Updated to support EC2 API version 2007-08-29 released on 2007-10-16

* Supports new instances type feature.  Specify an instance type to launch
  (m1.small, m1.large, m1.xlarge) when you call the 'run_instances' method.
  e.g. add to the params hash ':instance_type => "m1.small"'
* RunInstances and DescribeInstances now return the time when the Amazon EC2 instance was launched.
* Updated tests.
* Code Tidyup : Removed trailing spaces in code docs.


git-svn-id: svn+ssh://rubyforge.org/var/svn/amazon-ec2/trunk@103 2e111c3a-5438-44ef-9c14-21a960333853

amazon-ec2/History.txt

@@ -1,3 +1,12 @@
+=== 0.2.6 2007-10-16
+ * Updated to support EC2 API version 2007-08-29 released on 2007-10-16
+ * Supports new instances type feature.  Specify an instance type to launch
+   (m1.small, m1.large, m1.xlarge) when you call the 'run_instances' method.
+
+   e.g. add to the params hash ':instance_type => "m1.small"'
+  * RunInstances and DescribeInstances now return the time when the Amazon EC2 instance was launched.
+  * Code Tidyup : Removed trailing spaces in code docs.
+
 === 0.2.5 2007-09-26
  * Updated using Dr. Nic's newgem v. 0.13.5
  * Updated email address.

amazon-ec2/Manifest.txt

@@ -37,8 +37,6 @@ test/test_EC2_responses.rb
 test/test_EC2_security_groups.rb
 test/test_EC2_version.rb
 test/test_helper.rb
-website/announce.html
-website/announce.txt
 website/index.html
 website/index.txt
 website/javascripts/rounded_corners_lite.inc.js

amazon-ec2/README.txt

@@ -137,7 +137,7 @@ Try out the following bit of code.  This should walk through each image returned
 
 === Related Projects
 
-* Capazon : http://capazon.rubyforge.org
+* Capsize : http://capsize.rubyforge.org
 
 == Credits
 

amazon-ec2/lib/EC2.rb

@@ -12,25 +12,25 @@
 
 # Require any lib files that we have bundled with this Ruby Gem in the lib/EC2 directory.
 # Parts of the EC2 module and Base class are broken out into separate
-# files for maintainability and are organized by the functional groupings defined 
+# files for maintainability and are organized by the functional groupings defined
 # in the EC2 API developers guide.
 Dir[File.join(File.dirname(__FILE__), 'EC2/**/*.rb')].sort.each { |lib| require lib }
 
