN\">
+<html>
+  <title></title>
+  <body></body>
+</html>
+BODY
+      @handler.handle(Result.new(:response => response)).should == nil
+    end
+
+    it "rejects a document with errors" do
+      response = stub(:html? => true, :body => "<hotml>", :code => 200)
+      result = @handler.handle(Result.new(:response => response))
+      result.should.not.be nil
+      result.data.should =~ /Error: <hotml> is not recognized!/
+      result.description.should == "Bad HTML (Tidy)"
+    end
+
+    it "rejects a document with warnings" do
+      response = stub(:html? => true, :body => <<-BODY, :code => 200)
+<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 3.2//EN\">
+<html>
+</html>
+BODY
+      result = @handler.handle(Result.new(:response => response))
+      result.should.not.be nil
+      result.data.should =~ /Warning: inserting missing 'title' element/
+    end
+    
+  end
+  
+  describe "Relevance::Tarantula::TidyHandler with :show_warnings => false" do
+    before do
+      @handler = Relevance::Tarantula::TidyHandler.new(:show_warnings => false)
+    end
+
+    it "permits a document with warnings" do
+      response = stub(:html? => true, :body => <<-BODY, :code => 200)
+<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 3.2//EN\">
+<html>
+</html>
+BODY
+      result = @handler.handle(Result.new(:response => response))
+      result.should.be nil
+    end
+  end
+else
+  puts "TIDY_PATH not set. Tidy test will not run"
+end

data/spec/relevance/tarantula/transform_spec.rb

@@ -0,0 +1,20 @@
+require "spec_helper"
+
+describe "Relevance::Tarantula::Transform" do
+  it "can do a simple replace" do
+    t = Relevance::Tarantula::Transform.new(/\w/, ".")
+    t["hello world"].should == "..... ....."
+  end
+  
+  it "can do a replace with a block" do
+    t = Relevance::Tarantula::Transform.new(/([aeiou])/, Proc.new {|value| value.upcase})
+    t["hello world"].should == "hEllO wOrld"
+  end
+
+  # this is broken in Ruby?
+  it "cannot access groups from a block, despite Ruby docs" do
+    p = Proc.new {|value| $1.upcase}
+    t = Relevance::Tarantula::Transform.new(/([aeiou])/, p)
+    lambda {t["hello world"]}.should raise_error(NoMethodError)
+  end
+end

data/spec/relevance/tarantula_spec.rb

@@ -0,0 +1,23 @@
+require "spec_helper"
+
+describe Relevance::Tarantula do
+  include Relevance::Tarantula
+  attr_accessor :verbose
+  
+  it "writes to stdout if verbose" do
+    self.verbose = true
+    expects(:puts).with("foo")
+    log("foo")
+  end
+
+  it "swallows output if !verbose" do
+    self.verbose = false
+    expects(:puts).never
+    log("foo")
+  end
+  
+  it "puts RAILS_ROOT behind a method call" do
+    lambda{rails_root}.should raise_error(NameError)
+  end
+end
+

data/spec/spec_helper.rb

@@ -0,0 +1,43 @@
+if rails_version = ENV['RAILS_VERSION']
+  require 'rubygems'
+  gem "rails", rails_version
+end
+
+require "rails/version"
+puts "==== Testing with Rails #{Rails::VERSION::STRING} ===="
+
+require 'relevance/tarantula'
+require 'bundler'
+Bundler.require
+require 'ostruct'
+
+def test_output_dir
+  File.join(File.dirname(__FILE__), "..", "tmp", "test_output")
+end
+
+# TODO change puts/print to use a single method for logging, which will then make the stubbing cleaner
+def stub_puts_and_print(obj)
+  obj.stubs(:puts)
+  obj.stubs(:print)
+end
+
+def make_link(link, crawler=Relevance::Tarantula::Crawler.new, referrer=nil)
+  Relevance::Tarantula::Link.new(link, crawler, referrer)
+end
+
+def make_form(form, crawler=Relevance::Tarantula::Crawler.new, referrer=nil)
+  Relevance::Tarantula::Form.new(form, crawler, referrer)
+end
+
+def not_in_editor?
+  ['TM_MODE', 'EMACS', 'VIM'].all? { |k| !ENV.has_key?(k) }
+end
+
+RSpec.configure do |c|
+  c.alias_example_to :fit, :focused => true
+  c.alias_example_to :xit, :disabled => true
+  c.mock_with :mocha
+  c.color_enabled = not_in_editor?
+  c.filter_run :focused => true
+  c.run_all_when_everything_filtered = true
+end

