fix python XBMCAddon::xbmcgui::Window::onAction() action.getButtonCode()

[icr8ions]

Description

• changes to populate buttonCode for CAction such that it is available for python addons with custom controllers

Motivation and context

• addons such as Keymap Editor do not work with custom controllers as the code being returned is the default value 0

How has this been tested?

• successfully built on MacOS 11.5.2 and tested on MacOS 11.5.2 and 12.6 - functionality of Keymap Editor has been restored for custom controllers
• thread: https://forum.kodi.tv/showthread.php?tid=356730&pid=3166711#pid3166711
  not working snippet:
  2023-09-18 12:51:19.178 T:3131545   debug : EventClient: button code 194 pressed
  2023-09-18 12:51:19.178 T:3131545   debug : EventClient: button code 194 released
  2023-09-18 12:51:19.186 T:3131425   debug : EventServer: key 194 translated to action ActivateWindow(Programs)
  2023-09-18 12:51:19.187 T:3131425   debug : Activating window ID: 10001
  2023-09-18 12:51:19.187 T:5238372   debug : action.getButtonCode(): 0
  working snippet:
  2023-09-20 11:34:16.696 T:5759207   debug : EventClient: button code 194 pressed
  2023-09-20 11:34:16.697 T:5759207   debug : EventClient: button code 194 released
  2023-09-20 11:34:16.725 T:5758980   debug : EventServer: key 194 translated to action ActivateWindow(Programs)
  2023-09-20 11:34:16.725 T:5758980   debug : Activating window ID: 10001
  2023-09-20 11:34:16.725 T:5759731   debug : action.getButtonCode(): 194

What is the effect on users?

• should allow end-users to use Keymap Editor to regain functionality of their custom remotes and controllers

Screenshots (if appropriate):

Types of change

• Bug fix (non-breaking change which fixes an issue)
• Clean up (non-breaking change which removes non-working, unmaintained functionality)
• Improvement (non-breaking change which improves existing functionality)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that will cause existing functionality to change)
• Cosmetic change (non-breaking change that doesn't touch code)
• Student submission (PR was done for educational purposes and will be treated as such)
• None of the above (please explain below)

Checklist:

• My code follows the Code Guidelines of this project
• My change requires a change to the documentation, either Doxygen or wiki
• I have updated the documentation accordingly
• I have read the Contributing document
• I have added tests to cover my change
• All new and existing tests passed

[icr8ions]

Looks like the build is picking up old code - not the most recent push.
Action.h used to have two constructors with a similar signature on line 26 and 31. At this time there is no constructor on line 31 so I am wondering why stale pushes are sticking around for the build.

Why would the PR build be picking up stale code? The line numbers in the errors for a file do not match the actual commit.

Action.h snippet:
26 CAction(int actionID,
27 float amount1 = 1.0f,
28 float amount2 = 0.0f,
29 const std:: string& name = "",
30 unsigned int holdTime = 0,
31 unsigned int buttonCode = 0);
32 CAction(int actionID, wchar_t unicode);

error in recent builds:
/home/jenkins/jenkins-root/workspace/android-arm-docker@2/xbmc/input/actions/Action.h:26:3: note: candidate constructor
CAction(int actionID,
^
/home/jenkins/jenkins-root/workspace/android-arm-docker@2/xbmc/input/actions/Action.h:31:3: note: candidate constructor
CAction(int actionID,
^

[sundermann]

That can happen sometimes if you force push while jenkins is trying to build your PR

[sundermann]

Jenkins build this please

[icr8ions]

Any chances of getting this merged into Nexus as well?

[enen92]

Change makes sense and should cause no harm.
Better to use initializer lists but that's out of the scope of your fix attempt.