Age | Commit message (Collapse) | Author | Files | Lines |
|
The work Pavel did on spice.html did not get fully transferred
into spice_auto.html; in particular, we need to append the xfer
div to the main area in order for file transfer to work.
This fixes that, and now drag + drop works in spice_auto.html.
|
|
Avoid spamming log
|
|
For unknown channel types the type number is still used, e.g.:
"Channel type unknown-12 not implemented"
|
|
|
|
|
|
|
|
This helps video playback do a slightly better job of keeping up
in the browser. It's not a dramatic effect, but enough to start making
video playback almost tolerable.
|
|
|
|
|
|
|
|
|
|
|
|
MouseWheelEvent is deprecated and not working in Firefox.
|
|
The client should inform the agent about capabilities
when the connection is established. This avoid receiving unhandled
agent messages.
|
|
Stop reading a file when there are no tokens,
continue reading it when tokens arrive.
|
|
|
|
It is possible to transfer files from the client to the guest
using File API [0] when a spice vd agent is connected.
Methods for the transfer are based on spice-gtk implementation.
[0] http://www.w3.org/TR/file-upload/
|
|
Allowed size for SPICE_MSGC_MAIN_AGENT_DATA message is 2048 Bytes,
larger messages have to be splitted.
|
|
Messages that were not sent are stored in the queue.
They will be sent later when the client receives more agent tokens.
|
|
We should report that we handle the connected tokens message
by setting the SPICE_MAIN_CAP_AGENT_CONNECTED_TOKENS capability.
|
|
|
|
Current error message in SPICE-HTML5 error is poor (error-logging function
prints error.toString() which doesn't contain anything detailed).
This patch enhances this message in the following way:
- error event contains target attribute ([1]),
- the code in this patch checks for presence of 'url' attribute (which is
contained in WebSocket object[2]). If the target contains url then
meaningful message is logged.
[1]: https://developer.mozilla.org/en-US/docs/Web/API/Event
[2]: https://developer.mozilla.org/en-US/docs/Web/API/WebSocket
|
|
This keeps an XSpice server from spin looping, which it will do if
data is presented, but num_send_tokens == 0.
|
|
|
|
|
|
|
|
Hi,
Currently the websocket connection is only being established using
the ws scheme.
Adding support for wss scheme when using https protocol to allow
ssl connection.
|
|
|
|
Requires a browser with MediaSource extension support, and
Opus support for the source buffers. In practice, that is
Chrome and Firefox.
|
|
|
|
|
|
Also converts the display cache from an array to an object. This is to help enforce proper sparceness of the data as well as make it easier to reliably fully delete a given cache entry without affecting access to the remaining entries.
|
|
|
|
don't claim to have an infinitely large cache.
|
|
fairly massive memory leaks during video playback.
|
|
The failure to ack pings led to a nasty bug whereby an inactive client hangs.
The relatively recent change to monitor latency with ping exposed this bug.
|
|
If server runs on port 80 or 443 the default value of
window.location.port is an empty string
Signed-off-by: Jordan Pittier <jordan.pittier-ext@cloudwatt.com>
|
|
Signed-off-by: Alon Levy <alevy@redhat.com>
|
|
Hi,
I'm the Debian Developer working on spice-html5 (eg: I uploaded
spice-html5 recently in Debian Sid).
I noticed that the minus key on the alphanumeric block wasn't mapped.
This was particularly annoying on my laptop keyboard when using
spice-html5 to access remote servers (eg: many commands needs argument
with the minus sign), with the "no map for 173" alert box popping each
time...
The attached patch made it work with Firefox 21 (in fact, Iceweasel 21
in Debian). I don't know if it works on other browsers, but it worked
for me.
Cheers,
Thomas Goirand (zigo)
From ad9d69eb09f63538db06664a3d88d48ff56b2abb Mon Sep 17 00:00:00 2001
From: Thomas Goirand <thomas@goirand.fr>
Date: Sat, 25 May 2013 23:56:14 +0800
Subject: [PATCH] Adds missing mapping of the alphanumeric minus key.
|
|
|
|
The first approach is to use a moving image to simulate the cursor.
This does not work well, so a second approach, involving static
cursors is provided. That requires a site administrator to precreate
all the cursors that can be used by a particular application set.
A warning message strongly suggests using a different browser as well.
|
|
present.
Provides compatibility for IE10.
|
|
This allows IE10 to function. Note that we would normally subclass
this type, but ArrayBuffer is implemented in native code, and so is
difficult to subclass.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|