status
stringclasses 1
value | repo_name
stringclasses 13
values | repo_url
stringclasses 13
values | issue_id
int64 1
104k
| updated_files
stringlengths 11
1.76k
| title
stringlengths 4
369
| body
stringlengths 0
254k
β | issue_url
stringlengths 38
55
| pull_url
stringlengths 38
53
| before_fix_sha
stringlengths 40
40
| after_fix_sha
stringlengths 40
40
| report_datetime
unknown | language
stringclasses 5
values | commit_datetime
unknown |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
closed | godotengine/godot | https://github.com/godotengine/godot | 78,760 | ["platform/windows/display_server_windows.cpp"] | Vertical scrollbar can't be clicked and dragged in exported property dropdown selection | ### Godot version
v4.1.beta3.mono.official [ada712e06]
### System information
Godot v4.1.beta3.mono - Windows 10.0.19045 - Vulkan (Forward+) - dedicated NVIDIA GeForce GTX 1660 Ti with Max-Q Design (NVIDIA; 31.0.15.3179) - Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz (12 Threads)
### Issue description
Vertical scrollbar can't be clicked and dragged in exported property dropdown selection. When you hover over a scrollbar it will light up, but can't be clicked and dragged like other scrollbars.
### Steps to reproduce
Export a property that has a lot of options to select like `[Export] Resource resource;`
In the inspector open property selection dropdown then try to click and drag vertical scrollbar.
https://github.com/godotengine/godot/assets/8630984/60d8f85b-c72a-46b9-bcf7-1f28e680111a
### Minimal reproduction project
NA | https://github.com/godotengine/godot/issues/78760 | https://github.com/godotengine/godot/pull/85484 | c2d7cfe5387db5792a8832a58d7735969a827130 | 7b1a5de2841c98c81b98b09d5e7363fed078b341 | "2023-06-27T16:11:17Z" | c++ | "2023-12-04T22:15:18Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,750 | ["servers/physics_2d/godot_body_2d.cpp", "servers/physics_3d/godot_body_3d.cpp"] | Mass not set properly for RigidBody3D with custom inertia and center of mass | ### Godot version
v4.0.3.stable.official [5222a99f5]
### System information
Windows 11 - v4.0.3.stable - Forward+
### Issue description
EDIT: Applicable to RigidBody2D in the same way.
If a RigidBody3D uses both custom Center of Mass and custom Inertia and Mass != 1, all forces are scaled by the mass of the body. Using only one custom Center of Mass or custom Inertia results in correct and expected behavior.
For example, if we apply a force = 1N to a body with mass = 2kg it will accelerate with acceleration = 1 m/s^2.
Instead this body should accelerate with acceleration = 0.5 m/s^2.
Cause:
RigidBody3D uses 2 fields to store mass information: ```mass``` and ```_inv_mass```. The second is calculated as an inverse of the first in ```GodotBody3D::update_mass_properties()```. The problem arises when RigidBody3D uses both custom Center of Mass and custom Inertia. In such case the body is not added to the list of bodies requiring mass properties update in ```GodotBody3D::_mass_properties_changed()```.
This results in ```mass``` holding value set by the user and ```_inv_mass``` always equal 1, which in turn causes incorrect physics calculations in GodotBody3D::integrate_forces().
I conducted a simple experiment to investigate this behavior.
3 RigidBody3D instances are created:
1) Mass = 1kg, Auto Center of Mass, Auto Inertia
2) Mass = 2kg, Auto Center of Mass, Auto Inertia
3) Mass = 2kg, Custom Center of Mass (0, 0, 0), Custom Inertia (0.2, 0.2, 0.2)
On all 3 bodies acts a force of 1N upward.
All 3 bodies use gravity scale = 0.
Expected behavior would be that body 1 accelerates 2 times faster then 2 and 3. This is not observed! Body 1 does accelerate twice as fast as body 2, but body 3 accelerates equally fast as 1.
Attached images below represent state at the start of the experiment and after few seconds.
This can be confirmed by printing mass and inv_mass of a RigidBody3D during integrate_forces method.
### Steps to reproduce
Attached project contains a scene reproducing the experiment described in Issue description.
1. Open attached minimal project
2. Run RigidBodyMassIssue3D_apply_force.tscn
3. Observe how body with custom center of mass and custom inertia accelerates equally fast as body with mass = 1kg.
4. Change mass of body with custom center of mass and custom inertia and rerun the project, observe how the acceleration does not change and is always equal to the first body.
Attached project also contains an alternative scene where the same bodies are subject to gravity.
1. Open attached minimal project
2. Run RigidBodyMassIssue3D_gravity.tscn
3. Under gravity all bodies should fall at the same rate. Observe how the body with custom center of mass and custom inertia accelerates faster then the other
4. Change mass of the body with custom center of mass and custom inertia. Observe how the acceleration is proportional to the mass
### Minimal reproduction project
[GodotRigidBodyMassIssue.zip](https://github.com/godotengine/godot/files/11883011/GodotRigidBodyMassIssue.zip)
| https://github.com/godotengine/godot/issues/78750 | https://github.com/godotengine/godot/pull/78757 | 21cdedb79c718258d66242aa7645511830f296f0 | 1e4165ac603946afaf77c92c1a3d515432d3c747 | "2023-06-27T14:05:21Z" | c++ | "2023-09-26T11:44:36Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,749 | ["drivers/vulkan/rendering_device_vulkan.cpp", "drivers/vulkan/rendering_device_vulkan.h", "drivers/vulkan/vulkan_context.cpp"] | Occasional `attempting free on address which was not malloc()-ed` crash in CI | ### Godot version
master CI builds
### System information
Linux / Editor with doubles and GCC sanitizers
### Issue description
The `Linux / Editor with doubles and GCC sanitizers` build can fail with a `attempting free on address which was not malloc()-ed` crash in the `Open and close editor (Vulkan)` step.
https://github.com/godotengine/godot/actions/runs/5366302204/jobs/9735786296#step:15:138
https://github.com/godotengine/godot/actions/runs/5389199469/jobs/9783070708?pr=78740#step:15:138
The failure doesn't seem related to either PR (in fact one of them is just a docs change). I have only seen this issue these two times, still probably worth looking into.
### Steps to reproduce
Open and close https://github.com/godotengine/regression-test-project/archive/4.0.zip using the Vulkan renderer.
### Minimal reproduction project
https://github.com/godotengine/regression-test-project/archive/4.0.zip | https://github.com/godotengine/godot/issues/78749 | https://github.com/godotengine/godot/pull/80296 | 3f3f865f8a2b9396b16da28ce29c3c7470bf36d4 | 497ca8c4338b75e0c185858266b12fb0df38d720 | "2023-06-27T13:31:36Z" | c++ | "2023-09-02T22:20:40Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,748 | ["scene/gui/slider.cpp"] | Disabled Sliders shouldn't be highlightable | ### Godot version
v4.1.rc.custom_build [1b3bf4852]
### System information
Godot v4.1.rc (1b3bf4852) - Arch Linux #1 ZEN SMP PREEMPT_DYNAMIC Sat, 10 Jun 2023 00:35:19 +0000 - Vulkan (Compatibility) - AMD Radeon R9 380 Series (tonga, LLVM 15.0.7, DRM 3.52, 6.3.7-zen1-1-zen) () - AMD Ryzen 5 5600X 6-Core Processor (12 Threads)
### Issue description
Sliders that are set to *not* editable will still be highlighted on hover, even if you cannot interact with them.
This seems counterintuitive (I'd expect to be able to interact with something that lights up on hover), and inconsistent (disabled buttons do not get highlighted on hover).
Screencap of hover behaviour:
https://github.com/godotengine/godot/assets/1187609/af9466ad-37d5-4d13-ace5-b6268166e2b6
In Godot 3, uneditable sliders behaved as you would expect and did not get highlighted on hover, same as buttons.
### Steps to reproduce
Add HSlider or VSlider, hover over with mouse.
### Minimal reproduction project
None. | https://github.com/godotengine/godot/issues/78748 | https://github.com/godotengine/godot/pull/78776 | 7030ac555f6af936174097c4642e60c1a6b0877b | bbb11004e788c88a7315fbe414688eeaaa6c1c6f | "2023-06-27T12:50:02Z" | c++ | "2023-07-10T10:32:39Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,741 | ["doc/classes/TreeItem.xml", "scene/gui/tree.cpp", "scene/gui/tree.h"] | Scene Panel: node names are overlapping icons | ### Godot version
4.1 Beta 3
### System information
Win10
### Issue description
In the Godot 4.1. Beta3 Scene Panel, node names are overlapping the icons to the right. Like visibility icons, script icons ...
Especially apparent with longer node names:
https://github.com/godotengine/godot/assets/47016402/61cc8c75-efe5-4cfa-ac4f-5fcd4805140b
This is _not_ an issue in Godot 4.0.3:
https://github.com/godotengine/godot/assets/47016402/4035ea3f-80bd-479d-9d6f-761debdd6b49
### Steps to reproduce
Give long node names and resize the Scene Panel
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78741 | https://github.com/godotengine/godot/pull/78756 | fff32bb72aa5f0507db28c989e6c07eaeb6ce56c | 017b19638b69d8fad81fb4a2145c6da9780d161e | "2023-06-27T11:40:53Z" | c++ | "2023-08-25T12:58:27Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,726 | ["editor/export/editor_export_platform.cpp", "modules/zip/zip_packer.cpp"] | Naming files in Chinese, `ZIPPacker` and `ZIPReader` will generate Garbled code | ### Godot version
v4.0.3.stable.official [5222a99f5]
### System information
win10 21H2,Vulkan API 1.2.0
### Issue description
I'm for use `"Beta: Use Unicode UTF-8 for worldwide language support"`
The file name will display as normal, but `ZIPReader.read_file()` is still garbled
I tried using 7zip Packing files,
Disable `"Beta: Use Unicode UTF-8 for worldwide language support"`
Is still displayed normally, but `ZIPReader.read_file()` is still garbled
### Steps to reproduce
run node.tscn, look error reporting.
### Minimal reproduction project
[test_zip.zip](https://github.com/godotengine/godot/files/11873912/test_zip.zip)
| https://github.com/godotengine/godot/issues/78726 | https://github.com/godotengine/godot/pull/78732 | 654132cb9c6ff1f27be0bf325e348e74b3e49fba | 375d89ced067dab809ae43232143609ca90eb9ff | "2023-06-26T20:34:50Z" | c++ | "2023-12-05T12:05:02Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,715 | ["drivers/vulkan/rendering_device_vulkan.cpp", "drivers/vulkan/rendering_device_vulkan.h"] | Crash when using SoftBody on mobile | ### Godot version
4.1 beta 3, 4.1 beta 1
### System information
Windows 10 - 4.1 beta 3 - Android 12
### Issue description
Crash when using SoftBody on android mobile.
### Steps to reproduce
- Simple scene setup using a plane softbody, camera.
- Export to android mobile.
- Immediate crash when open the app.
### Minimal reproduction project
[SoftbodyMobileTest.zip](https://github.com/godotengine/godot/files/11871734/SoftbodyMobileTest.zip)
| https://github.com/godotengine/godot/issues/78715 | https://github.com/godotengine/godot/pull/84852 | 8a9e3ad8d5623bad204964795504e5cd10289dad | a9ba8695d4ac453c5edc8339371116ea28f403aa | "2023-06-26T16:58:21Z" | c++ | "2023-12-04T22:06:04Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,689 | ["core/string/translation.cpp", "core/string/translation.h"] | Double free or corruption involving ResourceLoader.load_threaded_request | ### Godot version
4.0.3
### System information
Linux Ubuntu 22.04
### Issue description
Program terminates with: `double free or corruption (fasttop)`
I captured this backtrace in GDB:
```
Thread 22 "godot.linuxbsd." received signal SIGABRT, Aborted.
[Switching to Thread 0x7fffb57fa640 (LWP 45613)]
__pthread_kill_implementation (no_tid=0, signo=6, threadid=140736238429760) at ./nptl/pthread_kill.c:44
44 ./nptl/pthread_kill.c: No such file or directory.
