fixed #137

- add skip_utf8_validation argument.
- if set this flag to true, utf-8 validation will be skipped.

ChangeLog

@@ -5,6 +5,7 @@ ChangeLog
 
   - Supporting http-basic auth in WebSocketApp (#143)
   - fix failer of test.testInternalRecvStrict(#141)
+  - skip utf8 validation by skip_utf8_validation argument (#137)
 
 - 0.23.0
 

websocket/_abnf.py

@@ -101,9 +101,10 @@ class ABNF(object):
         self.data = data
         self.get_mask_key = os.urandom
 
-    def validate(self):
+    def validate(self, skip_utf8_validation=False):
         """
         validate the ABNF frame.
+        skip_utf8_validation: skip utf8 validation.
         """
         if self.rsv1 or self.rsv2 or self.rsv3:
             raise WebSocketProtocolException("rsv is not implemented, yet")
@@ -120,7 +121,7 @@ class ABNF(object):
                 return
             if l == 1 or l >= 126:
                 raise WebSocketProtocolException("Invalid close frame.")
-            if l > 2 and not validate_utf8(self.data[2:]):
+            if l > 2 and not skip_utf8_validation and not validate_utf8(self.data[2:]):
                 raise WebSocketProtocolException("Invalid close frame.")
 
             code = 256*six.byte2int(self.data[0:1]) + six.byte2int(self.data[1:2])

websocket/_app.py

@@ -113,7 +113,8 @@ class WebSocketApp(object):
                 self.sock.ping()
 
     def run_forever(self, sockopt=None, sslopt=None, ping_interval=0, ping_timeout=None,
-        http_proxy_host=None, http_proxy_port=None, http_no_proxy=None, http_proxy_auth=None):
+        http_proxy_host=None, http_proxy_port=None, http_no_proxy=None, http_proxy_auth=None,
+        skip_utf8_validation=False):
         """
         run event loop for WebSocket framework.
         This loop is infinite loop and is alive during websocket is available.
@@ -126,6 +127,7 @@ class WebSocketApp(object):
         http_proxy_host: http proxy host name.
         http_proxy_port: http proxy port. If not set, set to 80.
         http_no_proxy: host names, which doesn't use proxy.
+        skip_utf8_validation: skip utf8 validation.
         """
 
         if not ping_timeout or ping_timeout<=0:
@@ -141,7 +143,8 @@ class WebSocketApp(object):
 
         try:
             self.sock = WebSocket(self.get_mask_key, sockopt=sockopt, sslopt=sslopt,
-                fire_cont_frame=self.on_cont_message and True or False)
+                fire_cont_frame=self.on_cont_message and True or False,
+                skip_utf8_validation=skip_utf8_validation)
             self.sock.settimeout(getdefaulttimeout())
             self.sock.connect(self.url, header=self.header, cookie=self.cookie,
                 http_proxy_host=http_proxy_host, http_proxy_port=http_proxy_port,

websocket/_core.py

@@ -256,13 +256,16 @@ def create_connection(url, timeout=None, **options):
              "sockopt" -> socket options
              "sslopt" -> ssl option
              "subprotocols" - array of available sub protocols. default is None.
+             "skip_utf8_validation" - skip utf8 validation.
     """
     sockopt = options.get("sockopt", [])
     sslopt = options.get("sslopt", {})
     fire_cont_frame = options.get("fire_cont_frame", False)
     enable_multithread = options.get("enable_multithread", False)
+    skip_utf8_validation = options.get("skip_utf8_validation", False)
     websock = WebSocket(sockopt=sockopt, sslopt=sslopt,
-        fire_cont_frame = fire_cont_frame, enable_multithread=enable_multithread)
+        fire_cont_frame = fire_cont_frame, enable_multithread=enable_multithread,
+        skip_utf8_validation=skip_utf8_validation)
     websock.settimeout(timeout if timeout is not None else default_timeout)
     websock.connect(url, **options)
     return websock
@@ -376,10 +379,11 @@ class WebSocket(object):
     sslopt: dict object for ssl socket option.
     fire_cont_frame: fire recv event for each cont frame. default is False
     enable_multithread: if set to True, lock send method.
+    skip_utf8_validation: skip utf8 validation.
     """
 
     def __init__(self, get_mask_key=None, sockopt=None, sslopt=None,
-        fire_cont_frame=False, enable_multithread=False):
+        fire_cont_frame=False, enable_multithread=False, skip_utf8_validation=False):
         """
         Initalize WebSocket object.
         """
@@ -394,6 +398,7 @@ class WebSocket(object):
         self.sslopt = sslopt
         self.get_mask_key = get_mask_key
         self.fire_cont_frame = fire_cont_frame
+        self.skip_utf8_validation = skip_utf8_validation
         # Buffers over the packets from the layer beneath until desired amount
         # bytes of bytes are received.
         self._recv_buffer = []
@@ -790,7 +795,7 @@ class WebSocket(object):
                     data = self._cont_data
                     self._cont_data = None
                     frame.data = data[1]
-                    if not self.fire_cont_frame and data[0] == ABNF.OPCODE_TEXT and not validate_utf8(frame.data):
+                    if not self.fire_cont_frame and data[0] == ABNF.OPCODE_TEXT and not self.skip_utf8_validation and not validate_utf8(frame.data):
                         raise WebSocketPayloadException("cannot decode: " + repr(frame.data))
                     return [data[0], frame]
 
@@ -839,7 +844,7 @@ class WebSocket(object):
         frame_buffer.clear()
 
         frame = ABNF(fin, rsv1, rsv2, rsv3, opcode, has_mask, payload)
-        frame.validate()
+        frame.validate(self.skip_utf8_validation)
 
         return frame