data/tarantula.gemspec

@@ -0,0 +1,25 @@
+# -*- encoding: utf-8 -*-
+$:.push File.expand_path("../lib", __FILE__)
+require "relevance/tarantula/version"
+
+Gem::Specification.new do |s|
+  s.name        = "codez-tarantula"
+  s.version     = Relevance::Tarantula::VERSION
+  s.authors     = ["Relevance, Inc.", "code!z"]
+  s.email       = ["spam@codez.ch"]
+  s.homepage    = "https://github.com/codez/tarantula"
+  s.summary     = %q{A big hairy fuzzy spider that crawls your site, wreaking havoc}
+  s.description = "Tarantula is a big fuzzy spider. It crawls your Rails 2.3 and 3.x applications, fuzzing data to see what breaks."
+
+  s.files         = `git ls-files`.split("\n")
+  s.test_files    = `git ls-files -- {examples,template}`.split("\n")
+
+  s.add_runtime_dependency "htmlentities", "~> 4.3.0"
+  s.add_runtime_dependency "hpricot", "~> 0.8.4"
+
+  s.add_development_dependency "rspec", "~> 2.12.0"
+  s.add_development_dependency 'rdoc', '~> 3.12.0'
+  s.add_development_dependency "log_buddy", "~> 0.6.0"
+  s.add_development_dependency "mocha", "~> 0.13.2"
+  s.add_development_dependency "rails", ">= 2.3.0"
+end

data/template/tarantula_test.rb

@@ -0,0 +1,22 @@
+require "test_helper"
+require "relevance/tarantula"
+
+class TarantulaTest < ActionController::IntegrationTest
+  # Load enough test data to ensure that there's a link to every page in your
+  # application. Doing so allows Tarantula to follow those links and crawl 
+  # every page.  For many applications, you can load a decent data set by
+  # loading all fixtures.
+  fixtures :all
+
+  def test_tarantula
+    # If your application requires users to log in before accessing certain 
+    # pages, uncomment the lines below and update them to allow this test to
+    # log in to your application.  Doing so allows Tarantula to crawl the 
+    # pages that are only accessible to logged-in users.
+    # 
+    #   post '/session', :login => 'quentin', :password => 'monkey'
+    #   follow_redirect!
+    
+    tarantula_crawl(self)
+  end
+end

data/vendor/xss-shield/MIT-LICENSE

@@ -0,0 +1,20 @@
+Copyright (c) 2007 Trampoline Systems
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

data/vendor/xss-shield/README

@@ -0,0 +1,76 @@
+FIXME: THIS README IS NOT UP-TO-DATE.
+
+This plugin provides XSS protection for views coded in HAML and RHTML.
+
+ERB templates are sometimes used for HTML, and sometimes for
+other kinds of languages (SQL, email templates, YAML etc.).
+XSS Shield protects only those templates with .rhtml extension,
+leaving templates with .erb extension unprotected.
+
+=== Quick start ===
+
+Assuming you're using HAML for all your templates.
+
+* Install plugin.
+* Edit all your layout files and change:
+    = @content_for_layout
+    = yield(:foo) # Foo being usually :js or :css
+  to:
+    = @content_for_layout.mark_as_xss_protected
+    = yield(:foo).mark_as_xss_protected
+* By this point your application should be runnanble,
+  but might need some tweaking here and there to avoid potential
+  double-escaping.
+
+=== How it works ===
+
+It works by subclassing String into SafeString.
+When HAML engine seems a "= foo" fragment it check if result of executing "foo"
+is a SafeString. If it is - it copies it to the output, if it's anything else
+(String, Integer, nil and so on) it HTML-escapes it first.
+
+To avoid double-escaping output of h is a SafeString, as is everything you
+mark as XSS-protected.
+  = h(@foo)
+  = @foo # fully equivalent to h(@foo)
+  = "X <br /> Y".mark_as_xss_protected
+
+It would be cumbersome to require mark_as_xss_protected every time you use
+some helper like render :partial or link_to, so some helpers are modified
+to return SafeString.
+
+  = render :partial => "foo"
+  = link_to "Bar", :action => :bar
+
+If you trust your helpers, make them as XSS-protected:
+
+  module Some::Module
+    mark_helpers_as_xss_protected :text_field, :check_box
+  end
+
+Because it is not possible to alter syntactic keywords like yield
+or instance variables like @content_for_layout to mark them automatically
+as secure, layout files need some manual tweaking.
+
+=== Other template engines ===
+
+If a templates uses some templating engine other than HAML or ERB,
+or it uses ERB but has extension .erb not .rhtml, XSS Shield does not protect it.
+
+However some helpers like link_to and button_to are patched by XSS Shield to
+make them more secure, and this extra security will be there even when used
+in an otherwise unprotected context.
+
+For example with XSS shield
+  link_to "A & B", "/foo"
+will return (marked as safe):
+  '<a href="/foo">A &amp; B</a>'
+not (plain String):
+  '<a href="/foo">A & B</a>'
+
+Also - RHTML protection only works with default ERB engine (erb.rb from Ruby base).
+If you use some alternative ERB engine it probably won't work.
+
+Adding support for alternative templating engine should be relatively straightforward.
+It's mostly a matter of changing to_s to to_s_xss_protected in a few places
+in their source.