Base: Add power requirements to the System monitor.

Base: Add power requirements to the System monitor.

The only thing we need in the short term is CPU_REQUIRED,
and that part is implemented for all platforms (as far as it
makes sense). The other options are implemented for Windows,
but given that currently there's no user, I expect the
implementation for other platforms to be written when there's
an actual need for it.

BUG=126591
TEST=none
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=137960

[rvargas (doing something else), 2012-05-16 20:34:11]

Please take a look.

[rvargas (doing something else), 2012-05-17 02:13:58]

Actually, I think I'll have to change the interface to allow multiple calls with
the same id, so no need to review this yet.

[rvargas (doing something else), 2012-05-17 19:34:38]

OK... changed the API and added unit tests. Please take a look.

[brettw, 2012-05-18 05:47:05]

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
File base/system_monitor/system_monitor.h (right):

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
base/system_monitor/system_monitor.h:206: std::map<std::string,
std::pair<HANDLE, int>> handles_;
">>" -> "> >"

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
File base/system_monitor/system_monitor_android.cc (right):

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
base/system_monitor/system_monitor_android.cc:11: if (requirement ==
CPU_REQUIRED)
It's not clear to me why Android & Posix early return for CPU_REQUIRED but warn
otherwise.

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
File base/system_monitor/system_monitor_win.cc (right):

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
base/system_monitor/system_monitor_win.cc:12: typedef std::map<std::string,
std::pair<HANDLE, int>> HanledMap;
Also space in >>

Hanled -> Handle

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
base/system_monitor/system_monitor_win.cc:20:
static_cast<POWER_REQUEST_TYPE>(PowerRequestExecutionRequired + 10);
Indent 2 more spaces.

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
base/system_monitor/system_monitor_win.cc:63: std::wstring wide_reason =
ASCIIToWide(reason);
Can you use string16 in new code?

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
base/system_monitor/system_monitor_win.cc:90: base::win::ScopedHandle
request_handle(handle);
I'd feel better if this was before the "< win8" early return since it will
ensure the handle is closed (although in practice the handle won't be valid so
it won't matter).

[rvargas (doing something else), 2012-05-18 18:11:04]

Thanks.

All should be fixed now.

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
File base/system_monitor/system_monitor_android.cc (right):

https://chromiumcodereview.appspot.com/10406012/diff/13001/base/system_monito...
base/system_monitor/system_monitor_android.cc:11: if (requirement ==
CPU_REQUIRED)
On 2012/05/18 05:47:05, brettw wrote:
> It's not clear to me why Android & Posix early return for CPU_REQUIRED but
warn
> otherwise.

I added a very short comment.

The reason is that (afaik), there's nothing to do in this case, given that we
are just coping with a Win8 issue. However, if this function is called with some
other requirement (like to avoid sleep), reaching this point means that said
functionality has to be implemented.

[brettw, 2012-05-18 18:19:46]

https://chromiumcodereview.appspot.com/10406012/diff/14007/base/system_monito...
File base/system_monitor/system_monitor_win.cc (right):

https://chromiumcodereview.appspot.com/10406012/diff/14007/base/system_monito...
base/system_monitor/system_monitor_win.cc:12: typedef std::map<std::string,
std::pair<HANDLE, int> > HanleMap;
How did you still manage to misspell this? :)

[rvargas (doing something else), 2012-05-18 18:44:52]

Once again :(

https://chromiumcodereview.appspot.com/10406012/diff/14007/base/system_monito...
File base/system_monitor/system_monitor_win.cc (right):

https://chromiumcodereview.appspot.com/10406012/diff/14007/base/system_monito...
base/system_monitor/system_monitor_win.cc:12: typedef std::map<std::string,
std::pair<HANDLE, int> > HanleMap;
On 2012/05/18 18:19:46, brettw wrote:
> How did you still manage to misspell this? :)

Never underestimate the power of the the brain auto-spell-check.

[brettw, 2012-05-18 19:28:10]

lgtm