Introduce the devices Mojo app

Introduce the devices Mojo app

It's a small Mojo app that currently only hosts USB device
services. This CL adds the app along with an apptest suite
(run manually with mojo_runner mojo:devices_apptests).

The app is hooked up to Chrome to run in the browser process,
though there are currently no production code paths which
cause the app to launch.

BUG=498557
Committed: https://crrev.com/ed800dc138edfa261aa766281b96f736ff9824f6
Cr-Commit-Position: refs/heads/master@{#333899}

[Ken Rockot(use gerrit already), 2015-06-09 23:59:49]

Patchset #1 (id:1) has been deleted

[Ken Rockot(use gerrit already), 2015-06-09 23:59:56]

Patchset #1 (id:20001) has been deleted

[Ken Rockot(use gerrit already), 2015-06-10 00:00:03]

Patchset #1 (id:40001) has been deleted

[Ken Rockot(use gerrit already), 2015-06-10 00:00:10]

Patchset #1 (id:60001) has been deleted

[Ken Rockot(use gerrit already), 2015-06-10 00:04:56]

rockot@chromium.org changed reviewers:
+ reillyg@chromium.org

[Ken Rockot(use gerrit already), 2015-06-10 00:07:24]

Reilly, please take a look at device/* and chrome_device_client.*

Changes to existing code are to do this stuff:
 - Remove the ConnectToUSBDeviceManager interface from
   DeviceClient
 - Clean up DeviceManagerImpl unittests (dead code)
 - Support DeviceManagerImpl living on a different thread 
   from UsbService

[Reilly Grant (use Gerrit), 2015-06-10 20:02:47]

https://codereview.chromium.org/1165223004/diff/80001/device/devices_app/devi...
File device/devices_app/devices_app.cc (right):

https://codereview.chromium.org/1165223004/diff/80001/device/devices_app/devi...
device/devices_app/devices_app.cc:46: // does presume not catch all common local
host strings.
does not presume to

https://codereview.chromium.org/1165223004/diff/80001/device/devices_app/devi...
device/devices_app/devices_app.cc:47: if (remote_url_.host() == "127.0.0.1" ||
remote_url_.host() == "localhost")
net::IsLocalHost() may be helpful here.

https://codereview.chromium.org/1165223004/diff/80001/device/usb/device_manag...
File device/usb/device_manager_impl.cc (right):

https://codereview.chromium.org/1165223004/diff/80001/device/usb/device_manag...
device/usb/device_manager_impl.cc:62: OPEN_DEVICE_ERROR_ACCESS_DENIED));
Why not base::Bind(callback, OPEN_DEVICE_ERROR_ACCESS_DENIED) here and below?

[Ken Rockot(use gerrit already), 2015-06-10 20:15:51]

https://codereview.chromium.org/1165223004/diff/80001/device/devices_app/devi...
File device/devices_app/devices_app.cc (right):

https://codereview.chromium.org/1165223004/diff/80001/device/devices_app/devi...
device/devices_app/devices_app.cc:46: // does presume not catch all common local
host strings.
On 2015/06/10 20:02:47, reillyg wrote:
> does not presume to

Done.

https://codereview.chromium.org/1165223004/diff/80001/device/devices_app/devi...
device/devices_app/devices_app.cc:47: if (remote_url_.host() == "127.0.0.1" ||
remote_url_.host() == "localhost")
On 2015/06/10 20:02:47, reillyg wrote:
> net::IsLocalHost() may be helpful here.

I was going to use that, but (despite the fact that other code does use it,
including extensions code) it is marked as NET_EXPORT_PRIVATE and isn't intended
to be consumed outside of //net.

https://codereview.chromium.org/1165223004/diff/80001/device/usb/device_manag...
File device/usb/device_manager_impl.cc (right):

https://codereview.chromium.org/1165223004/diff/80001/device/usb/device_manag...
device/usb/device_manager_impl.cc:62: OPEN_DEVICE_ERROR_ACCESS_DENIED));
On 2015/06/10 20:02:47, reillyg wrote:
> Why not base::Bind(callback, OPEN_DEVICE_ERROR_ACCESS_DENIED) here and below?

base::Bind unfortunately does not know how to rebind a mojo::Callback, and
probably never will since base can't depend on core mojo and core mojo can't
depend on base.

If this ends up being a frequent occurrence, we can add some utility code to
//mojo/common

[Reilly Grant (use Gerrit), 2015-06-10 21:33:17]

lgtm

https://codereview.chromium.org/1165223004/diff/80001/device/usb/device_manag...
File device/usb/device_manager_impl.cc (right):

https://codereview.chromium.org/1165223004/diff/80001/device/usb/device_manag...
device/usb/device_manager_impl.cc:62: OPEN_DEVICE_ERROR_ACCESS_DENIED));
On 2015/06/10 20:15:51, Ken Rockot wrote:
> On 2015/06/10 20:02:47, reillyg wrote:
> > Why not base::Bind(callback, OPEN_DEVICE_ERROR_ACCESS_DENIED) here and
below?
> 
> base::Bind unfortunately does not know how to rebind a mojo::Callback, and
> probably never will since base can't depend on core mojo and core mojo can't
> depend on base.
> 
> If this ends up being a frequent occurrence, we can add some utility code to
> //mojo/common

I'm going to put in a strong vote in favor of importing some equivalent of
base::Bind into Mojo. It doesn't have to support all the memory safety stuff
that base::Bind does though there may be equivalent concepts useful in Mojo.

[Ken Rockot(use gerrit already), 2015-06-10 23:09:13]

rockot@chromium.org changed reviewers:
+ sky@chromium.org

[Ken Rockot(use gerrit already), 2015-06-10 23:09:14]

sky could you please take a look at:

chrome/browser/BUILD.gn
chrome/browser/DEPS
chrome/browser/chrome_content_browser_client.cc
chrome/browser/chrome_content_browser_client.h

Thanks!

[sky, 2015-06-10 23:36:35]

LGTM

[Ken Rockot(use gerrit already), 2015-06-11 05:16:00]

The CQ bit was checked by rockot@chromium.org

[Ken Rockot(use gerrit already), 2015-06-11 05:16:00]

The patchset sent to the CQ was uploaded after l-g-t-m from
reillyg@chromium.org, sky@chromium.org
Link to the patchset: https://codereview.chromium.org/1165223004/#ps160001
(title: "move some deps to public_deps to satisfy new GN check ruels")

[commit-bot: I haz the power, 2015-06-11 05:16:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1165223004/160001

[commit-bot: I haz the power, 2015-06-11 05:54:04]

Committed patchset #5 (id:160001)

[commit-bot: I haz the power, 2015-06-11 05:55:07]

Patchset 5 (id:??) landed as
https://crrev.com/ed800dc138edfa261aa766281b96f736ff9824f6
Cr-Commit-Position: refs/heads/master@{#333899}

[kjellander_chromium, 2015-06-11 13:40:41]

A revert of this CL (patchset #5 id:160001) has been created in
https://codereview.chromium.org/1176383002/ by kjellander@chromium.org.

The reason for reverting is: This broke the Win 8 GN (dbg) and Win x64 GN (dbg)
bots:

https://build.chromium.org/p/chromium.win/builders/Win8%20GN%20%28dbg%29/buil...
https://build.chromium.org/p/chromium.win/builders/Win%20x64%20GN%20%28dbg%29....