 module EC2
-  
+
   # Which host FQDN will we connect to for all API calls to AWS?
   DEFAULT_HOST = 'ec2.amazonaws.com'
-  
+
   # This is the version of the API as defined by Amazon Web Services
-  API_VERSION = '2007-03-01'
-  
+  API_VERSION = '2007-08-29'
+
   # This release version is passed in with each request as part
-  # of the HTTP 'User-Agent' header.  Set this be the same value 
+  # of the HTTP 'User-Agent' header.  Set this be the same value
   # as what is stored in the lib/EC2/version.rb module constant instead.
-  # This way we keep it nice and DRY and only have to define the 
-  # version number in a single place. 
+  # This way we keep it nice and DRY and only have to define the
+  # version number in a single place.
   RELEASE_VERSION = EC2::VERSION::STRING
-  
+
   # Builds the canonical string for signing. This strips out all '&', '?', and '='
   # from the query string to be signed.
   #   Note:  The parameters in the path passed in must already be sorted in
@@ -42,7 +42,7 @@ def EC2.canonical_string(path)
     }
     return buf
   end
-  
+
   # Encodes the given string with the secret_access_key, by taking the
   # hmac-sha1 sum, and then base64 encoding it.  Optionally, it will also
   # url encode the result of that to protect the string if it's going to
@@ -52,20 +52,20 @@ def EC2.encode(secret_access_key, str, urlencode=true)
     b64_hmac =
       Base64.encode64(
         OpenSSL::HMAC.digest(digest, secret_access_key, str)).strip
-        
+
     if urlencode
       return CGI::escape(b64_hmac)
     else
       return b64_hmac
     end
   end
-  
-  
+
+
   #Introduction:
   #
   # The library exposes one main interface class, 'EC2::Base'.
-  # This class provides all the methods for using the EC2 service 
-  # including the handling of header signing and other security issues .  
+  # This class provides all the methods for using the EC2 service
+  # including the handling of header signing and other security issues .
   # This class uses Net::HTTP to interface with the EC2 Query API interface.
   #
   #Required Arguments:
@@ -79,27 +79,27 @@ def EC2.encode(secret_access_key, str, urlencode=true)
   # :server => String (default : 'ec2.amazonaws.com')
   #
   class Base
-    
+
     attr_reader :use_ssl, :server, :port
-    
+
     def initialize( options = {} )
-      
+
       options = { :access_key_id => "",
                   :secret_access_key => "",
                   :use_ssl => true,
                   :server => DEFAULT_HOST
                   }.merge(options)
-                  
+
       @server = options[:server]
       @use_ssl = options[:use_ssl]
-      
+
       raise ArgumentError, "No :access_key_id provided" if options[:access_key_id].nil? || options[:access_key_id].empty?
       raise ArgumentError, "No :secret_access_key provided" if options[:secret_access_key].nil? || options[:secret_access_key].empty?
       raise ArgumentError, "No :use_ssl value provided" if options[:use_ssl].nil?
       raise ArgumentError, "Invalid :use_ssl value provided, only 'true' or 'false' allowed" unless options[:use_ssl] == true || options[:use_ssl] == false
       raise ArgumentError, "No :server provided" if options[:server].nil? || options[:server].empty?
-      
-      
+
+
       # based on the :use_ssl boolean, determine which port we should connect to
       case @use_ssl
       when true
@@ -109,24 +109,24 @@ def initialize( options = {} )
         # http
         @port = 80
       end
-      
+
       @access_key_id = options[:access_key_id]
       @secret_access_key = options[:secret_access_key]
       @http = Net::HTTP.new(options[:server], @port)
       @http.use_ssl = @use_ssl
-      
+
       # Don't verify the SSL certificates.  Avoids SSL Cert warning in log on every GET.
       @http.verify_mode = OpenSSL::SSL::VERIFY_NONE
-      
+
     end
-    
-    
+
+
     private
-      
+
       # pathlist is a utility method which takes a key string and and array as input.
       # It converts the array into a Hash with the hash key being 'Key.n' where
-      # 'n' increments by 1 for each iteration.  So if you pass in args 
-      # ("ImageId", ["123", "456"]) you should get 
+      # 'n' increments by 1 for each iteration.  So if you pass in args
+      # ("ImageId", ["123", "456"]) you should get
       # {"ImageId.1"=>"123", "ImageId.2"=>"456"} returned.
       def pathlist(key, arr)
         params = {}
@@ -135,81 +135,81 @@ def pathlist(key, arr)
         end
         params
       end
-      
-      
-      # Make the connection to AWS EC2 passing in our request.  This is generally called from 
+
+
+      # Make the connection to AWS EC2 passing in our request.  This is generally called from
       # within a 'Response' class object or one of its sub-classes so the response is interpreted
       # in its proper context.  See lib/EC2/responses.rb
       def make_request(action, params, data='')
-        
+
         @http.start do
-          
+
           # remove any keys that have nil or empty values
           params.reject! { |key, value| value.nil? or value.empty?}
-          
+
           params.merge!( {"Action" => action,
                           "SignatureVersion" => "1",
                           "AWSAccessKeyId" => @access_key_id,
                           "Version" => API_VERSION,
                           "Timestamp"=>Time.now.getutc.iso8601} )
-          
+
           sigpath = "?" + params.sort_by { |param| param[0].downcase }.collect { |param| param.join("=") }.join("&")
-          
+
           sig = get_aws_auth_param(sigpath, @secret_access_key)
-          
+
           path = "?" + params.sort.collect do |param|
             CGI::escape(param[0]) + "=" + CGI::escape(param[1])
           end.join("&") + "&Signature=" + sig
-          
+
           req = Net::HTTP::Get.new("/#{path}")
-          
+
           # Ruby will automatically add a random content-type on some verbs, so
           # here we add a dummy one to 'supress' it.  Change this logic if having
           # an empty content-type header becomes semantically meaningful for any
           # other verb.
           req['Content-Type'] ||= ''
           req['User-Agent'] = "rubyforge-amazon-ec2-ruby-gem-query-api v-#{RELEASE_VERSION}"
-          
+
           #data = nil unless req.request_body_permitted?
           response = @http.request(req, nil)
-          
+
           # Make a call to see if we need to throw an error based on the response given by EC2
           # All error classes are defined in EC2/exceptions.rb
           ec2_error?(response)
-          
+
           return response
-          
+
         end
-        
+
       end
-      
+
       # Set the Authorization header using AWS signed header authentication
       def get_aws_auth_param(path, secret_access_key)
         canonical_string =  EC2.canonical_string(path)
         encoded_canonical = EC2.encode(secret_access_key, canonical_string)
       end
-      
+
       # allow us to have a one line call in each method which will do all of the work
       # in making the actual request to AWS.
       def response_generator( options = {} )
-        
+
         options = {
           :action => "",
           :params => {}
         }.merge(options)
-        
+
         raise ArgumentError, ":action must be provided to response_generator" if options[:action].nil? || options[:action].empty?
-        
+
         http_response = make_request(options[:action], options[:params])
         http_xml = http_response.body
         return Response.parse(:xml => http_xml)
-        
+
       end
-      
+
       # Raises the appropriate error if the specified Net::HTTPResponse object
       # contains an Amazon EC2 error; returns +false+ otherwise.
       def ec2_error?(response)
-        
+
         # return false if we got a HTTP 200 code,
         # otherwise there is some type of error (40x,50x) and
         # we should try to raise an appropriate exception
@@ -219,31 +219,31 @@ def ec2_error?(response)
 
