class AWS::S3::Base
AWS::S3::Base is the abstract super class of all classes who make requests against S3, such as the built in Service, Bucket and S3Object classes. It provides methods for making requests, inferring or setting response classes, processing request options, and accessing attributes from S3's response data.
Establishing a connection with the Base class is the entry point to using the library:
AWS::S3::Base.establish_connection!(:access_key_id => '...', :secret_access_key => '...')
The :access_key_id
and :secret_access_key
are the
two required connection options. More details can be found in the docs for
Connection::Management::ClassMethods.
Extensive examples can be found in the README.
Attributes
Public Class Methods
Called when a method which requires a bucket name is called without that bucket name specified. It will try to infer the current bucket by looking for it as the subdomain of the current connection's address. If no subdomain is found, CurrentBucketNotSpecified will be raised.
MusicBucket.establish_connection! :server => 'jukeboxzero.s3.amazonaws.com' MusicBucket.connection.server => 'jukeboxzero.s3.amazonaws.com' MusicBucket.current_bucket => 'jukeboxzero'
Rather than infering the current bucket from the subdomain, the current class' bucket can be explicitly set with set_current_bucket_to.
# File lib/aws/s3/base.rb, line 106 def current_bucket connection.subdomain or raise CurrentBucketNotSpecified.new(connection.http.address) end
Wraps the current connection's request method and picks the appropriate response class to wrap the response in. If the response is an error, it will raise that error as an exception. All such exceptions can be caught by rescuing their superclass, the ResponseError exception class.
It is unlikely that you would call this method directly. Subclasses of Base have convenience methods for each http request verb that wrap calls to request.
# File lib/aws/s3/base.rb, line 66 def request(verb, path, options = {}, body = nil, attempts = 0, &block) Service.response = nil process_options!(options, verb) response = response_class.new(connection.request(verb, path, options, body, attempts, &block)) Service.response = response Error::Response.new(response.response).error.raise if response.error? response # Once in a while, a request to S3 returns an internal error. A glitch in the matrix I presume. Since these # errors are few and far between the request method will rescue InternalErrors the first three times they encouter them # and will retry the request again. Most of the time the second attempt will work. rescue InternalError, RequestTimeout if attempts == 3 raise else attempts += 1 retry end end
If you plan on always using a specific bucket for certain files, you can skip always having to specify the bucket by creating a subclass of Bucket or S3Object and telling it what bucket to use:
class JukeBoxSong < AWS::S3::S3Object set_current_bucket_to 'jukebox' end
For all methods that take a bucket name as an argument, the current bucket will be used if the bucket name argument is omitted.
other_song = 'baby-please-come-home.mp3' JukeBoxSong.store(other_song, open(other_song))
This time we didn't have to explicitly pass in the bucket name, as the JukeBoxSong class knows that it will always use the 'jukebox' bucket.
“Astute readers”, as they say, may have noticed that we used the third parameter to pass in the content type, rather than the fourth parameter as we had the last time we created an object. If the bucket can be inferred, or is explicitly set, as we've done in the JukeBoxSong class, then the third argument can be used to pass in options.
Now all operations that would have required a bucket name no longer do.
other_song = JukeBoxSong.find('baby-please-come-home.mp3')
# File lib/aws/s3/base.rb, line 133 def set_current_bucket_to(name) raise ArgumentError, "`#{__method__}' must be called on a subclass of #{self.name}" if self == AWS::S3::Base instance_eval(" def current_bucket '#{name}' end ") end
Private Class Methods
# File lib/aws/s3/base.rb, line 178 def bucket_name(name) name || current_bucket end
# File lib/aws/s3/base.rb, line 149 def process_options!(options, verb) options.replace(RequestOptions.process(options, verb)) end
Using the conventions layed out in the response_class
works
for more than 80% of the time. There are a few edge cases though where we
want a given class to wrap its responses in different response classes
depending on which method is being called.
# File lib/aws/s3/base.rb, line 156 def respond_with(klass) eval(" def new_response_class #{klass} end class << self alias_method :old_response_class, :response_class alias_method :response_class, :new_response_class end ", binding, __FILE__, __LINE__) yield ensure # Restore the original version eval(" class << self alias_method :response_class, :old_response_class end ", binding, __FILE__, __LINE__) end
# File lib/aws/s3/base.rb, line 145 def response_class FindResponseClass.for(self) end
Private Instance Methods
# File lib/aws/s3/base.rb, line 216 def connection self.class.connection end
# File lib/aws/s3/base.rb, line 220 def http connection.http end
# File lib/aws/s3/base.rb, line 228 def method_missing(method, *args, &block) case when attributes.has_key?(method.to_s) attributes[method.to_s] when attributes.has_key?(method) attributes[method] else super end end
# File lib/aws/s3/base.rb, line 224 def request(*args, &block) self.class.request(*args, &block) end