Ozone: Add KeyMap support on Non-ChromeOS builds.

Ozone: Add KeyMap support on Non-ChromeOS builds.

SetCurrentLayoutByName has needed support for ChromeOS. It would be
nice to have needed non-ChromeOS support, so that other Ozone implementations
can re-use all this work. This patch adds needed support for retrieving Key
maps from the passed string.

Committed: https://crrev.com/c288ff747a395a18dc0103561c29f08c7d8de9ce
Cr-Commit-Position: refs/heads/master@{#310110}

[kalyank, 2014-12-26 06:26:21]

kalyan.kondapally@intel.com changed reviewers:
+ spang@chromium.org

[spang, 2015-01-02 16:35:49]

lgtm

[kalyank, 2015-01-05 20:33:34]

On 2015/01/02 16:35:49, spang wrote:
> lgtm

The build failures don't seem to be related, should this be committed ?

[spang, 2015-01-06 00:33:38]

On 2015/01/05 20:33:34, kalyank wrote:
> On 2015/01/02 16:35:49, spang wrote:
> > lgtm
> 
> The build failures don't seem to be related, should this be committed ?

That's fine, I think you'll need to rebase though.

[kalyank, 2015-01-06 10:54:15]

On 2015/01/06 00:33:38, spang wrote:
> On 2015/01/05 20:33:34, kalyank wrote:
> > On 2015/01/02 16:35:49, spang wrote:
> > > lgtm
> > 
> > The build failures don't seem to be related, should this be committed ?
> 
> That's fine, I think you'll need to rebase though.

Done. Can you PTAL?

[spang, 2015-01-06 18:00:11]

Patchset #2 (id:20001) has been deleted

[spang, 2015-01-06 18:00:15]

The CQ bit was checked by spang@chromium.org

[commit-bot: I haz the power, 2015-01-06 18:01:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/822383002/1

[commit-bot: I haz the power, 2015-01-06 18:13:19]

Committed patchset #1 (id:1)

[spang, 2015-01-06 18:14:39]

On 2015/01/06 18:13:19, I haz the power (commit-bot) wrote:
> Committed patchset #1 (id:1)

Landed first PS because the conflicting CL was reverted.

[commit-bot: I haz the power, 2015-01-06 18:15:07]

Patchset 1 (id:??) landed as
https://crrev.com/c288ff747a395a18dc0103561c29f08c7d8de9ce
Cr-Commit-Position: refs/heads/master@{#310110}

[spang, 2015-01-07 18:09:05]

spang@chromium.org changed reviewers:
+ kpschoedel@chromium.org

[spang, 2015-01-07 18:09:05]

https://codereview.chromium.org/822383002/diff/1/ui/events/ozone/layout/xkb/x...
File ui/events/ozone/layout/xkb/xkb_keyboard_layout_engine.cc (right):

https://codereview.chromium.org/822383002/diff/1/ui/events/ozone/layout/xkb/x...
ui/events/ozone/layout/xkb/xkb_keyboard_layout_engine.cc:678:
layout_name.c_str(),
How did you test this? I'm now confused.. xkb_map_new_from_string needs a keymap
file, not a name, right?

[spang, 2015-01-07 18:10:16]

On 2015/01/07 18:09:05, spang wrote:
>
https://codereview.chromium.org/822383002/diff/1/ui/events/ozone/layout/xkb/x...
> File ui/events/ozone/layout/xkb/xkb_keyboard_layout_engine.cc (right):
> 
>
https://codereview.chromium.org/822383002/diff/1/ui/events/ozone/layout/xkb/x...
> ui/events/ozone/layout/xkb/xkb_keyboard_layout_engine.cc:678:
> layout_name.c_str(),
> How did you test this? I'm now confused.. xkb_map_new_from_string needs a
keymap
> file, not a name, right?

Er, keymap file as a string.

[kpschoedel, 2015-01-07 18:17:08]

https://codereview.chromium.org/822383002/diff/1/ui/events/ozone/layout/xkb/x...
File ui/events/ozone/layout/xkb/xkb_keyboard_layout_engine.cc (right):

https://codereview.chromium.org/822383002/diff/1/ui/events/ozone/layout/xkb/x...
ui/events/ozone/layout/xkb/xkb_keyboard_layout_engine.cc:678:
layout_name.c_str(),
On 2015/01/07 18:09:05, spang wrote:
> How did you test this? I'm now confused.. xkb_map_new_from_string needs a
keymap
> file, not a name, right?

xkb_map_new_from_string() takes a layout description (like the output of
`setxkbmap -print | xkbcomp -xkb - -`). We don't currently have a spec for what
the ‘name’ in KeyboardLayoutEngine::SetCurrentLayoutByName() is, so for the
present it's functionally the same as IME SetCurrentKeyboardLayoutByName().