(gdb) bt
#0 __pthread_kill_implementation (no_tid=0, signo=6, threadid=140736238429760) at ./nptl/pthread_kill.c:44
#1 __pthread_kill_internal (signo=6, threadid=140736238429760) at ./nptl/pthread_kill.c:78
#2 __GI___pthread_kill (threadid=140736238429760, signo=signo@entry=6) at ./nptl/pthread_kill.c:89
#3 0x00007ffff7c42476 in __GI_raise (sig=sig@entry=6) at ../sysdeps/posix/raise.c:26
#4 0x00007ffff7c287f3 in __GI_abort () at ./stdlib/abort.c:79
#5 0x00007ffff7c896f6 in __libc_message (action=action@entry=do_abort, fmt=fmt@entry=0x7ffff7ddbb8c "%s\n") at ../sysdeps/posix/libc_fatal.c:155
#6 0x00007ffff7ca0d7c in malloc_printerr (str=str@entry=0x7ffff7dde768 "double free or corruption (fasttop)") at ./malloc/malloc.c:5664
#7 0x00007ffff7ca2a4a in _int_free (av=0x7fff90000030, p=0x7fff90006ac0, have_lock=0) at ./malloc/malloc.c:4539
#8 0x00007ffff7ca54d3 in __GI___libc_free (mem=<optimized out>) at ./malloc/malloc.c:3391
#9 0x0000555558435a03 in memdelete<HashMapElement<StringName, Ref<StyleBox> > > (p_class=0x7fff90006ae0) at ./core/os/memory.h:112
#10 DefaultTypedAllocator<HashMapElement<StringName, Ref<StyleBox> > >::delete_allocation (p_allocation=0x7fff90006ae0, this=0x555561b15514)
at ./core/os/memory.h:206
#11 HashMap<StringName, Ref<StyleBox>, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Ref<StyleBox> > > >::clear (this=0x555561b15508) at ./core/templates/hash_map.h:265
#12 HashMap<StringName, Ref<StyleBox>, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Ref<StyleBox> > > >::~HashMap (this=<optimized out>, this=<optimized out>) at ./core/templates/hash_map.h:582
#13 KeyValue<StringName, HashMap<StringName, Ref<StyleBox>, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Ref<StyleBox> > > > >::~KeyValue (this=<optimized out>, this=<optimized out>) at ./core/templates/pair.h:82
#14 HashMapElement<StringName, HashMap<StringName, Ref<StyleBox>, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Ref<StyleBox> > > > >::~HashMapElement (this=<optimized out>, this=<optimized out>) at ./core/templates/hash_map.h:55
#15 memdelete<HashMapElement<StringName, HashMap<StringName, Ref<StyleBox>, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Ref<StyleBox> > > > > > (p_class=0x555561b154f0) at ./core/os/memory.h:109
#16 DefaultTypedAllocator<HashMapElement<StringName, HashMap<StringName, Ref<StyleBox>, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Ref<StyleBox> > > > > >::delete_allocation (p_allocation=0x555561b154f0, this=0x555561b56c14)
at ./core/os/memory.h:206
#17 HashMap<StringName, HashMap<StringName, Ref<StyleBox>, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Ref<StyleBox> > > >, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, HashMap<StringName, Ref<StyleBox>, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Ref<StyleBox> > > > > > >::clear (this=0x555561b56c08) at ./core/templates/hash_map.h:265
#18 0x00005555584aac79 in Control::_invalidate_theme_cache (this=0x555561b56480) at scene/gui/control.cpp:2353
#19 0x00005555584e1c98 in Control::_notification (this=0x555561b56480, p_notification=<optimized out>) at scene/gui/control.cpp:3063
#20 0x0000555556c3f3ee in Control::_notificationv (p_reversed=false, p_notification=2010, this=0x555561b56480) at ./scene/gui/control.h:47
#21 Container::_notificationv (p_reversed=false, p_notification=2010, this=0x555561b56480) at ./scene/gui/container.h:37
#22 ScrollContainer::_notificationv (this=0x555561b56480, p_notification=2010, p_reversed=<optimized out>) at ./scene/gui/scroll_container.h:39
#23 0x0000555559fc5a38 in Object::notification (this=0x555561b56480, p_notification=2010, p_reversed=<optimized out>) at core/object/object.cpp:790
#24 0x0000555558331f0e in Node::propagate_notification (this=0x555561b56480, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1926
--Type <RET> for more, q to quit, c to continue without paging--c
#25 0x0000555558331f2f in Node::propagate_notification (this=0x555561b7bf00, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#26 0x0000555558331f2f in Node::propagate_notification (this=0x555561b88fe0, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#27 0x0000555558331f2f in Node::propagate_notification (this=0x555560c08e10, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#28 0x0000555558331f2f in Node::propagate_notification (this=0x5555608641b0, p_notification=2010) at scene/main/node.cpp:1929
#29 0x0000555559fc5a38 in Object::notification (this=0x555560862000, p_notification=2010, p_reversed=<optimized out>) at core/object/object.cpp:790
#30 0x0000555559d4f37c in TranslationLoaderPO::load_translation (f=..., r_error=<optimized out>) at core/io/translation_loader_po.cpp:336
#31 0x0000555559d51f1a in TranslationLoaderPO::load (this=<optimized out>, p_path=..., p_original_path=..., r_error=0x55555de35cd4, p_use_sub_threads=<optimized out>, r_progress=0x55555de35cc8, p_cache_mode=ResourceFormatLoader::CACHE_MODE_REUSE) at core/io/translation_loader_po.cpp:355
#32 0x0000555559d26a5b in ResourceLoader::_load (p_path=..., p_original_path=..., p_type_hint=..., p_cache_mode=ResourceFormatLoader::CACHE_MODE_REUSE, r_error=0x55555de35cd4, p_use_sub_threads=false, r_progress=0x55555de35cc8) at core/io/resource_loader.cpp:214
#33 0x0000555559d2718a in ResourceLoader::_thread_load_function (p_userdata=0x55555de35c98) at core/io/resource_loader.cpp:241
#34 0x0000555559bc6d66 in Thread::callback (p_caller_id=<optimized out>, p_settings=..., p_callback=0x555559d27060 <ResourceLoader::_thread_load_function(void*)>, p_userdata=0x55555de35c98) at core/os/thread.cpp:65
#35 0x000055555a8ac074 in execute_native_thread_routine ()
#36 0x00007ffff7c94b43 in start_thread (arg=<optimized out>) at ./nptl/pthread_create.c:442
#37 0x00007ffff7d26a00 in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:81
```
I'm also getting a segfault in another thread, which dereferences a null pointer in the theme cache:
```
Thread 19 "godot.linuxbsd." received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fffb6ffd640 (LWP 49020)]
0x00005555584b0a9c in HashMap<StringName, HashMap<StringName, Color, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Color> > >, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, HashMap<StringName, Color, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Color> > > > > >::_lookup_pos (r_pos=<synthetic pointer>: 0, p_key=..., this=0x555561f6abf0) at ./core/templates/hash_map.h:119
119 if (hashes[pos] == hash && Comparator::compare(elements[pos]->data.key, p_key)) {
(gdb) bt
#0 0x00005555584b0a9c in HashMap<StringName, HashMap<StringName, Color, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Color> > >, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, HashMap<StringName, Color, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Color> > > > > >::_lookup_pos (
r_pos=<synthetic pointer>: 0, p_key=..., this=0x555561f6abf0) at ./core/templates/hash_map.h:119
#1 HashMap<StringName, HashMap<StringName, Color, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Color> > >, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, HashMap<StringName, Color, HashMapHasherDefault, HashMapComparatorDefault<StringName>, DefaultTypedAllocator<HashMapElement<StringName, Color> > > > > >::has (p_key=...,
this=<optimized out>, this=<optimized out>, p_key=...) at ./core/templates/hash_map.h:310
#2 Control::get_theme_color (this=<optimized out>, p_name=..., p_theme_type=...) at scene/gui/control.cpp:2510
#3 0x000055555847141e in Button::_update_theme_item_cache (this=0x555561f6a3c0) at scene/gui/button.cpp:64
#4 0x00005555584e1ca5 in Control::_notification (this=0x555561f6a3c0, p_notification=<optimized out>) at scene/gui/control.cpp:3064
#5 0x00005555578daabe in Control::_notificationv (p_reversed=false, p_notification=2010, this=0x555561f6a3c0) at ./scene/gui/control.h:47
#6 BaseButton::_notificationv (p_reversed=false, p_notification=2010, this=0x555561f6a3c0) at ./scene/gui/base_button.h:40
#7 Button::_notificationv (this=0x555561f6a3c0, p_notification=2010, p_reversed=<optimized out>) at ./scene/gui/button.h:38
#8 0x0000555559fc5a38 in Object::notification (this=0x555561f6a3c0, p_notification=2010, p_reversed=<optimized out>) at core/object/object.cpp:790
#9 0x0000555558331f0e in Node::propagate_notification (this=0x555561f6a3c0, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1926
#10 0x0000555558331f2f in Node::propagate_notification (this=0x555561f66350, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#11 0x0000555558331f2f in Node::propagate_notification (this=0x555561f55400, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#12 0x0000555558331f2f in Node::propagate_notification (this=0x555561f29fc0, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#13 0x0000555558331f2f in Node::propagate_notification (this=0x555561f28990, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#14 0x0000555558331f2f in Node::propagate_notification (this=0x555561f27490, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#15 0x0000555558331f2f in Node::propagate_notification (this=0x555561f26170, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#16 0x0000555558331f2f in Node::propagate_notification (this=0x5555618c5e40, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#17 0x0000555558331f2f in Node::propagate_notification (this=0x555561aaa4c0, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#18 0x0000555558331f2f in Node::propagate_notification (this=0x555560e4ae00, p_notification=p_notification@entry=2010) at scene/main/node.cpp:1929
#19 0x0000555558331f2f in Node::propagate_notification (this=0x555560862130, p_notification=2010) at scene/main/node.cpp:1929
#20 0x0000555559fc5a38 in Object::notification (this=0x555560869770, p_notification=2010, p_reversed=<optimized out>) at core/object/object.cpp:790
#21 0x0000555559d4f37c in TranslationLoaderPO::load_translation (f=..., r_error=<optimized out>) at core/io/translation_loader_po.cpp:336
#22 0x0000555559d51f1a in TranslationLoaderPO::load (this=<optimized out>, p_path=..., p_original_path=..., r_error=0x555562efd154, p_use_sub_threads=<optimized out>, r_progress=0x555562efd148,
p_cache_mode=ResourceFormatLoader::CACHE_MODE_REUSE) at core/io/translation_loader_po.cpp:355
#23 0x0000555559d26a5b in ResourceLoader::_load (p_path=..., p_original_path=..., p_type_hint=..., p_cache_mode=ResourceFormatLoader::CACHE_MODE_REUSE, r_error=0x555562efd154, p_use_sub_threads=false, r_progress=0x555562efd148)
at core/io/resource_loader.cpp:214
#24 0x0000555559d2718a in ResourceLoader::_thread_load_function (p_userdata=0x555562efd118) at core/io/resource_loader.cpp:241
#25 0x0000555559bc6d66 in Thread::callback (p_caller_id=<optimized out>, p_settings=..., p_callback=0x555559d27060 <ResourceLoader::_thread_load_function(void*)>, p_userdata=0x555562efd118) at core/os/thread.cpp:65
#26 0x000055555a8ac074 in execute_native_thread_routine ()
#27 0x00007ffff7c94b43 in start_thread (arg=<optimized out>) at ./nptl/pthread_create.c:442
#28 0x00007ffff7d26a00 in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:81
```
From the backtrace, it seems that when Translations are loaded, a notification is sent to the tree (perhaps to update translations). This is probably the culprit as multithreaded access to the scene is not allowed IIRC
### Steps to reproduce
Reproduction steps are currently not 100% clear to me. I'm currently trying to create a MRP, but I cannot reproduce it outside of my project at the moment.
However, it seems to involve these components:
* Loading a translation in a background thread
* Freeing (`queue_free`) a gui node while the background thread is loading
### Minimal reproduction project
WIP | https://github.com/godotengine/godot/issues/78689 | https://github.com/godotengine/godot/pull/78747 | 92960b7a22dabe26bb724b7b0d00a5fd03324f3a | 7c204874eb079fbd401a13e6222878425b7287bf | "2023-06-25T21:47:43Z" | c++ | "2023-07-26T16:39:27Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,686 | ["doc/classes/Mutex.xml"] | Documentation issue with mutex | ### Godot version
v4.1.beta3.official [ada712e06]
### System information
Linux/BSD
### Issue description
Documentation issue: it says that mutexes will return false if the thread already has ownership of the mutex
### Steps to reproduce
```gdscript
func _ready():
var mtx:=Mutex.new()
mtx.lock()
var v = mtx.try_lock()
if v:
print("knew docs were wrong <",v,">")
mtx.unlock()
else:
print('the hell')
mtx.unlock()
```
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78686 | https://github.com/godotengine/godot/pull/78700 | 320711bc8be26d7c6c15b1203d028202bfa6444a | afc5fa14adce7676746662751e63eab2d2ad4420 | "2023-06-25T21:17:55Z" | c++ | "2023-06-26T08:13:53Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,674 | ["core/input/input_event.cpp"] | Xbox "Start" button name does not match the official one | ### Godot version
v3.6.beta2.official [68c507f59]
### System information
w10 64
### Issue description
This image is from the official documentation.
Where is the "Menu" button?
What is the "Menu" button supposed to be?
### Steps to reproduce
...
### Minimal reproduction project
... | https://github.com/godotengine/godot/issues/78674 | https://github.com/godotengine/godot/pull/78701 | e3dee8cea9e8e29d89aa34fa899f74e358d48ebb | 73ca58b45fe39de87e4b664f8c1c09416b18596e | "2023-06-25T15:45:38Z" | c++ | "2023-07-08T16:17:21Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,672 | ["editor/progress_dialog.cpp", "editor/progress_dialog.h", "editor/window_wrapper.cpp"] | Saving script from floating script window moves window focus back to the main editor window | ### Godot version
master (3710f06929)
### System information
Godot v4.1.beta (3710f0692) - Arch Linux #1 ZEN SMP PREEMPT_DYNAMIC Sat, 10 Jun 2023 00:35:19 +0000 - Vulkan (Forward+) - dedicated AMD Radeon R9 380 Series (RADV TONGA) () - AMD Ryzen 5 5600X 6-Core Processor (12 Threads)
### Issue description
Using i3wm, tiling window manager.
If I'm in the floating script window and I hit Ctrl+S to save, the window focus gets switched to the main Godot window.
I'd expect it to stay on the script window instead.
Note: this happens only when the script is associated with a scene. A standalone script does not trigger the focus switch.
### Steps to reproduce
Float script window, use ctrl+s to save, see focus returned to main godot editor window.
Following screencap of this exact sequence. I switch focus to the script window, type "pass" and press ctrl+s. and you can see the blue square around the windows showing current focus move back to the main window.
https://github.com/godotengine/godot/assets/1187609/8638d269-3f82-4f75-b559-f6ed8d148d65
### Minimal reproduction project
You only need a scene with a script attached. | https://github.com/godotengine/godot/issues/78672 | https://github.com/godotengine/godot/pull/83290 | c851a46065c92b9101800502ac8c00fb940cbb55 | d89c5313a5324185c56e804f2ef3715930093cf4 | "2023-06-25T14:21:16Z" | c++ | "2023-11-15T13:32:07Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,669 | ["modules/multiplayer/multiplayer_synchronizer.cpp"] | MultiplayerSynchronizer - "Watch" properties - on_delta_receive: Condition "props.size() == 0" is true. Returning: ERR_INVALID_DATA | ### Godot version
v4.1.beta2.mono.official [a2575cba4]
### System information
Godot v4.1.beta2.mono - Windows 10.0.19045 - Vulkan (Forward+) - dedicated GeForce RTX 2060 () - AMD Ryzen 5 2600X Six-Core Processor (12 Threads)
### Issue description
I'm using a Multiplayer Synchronizer to sync some variables of two instantiated players (each one has local multiplayer authority). Setting the "watch" property on the multiplayer synchronizer properties causes the player who receive the delta update to receive this error and to not store the updated value:
```
E 0:00:01:0604 on_delta_receive: Condition "props.size() == 0" is true. Returning: ERR_INVALID_DATA
<C++ Source> modules/multiplayer/scene_replication_interface.cpp:778 @ on_delta_receive()
```
I don't know if the problem is Mono-specific, but the error comes up even on Node3D default properties.
### Steps to reproduce
- Create a simple scene with a platform
- Create a "Player" scene, that contains a MultiplayerSynchronizer, and put some properties to sync with the "watch" property ON and the "sync" property to OFF
- Create a script to make one player create the server and the other to connect as a client to the server
- Make the server instantiate the player on connect
- Make a script to change the synced variables, to trigger the watched properties change
### Minimal reproduction project
[Test_delta.zip](https://github.com/godotengine/godot/files/11859717/Test_delta.zip)
[Test_delta-gdscript.zip](https://github.com/godotengine/godot/files/11859841/Test_delta-gdscript.zip)
| https://github.com/godotengine/godot/issues/78669 | https://github.com/godotengine/godot/pull/78709 | d08ba5f3dca8fb01e60bd369b9d5a77138659b50 | 3aaa7870949853d3f42a1f8b67d16f8f2602931c | "2023-06-25T12:44:11Z" | c++ | "2023-06-26T12:57:12Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,658 | ["doc/classes/NodePath.xml"] | get_node() not correctly getting grandparent nodes | ### Godot version
4.0.3 Stable
### System information
Windows 11
### Issue description
using get_node("...") should return the grandparent node of the node it is being called from however it instead throws an error forcing the user to have to use get_node("..").get_node("..")
### Steps to reproduce
create a node. create a child for this node. create a child for this child node. attach a script to the grandchild node and in one of the prebuilt functions add get_node("...").name . When you try to run the program it will throw an error that it cannot get the property name on null instance.
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78658 | https://github.com/godotengine/godot/pull/78660 | e0de078d65ef6225dab8355a5140a0ada2822bf9 | a93c671b7d59d9d9be76558a8cc53bd87a0f42d1 | "2023-06-25T00:26:04Z" | c++ | "2023-06-26T08:11:56Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,657 | ["editor/script_create_dialog.cpp"] | Groups of spaces in scripts automatically changing into tabs/indentation | ### Godot version
4.0.3.stable
### System information
Windows 11, ask for more info if needed, I'm not sure what else would be needed here.
