rtm_v2 api spins in infinite loop under gevent (#1246)

* Raise an exception when reading from socket and getting no data.
Returning and empty bytes object causes an infinite loop between this and the
fetch function which in some cases can cause 100% cpu utilization (this happens
under gevent). Also the exception handling here can't just return an empty bytes for the same
reason - the now dead connection cannot be replaced by code that uses this
code.

* Per https://docs.python.org/3/howto/sockets.html
No data means the other side has closed the connection:
"When a recv returns 0 bytes, it means the other side has closed (or is in the
process of closing) the connection. You will not receive any more data on this
connection. Ever. You may be able to send data successfully"

Author: mattbillenstein

slack_sdk/socket_mode/builtin/internals.py

@@ -1,4 +1,3 @@
-import errno
 import hashlib
 import itertools
 import os
@@ -24,6 +23,8 @@ def _parse_connect_response(sock: Socket) -> Tuple[Optional[int], str]:
         line = []
         while True:
             c = sock.recv(1)
+            if not c:
+                raise ConnectionError("Connection is closed")
             line.append(c)
             if c == b"\n":
                 break
@@ -114,7 +115,10 @@ def _establish_new_socket_connection(
 def _read_http_response_line(sock: ssl.SSLSocket) -> str:
     cs = []
     while True:
-        c: str = sock.recv(1).decode("utf-8")
+        b: bytes = sock.recv(1)
+        if not b:
+            raise ConnectionError("Connection is closed")
+        c: str = b.decode("utf-8")
         if c == "\r":
             break
         if c != "\n":
@@ -199,19 +203,12 @@ def _receive_messages(
     def receive(specific_buffer_size: Optional[int] = None):
         size = specific_buffer_size if specific_buffer_size is not None else receive_buffer_size
         with sock_receive_lock:
-            try:
-                received_bytes = sock.recv(size)
-                if all_message_trace_enabled:
-                    if len(received_bytes) > 0:
-                        logger.debug(f"Received bytes: {received_bytes}")
-                return received_bytes
-            except OSError as e:
-                # For Linux/macOS, errno.EBADF is the expected error for bad connections.
-                # The errno.ENOTSOCK can be sent when running on Windows OS.
-                if e.errno in (errno.EBADF, errno.ENOTSOCK):
-                    logger.debug("The connection seems to be already closed.")
-                    return bytes()
-                raise e
+            received_bytes = sock.recv(size)
+            if not received_bytes:
+                raise ConnectionError("Connection is closed")
+            if all_message_trace_enabled:
+                logger.debug(f"Received bytes: {received_bytes}")
+            return received_bytes
 
     return _fetch_messages(
         messages=[],