Hajime Hoshi
dc5869f607
internal/ui: bug fix: InitUnfocused option didn't work
...
Updates #2924
Closes #3099
2024-09-14 17:36:10 +09:00
Hajime Hoshi
daa85d17c2
internal/file: bug fix: VirtualFS.Open(".") should always return a new entry
...
Closes #3081
2024-09-06 10:46:35 +09:00
Hajime Hoshi
29946d037b
internal/gamepad: bug fix: crash when XInput and DInput gamepads were connected
...
Closes #3047
2024-07-20 22:30:56 +09:00
Hajime Hoshi
e5bb046a11
internal/gamepad: bug fix: GUID was not reliable as a unique identifier
...
Use a HID path as a gamepad unique identifier instead.
Closes #3046
2024-07-20 19:18:32 +09:00
cowboy.macke
bdb176a271
internal/ui: bug fix: crash by dropping an HTML element onto the canvas ( #3044 )
...
Closes #3043
2024-07-18 17:19:52 +09:00
Hajime Hoshi
71b7cedc5b
internal/graphicscommand: bug fix: buffered write pixel args might never be released
...
Closes #3036
2024-07-12 11:31:49 +09:00
Hajime Hoshi
9988a0c4e9
internal/ui: bug fix: the screen size must be initialized first even when the canvas is unfocused
...
Closes #3034
2024-07-04 16:19:10 +09:00
Hajime Hoshi
f6cdf401fc
internal/atlas: bug fix: compilation failure
2024-07-01 00:03:04 +09:00
Hajime Hoshi
001ac9a90e
internal/atlas: backends for unmanaged images were never released
...
Closes #3030
2024-06-30 23:47:46 +09:00
Hajime Hoshi
f4ceb961af
internal/ui: bug fix: the window size callback was not called on macOS
...
This is basically a revert of e150745598
.
As the window size callback was not called, ebiten.WindowSize didn't
work correctly for a while.
Even after this change, the issue #2257 could not be reproduced. This
is mysterious, but probably macOS itself has updated.
Updates #2257
Closes #3029
2024-06-30 16:02:23 +09:00
Hajime Hoshi
5e39cfec14
internal/ui: bug fix: wrong property names were specified
...
Closes #2975
2024-04-30 23:33:40 +09:00
Hajime Hoshi
17d75bfaad
internal/gamepaddb: bug fix: platform was not initialized correctly
...
After 6552ae1dbe
, the order of the init
function calls changed, and then the platform was not initialized
correctly.
This change fixes this issue by not relying on an init function to
get the platform.
Closes #2964
2024-04-18 13:30:31 +09:00
Hajime Hoshi
4d268f5ce4
internal/ui: bug fix: a window was shown unexpectedly at start-up
...
Closes #2951
2024-04-07 02:56:07 +09:00
Hajime Hoshi
3136de4958
internal/ui: bug fix: setting a size failed for an invisible window
...
Updates #2725
Closes #2951
2024-04-06 22:28:56 +09:00
Hajime Hoshi
5cab3cc88f
Revert "internal/graphicsdriver/directx: add 'tearing' for EBITENGINE_DIRECTX"
...
This reverts commit b6252a41f2
.
Reason: We need more investigation about tearing and vsync=off
Updates #2858
2024-03-30 20:09:01 +09:00
Hajime Hoshi
fd2c79398e
Revert "internal/atlas: refactoring: ensure ReadPixels to be processed in a frame"
...
This reverts commit 55702a7c28
.
Reason: This didn't work with the single-thread mode.
Updates #1704
Closes #2939
2024-03-26 13:10:00 +09:00
Hajime Hoshi
459ad709a6
internal/gamepaddb: omit the database for consoles
...
Updates #2936
2024-03-25 03:08:19 +09:00
Hajime Hoshi
04c4676b7c
internal/png: update with Go 1.22
2024-03-24 20:36:31 +09:00
Hajime Hoshi
6552ae1dbe
internal/gamepaddb: do not include the database for browsers
...
This reduces Wasm binary size by 525kb.
2024-03-24 20:16:11 +09:00
Hajime Hoshi
dc8baf96f9
internal/gamepaddb: update the database
2024-03-24 19:18:49 +09:00
Hajime Hoshi
d04d6271d0
internal/ui: bug fix: reset some input states on blur events on browsers
...
Closes #2874
2024-03-24 16:03:17 +09:00
Hajime Hoshi
f4029aaa77
ebiten: add (*Monitor).Size() to replace ScreenSizeInFullscreen()
...
Also, this change fixes redundant checks the case when a monitor
does not exist. Now Ebitengine checks a monitor existence at the
initialization.
Closes #2145
Closes #2795
2024-03-23 23:32:43 +09:00
Hajime Hoshi
b6252a41f2
internal/graphicsdriver/directx: add 'tearing' for EBITENGINE_DIRECTX
...
Now tearing happens only when 'tearing' is specified and vsync is off.