         # parse the XML document so we can walk through it
         doc = REXML::Document.new(response.body)
-        
+
         # Check that the Error element is in the place we would expect.
         # and if not raise a generic error exception
         unless doc.root.elements['Errors'].elements['Error'].name == 'Error'
           raise Error, "Unexpected error format. response.body is: #{response.body}"
         end
-        
+
         # An valid error response looks like this:
         # <?xml version="1.0"?><Response><Errors><Error><Code>InvalidParameterCombination</Code><Message>Unknown parameter: foo</Message></Error></Errors><RequestID>291cef62-3e86-414b-900e-17246eccfae8</RequestID></Response>
         # AWS EC2 throws some exception codes that look like Error.SubError.  Since we can't name classes this way
         # we need to strip out the '.' in the error 'Code' and we name the error exceptions with this
         # non '.' name as well.
         error_code    = doc.root.elements['Errors'].elements['Error'].elements['Code'].text.gsub('.', '')
         error_message = doc.root.elements['Errors'].elements['Error'].elements['Message'].text
-        
+
         # Raise one of our specific error classes if it exists.
         # otherwise, throw a generic EC2 Error with a few details.
         if EC2.const_defined?(error_code)
           raise EC2.const_get(error_code), error_message
         else
           raise Error, "This is an undefined error code which needs to be added to exceptions.rb : error_code => #{error_code} : error_message => #{error_message}"
         end
-        
+
       end
-      
+
   end
-  
+
 end

amazon-ec2/lib/EC2/console.rb

@@ -9,15 +9,15 @@
 #++
 
 module EC2
-  
+
   class Base
-    
+
     #Amazon Developer Guide Docs:
-    #  
+    #
     # The GetConsoleOutput operation retrieves console output that has been posted for the specified instance.
     #
-    # Instance console output is buffered and posted shortly after instance boot, reboot and once the instance 
-    # is terminated. Only the most recent 64 KB of posted output is available. Console output is available for 
+    # Instance console output is buffered and posted shortly after instance boot, reboot and once the instance
+    # is terminated. Only the most recent 64 KB of posted output is available. Console output is available for
     # at least 1 hour after the most recent post.
     #
     #Required Arguments:
@@ -29,16 +29,16 @@ class Base
     # none
     #
     def get_console_output( options ={} )
-      
+
       options = {:instance_id => ""}.merge(options)
-      
+
       raise ArgumentError, "No instance ID provided" if options[:instance_id].nil? || options[:instance_id].empty?
-      
+
       params = { "InstanceId" => options[:instance_id] }
-      
+
       return response_generator(:action => "GetConsoleOutput", :params => params)
-      
+
     end
   end
-  
+
 end