### Issue description
Issue also asked about on Godot Q&A (currently unresolved): https://ask.godotengine.org/155740/godot-automatically-creating-tabs-script-template-how-that
I have a script template that I use for all my scripts. In this template, I have an instance of a long line of spaces leading to a comment (this is my commenting method of choice, putting all comments on the right). Looks like (without underscores, would be spaces):
my code _ _ _ _ _ _ _ _ _ _ _ # my comment
It worked perfectly in prior versions, though after switching to Godot 4 it has started converting my groups of spaces into tabs. It now looks like this (with the indent carrots), despite me only including spaces there in my .gd file:
my code > > > > > > > > > # my comment
I cannot find a way to disable this "feature" and believe it to truly be a bug. Mr. J. Godfrey (a Godot contributor/Q&A admin I think?) believes so as well. I have taken a look at the indentation section of the editor settings but this is a different feature. I like tabs for indents, and am not trying to change that. I just don't want Godot to auto-change my spaces into indents when it sees a group of them. I understand others may, so I'm proposing there is an option created for that (or please inform me of which one it is if it exists). When I create/edit this template in Notepad, I'm able to verify that it is indeed just normal spaces.
### Steps to reproduce
1) Go to this link and download the script template I created (this program won't let me attach a .gd for some reason):
https://drive.google.com/file/d/1oXPKAn_3DulkF2oIFNYTFV3QKuNKdGz-/view?usp=sharing
2) Load it into a folder called "Objects" within your script templates folder (you may need to create this folder, found at *Godot master folder location*...Godot Engine\editor_data\script_templates\
3) Create a new script in Godot 4.0.3.stable and select this file as the template. You may need to create it on a node, type shouldn't matter.
4) Open the new script and see that the "extends" line has a bunch of tab indents when it should just be spaces.
### Minimal reproduction project
N/A, this is trivial | https://github.com/godotengine/godot/issues/78657 | https://github.com/godotengine/godot/pull/78675 | b156e24216f81aefcbc4f7983cdb7c6515d7ac76 | 2331eab9178fe59e7ed6706f41a86201e661f641 | "2023-06-25T00:03:22Z" | c++ | "2023-08-02T10:15:50Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,648 | ["editor/gui/scene_tree_editor.cpp", "editor/gui/scene_tree_editor.h"] | Scene editor doesn't grey out `process_mode = 4` nodes in opened scenes after editor start. | ### Godot version
4.1???- 4.1.beta1 - 4.1.beta3 - 030c1a950e8cee19e72104b4fb79caba17a2d678
### System information
Tested only Linux versions.
### Issue description
**Regression from 4.0.3.stable:**
In the editor UI set the `Node->process_mode` to `Disabled` or `When Paused`-> in the scene editor the subtree is grayed out. Save the scene and close the editor.
It doesn*t need to be the current scene.
Start it again, the subtree in the scene is no longer gray. The *.tscn file is updated correctly, e.g.:
```
[node name="Node2D" type="Node2D"]
+process_mode = 4
```
Workaround is to toggle the process mode again,
close and open or reload the scene,
(or disable the `interface/scene_tabs/restore_scenes_on_load` option ...)
### Image gallery
| https://github.com/godotengine/godot/issues/78648 | https://github.com/godotengine/godot/pull/81061 | cd5c0076286641b65cd95784eee3ac9e76b692b0 | b3811a3e4fca3a16e004257f327c3e540f5f05bb | "2023-06-24T14:57:35Z" | c++ | "2023-08-28T10:11:59Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,647 | ["core/object/class_db.cpp", "core/object/class_db.h", "editor/editor_help.cpp", "editor/editor_node.cpp", "editor/editor_settings.cpp", "editor/editor_settings.h"] | DocCache in the resource list | ### Godot version
4.1.beta3 (again)
### System information
4.1.beta3 - Ubuntu 22.04.2 LTS 22.04 - Vulkan (Forward+) - integrated Intel(R) HD Graphics 4400 (HSW GT2) () - Intel(R) Core(TM) i7-4600U CPU @ 2.10GHz (4 Threads)
### Issue description
DocCache is listed in resources.
If instantiated, it looks like this:
I can't find it in the Search Help menu. However, if I type it in the script editor and Ctrl+LMB, it leads to this documentation page:
This is weird, as I just showed the class has properties. Normally, undocumented classes have fields for each undocumented property or method. This makes me think this is unintended. An undocumented class with members normally looks like this:
### Steps to reproduce
Outlined above. F1 to open the Search Help menu. This button to create a resource:
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78647 | https://github.com/godotengine/godot/pull/78615 | c4e582262fa45483b1d68c5e7fa05351070355a0 | 92960b7a22dabe26bb724b7b0d00a5fd03324f3a | "2023-06-24T13:50:25Z" | c++ | "2023-07-26T16:39:22Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,628 | ["editor/editor_properties_array_dict.cpp"] | Unable to select other entry-field by mouse-click | ### Godot version
v4.1.beta.custom_build [afe65de11]
### System information
Debian Linux 12 Xfce
### Issue description
For an exported `PackedVector2Array` I am unable to select a different entry field with a mouse-click after an edit.
I have verified, that reverting #76711 fixes this issue.
### Steps to reproduce
1. Load and run MRP
2. select the main node in scene tree
3. Expand the `mypacked` variable in the inspector
4. click lmb on "Add Element"
5. click lmb in the x-entry-field
6. enter number "1"
7. click lmb in the y-entry-field
After 7 I would expect that the y-entry-field is selected, but no entry field is selected.
https://github.com/godotengine/godot/assets/6299227/997515ad-4e50-48a6-9e01-d9ba90e3114d
### Minimal reproduction project
https://github.com/godotengine/godot/files/11837624/empty.zip | https://github.com/godotengine/godot/issues/78628 | https://github.com/godotengine/godot/pull/78643 | 24e75022792c8389e5f2f0f3ca85bd6036054337 | 030c1a950e8cee19e72104b4fb79caba17a2d678 | "2023-06-23T21:03:31Z" | c++ | "2023-06-24T09:58:01Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,625 | ["modules/navigation/godot_navigation_server.cpp", "modules/navigation/nav_agent.cpp", "modules/navigation/nav_link.cpp", "modules/navigation/nav_region.cpp"] | Navigation Avoidance tries to avoid freed obstacles | ### Godot version
4.1 beta 3
### System information
Windows 11
### Issue description
When you free a navigation obstacle 3D (probably the same for 2D?), The navigation agents still treat them like they are there. I was able to find a workaround, which is to do `navigation_enabled = false` before you run `queue_free()`. However, this does NOT work when you use something like
```
func _notification(what):
if what == NOTIFICATION_PREDELETE:
avoidance_enabled = false
#This doesn't work ^^^
```
I would expect getting rid of a navigation obstacle to automatically stop it being considered by any navigation agents.
### Steps to reproduce
Set up some sort of navigation system with navigation avoidance turned on. Spawn in some obstacles and free them without turning off their avoidance_enabled(). As the navigation agent continues, it will still avoid nonexistent obstacles.
To use the MRP, play around with the exported variables "stop on predelete" and "stop before queue free" on the node3D in the main scene. Then run the project and wait for the obstacles to stop spawning in. If stop before queue free is off, then you should see the agent still trying to avoid obstacles even though they aren't there. If it is on, it should work as expected. The "stop on predelete" variable doesn't have any effect, despite turning off avoidance_enabled before the obstacle is freed.
### Minimal reproduction project
[Navigation Avoidance MRP.zip](https://github.com/godotengine/godot/files/11852775/Navigation.Avoidance.MRP.zip)
| https://github.com/godotengine/godot/issues/78625 | https://github.com/godotengine/godot/pull/78665 | a93c671b7d59d9d9be76558a8cc53bd87a0f42d1 | 710d2e7edf54c477a15c7bc3fb91ead394b51fd9 | "2023-06-23T20:31:16Z" | c++ | "2023-06-26T08:12:19Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,608 | ["drivers/gles3/storage/texture_storage.cpp"] | Prior working project now builds to a black screen using the Compatibility Renderer | ### Godot version
4.1 beta 3
### System information
Windows 11
### Issue description
Project builds and displays fine in all releases of Godot 4.x up until and including 4.1 beta 2.
When opening the project in 4.1 beta 3, it builds to a black screen, without crashing.
Switching from the Compatibility Renderer to Forward+ solves the issue.
Switching back to Compatibility breaks the build again
Reverting back to beta 2 also solves the issue, regardless of the renderer used.
### Steps to reproduce
- create a project in a version of Godot 4.x prior to beta 3, using the Compatibility renderer
- open and build the project in Godot 4.1 beta 3
- switch between different renderers to get the build to display
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78608 | https://github.com/godotengine/godot/pull/78620 | afe65de11001b392cb9666b1ee32ea6c4ecf10ff | 19e7490c9ad399f1951fc6b5863b45a3fa477d3e | "2023-06-23T10:19:14Z" | c++ | "2023-06-23T20:49:23Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,605 | ["drivers/gles3/storage/texture_storage.cpp"] | Problem with displaying the popup (except the first time) | ### Godot version
4.1 beta 3
The same project works fine on 4.1 **beta 1 and beta 2.**
### System information
Tested on MacOS, Fedora and HTML5
### Issue description
I have a problem in version 4.1 beta 3 with a popup (popup panel). The first time it always displays correctly. Every subsequent time, I only see a black spot.
[Nagranie ekranu z 2023-06-23 11-35-28.webm](https://github.com/godotengine/godot/assets/5132385/f7a7ad08-754c-4b4b-b594-b6880eff5f3d)
*The same if I show popup by methods like popup_centered.
### Steps to reproduce
I'm trying to reproduce it with a clean and minimal example, but unfortunately, without success. Maybe someone else has this issue or knows a way to reproduce it?
### Minimal reproduction project
I'm trying to create it. | https://github.com/godotengine/godot/issues/78605 | https://github.com/godotengine/godot/pull/78620 | afe65de11001b392cb9666b1ee32ea6c4ecf10ff | 19e7490c9ad399f1951fc6b5863b45a3fa477d3e | "2023-06-23T09:35:58Z" | c++ | "2023-06-23T20:49:23Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,590 | ["scene/main/viewport.cpp"] | Removing child on focus exit can cause a crash in certain cases | ### Godot version
v4.1.beta.custom_build [f23a12007]
### System information
macOS 13.3.0 - Vulkan (Forward+) - integrated Apple M1 Pro - Apple M1 Pro (8 Threads)
### Issue description
This was discovered through inspection of https://github.com/godotengine/godot/issues/78578#issuecomment-1603236210. This is happening when focus is requested by one node that ask for the previous focused node to release focus. But during focus release it remove the new focused node from the scene tree. This means the new focused node will first exit scene tree and then be focused. This means it will not be unfocused because release focus happens during exit of the tree.
### Steps to reproduce
1. run the MRP
2. click on one of the LineEdit
3. click on the other one
4. See Godot Crash
### Minimal reproduction project
[Test.zip](https://github.com/godotengine/godot/files/11841284/Test.zip)
| https://github.com/godotengine/godot/issues/78590 | https://github.com/godotengine/godot/pull/78591 | 68193bee9a1f77427a13b92508d611b1e2110217 | afe65de11001b392cb9666b1ee32ea6c4ecf10ff | "2023-06-22T23:42:15Z" | c++ | "2023-06-23T20:45:51Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,588 | ["editor/plugins/node_3d_editor_plugin.cpp"] | Grid view disappears | ### Godot version
v4.1.beta.custom_build [522a2ea3f]
### System information
Windows 10, Vulkan, GTX 960 536.23
### Issue description
When the grid is disabled in a scene and subsequently closed, it will not be displayed again and will remain invisible unless you refresh it from the 'view' option. This doesn't happen with the origin, but sometimes its colors may appear faded if you try to activate the grid again.
### Steps to reproduce
https://github.com/godotengine/godot/assets/48926464/c31770de-55c3-403a-8d91-07fc1184cae1
Note: It is not necessary to open or create a scene, just follow the steps shown in the video.
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78588 | https://github.com/godotengine/godot/pull/78694 | 02a4d099fd3371adb0bd3400db948b739dfa5e42 | e3dee8cea9e8e29d89aa34fa899f74e358d48ebb | "2023-06-22T22:15:26Z" | c++ | "2023-07-08T16:16:55Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,577 | ["editor/gui/scene_tree_editor.cpp", "editor/gui/scene_tree_editor.h"] | In the scene tree dock, children of the inherited scene and of instantiated scenes lose their yellow text color when viewed by clicking on the scene tab. | ### Godot version
4.1-beta3
### System information
Windows
### Issue description
When an inherited scene's source has children, or when an instantiated scene is expanded with the editable children option in the scene tree dock, they are usually colored yellow. In beta3 this yellow color disappears under certain conditions. I've found it to happen when another scene is switched to through the tab menu, and then the current scene is opened again. It also happens when first opening the project. The text is colored yellow again when the scene is opened from file, or when it is edited from the scene tree dock.
### Steps to reproduce
https://github.com/godotengine/godot/assets/76705401/ae4a6314-873c-490d-9cc0-d85766702772
### Minimal reproduction project
Trivial | https://github.com/godotengine/godot/issues/78577 | https://github.com/godotengine/godot/pull/81061 | cd5c0076286641b65cd95784eee3ac9e76b692b0 | b3811a3e4fca3a16e004257f327c3e540f5f05bb | "2023-06-22T17:36:13Z" | c++ | "2023-08-28T10:11:59Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,576 | ["modules/navigation/navigation_mesh_generator.cpp"] | Baking navigation mesh with GDScript produces error messages | ### Godot version
4.1.beta3
### System information
Godot v4.1.beta3 - macOS 13.2.1 - Vulkan (Forward+) - dedicated AMD Radeon Pro 580 - Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz (8 Threads)
### Issue description
When baking the mesh in GDScript after placing a Node3D (StaticBody3D with CollisionShape3D) these error message are displayed:
```
E 0:00:02:0733 update_polygons: Navigation map synchronization error. Attempted to update a navigation region with a navigation mesh that uses a different `cell_height` than the `cell_height` set on the navigation map.
<C++ Source> modules/navigation/nav_region.cpp:114 @ update_polygons()
E 0:00:19:0007 sync: Navigation map synchronization error. Attempted to merge a navigation mesh polygon edge with another already-merged edge. This is usually caused by crossing edges, overlapping polygons, or a mismatch of the NavigationMesh / NavigationPolygon baked 'cell_size' and navigation map 'cell_size'.