Closes #2858
2024-03-23 17:53:13 +09:00
Hajime Hoshi
ea6fce45ff
internal/shader: bug fix: crash on assignment mismatches
2024-03-23 15:30:49 +09:00
Hajime Hoshi
66667fe877
internal/shader: bug fix: test failures
...
Updates #2934
2024-03-23 15:15:33 +09:00
Hajime Hoshi
7842942b24
internal/shader: bug fix: need to covert constant type correctly for assignments
...
Closes #2934
2024-03-23 15:08:34 +09:00
Hajime Hoshi
e7bb66bb2f
internal/shader: bug fix: ++/-- statements didn't work for vec2 on browsers
...
Closes #2933
2024-03-22 23:15:33 +09:00
Mykhailo Lohachov
1586c6764a
internal/shaderir: Kage update support for shift operators ( #2916 )
...
Closes #2755
2024-03-22 18:14:03 +09:00
Hajime Hoshi
b7dd45c0e4
internal/gamepad: ignore the very first MotionEvent with 0 value for Android
...
On Android, MotionEvent with 0 values might come for axes when connecting
a gamepad, even though a user didn't touch any axes. This is problematic
especially for tirgger axes, where the default value should be -1.
This change fixes the issue by adding a new state `axesReady` to check
if an axis is really touched or not. If an axis is not touched yet,
a button value for a standard (trigger) button always returns 0.
This change also removes an old hack to initialize axis values for
triggers.
Closes #2598
2024-03-21 22:28:48 +09:00
Hajime Hoshi
9faa3f4601
internal/gamepaddb: refactoring
2024-03-16 17:57:52 +09:00
Hajime Hoshi
696938987d
internal/gamepad: use locks for consistency
...
Perhaps Gamepad's m might not be needed, but let's use the lock for
consistency for the current situation.
2024-03-16 16:31:49 +09:00
Hajime Hoshi
209dc50f72
internal/gamepaddb: refactoring
2024-03-16 15:37:42 +09:00
Hajime Hoshi
047858aa59
internal/gamepaddb: rename functions
2024-03-16 15:16:29 +09:00
Hajime Hoshi
bb6430d3ba
internal/shader: bug fix: unexpected crash for out of range
...
Closes #2926
2024-03-13 11:37:00 +09:00
Hajime Hoshi
7389f9ddb2
ebiten: add KeyIntlBackslash
...
Updates #2921
2024-03-12 12:49:02 +09:00
Hajime Hoshi
63e97c7064
internal/shader: bug fix: needed to resolve const and non-const types
...
Closes #2922
2024-03-10 19:49:19 +09:00
Hajime Hoshi
c9a973c6c1
internal/ui: bug fix: needed to focus the window at launch
...
Updates #2725
Closes #2924
2024-03-10 12:44:28 +09:00
Hajime Hoshi
9a7dcb1077
internal/shader: bug fix: failed to return an array in HLSL
...
Closes #2923
2024-03-10 11:59:50 +09:00
Hajime Hoshi
927e025982
internal/shader: bug fix: wrong type conversion for min, max, and clamp
...
Closes #2922
2024-03-10 11:30:06 +09:00
Hajime Hoshi
c0d9954b3e
exp/textinput: use native pixels for a candidate window position
2024-03-03 23:35:34 +09:00
Hajime Hoshi
3e4c47eb70
internal/ui: refactoring
2024-03-03 23:27:02 +09:00
Hajime Hoshi
cc3db584f2
internal/shader: allow integer types for clamp
...
Closes #2788
2024-02-25 21:21:27 +09:00
Hajime Hoshi
4b9508c523
internal/shader: bug fix: wrong type checks on min/max
...
Updates #2788
2024-02-25 21:17:54 +09:00
Hajime Hoshi
e8cdf885c0
internal/shader: allow integer types for min and max
...
Updates #2788
2024-02-25 20:18:19 +09:00
Hajime Hoshi
face5fd870
internal/shader: allow integer types for abs and sign
...
Updates #2788
2024-02-25 18:16:44 +09:00
Mykhailo Lohachov
012fe52b6f
internal/atlas: use bit manipulation for function power of 2 -like functions ( #2915 )
...
Closes #2914
2024-02-25 02:06:52 +09:00
Hajime Hoshi
64cb6cf8a9
internal/atlas: reduce test flakiness
2024-02-24 23:03:12 +09:00
Hajime Hoshi
732eedf2e0
all: use github.com/ebitengine/gomobile instead of golang.org/x/mobile
...
Closes #2748
Closes #2899
2024-02-24 22:07:27 +09:00
Hajime Hoshi
40f3cbe1e3
internal/ui: bug fix: DeviceScaleFactor should be initialized asap on Android
...
Closes #2913
2024-02-24 19:48:20 +09:00
Hajime Hoshi
df68cbfc5b
internal/gamepaddb: update the database
2024-02-21 20:36:46 +09:00