Fixed a race condition when attaching to an existing socket

This is an error that presents itself with RTCDataChannel's, I suspect this could not
happen with a pre-existing WebSocket.

If the remote connection creates a data channel then the local (VNC) side gets a channel
created callback. It may also be the case that in that very same tick the socket is also
opened and buffered data received. This meant that (in my tests) about 1/3 of the time
noVNC would fail to respond to the initial message from the server because it was received
and subsequently not handled during that initial tick.

Also made the documentation reflect this new behaviour and document the existing behaviour.

1 files changed, 11 insertions, 2 deletions

diff --git a/core/rfb.js b/core/rfb.js
index 876255b..f8eeb51 100644
--- a/core/rfb.js
+++ b/core/rfb.js
@@ -286,8 +286,17 @@ export default class RFB extends EventTargetMixin {
         this._sock.on('error', e => Log.Warn("WebSocket on-error event"));
 
         // Slight delay of the actual connection so that the caller has
-        // time to set up callbacks
-        setTimeout(this._updateConnectionState.bind(this, 'connecting'));
+        // time to set up callbacks.
+        // This it not possible when a pre-existing socket is passed in and is just opened.
+        // If the caller creates this object in the open() callback of a socket and there's
+        // data pending doing it next tick causes a packet to be lost.
+        // This is particularly noticable for RTCDataChannel's where the other end creates
+        // the channel and the client, this end, gets notified it exists.
+        if (typeof urlOrChannel === 'string') {
+            setTimeout(this._updateConnectionState.bind(this, 'connecting'));
+        } else {
+            this._updateConnectionState('connecting');
+        }
 
         Log.Debug("<< RFB.constructor");