<C++ Source> modules/navigation/nav_map.cpp:770 @ sync()
```
**In Godot v4.1.beta2 it worked,** i.e. no error message
### Steps to reproduce
GDScript:
```
@onready var _navigationRegion3D: NavigationRegion3D = %NavigationRegion3D
:
_navigationRegion3D.bake_navigation_mesh(false)
```
Properties NavigationRegion3D:
<img width="434" alt="CleanShot 2023-06-22 at 19 15 44@2x" src="https://github.com/godotengine/godot/assets/1539838/30410b5f-99f1-40a4-82d4-2d6d66f673d5">
### Minimal reproduction project
[RTS 3D BakeIssue.zip](https://github.com/godotengine/godot/files/11839898/RTS.3D.BakeIssue.zip)
| https://github.com/godotengine/godot/issues/78576 | https://github.com/godotengine/godot/pull/78596 | 8d65fb9ec7ac0d21c196d966c3b94c91ca895727 | fee49f4f9d9603a041cc30e85e33712a72fb8fc4 | "2023-06-22T17:29:57Z" | c++ | "2023-06-23T06:49:06Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,572 | ["scene/3d/node_3d.cpp"] | 'Top Level' property in Inspector resets on project load | ### Godot version
4.1 Beta 3
### System information
Win 10, RTX 2070 mobile
### Issue description
If you set a body to be Top Level using the check box in inspector and save your project, next time you open the project that checkbox is unchecked. Looks like 'project save' somehow ignores that property and doesn't save it,
### Steps to reproduce
1. start a new 3d scene
2. add a 3d node, say, a static body
3. set the node to Top Level by checking the Top Level box in the inspector with the node selected.
4. Save your project
5. Reload project (not scene, project)
6. The box is unchecked
### Minimal reproduction project
n/a | https://github.com/godotengine/godot/issues/78572 | https://github.com/godotengine/godot/pull/78574 | d87bdef2a4f7c83c6a236f8916ebac35ae4fd176 | 120aa3fca8ea31e1f55f5c718c59f125ee3c93a7 | "2023-06-22T16:03:27Z" | c++ | "2023-06-22T18:44:36Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,569 | ["servers/rendering/renderer_scene_cull.cpp"] | VoxelGI not showing at runtime. Only manually activating .bake() in code makes it visible. | ### Godot version
4.0.3, 4.1 beta 3
### System information
macOS Ventura 13.0
### Issue description
As my scene grown complex VoxelGI started showing some weird behaviour.
In editor VoxelGI appears way too bright, and in game - "normal amount".
At first I worked almost 3 weeks on a scene thinking "VoxelGI is just exaggerated in the editor, should keep checking lighting at runtime after scene changes".
To my horror, I noticed something today. Just for the sake of experiment while playing the scene I unchecked and checked VoxelGI node visibility in the editor.
It appears, VoxelGI was not even turned on at runtime (despite the fact that it is visible in editor in local and remote tabs, eye icon is lit). Only unchecking and checking in editor, while game was up, connected to the editor, I learned that I worked on a "usual" lighting without VoxelGI all that time!
No errors in console, nothing. At the moment I have 300+ objects when VoxelGI is baking, it takes approximately 3-4 seconds to bake. Objects of a different size.
Only baking at runtime or checking in editor while game is up - helps.
Video of me baking manually in code at runtime, difference is visible (freeze is the moment of baking at runtime):
https://imgur.com/a/25KSZ4X
This seems like some kind of elusive bug, since there are no errors in the console. I hope someone else would have more information about this behaviour (to confirm it), and at the moment all I can do is bake at runtime.
Workaround:
Place this script on VoxelGI:
```gdscript
extends VoxelGI
# Called when the node enters the scene tree for the first time.
func _ready():
get_tree().create_timer(2.0).connect(
"timeout",
func():
print("Making GI visible!")
self.bake()
)
```
### Steps to reproduce
Unfortunately I cannot give steps to reproduce, since it happens only in my scene and basically you will need to do a lot of different kind of actions to achieve the problematic behaviour.
### Minimal reproduction project
Again, I can't give minimal reproduction project. I tried to reproduce it with new scene and CSGBoxes, a lot of them - no luck. | https://github.com/godotengine/godot/issues/78569 | https://github.com/godotengine/godot/pull/81124 | d351d40e2902b05597123eaf6dd5cf15bda2caa5 | f2ba8ec6ca2e4ff493cba11f95621d10ab8bb363 | "2023-06-22T15:44:46Z" | c++ | "2023-10-05T22:24:29Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,555 | ["scene/gui/graph_edit.cpp"] | Hotzones of visual shader nodes are too small | ### Godot version
4.1
### System information
Fedora, RX 590
### Issue description
Between 4.0.2 release and 4.0.3 RC1 the hotzone of the visual shader nodes have shrunk on the side that overlaps the node itself.
### Steps to reproduce
Download 4.0.2 release and 4.0.3-rc1.
Open MRP, navigate to hotzone.tres shader.
Zoom out to this level of zoom:
Try to extract a pin from ColorOp node.
In 4.0.3 you can observe how if you're not very precise you will drag the node around.
### Minimal reproduction project
[hotzone.zip](https://github.com/godotengine/godot/files/11832820/hotzone.zip)
| https://github.com/godotengine/godot/issues/78555 | https://github.com/godotengine/godot/pull/78556 | 1d41ce76ca605b4f307043f2dee8f5baa85e5c8a | f24f615b0458b6549d2d4488e4d1be2e495856ff | "2023-06-22T10:16:15Z" | c++ | "2023-06-23T06:47:56Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,545 | ["core/variant/callable_bind.cpp", "core/variant/callable_bind.h"] | Nonexistent function 'rpc' in base 'Callable' | ### Godot version
v4.1.dev3.official [a67d37f7c]
### System information
Linux 5.4.0-135-generic
### Issue description
I am trying to call Callable.rpc() and Callable.rpc_id(...) method but it doesn't work with some Callable.
If I call the .bind method and then call .rpc, I get the error "Invalid call. Nonexistent function 'rpc' in base 'Callable'.", execution aborts.
Issue is that we know for sure that the .rpc method in the Callable type is always present, and we would get a different error, but definitely not this one.
### Steps to reproduce
Create an empty project, a new scene with Node at the root, and attach the following script to it:
```GDScript
extends Node
@rpc
func foo(value = null):
print("foo(%s)" % value)
func _ready():
rpc("foo") # ok
rpc("foo", 123) # ok
foo.rpc() # ok
foo.rpc(123) # ok
foo.bind(123).rpc() # error: "invalid call. Nonexistent function 'rpc' in base 'Callable'."
foo.bindv([123]).rpc() # error: "invalid call. Nonexistent function 'rpc' in base 'Callable'."
foo.bind(123).unbind(1).rpc() # error: "invalid call. Nonexistent function 'rpc' in base 'Callable'."
foo.bindv([123]).unbind(1).rpc() # error: "invalid call. Nonexistent function 'rpc' in base 'Callable'."
```
Run the scene and observe the error. After an error, execution is interrupted, so comment out the found line to find the next error. Run the scene again.
See all available function values for these Callables. Except Callable.get_method() but that's a whole other issue.
```GDScript
for callable in [foo, foo.bind(123), foo.bindv([123]), foo.bind(123).unbind(1), foo.bindv([123]).unbind(1)]:
print("%6s ".repeat(11) % [
callable,
callable.get_bound_arguments(), callable.get_bound_arguments_count(), "callable.get_method()", callable.get_object(), callable.get_object_id(),
callable.hash(), callable.is_custom(), callable.is_null(), callable.is_standard(), callable.is_valid()
])
```
### Minimal reproduction project
N / A | https://github.com/godotengine/godot/issues/78545 | https://github.com/godotengine/godot/pull/78551 | 91f029e9126bf908b6fb27628ce38b3f80241a72 | 0c1442f82a4fd5020ed621731e479a71e89b4cb3 | "2023-06-22T06:19:14Z" | c++ | "2023-07-08T16:15:41Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,535 | ["editor/editor_file_system.cpp"] | Non lowercase extension triggers silent error when importing FBX files | ### Godot version
v4.1.beta3.official [ada712e06]
### System information
Windows 10.0.22621 - Vulkan (Forward+) - dedicated AMD Radeon RX 6700S (Advanced Micro Devices, Inc.; 31.0.12024.2005) - AMD Ryzen 9 6900HS with Radeon Graphics (16 Threads)
### Issue description
When importing an FBX file if the path to FBX2gltF is not set Godot normally shows a popup indicating that, together with a link to download the converter.
If the extension is uppercase (as is the case when exporting from Unreal) or anything other than lowercase, instead of the popup you get the following error.
```
- Could not create child process: --pbr-metallic-roughness --input "C:/repos/FBX Import Fail/Room #1.fBx" --output "C:/repos/FBX Import Fail/.godot/imported/Room #1-408dbc91e4717f19e6488d2025606d51.glb" --binary
modules/gltf/editor/editor_scene_importer_fbx.cpp:84
- FBX conversion to glTF failed with error: 32758.
- Error importing 'res://Room #1.fBx'.
```
### Steps to reproduce
- Create a new project
- Add an an FBX file making sure it ends with .FBX and not .fbx
- Go back to Godot to trigger importing and getting the error
- Rename the file to .fbx to get the popup
Any capitalization other than lowercase triggers the bug
### Minimal reproduction project
The fbx file makes the mrp go over the maximum sized allowed by Github.
You can download it here https://free3d.com/3d-model/low-poly-isometric-room-1-704614.html and add it to an empty project. Make sure to rename it so the extension is uppercase | https://github.com/godotengine/godot/issues/78535 | https://github.com/godotengine/godot/pull/78567 | ff689003a576f0fdf854a7756fb4ceb64f39ae7b | 7aff59f0b6920b970454c7608148845619d6cd60 | "2023-06-21T19:30:10Z" | c++ | "2023-07-12T19:02:42Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,532 | ["modules/gdscript/gdscript_analyzer.cpp", "modules/gdscript/tests/scripts/analyzer/errors/virtual_method_not_implemented.gd", "modules/gdscript/tests/scripts/analyzer/errors/virtual_method_not_implemented.out", "modules/gdscript/tests/scripts/analyzer/errors/virtual_super_not_implemented.gd", "modules/gdscript/tests/scripts/analyzer/errors/virtual_super_not_implemented.out", "modules/gdscript/tests/scripts/analyzer/features/virtual_method_implemented.gd", "modules/gdscript/tests/scripts/analyzer/features/virtual_method_implemented.out", "modules/gdscript/tests/scripts/analyzer/features/virtual_super_implemented.gd", "modules/gdscript/tests/scripts/analyzer/features/virtual_super_implemented.out"] | [4.1.beta3][GDScript]Β Cannot call virtual function because it hasn't been defined. | ### Godot version
4.1.beta3
### System information
Fedora Workstation 38 - Godot 4.1.beta3 - Vulkan (Forward+)
### Issue description
**This is an issue created by the production team, we're working on it**
Currently, it's impossible to call virtual functions within GDScript.
This currently returns an error:
```gdscript
# a.gd
extends Node
func _process(delta: float) -> void:
pass
```
```gdscript
# main.gd
extends Node
var my_node: Node = preload("res://a.gd").new()
func _process(delta: float) -> void:
my_node._process(delta) # Cannot call virtual function "_process()" because it hasn't been defined.
```
### Steps to reproduce
1. Create a script (`a.gd`) based on `Node` that declares `_process()`;
2. Create a main node;
3. Attach a script to the main node;
4. Initialize a variable with a new `a.gd` instance;
5. Try to run `_process()` on that node;
6. Find out that it returns an error.
### Minimal reproduction project
[virtual_functions.zip (updated)](https://github.com/godotengine/godot/files/11823992/virtual_functions.zip)
| https://github.com/godotengine/godot/issues/78532 | https://github.com/godotengine/godot/pull/78533 | 28a60b3de02e8cfcc742a248078e6b7ee8f728b1 | 81a0199be46679213bd2a46cb75394c05eb52c30 | "2023-06-21T18:37:42Z" | c++ | "2023-06-21T19:50:33Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,531 | ["core/input/input.cpp"] | Engine freezes when calling Sprite.new() from joy_connection_changed callback | ### Godot version
v3.5.stable.official [991bb6ac7]
### System information
Debian 11 - Intel Ivybridge Mobile - GLES3
### Issue description
Pretty much what the title says, the engine freezes when I call Sprite.new() from a method that I previously connected to `joy_connection_changed` signal. No errors are shown.
### Steps to reproduce
1. Run the minimal reproduction project, you will see the Godot icon moving around.
2. Connect or disconnect a gamepad and you will see the icon stops moving due to engine freeze.
### Minimal reproduction project
[minimal-reproduction-project.zip](https://github.com/godotengine/godot/files/11821885/minimal-reproduction-project.zip)
| https://github.com/godotengine/godot/issues/78531 | https://github.com/godotengine/godot/pull/80432 | 1b132b7c231453fa3c321c41013c3a8da3954b30 | 5cfa9a0d7b0d777d550662792e28dcfd6e84d459 | "2023-06-21T17:55:09Z" | c++ | "2023-08-09T15:46:37Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,530 | ["platform/windows/display_server_windows.cpp", "scene/main/window.cpp"] | Window node fullscreen mode not working | ### Godot version
4.0.3.stable
### System information
windows 11
### Issue description
fullscreen on the window node is not working and will show up as a windowed.
### Steps to reproduce
disable Embed Subwindows in Project Settings/Display/Window.
create a window node and set its mode to fullscreen.
### Minimal reproduction project
[fullscreen.zip](https://github.com/godotengine/godot/files/11821834/fullscreen.zip) | https://github.com/godotengine/godot/issues/78530 | https://github.com/godotengine/godot/pull/79016 | 7cd92ad02322815fd4196cc3fdcd4d2dd48e6313 | 988abb2161f73a56795e4f82a93e32b945ce5c8e | "2023-06-21T17:48:25Z" | c++ | "2023-07-08T16:21:04Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,520 | ["doc/classes/RichTextLabel.xml", "editor/editor_log.cpp", "scene/gui/rich_text_label.cpp", "scene/gui/rich_text_label.h"] | Printing with print_rich a lot of colored text breaks formating after output overflow. | ### Godot version
4.1
### System information
Win 10
### Issue description
When causing _"output oveflow"_ by printing a lot of text, formating breaks. Possible because enclosing tag get flushed.
Suggested solution: clear text formating after _output oveflow_ warning.
### Steps to reproduce
Print a lot of text with different text tags (like [color] or [bgcolor]) to console. For example in _process function.
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78520 | https://github.com/godotengine/godot/pull/79011 | bb6879e7cc17677876bda055ee951e46670e362b | 5c56206e6c84e56d9ad229493c751690742e9d44 | "2023-06-21T13:13:54Z" | c++ | "2023-07-14T16:49:21Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,500 | ["servers/rendering/renderer_rd/shaders/environment/volumetric_fog_process.glsl"] | SpotLight3D turns negative, and creates black boxes on small angles with volumetric fog | ### Godot version
v4.0.stable.official [92bee43ad]
### System information
Windows 10, AMD 5 5600G, RTX 3060, Forward+ Shading
### Issue description
When creating a SpotLight3D in a scene with fog, the light will first invert (e.g. opposite colours) and then create black splotches as seen in the screenshots below:
Angle of 2.0 degrees, range 15.0
Angle of 1.0 degrees, range 15.0,
### Steps to reproduce
Create a spotlight with the following settings:
range: 15
angle: 1.0
Colour: ff3e00
Energy, Indirect Energy, and Volumetric Fog Energy: 500
Size: 1.0 m
And a world environment with volumetric fog on set to defaults.
Note that any angle below 3 seems to act strangely.
### Minimal reproduction project
Note for opening the project: The scene opens with the camera close to the center of the scene.
Make sure to toggle the volumetric fog off and on to reset the dark spots.
