Implement test-echo extension message and reply.

Implement test-echo extension message and reply.

TEST=Run the following command in the JS console once connected: remoting.clientSession.plugin.sendExtensionMessage('test-echo', <thing to be echoed>)

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=220802

[Jamie, 2013-08-29 21:41:12]

ptal

https://codereview.chromium.org/23491004/diff/3001/remoting/webapp/client_plu...
File remoting/webapp/client_plugin_async.js (right):

https://codereview.chromium.org/23491004/diff/3001/remoting/webapp/client_plu...
remoting/webapp/client_plugin_async.js:324: console.log('Got echo reply: ' +
message.data['data']);
We'll want something more testable once I get around to writing a browser test
for this, but this will do for now.

[Wez, 2013-08-30 01:18:10]

https://codereview.chromium.org/23491004/diff/3001/remoting/host/client_sessi...
File remoting/host/client_session.cc (right):

https://codereview.chromium.org/23491004/diff/3001/remoting/host/client_sessi...
remoting/host/client_session.cc:193: if (message.type() == "test-echo") {
This makes it possible for a client to connect & soak up lots of bandwidth to a
host by sending it big echo messages, which doesn't seem ideal.

Could we make this DEBUG-only, or controlled by a runtime flag?

https://codereview.chromium.org/23491004/diff/3001/remoting/webapp/client_plu...
File remoting/webapp/client_plugin_async.js (right):

https://codereview.chromium.org/23491004/diff/3001/remoting/webapp/client_plu...
remoting/webapp/client_plugin_async.js:324: console.log('Got echo reply: ' +
message.data['data']);
On 2013/08/29 21:41:12, Jamie wrote:
> We'll want something more testable once I get around to writing a browser test
> for this, but this will do for now.

Could you pass test-echo-reply messages to an onTestEchoReply handler, and have
the browser-tests inject that into the client app?

[Jamie, 2013-08-30 02:14:04]

https://codereview.chromium.org/23491004/diff/3001/remoting/host/client_sessi...
File remoting/host/client_session.cc (right):

https://codereview.chromium.org/23491004/diff/3001/remoting/host/client_sessi...
remoting/host/client_session.cc:193: if (message.type() == "test-echo") {
On 2013/08/30 01:18:11, Wez wrote:
> This makes it possible for a client to connect & soak up lots of bandwidth to
a
> host by sending it big echo messages, which doesn't seem ideal.
> 
> Could we make this DEBUG-only, or controlled by a runtime flag?

The extension protocol already allows this: even if the host doesn't recognise a
message, it stills receives it. This allows a client to soak up lots of
bandwidth *from* the host, but I could address this by echoing at most, say 100
characters. WDYT?

https://codereview.chromium.org/23491004/diff/3001/remoting/webapp/client_plu...
File remoting/webapp/client_plugin_async.js (right):

https://codereview.chromium.org/23491004/diff/3001/remoting/webapp/client_plu...
remoting/webapp/client_plugin_async.js:324: console.log('Got echo reply: ' +
message.data['data']);
On 2013/08/30 01:18:11, Wez wrote:
> On 2013/08/29 21:41:12, Jamie wrote:
> > We'll want something more testable once I get around to writing a browser
test
> > for this, but this will do for now.
> 
> Could you pass test-echo-reply messages to an onTestEchoReply handler, and
have
> the browser-tests inject that into the client app?

Yes, I'll probably want something along those lines, but I think it's best left
to the CL that introduces the browser test (which could be this one if I wait
for Weitao's e2e test framework).

[Jamie, 2013-08-30 02:15:44]

https://codereview.chromium.org/23491004/diff/3001/remoting/host/client_sessi...
File remoting/host/client_session.cc (right):

https://codereview.chromium.org/23491004/diff/3001/remoting/host/client_sessi...
remoting/host/client_session.cc:193: if (message.type() == "test-echo") {
On 2013/08/30 02:14:05, Jamie wrote:
> On 2013/08/30 01:18:11, Wez wrote:
> > This makes it possible for a client to connect & soak up lots of bandwidth
to
> a
> > host by sending it big echo messages, which doesn't seem ideal.
> > 
> > Could we make this DEBUG-only, or controlled by a runtime flag?
> 
> The extension protocol already allows this: even if the host doesn't recognise
a
> message, it stills receives it. This allows a client to soak up lots of
> bandwidth *from* the host, but I could address this by echoing at most, say
100
> characters. WDYT?

Also, an authenticated client can already do just about anything it wants, so
I'm not sure it's worth worrying about this case.

[Wez, 2013-08-31 02:06:20]

On 2013/08/30 02:15:44, Jamie wrote:
>
https://codereview.chromium.org/23491004/diff/3001/remoting/host/client_sessi...
> File remoting/host/client_session.cc (right):
> 
>
https://codereview.chromium.org/23491004/diff/3001/remoting/host/client_sessi...
> remoting/host/client_session.cc:193: if (message.type() == "test-echo") {
> On 2013/08/30 02:14:05, Jamie wrote:
> > On 2013/08/30 01:18:11, Wez wrote:
> > > This makes it possible for a client to connect & soak up lots of bandwidth
> to
> > a
> > > host by sending it big echo messages, which doesn't seem ideal.
> > > 
> > > Could we make this DEBUG-only, or controlled by a runtime flag?
> > 
> > The extension protocol already allows this: even if the host doesn't
recognise
> a
> > message, it stills receives it. This allows a client to soak up lots of
> > bandwidth *from* the host, but I could address this by echoing at most, say
> 100
> > characters. WDYT?
> 
> Also, an authenticated client can already do just about anything it wants, so
> I'm not sure it's worth worrying about this case.

It's more that a client can cause host->client bandwidth to be consumed, as well
as just client->host; it's not necessarily a problem but it seems unnecessarily
permissive for a test-only feature!

[Jamie, 2013-08-31 02:24:48]

ptal

[Wez, 2013-08-31 02:46:11]

lgtm

[commit-bot: I haz the power, 2013-08-31 03:19:27]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/23491004/12001

[commit-bot: I haz the power, 2013-08-31 11:55:54]

Retried try job too often on win7_aura for step(s) browser_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win7_aura&...

[commit-bot: I haz the power, 2013-09-01 15:45:05]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/23491004/12001

[commit-bot: I haz the power, 2013-09-02 05:27:03]

Change committed as 220802