Revise uses of encoding APIs.

When originally implemented, we still supported Ruby 1.8, which
necessitated checking for encoding methods and using a regex to validate
UTF-8. These checks are now gone.

We tagged many strings as binary when not strictly necessary, either
because we were just going to iterate their bytes or because we were
going to hand them off to the caller which should just write them
directly to a socket. Strings used as buffers to accumulate streaming
input are still tagged as binary to avoid encoding
collision/conversion.

The places where we do need to tag as UTF-8 (i.e. just before emitting
to the application) remain, but copy the string if necessary. This
allows us to work with frozen strings.

Finally, strings passed in via the Driver#text method should be
*transcoded* to UTF-8 if necessary, not merely tagged. The Ruby
String#encode method produces a new string so this should also be safe
with frozen strings.

lib/websocket/driver.rb

@@ -33,13 +33,12 @@ module WebSocket
       end
     end
 
-    unless String.instance_methods.include?(:force_encoding)
-      require root + '/utf8_match'
-    end
-
     MAX_LENGTH = 0x3ffffff
     STATES     = [:connecting, :open, :closing, :closed]
 
+    BINARY  = 'ASCII-8BIT'
+    UNICODE = 'UTF-8'
+
     ConnectEvent = Struct.new(nil)
     OpenEvent    = Struct.new(nil)
     MessageEvent = Struct.new(:data)
@@ -94,13 +93,14 @@ module WebSocket
       return false unless @ready_state == 0
       response = handshake_response
       return false unless response
-      @socket.write(Driver.encode(response, :binary))
+      @socket.write(response)
       open unless @stage == -1
       true
     end
 
     def text(message)
-      frame(message)
+      message = message.encode(UNICODE) unless message.encoding.name == UNICODE
+      frame(message, :text)
     end
 
     def binary(message)
@@ -159,14 +159,15 @@ module WebSocket
       case string
         when Array then
           string = string.pack('C*')
-          encoding ||= :binary
+          encoding ||= BINARY
         when String then
-          encoding ||= :utf8
+          encoding ||= UNICODE
       end
-      encodings = {:utf8 => 'UTF-8', :binary => 'ASCII-8BIT'}
-      string.force_encoding(encodings[encoding]) if string.respond_to?(:force_encoding)
-      return nil if encoding == :utf8 and not valid_utf8?(string)
-      string
+      unless string.encoding.name == encoding
+        string = string.dup if string.frozen?
+        string.force_encoding(encoding)
+      end
+      string.valid_encoding? ? string : nil
     end
 
     def self.validate_options(options, valid_keys)
@@ -177,14 +178,6 @@ module WebSocket
       end
     end
 
-    def self.valid_utf8?(string)
-      if defined?(UTF8_MATCH)
-        UTF8_MATCH =~ string ? true : false
-      else
-        string.valid_encoding?
-      end
-    end
-
     def self.websocket?(env)
       connection = env['HTTP_CONNECTION'] || ''
       upgrade    = env['HTTP_UPGRADE']    || ''

lib/websocket/driver/client.rb

@@ -53,7 +53,7 @@ module WebSocket
 
       def start
         return false unless @ready_state == -1
-        @socket.write(Driver.encode(handshake_request, :binary))
+        @socket.write(handshake_request)
         @ready_state = 0
         true
       end

lib/websocket/driver/draft75.rb

@@ -56,7 +56,7 @@ module WebSocket
             when 2 then
               if octet == 0xFF
                 @stage = 0
-                emit(:message, MessageEvent.new(Driver.encode(@buffer, :utf8)))
+                emit(:message, MessageEvent.new(Driver.encode(@buffer, UNICODE)))
               else
                 if @length
                   @skipped += 1

lib/websocket/driver/draft76.rb

@@ -8,7 +8,7 @@ module WebSocket
         super
         input  = @socket.env['rack.input']
         @stage = -1
-        @body  = Driver.encode(input ? input.read : '', :binary)
+        @body  = (input ? input.read : String.new('')).force_encoding(BINARY)
 
         @headers.clear
         @headers['Upgrade'] = 'WebSocket'
@@ -70,7 +70,7 @@ module WebSocket
 
       def send_handshake_body
         return unless signature = handshake_signature
-        @socket.write(Driver.encode(signature, :binary))
+        @socket.write(signature)
         @stage = 0
         open
         parse(@body[BODY_SIZE..-1]) if @body.bytesize > BODY_SIZE

lib/websocket/driver/hybi.rb

@@ -129,10 +129,6 @@ module WebSocket
         end
       end
 
-      def text(message)
-        frame(message, :text)
-      end
-
       def binary(message)
         frame(message, :binary)
       end
@@ -356,7 +352,7 @@ module WebSocket
 
           when OPCODES[:close] then
             code   = (bytesize >= 2) ? payload.unpack(PACK_FORMATS[2]).first : nil
-            reason = (bytesize > 2)  ? Driver.encode(bytes[2..-1] || [], :utf8) : nil
+            reason = (bytesize > 2)  ? Driver.encode(bytes[2..-1] || [], UNICODE) : nil
 