[SpotLight3DBugRepro.zip](https://github.com/godotengine/godot/files/11811412/SpotLight3DBugRepro.zip)
| https://github.com/godotengine/godot/issues/78500 | https://github.com/godotengine/godot/pull/82546 | 1edf0f35b151ab3a8e3140aeaef08609e011ebdb | 950139e489ccaec03e744052ccdeb3adace18e57 | "2023-06-21T01:43:52Z" | c++ | "2023-10-05T20:50:12Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,487 | ["drivers/vulkan/vulkan_context.cpp"] | Godot editor hangs on opening a menu on Linux using wayland | ### Godot version
v4.1.beta.custom_build [d7af287ce]
### System information
EndeavourOS Linux, KDE plasma Wayland
### Issue description
When opening any menu or hovering above nodes or properties the editor just freezes. The console doesn't print anything when this happens. On X11 works fine. It doesn't matter if its a new project or an old one. On the beta 2 also works as normal, so it's a regression between beta 2 and now.
### Steps to reproduce
Using Wayland, create a new project and open any menu and the editor freezes.
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78487 | https://github.com/godotengine/godot/pull/79143 | 83cc5d4914a6bff76069ac19191192337e4df3de | 1453dc9d5dea8934f26d157b78aa8ed3b7a8a500 | "2023-06-20T19:21:39Z" | c++ | "2023-07-09T10:24:07Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,485 | ["scene/resources/tile_set.cpp"] | Removing a custom data layer does not update custom data name-to-layer mappings. | ### Godot version
v4.0.3.stable.mono.official [5222a99f5]
### System information
macOS Monterey 12.6.6 (21G646)
### Issue description
When creating a tileset by code with custom data layers, removing a layer with `remove_custom_data_layer` does not update the name-to-layer mapping; `get_custom_data_layer_by_name` will continue to return that layer's original index.
This has a downstream effect of causing TileData objects attempting to use `get_custom_data('Layer B')` to return invalid data (whatever custom data is now in Layer B's old index; e.g. Layer C).
I would expect that after using `remove_custom_data_layer`, I would be able to still use `get_custom_data_layer_by_name` on that tileset.
### Steps to reproduce
Run this code to see that getting Layer B by name after removing the first layer continues to return its original index (1) when in reality it is now at index 0.
```
var tileset:TileSet = TileSet.new()
tileset.add_custom_data_layer()
tileset.set_custom_data_layer_name(0, 'Layer A')
tileset.set_custom_data_layer_type(0, TYPE_INT)
tileset.add_custom_data_layer()
tileset.set_custom_data_layer_name(1, 'Layer B')
tileset.set_custom_data_layer_type(1, TYPE_STRING)
print('Layer index 0 name : ', tileset.get_custom_data_layer_name(0)) # Layer A
print('Layer index 1 name : ', tileset.get_custom_data_layer_name(1)) # Layer B
print('Layer B is on index : ', tileset.get_custom_data_layer_by_name("Layer B")) # 1
print('----- removing custom data layer 0 -----')
tileset.remove_custom_data_layer(0)
print('Layer index 0 name : ', tileset.get_custom_data_layer_name(0)) # Layer B
print('Layer index 1 name : ', tileset.get_custom_data_layer_name(1)) # prints an error, as expected
print('Layer B is on index : ', tileset.get_custom_data_layer_by_name("Layer B")) # returns 1 still! should return 0.
```
### Minimal reproduction project
[RemoveCustomDataLayer.zip](https://github.com/godotengine/godot/files/11804762/RemoveCustomDataLayer.zip)
| https://github.com/godotengine/godot/issues/78485 | https://github.com/godotengine/godot/pull/78492 | 4fc045e9b53becf063b50a457709ade403235d80 | ebd44641c53ecfafb3fa0bd5f0f6abcbc3a25e05 | "2023-06-20T18:10:04Z" | c++ | "2023-06-21T09:19:36Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,483 | ["drivers/gles3/storage/material_storage.cpp", "drivers/gles3/storage/particles_storage.cpp", "drivers/gles3/storage/texture_storage.cpp", "scene/2d/gpu_particles_2d.cpp", "scene/3d/gpu_particles_3d.cpp", "scene/resources/particle_process_material.cpp"] | Vertex Shader complation failed for GPUParticles2D and sub emiters | ### Godot version
v4.1.beta1.mono.official [828ec2c5d]
### System information
Kubuntu 22.04 Compatability
### Issue description
<Details>
```
Set emitting
Set emitting
Set sub_emitter_mode
Set sub_emitter_mode
drivers/gles3/shader_gles3.cpp:242 - ParticlesShaderGLES3: Vertex shader compilation failed:
0:679(16): error: `EMISSION_FLAG_HAS_POSITION' undeclared
0:679(45): error: `EMISSION_FLAG_HAS_ROTATION_SCALE' undeclared
0:679(16): error: LHS of `|' must be an integer
0:682(11): error: `EMISSION_FLAG_HAS_VELOCITY' undeclared
0:684(2): error: no function with name 'emit_subparticle'
drivers/gles3/shader_gles3.cpp:336 - Method/function failed.
1: #version 330
2: #define USE_GLES_OVER_GL
3: #define MATERIAL_UNIFORMS_USED
4: #define START_CODE_USED
5: #define PROCESS_CODE_USED
6:
7: #define MAX_GLOBAL_SHADER_UNIFORMS 256
8:
9:
10: #ifdef USE_MULTIVIEW
11: #if defined(GL_OVR_multiview2)
12: #extension GL_OVR_multiview2 : require
13: #elif defined(GL_OVR_multiview)
14: #extension GL_OVR_multiview : require
15: #endif
16: layout(num_views=2) in;
17: #define ViewIndex gl_ViewID_OVR
18: #define MAX_VIEWS 2
19: #else
20: #define ViewIndex uint(0)
21: #define MAX_VIEWS 1
22: #endif
23: precision highp float;
24: precision highp int;
25:
26: #define SDF_MAX_LENGTH 16384.0
27:
28: layout(std140) uniform GlobalShaderUniformData { //ubo:1
29: vec4 global_shader_uniforms[MAX_GLOBAL_SHADER_UNIFORMS];
30: };
31:
32: // This needs to be outside clang-format so the ubo comment is in the right place
33: #ifdef MATERIAL_UNIFORMS_USED
34: layout(std140) uniform MaterialUniforms{ //ubo:2
35:
36: highp vec3 m_direction;
37: highp float m_spread;
38: highp float m_flatness;
39: highp float m_initial_linear_velocity_min;
40: highp float m_initial_angle_min;
41: highp float m_angular_velocity_min;
42: highp float m_orbit_velocity_min;
43: highp float m_linear_accel_min;
44: highp float m_radial_accel_min;
45: highp float m_tangent_accel_min;
46: highp float m_damping_min;
47: highp float m_scale_min;
48: highp float m_hue_variation_min;
49: highp float m_anim_speed_min;
50: highp float m_anim_offset_min;
51: highp float m_initial_linear_velocity_max;
52: highp float m_initial_angle_max;
53: highp float m_angular_velocity_max;
54: highp float m_orbit_velocity_max;
55: highp float m_linear_accel_max;
56: highp float m_radial_accel_max;
57: highp float m_tangent_accel_max;
58: highp float m_damping_max;
59: highp float m_scale_max;
60: highp float m_hue_variation_max;
61: highp float m_anim_speed_max;
62: highp float m_anim_offset_max;
63: highp float m_lifetime_randomness;
64: highp float m_emission_sphere_radius;
65: int m_sub_emitter_amount_at_end;
66: bool m_sub_emitter_keep_velocity;
67: highp vec4 m_color_value;
68: highp vec3 m_gravity;
69:
70: };
71: #endif
72:
73: /* clang-format on */
74:
75: #define MAX_ATTRACTORS 32
76:
77: #define ATTRACTOR_TYPE_SPHERE uint(0)
78: #define ATTRACTOR_TYPE_BOX uint(1)
79: #define ATTRACTOR_TYPE_VECTOR_FIELD uint(2)
80:
81: struct Attractor {
82: mat4 transform;
83: vec4 extents; // Extents or radius. w-channel is padding.
84:
85: uint type;
86: float strength;
87: float attenuation;
88: float directionality;
89: };
90:
91: #define MAX_COLLIDERS 32
92:
93: #define COLLIDER_TYPE_SPHERE uint(0)
94: #define COLLIDER_TYPE_BOX uint(1)
95: #define COLLIDER_TYPE_SDF uint(2)
96: #define COLLIDER_TYPE_HEIGHT_FIELD uint(3)
97: #define COLLIDER_TYPE_2D_SDF uint(4)
98:
99: struct Collider {
100: mat4 transform;
101: vec4 extents; // Extents or radius. w-channel is padding.
102:
103: uint type;
104: float scale;
105: float pad0;
106: float pad1;
107: };
108:
109: layout(std140) uniform FrameData { //ubo:0
110: bool emitting;
111: uint cycle;
112: float system_phase;
113: float prev_system_phase;
114:
115: float explosiveness;
116: float randomness;
117: float time;
118: float delta;
119:
120: float particle_size;
121: float pad0;
122: float pad1;
123: float pad2;
124:
125: uint random_seed;
126: uint attractor_count;
127: uint collider_count;
128: uint frame;
129:
130: mat4 emission_transform;
131:
132: Attractor attractors[MAX_ATTRACTORS];
133: Collider colliders[MAX_COLLIDERS];
134: };
135:
136: #define PARTICLE_FLAG_ACTIVE uint(1)
137: #define PARTICLE_FLAG_STARTED uint(2)
138: #define PARTICLE_FLAG_TRAILED uint(4)
139: #define PARTICLE_FRAME_MASK uint(0xFFFF)
140: #define PARTICLE_FRAME_SHIFT uint(16)
141:
142: // ParticleData
143: layout(location = 0) in highp vec4 color;
144: layout(location = 1) in highp vec4 velocity_flags;
145: layout(location = 2) in highp vec4 custom;
146: layout(location = 3) in highp vec4 xform_1;
147: layout(location = 4) in highp vec4 xform_2;
148: #ifdef MODE_3D
149: layout(location = 5) in highp vec4 xform_3;
150: #endif
151: #ifdef USERDATA1_USED
152: layout(location = 6) in highp vec4 userdata1;
153: #endif
154: #ifdef USERDATA2_USED
155: layout(location = 7) in highp vec4 userdata2;
156: #endif
157: #ifdef USERDATA3_USED
158: layout(location = 8) in highp vec4 userdata3;
159: #endif
160: #ifdef USERDATA4_USED
161: layout(location = 9) in highp vec4 userdata4;
162: #endif
163: #ifdef USERDATA5_USED
164: layout(location = 10) in highp vec4 userdata5;
165: #endif
166: #ifdef USERDATA6_USED
167: layout(location = 11) in highp vec4 userdata6;
168: #endif
169:
170: out highp vec4 out_color; //tfb:
171: out highp vec4 out_velocity_flags; //tfb:
172: out highp vec4 out_custom; //tfb:
173: out highp vec4 out_xform_1; //tfb:
174: out highp vec4 out_xform_2; //tfb:
175: #ifdef MODE_3D
176: out highp vec4 out_xform_3; //tfb:MODE_3D
177: #endif
178: #ifdef USERDATA1_USED
179: out highp vec4 out_userdata1; //tfb:USERDATA1_USED
180: #endif
181: #ifdef USERDATA2_USED
182: out highp vec4 out_userdata2; //tfb:USERDATA2_USED
183: #endif
184: #ifdef USERDATA3_USED
185: out highp vec4 out_userdata3; //tfb:USERDATA3_USED
186: #endif
187: #ifdef USERDATA4_USED
188: out highp vec4 out_userdata4; //tfb:USERDATA4_USED
189: #endif
190: #ifdef USERDATA5_USED
191: out highp vec4 out_userdata5; //tfb:USERDATA5_USED
192: #endif
193: #ifdef USERDATA6_USED
194: out highp vec4 out_userdata6; //tfb:USERDATA6_USED
195: #endif
196:
197: uniform sampler2D height_field_texture; //texunit:0
198:
199: uniform float lifetime;
200: uniform bool clear;
201: uniform uint total_particles;
202: uniform bool use_fractional_delta;
203:
204: uint hash(uint x) {
205: x = ((x >> uint(16)) ^ x) * uint(0x45d9f3b);
206: x = ((x >> uint(16)) ^ x) * uint(0x45d9f3b);
207: x = (x >> uint(16)) ^ x;
208: return x;
209: }
210:
211: vec3 safe_normalize(vec3 direction) {
212: const float EPSILON = 0.001;
213: if (length(direction) < EPSILON) {
214: return vec3(0.0);
215: }
216: return normalize(direction);
217: }
218:
219: // Needed whenever 2D sdf texture is read from as it is packed in RGBA8.
