This way the tests can be run directly from within the tests directory
and they might also have a chance of running on Windows (where the
path separator is different).
TCP_KEEPALIVE is now enabled by default. Settings for
KEEPCNT, KEEPINTVL and KEEPIDLE can be supplied when
creating WebSocketServer and KEEPALIVE can also be
disabled if required.
Also adds new unit test for testing.
Co-authored-by: natsume.takashi@lab.ntt.co.jp
websocket.py:
* With echo.py, which doesn't need any server configuration, we can
just switch over our application class to inherit from
WebSocketRequestHandler instead of WebSocketServer. Also, need to
use the new method name new_websocket_client.
* With load.py, since we have the "delay" configuration, we need both
a server class and a request handler.
Note that for both tests, I've removed the raising of
self.EClose(closed). This is incorrect. First of all, it's described
as "An exception before the WebSocket connection was established", so
not suitable for our case. Second, it will cause send_close to be
called twice. Finally, self.EClose is now in the WebSocketServer
class, and not a member of the request handler.
- Add initial IETF-07 (HyBi-07) protocol version support. This version
still uses base64 encoding since the API for binary support is not
yet finalized.
- Move socket send and recieve functions into the WebSocketServer
class instead of having the sub-class do this. This simplifies
sub-classes somewhat. The send_frame routine now returns the number
of frames that were unable to be sent. If this value is non-zero
then the sub-class should call again when the socket is ready until
the pending frames count is 0.
- Do traffic reporting in the main class instead.
- When the client is HyBi style (i.e. IETF-07) then use the
sub-protocol header to select whether to do base64 encoding or
simply send the frame data raw (binary). Update include/websock.js
to send a 'base64' protocol selector. Once the API support binary,
then the client will need to detect this and set the protocol to
'binary'.
Convert latency test to use include/websock.js instead of direct
WebSockets handling.
Add support for configuring the maximum bufferedAmount. This allows us
to configure it high enough to get around a bug in bufferedAmount
reporting in Opera.
Add some latency test results for Opera 11 with WebSockets turned on.
- add dup2 functionality. This requires adding a ref cnt to the
_WS_connections structure so that we only free the structure once
all dup'd referenced are closed. Also, refactor malloc and free of
connection structure into _WS_alloc and _WS_free.
- allow select to accept a NULL timeout value which means sleep
forever instead of segfaulting.
- fix some compile warnings related to ppoll definition.
- move some WebSockets related html test pages into utils and symlink
them from tests.
Moved websocket.py code into a class WebSocketServer. WebSockets
server implementations will sub-class and define a handler() method
which is passed the client socket after. Global variable settings have been
changed to be parameters for WebSocketServer when created.
Subclass implementations still have to handle queueing and sending but
the parent class handles everything else (daemonizing, websocket
handshake, encode/decode, etc). It would be better if the parent class
could handle queueing and sending. This adds some buffering and
polling complexity to the parent class but it would be better to do so
at some point. However, the result is still much cleaner as can be
seen in wsecho.py.
Refactored wsproxy.py and wstest.py (formerly ws.py) to use the new
class. Added wsecho.py as a simple echo server.
- rename tests/ws.py to utils/wstest.py and add a symlink from
tests/wstest.py
- rename tests/ws.html to tests/wstest.html to match utils/wstest.py.
- add utils/wsecho.py
- add tests/wsecho.html which communicates with wsecho.py and simply
sends periodic messages and shows what is received.
Rename the $() selector to $D() so that it doesn't collide with
the jQuery name.
The API change is that the 'target' option for Canvas and RFB objects
must now be a DOM Canvas element. A string is no longer accepted
because this requires that a DOM lookup is done and the Canvas and RFB
should have no UI code in them. Modularity.
- util.js that contains essential functions
- webutils.js that contains the GUI utility function.js
this helps to include noVNC in other project, especially Cappuccino Application
i
- include/rfb.js: Keep track of the number of rects of each encoding
type and print them out when we close a connection (if 'info'
logging level).
- tests/vnc_perf.html: first pass at a noVNC based performance
benchmark.
- utils/wsproxy.py: Fix the output of the record filename.
- include/canvas.js: When 'debug' logging, show browser detection
values.
- test/canvas.html: Only restore the canvas to it's starting state if
the logging level is not 'debug'.
- wsproxy.py: Append the session number to the record filename so that
multiple sessions don't stomp on each other.
Generally, most servers send hextile updates as single updates
containing many rects. Some servers send hextile updates as many small
framebuffer updates with a few rects each (such as QEMU). This latter
cases revealed that shifting off the beginning of the receive queue
(which happens after each hextile FBU) performs poorly.
This change switches to using an indexed receive queue (instead of
actually shifting off the array). When the receive queue has grown to
a certain size, then it is compacted all at once.
The code is not as clean, but this change results in more than 2X
speedup under Chrome for the pessimal case and 10-20% in firefox.
Turns out when Windows is running in QEMU and a window scroll happens,
there are lots of little hextile rects sent. This is slow in noVNC.
- Some recording/playback improvement.
- Add test harness to drive playback of recordings.