             unless (bytesize == 0) or
                    (code && code >= MIN_RESERVED_ERROR && code <= MAX_RESERVED_ERROR) or
@@ -374,7 +370,7 @@ module WebSocket
             frame(payload, :pong)
 
           when OPCODES[:pong] then
-            message = Driver.encode(payload, :utf8)
+            message = Driver.encode(payload, UNICODE)
             callback = @ping_callbacks[message]
             @ping_callbacks.delete(message)
             callback.call if callback
@@ -391,7 +387,7 @@ module WebSocket
 
         case message.opcode
           when OPCODES[:text] then
-            payload = Driver.encode(payload, :utf8)
+            payload = Driver.encode(payload, UNICODE)
           when OPCODES[:binary]
             payload = payload.bytes.to_a
         end

lib/websocket/driver/hybi/message.rb

@@ -14,7 +14,7 @@ module WebSocket
           @rsv2   = false
           @rsv3   = false
           @opcode = nil
-          @data   = Driver.encode('', :binary)
+          @data   = String.new('').force_encoding(BINARY)
         end
 
         def <<(frame)

lib/websocket/driver/proxy.rb

@@ -44,7 +44,7 @@ module WebSocket
         start   = "CONNECT #{@origin.host}:#{port} HTTP/1.1"
         headers = [start, @headers.to_s, '']
 
-        @socket.write(Driver.encode(headers.join("\r\n"), :binary))
+        @socket.write(headers.join("\r\n"))
         true
       end
 

lib/websocket/driver/server.rb

@@ -58,7 +58,7 @@ module WebSocket
       end
 
       def write(buffer)
-        @socket.write(Driver.encode(buffer, :binary))
+        @socket.write(buffer)
       end
 
     private

lib/websocket/driver/stream_reader.rb

@@ -6,13 +6,13 @@ module WebSocket
       MINIMUM_AUTOMATIC_PRUNE_OFFSET = 128
 
       def initialize
-        @buffer = Driver.encode('', :binary)
+        @buffer = String.new('').force_encoding(BINARY)
         @offset = 0
       end
 
       def put(chunk)
         return unless chunk and chunk.bytesize > 0
-        @buffer << Driver.encode(chunk, :binary)
+        @buffer << chunk.force_encoding(BINARY)
       end
 
       # Read bytes from the data:
@@ -42,7 +42,7 @@ module WebSocket
         buffer_size = @buffer.bytesize
 
         if @offset > buffer_size
-          @buffer = Driver.encode('', :binary)
+          @buffer = String.new('').force_encoding(BINARY)
         else
           @buffer = @buffer.byteslice(@offset, buffer_size - @offset)
         end

lib/websocket/driver/utf8_match.rb

@@ -1,6 +0,0 @@
-module WebSocket
-  class Driver
-    # http://www.w3.org/International/questions/qa-forms-utf-8.en.php
-    UTF8_MATCH = /^([\x00-\x7F]|[\xC2-\xDF][\x80-\xBF]|\xE0[\xA0-\xBF][\x80-\xBF]|[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}|\xED[\x80-\x9F][\x80-\xBF]|\xF0[\x90-\xBF][\x80-\xBF]{2}|[\xF1-\xF3][\x80-\xBF]{3}|\xF4[\x80-\x8F][\x80-\xBF]{2})*$/
-  end
-end

spec/websocket/driver/draft75_spec.rb

@@ -90,7 +90,7 @@ describe WebSocket::Driver::Draft75 do
             "WebSocket-Origin: http://www.example.com\r\n" +
             "WebSocket-Location: ws://www.example.com/socket\r\n" +
             "\r\n")
-        expect(socket).to receive(:write).with(WebSocket::Driver.encode "\x00Hi\xFF", :binary)
+        expect(socket).to receive(:write).with(WebSocket::Driver.encode "\x00Hi\xFF", WebSocket::Driver::BINARY)
 
         driver.frame("Hi")
         driver.start

spec/websocket/driver/draft76_spec.rb

@@ -137,7 +137,7 @@ describe WebSocket::Driver::Draft76 do
             "Sec-WebSocket-Location: ws://www.example.com/socket\r\n" +
             "\r\n")
         expect(socket).to receive(:write).with(response)
-        expect(socket).to receive(:write).with(WebSocket::Driver.encode "\x00Hi\xFF", :binary)
+        expect(socket).to receive(:write).with(WebSocket::Driver.encode "\x00Hi\xFF", WebSocket::Driver::BINARY)
 
         driver.frame("Hi")
         driver.start
@@ -191,7 +191,7 @@ describe WebSocket::Driver::Draft76 do
 
           it "sends any frames queued before the handshake was complete" do
             expect(socket).to receive(:write).with(response)
-            expect(socket).to receive(:write).with(WebSocket::Driver.encode "\x00hello\xFF", :binary)
+            expect(socket).to receive(:write).with(WebSocket::Driver.encode "\x00hello\xFF", WebSocket::Driver::BINARY)
             driver.frame("hello")
             driver.parse(body)
             expect(@bytes).to eq [0, 104, 101, 108, 108, 111, 255]