exo: Add initial support for service side decorations.

exo: Add initial support for service side decorations.

This adds the ability to use service side window decorations
for shell surfaces. XDG shell surfaces require client side
decorations like before but wl_shell_surfaces are now using
service side decorations.

The surface bounds logic has been improved to support a
client view that is not at the origin of the widget and
resize logic is now working correctly when interactive
resize is not initiated by the client.

BUG=
TEST=exo_unittests

Committed: https://crrev.com/ae5575e315a97682ce1b3fc881cfd3b828d39941
Cr-Commit-Position: refs/heads/master@{#437391}

[reveman, 2016-12-03 00:14:45]

reveman@chromium.org changed reviewers:
+ dcastagna@chromium.org

[Daniele Castagna, 2016-12-08 20:19:13]

LGTM

https://codereview.chromium.org/2548653005/diff/40001/components/exo/pointer_...
File components/exo/pointer_unittest.cc (right):

https://codereview.chromium.org/2548653005/diff/40001/components/exo/pointer_...
components/exo/pointer_unittest.cc:179: false,
ash::kShellWindowId_DefaultContainer));
supernit: Do you want to add /* can_minimize */ as you did in display.cc?

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
File components/exo/shell_surface.cc (left):

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
components/exo/shell_surface.cc:1072:
ash::wm::ToWindowShowState(ash::wm::WINDOW_STATE_TYPE_AUTO_POSITIONED) ==
:( the animations were cool.

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
File components/exo/shell_surface.cc (right):

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
components/exo/shell_surface.cc:1095: // Notify client of initial state.
This can go away.

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
File components/exo/shell_surface.h (right):

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
components/exo/shell_surface.h:54: bool can_minimize,
nit: Should this be minimizable to stay consistent with activatable that is an
adjective?

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
components/exo/shell_surface.h:287: const bool can_minimize_;
nit: use an adjective instead?

[reveman, 2016-12-08 21:33:53]

The CQ bit was checked by reveman@chromium.org

[reveman, 2016-12-08 21:33:53]

The patchset sent to the CQ was uploaded after l-g-t-m from
dcastagna@chromium.org
Link to the patchset: https://codereview.chromium.org/2548653005/#ps60001
(title: "fix typo")

[commit-bot: I haz the power, 2016-12-08 21:34:49]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-08 23:01:16]

CQ is committing da patch.

Bot data: {"patchset_id": 60001, "attempt_start_ts": 1481232833594990,
"parent_rev": "d686559049ca586a3fe56cf26d2a006cdc23dc6d", "commit_rev":
"8c8102a266ac5b104b38be50be656b7a981c4e68"}

[commit-bot: I haz the power, 2016-12-08 23:01:44]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2016-12-08 23:03:27]

Description was changed from

==========
exo: Add initial support for service side decorations.

This adds the ability to use service side window decorations
for shell surfaces. XDG shell surfaces require client side
decorations like before but wl_shell_surfaces are now using
service side decorations.

The surface bounds logic has been improved to support a
client view that is not at the origin of the widget and
resize logic is now working correctly when interactive
resize is not initiated by the client.

BUG=
TEST=exo_unittests
==========

to

==========
exo: Add initial support for service side decorations.

This adds the ability to use service side window decorations
for shell surfaces. XDG shell surfaces require client side
decorations like before but wl_shell_surfaces are now using
service side decorations.

The surface bounds logic has been improved to support a
client view that is not at the origin of the widget and
resize logic is now working correctly when interactive
resize is not initiated by the client.

BUG=
TEST=exo_unittests

Committed: https://crrev.com/ae5575e315a97682ce1b3fc881cfd3b828d39941
Cr-Commit-Position: refs/heads/master@{#437391}
==========

[commit-bot: I haz the power, 2016-12-08 23:03:28]

Patchset 4 (id:??) landed as
https://crrev.com/ae5575e315a97682ce1b3fc881cfd3b828d39941
Cr-Commit-Position: refs/heads/master@{#437391}

[reveman, 2016-12-11 07:14:05]

hm, I guess these replies never got sent out..

https://codereview.chromium.org/2548653005/diff/40001/components/exo/pointer_...
File components/exo/pointer_unittest.cc (right):

https://codereview.chromium.org/2548653005/diff/40001/components/exo/pointer_...
components/exo/pointer_unittest.cc:179: false,
ash::kShellWindowId_DefaultContainer));
On 2016/12/08 at 20:19:13, Daniele Castagna wrote:
> supernit: Do you want to add /* can_minimize */ as you did in display.cc?

we don't have it for tests (see "true" above) and I'd rather not change that all
over as part of this patch.

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
File components/exo/shell_surface.cc (left):

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
components/exo/shell_surface.cc:1072:
ash::wm::ToWindowShowState(ash::wm::WINDOW_STATE_TYPE_AUTO_POSITIONED) ==
On 2016/12/08 at 20:19:13, Daniele Castagna wrote:
> :( the animations were cool.

sorry. maybe we'll enable them again in the future.

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
File components/exo/shell_surface.cc (right):

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
components/exo/shell_surface.cc:1095: // Notify client of initial state.
On 2016/12/08 at 20:19:13, Daniele Castagna wrote:
> This can go away.

Done.

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
File components/exo/shell_surface.h (right):

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
components/exo/shell_surface.h:54: bool can_minimize,
On 2016/12/08 at 20:19:13, Daniele Castagna wrote:
> nit: Should this be minimizable to stay consistent with activatable that is an
adjective?

The name is chosen based on the window state or widget function it's returned
from. That is currently CanMinimize.

https://codereview.chromium.org/2548653005/diff/40001/components/exo/shell_su...
components/exo/shell_surface.h:287: const bool can_minimize_;
On 2016/12/08 at 20:19:13, Daniele Castagna wrote:
> nit: use an adjective instead?

some reply as above