- By pulling off the rect header in one chunk we get a 3X speedup in
Chrome and a 20% speedup in firefox (specifically for the scroll
test).
- Also, get rid of some noise from creating timers for handle_message.
Check to make sure there isn't already a pending timer first.
New API:
To use the RFB object, you now must instantiate it (this allows more
than one instance of it on the same page).
rfb = new RFB(settings);
The 'settings' variable is a namespace that contains initial default
settings. These can also be set and read using 'rfb.set_FOO()' and
'rfb.get_FOO()' where FOO is the setting name. The current settings
are (and defaults) are:
- target: the DOM Canvas element to use ('VNC_canvas').
- encrypt: whether to encrypt the connection (false)
- true_color: true_color or palette (true)
- b64encode: base64 encode the WebSockets data (true)
- local_cursor: use local cursor rendering (true if supported)
- connectTimeout: milliseconds to wait for connect (2000)
- updateState: callback when RFB state changes (none)
- clipboardReceive: callback when clipboard data received (none)
The parameters to the updateState callback have also changed. The
function spec is now updateState(rfb, state, oldstate, msg):
- rfb: the RFB object that this state change is for.
- state: the new state
- oldstate: the previous state
- msg: a message associate with the state (not always set).
The clipboardReceive spec is clipboardReceive(rfb, text):
- rfb: the RFB object that this text is from.
- text: the clipboard text received.
Changes:
- The RFB and Canvas namespaces are now more proper objects. Private
implementation is no longer exposed and the public API has been made
explicit. Also, instantiation allows more than one VNC connection
on the same page (to complete this, DefaultControls will also need
this same refactoring).
- Added 'none' logging level.
- Removed automatic stylesheet selection workaround in util.js and
move it to defaultcontrols so that it doesn't interfere with
intergration.
- Also, some major JSLinting.
- Fix input, canvas, and cursor tests to work with new model.
The following API changes may affect integrators:
- Settings have been moved out of the RFB.connect() call. Each
setting now has it's own setter function: setEncrypt, setBase64,
setTrueColor, setCursor.
- Encrypt and cursor settings now default to on.
- CSS changes:
- VNC_status_bar for input buttons switched to a element class.
- VNC_buttons split into VNC_buttons_right and
VNC_buttons_left
- New id styles for VNC_settings_menu and VNC_setting
Note: the encrypt, true_color and cursor, logging setting can all be
set on load using query string variables (in addition to host, port
and password).
Client cursor (cursor pseudo-encoding) support has been polished and
activated.
The RFB settings are now presented as radio button list items in
a drop-down "Settings" menu when using the default controls.
Also, in the settings menu is the ability to select between alternate
style-sheets.
Cookie and stylesheet selection support added to util.js.
Interestingly it turns out that using the native base64 routines does
not improve performance. Likely because the actual time is in
marshalling/unmarshalling between strings and arrays (and associated
garbage collection overhead) which has to be done either way.
Looks like disabling web-socket-js debug messages by default that we
get a minor speedup.
Python proxy should support both 75 and 76 (00) modes. Also, update ws
test to more reliably hit the WebSockets ordering/drop issue.
Also add a wsencoding test client/server program to test send a set of
values between client and server and vice-versa to test encodings.
Not turned on by default.
Add support for encode/decode of UTF-8 in the proxy. This leverages
the browser for decoding the WebSocket stream directly instead of
doing base64 decode in the browser itself.
Unfortunately, in Chrome this has negligible impact (round-trip time
is increased slightly likely due to extra python processing).
In firefox, due to the use of the flash WebSocket emulator the
performance is even worse. This is because it's really annoying to get
the flash WebSocket emulator to properly decode a UTF-8 bytestream.
The problem is that the readUTFBytes and readMultiByte methods of an
ActionScript ByteArray don't treat 0x00 correctly. They return
a string that ends at the first 0x00, but the index into the ByteArray
has been advanced by however much you requested.
This is very silly for two reasons: ActionScript (and Javascript)
strings can contain 0x00 (they are not null terminated) and second,
UTF-8 can legitimately contain 0x00 values. Since UTF-8 is not
constant width there isn't a great way to determine if those methods
in fact did encounter a 0x00 or they just read the number of bytes
requested.
Doing manual decoding using readUTFByte one character at a time slows
things down quite a bit. And to top it all off, those methods don't
support the alternate UTF-8 encoding for 0x00 ("\xc0\x80"). They also
just treat that encoding as the end of string too.
So to get around this, for now I'm encoding zero as 256 ("\xc4\x80")
and then doing mod 256 in Javascript. Still doesn't result in much
benefit in firefox.
But, it's an interesting approach that could use some more exploration
so I'm leaving in the code in both places.
- By dereferencing the 'data' field of the imageData object before the
loop, the hextile performance on Chrome is down to 140ms or so for
a full 800x600 update. Still have to fall back to Canvas operations
for firefox.
- Fix RQ empty after reorder bug.
Joel Martin
Edward Hope-Morley
chrislee35
Peter Åstrand (astrand)
Antoine Mercadal
Joel Martin