220: float vec4_to_float(vec4 p_vec) {
221: return dot(p_vec, vec4(1.0 / (255.0 * 255.0 * 255.0), 1.0 / (255.0 * 255.0), 1.0 / 255.0, 1.0)) * 2.0 - 1.0;
222: }
223:
224:
225: uint m_hash(uint m_x)
226: {
227: m_x=(((m_x >> 16u) ^ m_x) * 73244475u);
228: m_x=(((m_x >> 16u) ^ m_x) * 73244475u);
229: m_x=((m_x >> 16u) ^ m_x);
230: return m_x; }
231:
232: float m_rand_from_seed(inout uint m_seed)
233: {
234: int m_k;
235: int m_s=int(m_seed);
236: if ((m_s == 0))
237: {
238: m_s=305420679;
239: }
240: m_k=(m_s / 127773);
241: m_s=((16807 * (m_s - (m_k * 127773))) - (2836 * m_k));
242: if ((m_s < 0))
243: {
244: m_s+=2147483647;
245: }
246: m_seed=uint(m_s);
247: return (float((m_seed % 65536u)) / 65535.0); }
248:
249: float m_rand_from_seed_m1_p1(inout uint m_seed)
250: {
251: return ((m_rand_from_seed(m_seed) * 2.0) - 1.0); }
252:
253: void main() {
254: bool apply_forces = true;
255: bool apply_velocity = true;
256: float local_delta = delta;
257:
258: float mass = 1.0;
259:
260: bool restart = false;
261:
262: bool restart_position = false;
263: bool restart_rotation_scale = false;
264: bool restart_velocity = false;
265: bool restart_color = false;
266: bool restart_custom = false;
267:
268: mat4 xform = mat4(1.0);
269: uint flags = 0u;
270:
271: if (clear) {
272: out_color = vec4(1.0);
273: out_custom = vec4(0.0);
274: out_velocity_flags = vec4(0.0);
275: } else {
276: out_color = color;
277: out_velocity_flags = velocity_flags;
278: out_custom = custom;
279: xform[0] = xform_1;
280: xform[1] = xform_2;
281: #ifdef MODE_3D
282: xform[2] = xform_3;
283: #endif
284: xform = transpose(xform);
285: flags = floatBitsToUint(velocity_flags.w);
286: }
287:
288: //clear started flag if set
289: flags &= ~PARTICLE_FLAG_STARTED;
290:
291: bool collided = false;
292: vec3 collision_normal = vec3(0.0);
293: float collision_depth = 0.0;
294:
295: vec3 attractor_force = vec3(0.0);
296:
297: #if !defined(DISABLE_VELOCITY)
298:
299: if (bool(flags & PARTICLE_FLAG_ACTIVE)) {
300: xform[3].xyz += out_velocity_flags.xyz * local_delta;
301: }
302: #endif
303: uint index = uint(gl_VertexID);
304: if (emitting) {
305: float restart_phase = float(index) / float(total_particles);
306:
307: if (randomness > 0.0) {
308: uint seed = cycle;
309: if (restart_phase >= system_phase) {
310: seed -= uint(1);
311: }
312: seed *= uint(total_particles);
313: seed += index;
314: float random = float(hash(seed) % uint(65536)) / 65536.0;
315: restart_phase += randomness * random * 1.0 / float(total_particles);
316: }
317:
318: restart_phase *= (1.0 - explosiveness);
319:
320: if (system_phase > prev_system_phase) {
321: // restart_phase >= prev_system_phase is used so particles emit in the first frame they are processed
322:
323: if (restart_phase >= prev_system_phase && restart_phase < system_phase) {
324: restart = true;
325: if (use_fractional_delta) {
326: local_delta = (system_phase - restart_phase) * lifetime;
327: }
328: }
329:
330: } else if (delta > 0.0) {
331: if (restart_phase >= prev_system_phase) {
332: restart = true;
333: if (use_fractional_delta) {
334: local_delta = (1.0 - restart_phase + system_phase) * lifetime;
335: }
336:
337: } else if (restart_phase < system_phase) {
338: restart = true;
339: if (use_fractional_delta) {
340: local_delta = (system_phase - restart_phase) * lifetime;
341: }
342: }
343: }
344:
345: if (restart) {
346: flags = emitting ? (PARTICLE_FLAG_ACTIVE | PARTICLE_FLAG_STARTED | (cycle << PARTICLE_FRAME_SHIFT)) : 0u;
347: restart_position = true;
348: restart_rotation_scale = true;
349: restart_velocity = true;
350: restart_color = true;
351: restart_custom = true;
352: }
353: }
354:
355: bool particle_active = bool(flags & PARTICLE_FLAG_ACTIVE);
356:
357: uint particle_number = (flags >> PARTICLE_FRAME_SHIFT) * uint(total_particles) + index;
358:
359: if (restart && particle_active) {
360: {
361: uint m_base_number=particle_number;
362: uint m_alt_seed=m_hash(((m_base_number + 1u) + random_seed));
363: float m_angle_rand=m_rand_from_seed(m_alt_seed);
364: float m_scale_rand=m_rand_from_seed(m_alt_seed);
365: float m_hue_rot_rand=m_rand_from_seed(m_alt_seed);
366: float m_anim_offset_rand=m_rand_from_seed(m_alt_seed);
367: float m_pi=3.14159;
368: float m_degree_to_rad=(m_pi / 180.0);
369: float m_tex_angle=1.0;
370: float m_tex_anim_offset=1.0;
371: float m_spread_rad=(m_spread * m_degree_to_rad);
372: if (restart_velocity)
373: {
374: {
375: float m_tex_linear_velocity=1.0;
376: {
377: float m_angle1_rad=(m_rand_from_seed_m1_p1(m_alt_seed) * m_spread_rad);
378: m_angle1_rad+=((m_direction.x != 0.0)?atan(m_direction.y, m_direction.x):(sign(m_direction.y) * (m_pi / 2.0)));
379: vec3 m_rot=vec3(cos(m_angle1_rad), sin(m_angle1_rad), 0.0);
380: out_velocity_flags.xyz=(m_rot * mix(m_initial_linear_velocity_min, m_initial_linear_velocity_max, m_rand_from_seed(m_alt_seed)));
381: }
382: ;
383: }
384: ;
385: }
386: float m_base_angle=(m_tex_angle * mix(m_initial_angle_min, m_initial_angle_max, m_angle_rand));
387: out_custom.x=(m_base_angle * m_degree_to_rad);
388: out_custom.y=0.0;
389: out_custom.w=(1.0 - (m_lifetime_randomness * m_rand_from_seed(m_alt_seed)));
390: out_custom.z=(m_tex_anim_offset * mix(m_anim_offset_min, m_anim_offset_max, m_anim_offset_rand));
391: if (restart_position)
392: {
393: {
394: float m_s=((m_rand_from_seed(m_alt_seed) * 2.0) - 1.0);
395: float m_t=((m_rand_from_seed(m_alt_seed) * 2.0) * m_pi);
396: float m_p=m_rand_from_seed(m_alt_seed);
397: float m_radius=(m_emission_sphere_radius * sqrt((1.0 - (m_s * m_s))));
398: xform[3].xyz=mix(vec3(0.0,0.0,0.0), vec3((m_radius * cos(m_t)), (m_radius * sin(m_t)), (m_emission_sphere_radius * m_s)), m_p);
399: if (restart_velocity)
400: {
401: out_velocity_flags.xyz=(emission_transform * vec4(out_velocity_flags.xyz, 0.0)).xyz;
402: }
403: xform=(emission_transform * xform);
404: out_velocity_flags.xyz.z=0.0;
405: xform[3].z=0.0;
406: }
407: ;
408: }
409: }
410: }
411:
412: if (particle_active) {
413: for (uint i = 0u; i < attractor_count; i++) {
414: vec3 dir;
415: float amount;
416: vec3 rel_vec = xform[3].xyz - attractors[i].transform[3].xyz;
417: vec3 local_pos = rel_vec * mat3(attractors[i].transform);
418:
419: if (attractors[i].type == ATTRACTOR_TYPE_SPHERE) {
420: dir = safe_normalize(rel_vec);
421: float d = length(local_pos) / attractors[i].extents.x;
422: if (d > 1.0) {
423: continue;
424: }
425: amount = max(0.0, 1.0 - d);
426: } else if (attractors[i].type == ATTRACTOR_TYPE_BOX) {
427: dir = safe_normalize(rel_vec);
428:
429: vec3 abs_pos = abs(local_pos / attractors[i].extents.xyz);
430: float d = max(abs_pos.x, max(abs_pos.y, abs_pos.z));
431: if (d > 1.0) {
432: continue;
433: }
434: amount = max(0.0, 1.0 - d);
435: } else if (attractors[i].type == ATTRACTOR_TYPE_VECTOR_FIELD) {
436: }
437: amount = pow(amount, attractors[i].attenuation);
438: dir = safe_normalize(mix(dir, attractors[i].transform[2].xyz, attractors[i].directionality));
439: attractor_force -= amount * dir * attractors[i].strength;
440: }
441:
442: float particle_size = particle_size;
443:
444: #ifdef USE_COLLISION_SCALE
445:
446: particle_size *= dot(vec3(length(xform[0].xyz), length(xform[1].xyz), length(xform[2].xyz)), vec3(0.33333333333));
447:
448: #endif
449:
450: if (collider_count == 1u && colliders[0].type == COLLIDER_TYPE_2D_SDF) {
451: //2D collision
452:
453: vec2 pos = xform[3].xy;
454: vec4 to_sdf_x = colliders[0].transform[0];
455: vec4 to_sdf_y = colliders[0].transform[1];
456: vec2 sdf_pos = vec2(dot(vec4(pos, 0, 1), to_sdf_x), dot(vec4(pos, 0, 1), to_sdf_y));
457:
458: vec4 sdf_to_screen = vec4(colliders[0].extents.xyz, colliders[0].scale);
459:
460: vec2 uv_pos = sdf_pos * sdf_to_screen.xy + sdf_to_screen.zw;
461:
462: if (all(greaterThan(uv_pos, vec2(0.0))) && all(lessThan(uv_pos, vec2(1.0)))) {
463: vec2 pos2 = pos + vec2(0, particle_size);
464: vec2 sdf_pos2 = vec2(dot(vec4(pos2, 0, 1), to_sdf_x), dot(vec4(pos2, 0, 1), to_sdf_y));
465: float sdf_particle_size = distance(sdf_pos, sdf_pos2);
466:
467: float d = vec4_to_float(texture(height_field_texture, uv_pos)) * SDF_MAX_LENGTH;
468:
469: d -= sdf_particle_size;
470:
471: if (d < 0.0) {
472: const float EPSILON = 0.001;
473: vec2 n = normalize(vec2(
474: vec4_to_float(texture(height_field_texture, uv_pos + vec2(EPSILON, 0.0))) - vec4_to_float(texture(height_field_texture, uv_pos - vec2(EPSILON, 0.0))),
475: vec4_to_float(texture(height_field_texture, uv_pos + vec2(0.0, EPSILON))) - vec4_to_float(texture(height_field_texture, uv_pos - vec2(0.0, EPSILON)))));
476:
477: collided = true;
478: sdf_pos2 = sdf_pos + n * d;
479: pos2 = vec2(dot(vec4(sdf_pos2, 0, 1), colliders[0].transform[2]), dot(vec4(sdf_pos2, 0, 1), colliders[0].transform[3]));
480:
481: n = pos - pos2;
482:
483: collision_normal = normalize(vec3(n, 0.0));
484: collision_depth = length(n);
485: }
486: }
487:
488: } else {
489: for (uint i = 0u; i < collider_count; i++) {
490: vec3 normal;
491: float depth;
492: bool col = false;
493:
494: vec3 rel_vec = xform[3].xyz - colliders[i].transform[3].xyz;
495: vec3 local_pos = rel_vec * mat3(colliders[i].transform);
496:
497: if (colliders[i].type == COLLIDER_TYPE_SPHERE) {
498: float d = length(rel_vec) - (particle_size + colliders[i].extents.x);
499:
500: if (d < 0.0) {
501: col = true;
502: depth = -d;
503: normal = normalize(rel_vec);
504: }
505: } else if (colliders[i].type == COLLIDER_TYPE_BOX) {
506: vec3 abs_pos = abs(local_pos);
507: vec3 sgn_pos = sign(local_pos);
508:
509: if (any(greaterThan(abs_pos, colliders[i].extents.xyz))) {
510: //point outside box
511:
512: vec3 closest = min(abs_pos, colliders[i].extents.xyz);
513: vec3 rel = abs_pos - closest;
514: depth = length(rel) - particle_size;
515: if (depth < 0.0) {
516: col = true;
517: normal = mat3(colliders[i].transform) * (normalize(rel) * sgn_pos);
518: depth = -depth;
519: }
520: } else {
521: //point inside box
522: vec3 axis_len = colliders[i].extents.xyz - abs_pos;
523: // there has to be a faster way to do this?
524: if (all(lessThan(axis_len.xx, axis_len.yz))) {
525: normal = vec3(1, 0, 0);
526: } else if (all(lessThan(axis_len.yy, axis_len.xz))) {
527: normal = vec3(0, 1, 0);
528: } else {
529: normal = vec3(0, 0, 1);
530: }
531:
532: col = true;
533: depth = dot(normal * axis_len, vec3(1)) + particle_size;
534: normal = mat3(colliders[i].transform) * (normal * sgn_pos);
535: }
536: } else if (colliders[i].type == COLLIDER_TYPE_SDF) {
537: } else if (colliders[i].type == COLLIDER_TYPE_HEIGHT_FIELD) {
538: vec3 local_pos_bottom = local_pos;
539: local_pos_bottom.y -= particle_size;
540:
541: if (any(greaterThan(abs(local_pos_bottom), colliders[i].extents.xyz))) {
542: continue;
543: }
544: const float DELTA = 1.0 / 8192.0;
545:
546: vec3 uvw_pos = vec3(local_pos_bottom / colliders[i].extents.xyz) * 0.5 + 0.5;
547:
548: float y = 1.0 - texture(height_field_texture, uvw_pos.xz).r;
549:
550: if (y > uvw_pos.y) {
551: //inside heightfield
552:
553: vec3 pos1 = (vec3(uvw_pos.x, y, uvw_pos.z) * 2.0 - 1.0) * colliders[i].extents.xyz;
554: vec3 pos2 = (vec3(uvw_pos.x + DELTA, 1.0 - texture(height_field_texture, uvw_pos.xz + vec2(DELTA, 0)).r, uvw_pos.z) * 2.0 - 1.0) * colliders[i].extents.xyz;
555: vec3 pos3 = (vec3(uvw_pos.x, 1.0 - texture(height_field_texture, uvw_pos.xz + vec2(0, DELTA)).r, uvw_pos.z + DELTA) * 2.0 - 1.0) * colliders[i].extents.xyz;
556:
557: normal = normalize(cross(pos1 - pos2, pos1 - pos3));
558: float local_y = (vec3(local_pos / colliders[i].extents.xyz) * 0.5 + 0.5).y;
559:
560: col = true;
561: depth = dot(normal, pos1) - dot(normal, local_pos_bottom);
562: }
563: }
564:
565: if (col) {
566: if (!collided) {
567: collided = true;
568: collision_normal = normal;
569: collision_depth = depth;
570: } else {
571: vec3 c = collision_normal * collision_depth;
572: c += normal * max(0.0, depth - dot(normal, c));
573: collision_normal = normalize(c);
574: collision_depth = length(c);
575: }
576: }
577: }
578: }
579: }
580:
581: if (particle_active) {
582: {
583: uint m_base_number=particle_number;
584: uint m_alt_seed=m_hash(((m_base_number + 1u) + random_seed));
585: float m_angle_rand=m_rand_from_seed(m_alt_seed);
586: float m_scale_rand=m_rand_from_seed(m_alt_seed);
587: float m_hue_rot_rand=m_rand_from_seed(m_alt_seed);
588: float m_anim_offset_rand=m_rand_from_seed(m_alt_seed);
589: float m_pi=3.14159;
590: float m_degree_to_rad=(m_pi / 180.0);
591: out_custom.y+=(local_delta / lifetime);
592: float m_tv=(out_custom.y / out_custom.w);
593: float m_tex_linear_velocity=1.0;
594: float m_tex_orbit_velocity=1.0;
595: float m_tex_angular_velocity=1.0;
596: float m_tex_linear_accel=1.0;
597: float m_tex_radial_accel=1.0;
598: float m_tex_tangent_accel=1.0;
599: float m_tex_damping=1.0;
600: float m_tex_angle=1.0;
601: float m_tex_anim_speed=1.0;
602: float m_tex_anim_offset=1.0;
603: vec3 m_force=m_gravity;
604: vec3 m_pos=xform[3].xyz;
605: m_pos.z=0.0;
606: m_force+=((length(out_velocity_flags.xyz) > 0.0)?((normalize(out_velocity_flags.xyz) * m_tex_linear_accel) * mix(m_linear_accel_min, m_linear_accel_max, m_rand_from_seed(m_alt_seed))):vec3(0.0,0.0,0.0));
607: vec3 m_org=emission_transform[3].xyz;
608: vec3 m_diff=(m_pos - m_org);
609: m_force+=((length(m_diff) > 0.0)?((normalize(m_diff) * m_tex_radial_accel) * mix(m_radial_accel_min, m_radial_accel_max, m_rand_from_seed(m_alt_seed))):vec3(0.0,0.0,0.0));
610: float m_tangent_accel_val=(m_tex_tangent_accel * mix(m_tangent_accel_min, m_tangent_accel_max, m_rand_from_seed(m_alt_seed)));
