| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
Only Vulkan is fully implemented for now, so OpenGL isn't available
in the project manager yet.
This also makes the rendering driver checks use lowercase names
everywhere for consistency.
|
|
buffer
|
|
clang-format keeps breaking the way it handles break *after* ternary operators,
so I give up and go with the only style they seem to actually test.
|
|
All reviewed manually and occasionally rewritten to avoid bad auto formatting.
|
|
Sets `AlignOperands` to `DontAlign`.
`clang-format` developers seem to mostly care about space-based indentation and
every other version of clang-format breaks the bad mismatch of tabs and spaces
that it seems to use for operand alignment. So it's better without, so that it
respects our two-tabs `ContinuationIndentWidth`.
|
|
Doesn't change the default behavior, but allows plugins to add their
view behind the main view, which gives more control over what happens
with inputs and can be useful along with transparent rendering.
|
|
|
|
|
|
Godot host.
|
|
This is done by providing API access to app specific directories which don't have any limitations and allows us to bump the target sdk version to 30.
In addition, we're also bumping the min sdk version to 19 as version 18 is no longer supported by Google Play Services and only account of 0.3% of Android devices.
|
|
Key, touch and joystick events will be passed directly from the UI thread to Godot, so they can benefit from agile input flushing.
As another consequence of this new way of passing events, less Java object are created at runtime (`Runnable`), which is good since the garbage collector needs to run less.
`AndroidInputHandler` is introduced to have a smaller cross-thread surface. `main_loop_request_go_back()` is removed in favor just inline calling `send_window_event()` at the most caller's convenience (i.e., leveraging the new `p_deferred`` parameter as appropriate).
Lastly, `get_mouse_position()` and `get_mouse_button_state()` now just call through `Input` to avoid the need of sync of mouse data tracked on the UI thread.
|
|
This was caused by the fact that a new instance of Godot was created at resume while a previous instance already existed.
The previous instance would then go through its cleanup lifecycle, and would thus attempt to close the entire app, leading to the system to restart the app, thus starting the cycle anew.
The fix involves reusing the previous instance of Godot if one is available instead of creating a new one, as well as giving control to the host activity for how the process should be terminated.
|
|
|
|
Found via `codespell -q 3 -S ./thirdparty,*.po,./DONORS.md -L ackward,ang,ans,ba,beng,cas,childs,childrens,dof,doubleclick,fave,findn,hist,inout,leapyear,lod,nd,numer,ois,ony,paket,seeked,sinc,switchs,te,uint`
|
|
Queue the calls to GodotLib.key when Android virtual done is pressed
|
|
|
|
|
|
Replace single method anonymous classes with lambdas in Godot Java code
|
|
|
|
|
|
Use StringBuilder instead StringBuffer in Godot Java code
|
|
Fix raw use of parameterized Class
|
|
|
|
|
|
|
|
|
|
|
|
Use Java array declarations not C-style declarations in Android Java code
|
|
Remove redundant interface modifiers from Android Java code
|
|
|
|
|
|
|
|
|
|
Add OS.get_external_data_dir() to get Android external directory
|
|
|
|
|
|
Or after punctuation. Tried to leave third-party stuff alone, unless it has
been heavily modified for Godot.
|
|
|
|
|
|
|
|
`requestCode` and deprecate support for forwarding callbacks from Godot's parent activity.
|
|
- creating Vulkan context instead of OpenGL
- checking for validity of ENV in wrapper classes
- fix for access to JavaVM from threads
|
|
|
|
plugin methods.
|
|
The previous logic had the side effect of imposing a limit of one plugin per `aar` binary. The update lifts that restriction.
|
|
consistent with the previous activity based timeline.
|
|
This enables creation and use of a plugin like class by composition rather than inheritance.
|
|
Issues addressed:
a) Axis mappings were including virtual mouse axes on NVIDIA Shield TV.
The virtual mouse axes have the same axis numbers as the normal analog stick numbers. This was completely breaking joypad support on NVIDIA Shield TV.
b) Joypads were being tracked in a List with the index in the list being treated as the Godot device id.
If a device were to be removed, any device later in the list would be shifted, potentially causing future events with the shifted joypads to have incorrect IDs according to the Godot engine.
c) Unnecessary events were being sent to the Godot engine.
A check was added (per Joystick) that will prevent sending events for all axes when only a single axis value changed.
A similar check was added for "HATs".
See #45712
|