Merge lp:~dandrader/unity-api/app-state-handling into lp:unity-api
| Status: | Merged |
|---|---|
| Approved by: | Gerry Boland on 2015-05-12 |
| Approved revision: | no longer in the source branch. |
| Merged at revision: | 179 |
| Proposed branch: | lp:~dandrader/unity-api/app-state-handling |
| Merge into: | lp:unity-api |
| Diff against target: |
311 lines (+48/-71) 9 files modified
debian/changelog (+10/-2) include/unity/shell/application/ApplicationInfoInterface.h (+21/-4) include/unity/shell/application/ApplicationManagerInterface.h (+0/-29) include/unity/shell/application/CMakeLists.txt (+1/-1) test/qmltest/mocks/plugins/Unity/Application/Mocks/MockApplicationInfo.cpp (+9/-0) test/qmltest/mocks/plugins/Unity/Application/Mocks/MockApplicationInfo.h (+3/-4) test/qmltest/mocks/plugins/Unity/Application/Mocks/MockApplicationManager.cpp (+1/-21) test/qmltest/mocks/plugins/Unity/Application/Mocks/MockApplicationManager.h (+1/-7) test/qmltest/unity/shell/application/tst_Application.qml (+2/-3) |
| To merge this branch: | bzr merge lp:~dandrader/unity-api/app-state-handling |
| Related bugs: |
| Reviewer | Review Type | Date Requested | Status |
|---|---|---|---|
| Gerry Boland | 2015-05-08 | Approve on 2015-06-29 | |
| PS Jenkins bot | continuous-integration | Approve on 2015-05-11 | |
|
Review via email:
|
|||
Commit Message
Let users of the API suspend & resume individual apps at will
API Changes:
- Removed ApplicationMana
- Removed ApplicationMana
- Added ApplicationInfo
Description of the Change
Needed for having unity8 handling the application lifecycle (suspending&
| Gerry Boland (gerboland) wrote : | # |
+ Q_PROPERTY(bool active READ active WRITE setActive NOTIFY activeChanged)
FYI: on Thursday at UOS, we had a session on desktop app lifecycle stuff. There we decided that there will be more states than running/suspended. For the desktop we will want a "throttled" state, i.e. application has throttled CPU usage.
It will be up to unity8 to have that policy, so qtmir needs to export that capability.
In that case, we would prefer a property supporting an enum: Running/
Could you change this property to have Running/Suspended states, with a view to supporting more in future?
| Daniel d'Andrada (dandrader) wrote : | # |
> + Q_PROPERTY(bool active READ active WRITE setActive NOTIFY activeChanged)
> FYI: on Thursday at UOS, we had a session on desktop app lifecycle stuff.
> There we decided that there will be more states than running/suspended. For
> the desktop we will want a "throttled" state, i.e. application has throttled
> CPU usage.
>
> It will be up to unity8 to have that policy, so qtmir needs to export that
> capability.
>
> In that case, we would prefer a property supporting an enum:
> Running/
>
> Could you change this property to have Running/Suspended states, with a view
> to supporting more in future?
Done.
| PS Jenkins bot (ps-jenkins) wrote : | # |
PASSED: Continuous integration, rev:180
http://
Executed test runs:
SUCCESS: http://
SUCCESS: http://
deb: http://
SUCCESS: http://
Click here to trigger a rebuild:
http://
| Gerry Boland (gerboland) wrote : | # |
+ * @brief The aplication's requested state
typo. Also "state" is vague in this context, would elaborate slightly to "application's requested running state"
+ virtual void setRequestedSta
+ void requestedStateC
"const RequestedState" would do no harm, and would help compiler slightly
Looks good otherwise. Thanks for the API change
| Daniel d'Andrada (dandrader) wrote : | # |
On 11/05/15 08:07, Gerry Boland wrote:
> Review: Needs Fixing
>
> + * @brief The aplication's requested state
> typo. Also "state" is vague in this context, would elaborate slightly to "application's requested running state"
Done.
>
> + virtual void setRequestedSta
> + void requestedStateC
> "const RequestedState" would do no harm, and would help compiler slightly
I does do some harm by making the signature longer, and the benefit of
having a "const int" parameter (as merely a hint to the reader) is
arguable in my opinion. Would like a third vote on this.
| PS Jenkins bot (ps-jenkins) wrote : | # |
PASSED: Continuous integration, rev:181
http://
Executed test runs:
SUCCESS: http://
SUCCESS: http://
deb: http://
SUCCESS: http://
Click here to trigger a rebuild:
http://
| Michael Zanetti (mzanetti) wrote : | # |
> > + virtual void setRequestedSta
> > + void requestedStateC
> > "const RequestedState" would do no harm, and would help compiler slightly
>
> I does do some harm by making the signature longer, and the benefit of
> having a "const int" parameter (as merely a hint to the reader) is
> arguable in my opinion. Would like a third vote on this.
As Gerry said earlier, we are in Bikeshed land here, but as you asked for a third opinion, I'm -1 for const int and agree with Daniel here.
- 173. By Daniel d'Andrada on 2015-06-11
-
Add supportedOrient
ations and rotatesWindowCo ntents to ApplicationInfo
Approved by: PS Jenkins bot, Gerry Boland - 174. By CI Train Bot Account on 2015-06-11
-
Releasing 7.97+15.
10.20150611- 0ubuntu1 - 175. By Daniel d'Andrada on 2015-06-19
-
Changes to application lifecycle APIs
* Removed ApplicationMana
gerInterface. forceDashActive
* Removed ApplicationManagerInterface. suspended
* Added ApplicationInfoInterface. requestedState
- 176. By Daniel d'Andrada on 2015-08-03
-
Merge trunk
[ Mirco Müller (MacSlow) ]
* added alerting/setAlerting API to LauncherModel and LauncherItem interfaces
[ Michi Henning ]
* Remove dependency on gcc 4.9. (LP: #1452342)

PASSED: Continuous integration, rev:179 jenkins. qa.ubuntu. com/job/ unity-api- ci/297/ jenkins. qa.ubuntu. com/job/ unity-api- vivid-amd64- ci/17 jenkins. qa.ubuntu. com/job/ unity-api- vivid-armhf- ci/17 jenkins. qa.ubuntu. com/job/ unity-api- vivid-armhf- ci/17/artifact/ work/output/ *zip*/output. zip jenkins. qa.ubuntu. com/job/ unity-api- vivid-i386- ci/17
http://
Executed test runs:
SUCCESS: http://
SUCCESS: http://
deb: http://
SUCCESS: http://
Click here to trigger a rebuild: s-jenkins. ubuntu- ci:8080/ job/unity- api-ci/ 297/rebuild
http://