611: m_force+=((length(m_diff.yx) > 0.0)?(vec3(normalize((m_diff.yx * vec2(-1.0,1.0))), 0.0) * m_tangent_accel_val):vec3(0.0,0.0,0.0));
612: m_force+=attractor_force;
613: out_velocity_flags.xyz+=(m_force * local_delta);
614: float m_orbit_amount=(m_tex_orbit_velocity * mix(m_orbit_velocity_min, m_orbit_velocity_max, m_rand_from_seed(m_alt_seed)));
615: if ((m_orbit_amount != 0.0))
616: {
617: {
618: float m_ang=(((m_orbit_amount * local_delta) * m_pi) * 2.0);
619: mat2 m_rot=mat2(vec2(cos(m_ang), -sin(m_ang)), vec2(sin(m_ang), cos(m_ang)));
620: xform[3].xy-=m_diff.xy;
621: xform[3].xy+=(m_rot * m_diff.xy);
622: }
623: ;
624: }
625: float m_dmp=mix(m_damping_min, m_damping_max, m_rand_from_seed(m_alt_seed));
626: if (((m_dmp * m_tex_damping) > 0.0))
627: {
628: {
629: float m_v=length(out_velocity_flags.xyz);
630: float m_damp=(m_tex_damping * m_dmp);
631: m_v-=(m_damp * local_delta);
632: if ((m_v < 0.0))
633: {
634: {
635: out_velocity_flags.xyz=vec3(0.0,0.0,0.0);
636: }
637: ;
638: }
639: else
640: {
641: {
642: out_velocity_flags.xyz=(normalize(out_velocity_flags.xyz) * m_v);
643: }
644: ;
645: }
646: }
647: ;
648: }
649: float m_base_angle=(m_tex_angle * mix(m_initial_angle_min, m_initial_angle_max, m_angle_rand));
650: m_base_angle+=(((out_custom.y * lifetime) * m_tex_angular_velocity) * mix(m_angular_velocity_min, m_angular_velocity_max, m_rand_from_seed(m_alt_seed)));
651: out_custom.x=(m_base_angle * m_degree_to_rad);
652: out_custom.z=((m_tex_anim_offset * mix(m_anim_offset_min, m_anim_offset_max, m_rand_from_seed(m_alt_seed))) + ((m_tv * m_tex_anim_speed) * mix(m_anim_speed_min, m_anim_speed_max, m_rand_from_seed(m_alt_seed))));
653: vec3 m_tex_scale=vec3(1.0,1.0,1.0);
654: float m_tex_hue_variation=1.0;
655: float m_hue_rot_angle=(((m_tex_hue_variation * m_pi) * 2.0) * mix(m_hue_variation_min, m_hue_variation_max, m_rand_from_seed(m_alt_seed)));
656: float m_hue_rot_c=cos(m_hue_rot_angle);
657: float m_hue_rot_s=sin(m_hue_rot_angle);
658: mat4 m_hue_rot_mat=((mat4(0.299,0.587,0.114,0.0,0.299,0.587,0.114,0.0,0.299,0.587,0.114,0.0,0.0,0.0,0.0,1.0) + (mat4(0.701,-0.587,-0.114,0.0,-0.299,0.413,-0.114,0.0,-0.3,-0.588,0.886,0.0,0.0,0.0,0.0,0.0) * m_hue_rot_c)) + (mat4(0.168,0.33,-0.497,0.0,-0.328,0.035,0.292,0.0,1.25,-1.05,-0.203,0.0,0.0,0.0,0.0,0.0) * m_hue_rot_s));
659: out_color=(m_hue_rot_mat * m_color_value);
660: xform[0]=vec4(cos(out_custom.x), -sin(out_custom.x), 0.0, 0.0);
661: xform[1]=vec4(sin(out_custom.x), cos(out_custom.x), 0.0, 0.0);
662: xform[2]=vec4(0.0,0.0,1.0,0.0);
663: xform[3].z=0.0;
664: float m_base_scale=mix(m_scale_min, m_scale_max, m_scale_rand);
665: m_base_scale=(sign(m_base_scale) * max(abs(m_base_scale), 0.001));
666: xform[0].xyz*=((m_base_scale * sign(m_tex_scale.r)) * max(abs(m_tex_scale.r), 0.001));
667: xform[1].xyz*=((m_base_scale * sign(m_tex_scale.g)) * max(abs(m_tex_scale.g), 0.001));
668: xform[2].xyz*=((m_base_scale * sign(m_tex_scale.b)) * max(abs(m_tex_scale.b), 0.001));
669: int m_emit_count=0;
670: float m_unit_delta=(local_delta / lifetime);
671: float m_end_time=(out_custom.w * 0.95);
672: if (((out_custom.y < m_end_time) && ((out_custom.y + m_unit_delta) >= m_end_time)))
673: {
674: m_emit_count=m_sub_emitter_amount_at_end;
675: }
676: for (int m_i=0;(m_i < m_emit_count);m_i++)
677: {
678: {
679: uint m_flags=(EMISSION_FLAG_HAS_POSITION | EMISSION_FLAG_HAS_ROTATION_SCALE);
680: if (m_sub_emitter_keep_velocity)
681: {
682: m_flags|=EMISSION_FLAG_HAS_VELOCITY;
683: }
684: emit_subparticle(xform, out_velocity_flags.xyz, vec4(0.0,0.0,0.0,0.0), vec4(0.0,0.0,0.0,0.0), m_flags);
685: }
686: ;
687: }
688: if ((out_custom.y > out_custom.w))
689: {
690: {
691: particle_active=false;
692: }
693: ;
694: }
695: }
696: }
697:
698: flags &= ~PARTICLE_FLAG_ACTIVE;
699: if (particle_active) {
700: flags |= PARTICLE_FLAG_ACTIVE;
701: }
702:
703: xform = transpose(xform);
704: out_xform_1 = xform[0];
705: out_xform_2 = xform[1];
706: #ifdef MODE_3D
707: out_xform_3 = xform[2];
708: #endif
709: out_velocity_flags.w = uintBitsToFloat(flags);
710: }
711:
712: /* clang-format off */
713:
714:
```
</Details>
### Steps to reproduce
- Create two GPUParticles2D Effect1 and Effect2
- Set Effect1 Amount to 24
- Set Effect1 Sub Emitter to Effect2
- Set Effect1 Process Material to a new ParticleProcessMaterial
- Set Effect1 Process Material to use emission shape Sphere
- Set Effect1 Gravity to 0, 98, 0
- Set Effect1 Animation SpeedMin to 7
- Set Effect1 Animation SpeedMax to 7
- Set Effect1 Sub Emitter to Mode => AtEnd
- Set Effect1 Amount at end to 2
- Set Effect1 Texture to a 7 frame 8x8 animation
- Set Effect1 Lifetime to 2
- Set Effect1 Oneshot to true
- Set Effect1 Material to a new CanvasItemMaterial
- Set Effect1 Material to use Particle Animation
- Set Effect1 Anim H Frames to 7
--------------------------------
- Set Effect2 Amount to 1
- Set Effect2 Process Material to a new ParticleProcessMaterial
- Set Effect2 Gravity to 0, -64, 0
- Set Effect2 Animation SpeedMin to 5
- Set Effect2 Animation SpeedMax to 5
- Set Effect2 Texture to a 5 frame 8x8 animation
- Set Effect2 Lifetime to 4
- Set Effect2 Oneshot to true
- Set Effect2 Material to a new CanvasItemMaterial
- Set Effect2 Material to use Particle Animation
- Set Effect2 Anim H Frames to 5
------------------------------
- Now enable Effect1 Emitting
### Minimal reproduction project
[SubEmiiterBug.zip](https://github.com/godotengine/godot/files/11804736/SubEmiiterBug.zip)
| https://github.com/godotengine/godot/issues/78483 | https://github.com/godotengine/godot/pull/78490 | 88ce76537e771527561c592660b70872976078b7 | cb73a6e9f9046229a16b6b298c9086f03548e186 | "2023-06-20T17:55:50Z" | c++ | "2023-06-22T19:02:38Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,482 | ["scene/resources/shader.cpp"] | Disabling shader pre-processor displays editor-only error about unknown built-ins | ### Godot version
4.0.3.stable
### System information
Linux - Godot v4.0.3.stable - Vulkan (Forward+)
### Issue description
Disabling the shader pre-processor with `#pragma disable_preprocessor` causes errors to be displayed about built-ins being unknown. However, the shader will correctly compile and function as normal - the error is only in the shader editor.
Example screenshot:
### Steps to reproduce
1. Open project
2. Open `error-example.gdshader` in shader editor
3. Observe error
### Minimal reproduction project
[godot-4-shader-preprocessor-editor-error-minimal-reproduction-project-main.zip](https://github.com/godotengine/godot/files/11804403/godot-4-shader-preprocessor-editor-error-minimal-reproduction-project-main.zip)
Can also be found at: https://github.com/MenacingMecha/godot-4-shader-preprocessor-editor-error-minimal-reproduction-project | https://github.com/godotengine/godot/issues/78482 | https://github.com/godotengine/godot/pull/79287 | de14f09c1d903c808a50aab7b04574ca44e6016d | ead69eba96e51e377029c5a942aac0ecaa53ce1a | "2023-06-20T17:33:48Z" | c++ | "2023-07-14T16:49:44Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,470 | ["scene/main/scene_tree.cpp"] | Skeleton causes error and project stops working when played when a camera3D is in the scene. | ### Godot version
4.1 b2
### System information
Windows 10, Nvidia GTX 1660Ti, Vulkan
### Issue description
As you can see in the image:
In the editor I have an imported model, no errors. The model has a skeleton. Moving bones moves the mesh accordingly no errors and nothing seems to be wrong as far as I can tell.
When I play the project without the camera I get no errors.
When I play the project with a Camera3D the project throws 12 errors (see the image above) and the whole project freezes without dispalying anything as you can see in the image it's just a gray window.
This model and import works just fine in 4.0.3, this is also not the only model I have issues with so it's not just this model.
I can't provide a test scene or models because these models are purchased but I hope the fact it works in 4.0.3 and the error could help?
I'll try anything you ask me to do to help to resolve this - hopefully before 4.1 get released.
NOTE: Same error and the following line as in the screenshot appear in the windows Console, no other errors.
NOTE2: This project was initiated from 4.1beta as brand new, nothing else is in as you can see.
NOTE3: I'll need to double check but I had some models working before so it might not be with every model but I'll check later on today.
### Steps to reproduce
Hard to tell, if you have a model with a skeleton give it a try, add the model t othe scene, camera3d and then see if it happens?
### Minimal reproduction project
~~For now check this one from another report: https://github.com/godotengine/godot/issues/77780~~
[TestMeshWithSkinGD.zip](https://github.com/godotengine/godot/files/11851108/TestMeshWithSkinGD.zip)
EDIT:
Related line where the editor takes me to: https://github.com/godotengine/godot/blob/a2575cba48121a9e31c3a550ebd29398a7facf3f/servers/rendering/renderer_rd/storage_rd/mesh_storage.h#L693
**Production team edit:** Added the non-mono MRP (reduced to fit Github's size limit) to the issue. | https://github.com/godotengine/godot/issues/78470 | https://github.com/godotengine/godot/pull/78713 | 3aaa7870949853d3f42a1f8b67d16f8f2602931c | 61937011c93e4baff879548932c4406406114ff5 | "2023-06-20T11:13:36Z" | c++ | "2023-06-26T15:28:31Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,459 | ["doc/classes/String.xml"] | String.get_slice return Raw String | ### Godot version
v4.1.beta2.official [a2575cba4]
### System information
Windows 10, 64bit
### Issue description
`String.get_slice()` returns an raw string if the slice does not exist, but the docs say:
### Steps to reproduce
/
### Minimal reproduction project
/ | https://github.com/godotengine/godot/issues/78459 | https://github.com/godotengine/godot/pull/78463 | ddaefb3cd680d19db6bacd03ffc38ac7a9ef0ec8 | dbe8712d4eda555829283624df7c79faa4bc02ea | "2023-06-20T06:46:48Z" | c++ | "2023-06-20T11:14:26Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,445 | ["drivers/vulkan/rendering_device_vulkan.cpp"] | problem with ImageTexture created from image with FORMAT_RH | ### Godot version
4.0
### System information
Linux manjaro gpu rx 6900 xt
### Issue description
I am working on a terrain system which divide heightmap image
Also it change the resolution of heightmap depend on LOD
After converting the height map to texture with ImageTexture some of region have some white and black point at the end of texture
This will happen only when I use FORMAT_RH format, when I use full float or FORMAT_RF, there is no problem and everything will work
Also this will happen for some of the regions not for all of them
This is a photo of what happening, you can see at the end of photo this strange pixels will apeare
and this is my terrain, as you can see with this pixels I can not use half float images for my terrain
and for now I just use full float images
### Steps to reproduce
Create some half float images with FORMAT_RH format
resize them
use ImageTexture::create_from_image(image) to create a texture
Some of the texture that you have created has some black and white pixels at the end of texture
Also I am using GDExtension for this
I just generating different heightmap with PackedByteArray, and resize them with my algorithm to optimize GPU memory usage, and I have this problem
But I test this also with Image resize function and all of them get the same result when using FORMAT_RH format
As soon as I change the format to FORMAT_RF everything work again
Also I think the problem come from this: ImageTexture::create_from_image(image)
because I saved those images to see if the problem is from image class but all images were ok
### Minimal reproduction project
N/A
[empty.zip](https://github.com/godotengine/godot/files/11792662/empty.zip)
| https://github.com/godotengine/godot/issues/78445 | https://github.com/godotengine/godot/pull/80781 | cc01b288866a577a270f1d978fcd0f210563c47b | fef62a6860a13b31f5df10533552d41be6ea04a3 | "2023-06-19T19:40:41Z" | c++ | "2023-08-21T17:32:12Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,434 | ["scene/gui/rich_text_label.cpp"] | Editor Output window regression, RichTextLabel does not always update its scrollbar properly | ### Godot version
4.1.dev 116f783db73f4bf7e9e96ae54dd3d0a20337cc8a
### System information
Godot v4.1.beta (116f783db) - Windows 10.0.19045 - Vulkan (Forward+) - dedicated NVIDIA GeForce GTX 970 (NVIDIA; 31.0.15.1694) - Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz (8 Threads)
### Issue description
Editor Output window RichTextLabel does not always update its scrollbar properly after print() calls. Looks like a regression from https://github.com/godotengine/godot/pull/78241, reverting it fixes this issue (but of course restores the bug it was fixing). CC @bruvzg
### Steps to reproduce
1) Open the demo project
2) Make the Output window visible in the bottom view, press "Clear Output"
2) Run the project, notice how the output scroll does not follow the actual content added. In fact, the vertical scroll bar is not even visible. However, manually resizing the bottom view updates the layout and the scrollbar becomes visible and "jumps" to the end as expected.
Looks like printing a single line in _process() doesn't always update the scrolling. However, uncommenting the line in _ready() to print many lines in a single frame seems to be enough to update the scrolling and it starts working for the _process() prints, too.
