Monitor udev for hidraw devices instead of hid devices.

Monitor udev for hidraw devices instead of hid devices.

Watching for new hid devices is racy because the hidraw device rules are
still running. Fix the race by explicitly waiting for the hidraw device
add event. The properties needed from the hid device can be gotten by
fetching the parent of the hidraw device from the sysfs tree.

BUG=394200
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=283810

[mschilder1, 2014-07-17 01:52:38]

style nits.

Manually tested to fix the reported problem.

https://codereview.chromium.org/396993010/diff/1/device/hid/hid_service_linux.cc
File device/hid/hid_service_linux.cc (right):

https://codereview.chromium.org/396993010/diff/1/device/hid/hid_service_linux...
device/hid/hid_service_linux.cc:30: const char kHidrawSubSystem[] = "hidraw";
as const, should this be kHIDRAWSUBSYSTEM?

https://codereview.chromium.org/396993010/diff/1/device/hid/hid_service_linux...
device/hid/hid_service_linux.cc:85: if (!parent) {
inconsistent if(){ expr; } vs. if () expr; in this file / method.
What do the rules say again?
Certainly not tolerate inconsistency ;-)

[Reilly Grant (use Gerrit), 2014-07-17 03:22:53]

https://codereview.chromium.org/396993010/diff/1/device/hid/hid_service_linux.cc
File device/hid/hid_service_linux.cc (right):

https://codereview.chromium.org/396993010/diff/1/device/hid/hid_service_linux...
device/hid/hid_service_linux.cc:30: const char kHidrawSubSystem[] = "hidraw";
On 2014/07/17 01:52:38, mschilder1 wrote:
> as const, should this be kHIDRAWSUBSYSTEM?

Google C++ style says kHidrawSubSystem (or kHidrawSubsystem) and Chromium style
says HIDRAW_SUBSYSTEM. I've seen the Google style in a bunch of places so for
consistency I think this should be left as is.

https://codereview.chromium.org/396993010/diff/1/device/hid/hid_service_linux...
device/hid/hid_service_linux.cc:85: if (!parent) {
On 2014/07/17 01:52:38, mschilder1 wrote:
> inconsistent if(){ expr; } vs. if () expr; in this file / method.
> What do the rules say again?
> Certainly not tolerate inconsistency ;-)

The rules say it is my choice. I dislike inconsistency but also have a strong
inclination to avoid single-line ifs.

[Ken Rockot(use gerrit already), 2014-07-17 14:46:08]

LGTM, with a nit: I have a strong preference for always making blocks explicit,
even for single-line ifs.

[Reilly Grant (use Gerrit), 2014-07-17 16:11:21]

The CQ bit was checked by reillyg@chromium.org

[commit-bot: I haz the power, 2014-07-17 16:12:39]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/reillyg@chromium.org/396993010/20001

[commit-bot: I haz the power, 2014-07-17 17:53:06]

Change committed as 283810