Curtain mode implementation for Mac OS X.

Curtain mode implementation for Mac OS X.

This implements the basic curtain-mode functionality:
* Switch out the active session on connection (if necessary--it doesn't do this if the session is at the login screen or already switched out)
* Close all connections if the session becomes active again (for example, if someone logs into it at the console).

It doesn't solve the problem of launchd starting a new instance of the server when we switch out. There will be a follow-up CL to address that.

BUG=110111
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=150746

[Jamie, 2012-08-07 21:51:42]

ptal

[Lambros, 2012-08-08 00:48:29]

dcaiafa: I can't see any problems myself, but please could you check the usage
of the CF APIs looks OK?

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
File remoting/host/curtain_mode_mac.cc (right):

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:14: static const char* kCGSessionPath =
nit: don't need static in anon namespace

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:53: execl(kCGSessionPath, kCGSessionPath,
"-suspend", NULL);
Is there a C API equivalent of this command-line tool?
Do we care about logging errors if fork/exec fails?
Will this still work if Fast User Switching is not enabled?

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:54: exit(1);
Use _exit instead of exit (after a fork with failed exec).

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:65: CurtainMode* self =
reinterpret_cast<CurtainMode*>(user_data);
nit: Use static_cast<> ?

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:76: }
nit: // namespace remoting

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
File remoting/host/curtain_mode_mac.h (right):

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.h:8: #include <Carbon/Carbon.h>
nit: blank line between C and C++ headers

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/remot...
File remoting/host/remoting_me2me_host.cc (right):

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/remot...
remoting/host/remoting_me2me_host.cc:440: #if defined(OS_MACOSX)
Please could you add a comment explaining why we want to disconnect the user?

[Jamie, 2012-08-08 01:39:42]

ptal

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
File remoting/host/curtain_mode_mac.cc (right):

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:14: static const char* kCGSessionPath =
On 2012/08/08 00:48:29, Lambros wrote:
> nit: don't need static in anon namespace

Done.

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:53: execl(kCGSessionPath, kCGSessionPath,
"-suspend", NULL);
On 2012/08/08 00:48:29, Lambros wrote:
> Is there a C API equivalent of this command-line tool?

I couldn't find one last time I looked, but I agree that it would be a lot nicer
if such an API existed.

> Do we care about logging errors if fork/exec fails?

That's a good point. I think for now we don't care, but once I add in policy
support, we'll want to close the connection if it can't be curtained.

> Will this still work if Fast User Switching is not enabled?

Another good point. I don't know how to turn off FUS to test it. Removing the
menu doesn't prevent it from working.

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:54: exit(1);
On 2012/08/08 00:48:29, Lambros wrote:
> Use _exit instead of exit (after a fork with failed exec).

Done.

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:65: CurtainMode* self =
reinterpret_cast<CurtainMode*>(user_data);
On 2012/08/08 00:48:29, Lambros wrote:
> nit: Use static_cast<> ?

Done.

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.cc:76: }
On 2012/08/08 00:48:29, Lambros wrote:
> nit: // namespace remoting

Done.

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
File remoting/host/curtain_mode_mac.h (right):

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/curta...
remoting/host/curtain_mode_mac.h:8: #include <Carbon/Carbon.h>
On 2012/08/08 00:48:29, Lambros wrote:
> nit: blank line between C and C++ headers

Done.

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/remot...
File remoting/host/remoting_me2me_host.cc (right):

https://chromiumcodereview.appspot.com/10823215/diff/4001/remoting/host/remot...
remoting/host/remoting_me2me_host.cc:440: #if defined(OS_MACOSX)
On 2012/08/08 00:48:29, Lambros wrote:
> Please could you add a comment explaining why we want to disconnect the user?

I've added it to the header file, which seemed a more appropriate place.

[dcaiafa, 2012-08-08 17:07:15]

lgtm

[Lambros, 2012-08-08 18:31:56]

lgtm

[commit-bot: I haz the power, 2012-08-08 21:49:39]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamiewalch@chromium.org/10823215/10002

[commit-bot: I haz the power, 2012-08-08 23:10:27]

Try job failure for 10823215-10002 (retry) on mac_rel for step "browser_tests".
It's a second try, previously, step "browser_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac_rel&nu...

[commit-bot: I haz the power, 2012-08-08 23:13:51]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamiewalch@chromium.org/10823215/10002

[commit-bot: I haz the power, 2012-08-08 23:43:02]

Try job failure for 10823215-10002 (retry) on android for steps "compile, build"
(clobber build).
It's a second try, previously, steps "compile, build" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android&nu...

[commit-bot: I haz the power, 2012-08-09 00:01:50]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamiewalch@chromium.org/10823215/10002

[commit-bot: I haz the power, 2012-08-09 02:07:21]

Try job failure for 10823215-10002 (previous was lost) (retry) on win_rel for
step "nacl_integration".
It's a second try, previously, step "nacl_integration" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[commit-bot: I haz the power, 2012-08-09 04:01:53]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamiewalch@chromium.org/10823215/10002

[commit-bot: I haz the power, 2012-08-09 05:31:24]

Change committed as 150746