### Minimal reproduction project
[OutputScrollBug.zip](https://github.com/godotengine/godot/files/11786216/OutputScrollBug.zip)
| https://github.com/godotengine/godot/issues/78434 | https://github.com/godotengine/godot/pull/78833 | ab430b0b4a2d4434eabf54f8704459b37cd19dee | 46424488edc341b65467ee7fd3ac423e4d49ad34 | "2023-06-19T08:54:28Z" | c++ | "2023-06-29T13:43:03Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,411 | ["servers/rendering/shader_language.cpp"] | Using `hint_normal_roughness_texture` on the forward mobile renderer causes shader compilation failure | ### Godot version
Tested on 4.0.3 and 4.1 beta 2
### System information
Arch Linux
### Issue description
In a spatial shader, using the hint `hint_normal_roughness_texture` and then trying to sample that uniform in the `fragment` function, as outlined in the screen reading shader docs, while using the mobile renderer causes shader compilation failure. It can also freeze the editor requiring a SIGKILL to end the process if the shader is applied to a mesh and run.
```servers/rendering/renderer_rd/forward_mobile/scene_shader_forward_mobile.cpp:182 - Condition "!shader_singleton->shader.version_is_valid(version)" is true.```
It seems the shader compiler still validates the gdshader, generates the shader code, which references `normal_roughness_buffer`, but this is only defined in the glsl for the forward clustered renderer and is left undefined in mobile, causing an "undeclared identifier" error in the generated shader code.
The compatibility shader, by comparison, explicitly errors in the shader editor and will not try to compile the shader. If the mobile renderer is not intended to support this, it should probably use the same behaviour, and compatibility with the renderers explained in the docs.
### Steps to reproduce
Create a shader:
```
shader_type spatial;
uniform sampler2D normal_roughness_texture : hint_normal_roughness_texture, repeat_disable, filter_nearest;
void fragment() {
vec3 screen_normal = texture(normal_roughness_texture, SCREEN_UV).xyz;
}
```
Switch to mobile renderer.
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78411 | https://github.com/godotengine/godot/pull/78839 | 6ab727edc2b0bf51ca3458237505ff8763f20cf9 | 46cd84b36212d5932c6cddfc6ed5f50ff60030bc | "2023-06-18T16:19:02Z" | c++ | "2023-07-08T16:19:49Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,406 | ["drivers/gles3/storage/mesh_storage.cpp", "servers/rendering/renderer_rd/storage_rd/mesh_storage.cpp"] | MeshInstance3D and BlendShapes: Exceptions when closing app with node in scene + on "Create Inherited Scene" [rendering_device_vulkan.cpp:8405] | ### Godot version
Godot v4.1.beta2.mono
### System information
Godot v4.1.beta2.mono - Windows 10.0.19045 - Vulkan (Forward+) - dedicated NVIDIA GeForce RTX 3080 (NVIDIA; 31.0.15.1694) - AMD Ryzen 7 3800X 8-Core Processor (16 Threads)
### Issue description
The error message/exception of this bug can only be seen in the console (Godot_v4.1-beta2_mono_win64_console.exe) and will not show up in the in-engine console
**The Problem**
When closing an application/game that contains one or more MeshInstance3D objects that have BlendShapes, Godot will throw an exception for each of them in your scene
ERROR: Attempted to free invalid ID: 30150670417926
at: _free_internal (drivers/vulkan/rendering_device_vulkan.cpp:8405)
This same bug seems to occur also under certain circumstances when you have a imported .glb as child object in your Scene hierachy and you then make a inherited scene out of it from the FileSystem and save it.
### Steps to reproduce
(Since this exception involves Blender/3D, i've added also screenshots from blender to make it as accessable as possible for everyone)
1) Launch the Godot_v4.1-beta2_mono_win64_console.exe and create a new project using the Forward+ renderer
2) Go to [blender.org](https://www.blender.org/download/) and download Blender 3.5.1 (Windows)
3) Open up Blender 3.5.1 and create a new General File / new Project.
Select the Default Cube by Left-Clicking on it in the 3D-Viewport or in the hierachy on the right side of the screen.
4) With the Cube selected - Click on the green triangle-icon on the vertical sidebar, right of the 3d-viewport. It displays "Object Data Properties" when hovering over it. Click on it.
5) Open up the section "Shape Keys" and click twice on the [+] button right to it to add both the "Basis" and "Key1" Shape-Keys (Blend-Shapes)
6) Export the file!
File > Export > gITF 2.0 (.glb / .gitf) and make sure to "Include" > "Limit To: Selected Objects" and saving it inside your Godot 4.1 Beta 2 project.
7) Go back into Godot Engine and create a new scene based on Node or Node3D and then drag the imported model onto your scene's root node.
8) Hit "F5" to start the application and wait until the application window shows up.
9) Then hit "F8" or use the [X] at the top right to close it.
As a result of this, you'll see the exception NOT in your in-engine console, but in the seperate console window from which the engine launched
ERROR: Attempted to free invalid ID: 4127463571469
at: _free_internal (drivers/vulkan/rendering_device_vulkan.cpp:8405)
A second way to reproduce this same exception is to
10) In the FileSystem - create a "Inherited Scene" from the imported .glb
11) The moment you open up the inherited scene and you try to save it using Ctrl+S, you get the exception also inside the in-engine console
While in the launched console window you'll see this
12) Navigate back to your main scene where the .glb is a child node. Now the MeshInstance3D is still there, but it visually either disappears or looks totally messed up! it wont fix itsself unless you "Reload Current Project". But then you get a different exception
13) If you delete the created .tscn file and repeat the same process from 10 to 12, you'll always get the same order of exceptions and the same problem will re-occur
### Minimal reproduction project
[rendering_device_vulkan_8405.zip](https://github.com/godotengine/godot/files/11781570/rendering_device_vulkan_8405.zip)
Extract the .zip
Launch the Godot_v4.1-beta2_mono_win64_console.exe
Include the project into Godot Engine 4.1 Beta 2
Start the project with node_3d.tscn as Main Scene (or Current Scene)
Close it
Checkout the Console Window | https://github.com/godotengine/godot/issues/78406 | https://github.com/godotengine/godot/pull/78433 | 914cc1d6f55ce2c3c2964f5b4f25143b3e103933 | 7188ed5f9140f6e2a38a24ac3364bba741b45135 | "2023-06-18T14:00:19Z" | c++ | "2023-06-19T08:38:24Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,399 | ["scene/2d/navigation_agent_2d.cpp"] | NavigationAgent2D debug path not showing | ### Godot version
4.0.3.stable.mono, 4.1.beta2.mono
### System information
Godot v4.1.beta2.mono - Windows 10.0.22621 - Vulkan (Forward+) - dedicated NVIDIA GeForce RTX 3070 Ti (NVIDIA; 31.0.15.2737) - 12th Gen Intel(R) Core(TM) i9-12900K (24 Threads)
### Issue description
NavigationAgent2D debug path is not appearing when the debug option is enabled. This used to draw fine when I first started working on my current project. I'm not sure what happened or what setting I've changed, but debug paths are no longer being drawn for any agent. Even when the custom path colour and size values are used, no path is visible.
Strangely, in the minimal reproduction project I've attached, adding a call to the nav agent `is_target_reachable()` seems to cause the path to be drawn. However, this still does not work in my main project.
It may be that I'm missing something here and would appreciate any help or insight into whats gone wrong as the debug path functionality was suuuuper useful when it was working previously. I can't seem to find any other posts online about this issue.
### Steps to reproduce
- Setup a navigation agent 2D as described in the Godot docs https://docs.godotengine.org/en/stable/tutorials/navigation/navigation_using_navigationagents.html
- Enable the debug option.
<img width="349" alt="image" src="https://github.com/godotengine/godot/assets/19389664/44509b1e-6b6e-480a-a03b-c0c8eb529e35">
When the nav agent target is set by code, this should trigger a path to be drawn from the nav agent position to the target position yet no line is drawn.
### Minimal reproduction project
Run the attached project and click around the screen. No debug path is drawn.
Go into one of the attached scripts, NavigationDebug.cs or NavigationDebug.gd and uncomment the line that prints `is_target_reachable()`, rerun the project and click aorund the screen again. For some reason, this causes the debug line to be drawn.
[TestNavDebug.zip](https://github.com/godotengine/godot/files/11781387/TestNavDebug.zip)
I've tried adding in this function call to my main project in the hopes that it would begin drawing the debug paths again but no luck. | https://github.com/godotengine/godot/issues/78399 | https://github.com/godotengine/godot/pull/78438 | 58c3f24d96a052b3486ea74a47a7ced4e5c5c1f8 | ea5d02b5823cd3338a94e9eddb4e032d2f6fbec2 | "2023-06-18T12:16:23Z" | c++ | "2023-06-19T13:51:16Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,395 | ["editor/export/editor_export.cpp", "editor/export/editor_export_platform.h", "editor/export/project_export.cpp", "editor/export/project_export.h", "platform/android/export/export_plugin.cpp", "platform/ios/export/export_plugin.cpp", "platform/macos/export/export_plugin.cpp", "platform/web/export/export_plugin.cpp"] | Incorrect (?) warning about enabling S3TC BPTC texture compression on macOS export | ### Godot version
4.1.beta2
### System information
macOS 13.3.1 mobile renderer
### Issue description
When setting up an export preset for macOS in a project where the Import S3TC BPTC setting is not enabled you get a warning that this platform requires this to be enabled.
However, if you check the documentation for this project setting, it seems that this is not required at all, and it's actually very unclear (to me) what enabling it will actually change.
As I understand this description the platform will use the format if it requires it even if the setting is disabled (so it doesn't matter). In this case the macOS export warning should be removed. If however this is required the description text also seems to note that just enabling this will do nothing, you also need to re-import all your textures for this to take effect. So just enabling the setting wouldn't be enough?
The setting is for the full project, so if you have a project with exports for both mobile and desktop I guess you need to have it turned off since it's not supported on mobile? So if it is indeed required on some desktop systems perhaps there's need for a mobile version of this setting to have it disabled there? Or just get rid of the setting entirely if this is handled automatically anyway to avoid this confusion?
### Steps to reproduce
- Have a project where the Import S3TC BPTC setting is not enabled in the project settings
- Set up a macOS export preset
- Observe warning (and read setting documentation)
### Minimal reproduction project
[Simple macOS export project](https://www.lostminds.com/godot/MacExportPermissionsTest.zip) | https://github.com/godotengine/godot/issues/78395 | https://github.com/godotengine/godot/pull/78457 | aa9e4b8aceee79122cc336f616e48e56588ebfc1 | 229af8e5731030064deb5777f6adf3838618df54 | "2023-06-18T11:01:32Z" | c++ | "2023-08-17T09:20:08Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,432 | ["editor/icons/GizmoCPUParticles3D.svg", "editor/icons/GizmoDecal.svg", "editor/icons/GizmoFogVolume.svg", "editor/icons/GizmoLightmapProbe.svg", "editor/plugins/gizmos/decal_gizmo_plugin.cpp", "editor/plugins/gizmos/fog_volume_gizmo_plugin.cpp", "editor/plugins/gizmos/lightmap_probe_gizmo_plugin.cpp"] | LightmapProbes not selectable in the 3D editor viewport | ### Describe the project you are working on
N/A
### Describe the problem or limitation you are having in your project
As of currently, you cannot click-select them in 3D view and have to select them from scene tree. Considering that manually placing down light probes is already somewhat laborious(and using something to place them on a subdivided grid may not be enough), not being able to click-select them would be a huge pain.
### Describe the feature / enhancement and how it helps to overcome the problem or limitation
It helps speed up the process of manually placing `LightmapProbe`s and moving them about. A huge QoL improvement in my opinion.
### Describe how your proposal will work, with code, pseudo-code, mock-ups, and/or diagrams
When the user clicks on a `LightmapProbe`'s gizmo(see the circular thing below), the `LightmapProbe` it belongs to will be selected and moved around easily. That's literally it.
### If this enhancement will not be used often, can it be worked around with a few lines of script?
`LightmapProbe` is implemented in core.
### Is there a reason why this should be core and not an add-on in the asset library?
See above. | https://github.com/godotengine/godot/issues/78432 | https://github.com/godotengine/godot/pull/81554 | 51f67ea4c51109d23eda29d68ae83e532fb01cbc | 7ca579cb2742e3a7f758903736cc53b9c8ae2c2d | "2023-06-18T08:36:09Z" | c++ | "2023-09-16T19:22:54Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,385 | ["modules/mono/editor/GodotTools/GodotTools/Export/ExportPlugin.cs", "modules/mono/godotsharp_dirs.cpp"] | Linux + C# : "Failed to initialize context for app" on exported project | ### Godot version
Godot Engine v4.0.3.stable.mono.official.5222a99f5
### System information
Ubuntu 22.04.2 LTS
### Issue description
when exported project is launch it crash :
```
Godot Engine v4.0.3.stable.mono.official.5222a99f5 - https://godotengine.org
Vulkan API 1.3.236 - Forward+ - Using Vulkan Device #0: NVIDIA - NVIDIA GeForce GTX 1650
Could not resolve CoreCLR path. For more details, enable tracing by setting COREHOST_TRACE environment variable to 1
Failed to initialize context for app: /mnt/nova/secure/workspace/wip/godot4/empty/build/data_empty_x86_64/empty.dll. Error code: 0x80008087
ERROR: hostfxr_initialize_for_dotnet_command_line failed with code: -2147450745
at: initialize_hostfxr_self_contained (modules/mono/mono_gd/gd_mono.cpp:239)
ERROR: Parameter "load_assembly_and_get_function_pointer" is null.
at: initialize_hostfxr_and_godot_plugins (modules/mono/mono_gd/gd_mono.cpp:316)
ERROR: Parameter "godot_plugins_initialize" is null.
at: initialize (modules/mono/mono_gd/gd_mono.cpp:411)
Erreur de segmentation (core dumped)
```
### Steps to reproduce
Linux export a simple project and run the .x86_64 binary
### Minimal reproduction project
Only one scene with one node with just one empty C# script | https://github.com/godotengine/godot/issues/78385 | https://github.com/godotengine/godot/pull/78846 | dc6ea03101c12452ae1683c3df4a241e2470583e | c244903c3dbc7b8178a735d1e0cc938ba91312ce | "2023-06-18T02:22:02Z" | c++ | "2023-08-04T19:01:38Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,381 | ["editor/scene_tree_dock.cpp"] | Move multiple nodes to the end of their parent by dragging will actually move some of them to the beginning | ### Godot version
4.1-beta2
### System information
Godot v4.1.beta2.mono - Ubuntu 23.04 23.04 - Vulkan (Forward+) - dedicated NVIDIA GeForce GTX 1060 6GB (nvidia; 515.105.01) - Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz (8 Threads)
### Issue description
If multiple nodes are selected in the editor and drag them to the end of their parent in the scene tree, every node except the first one will be moved to the beginning of its parent.
Right click selected nodes and select move down will give expected result
https://github.com/godotengine/godot/assets/6362922/beab3a3d-0971-448f-9970-afe564d8b7f6
### Steps to reproduce
Create a parent node
Add few children nodes
Select some nodes and drag them to the end of the parent node
### Minimal reproduction project
[NodeArrangementReproduce.zip](https://github.com/godotengine/godot/files/11780108/NodeArrangementReproduce.zip)
| https://github.com/godotengine/godot/issues/78381 | https://github.com/godotengine/godot/pull/78430 | cfae05094adf118a5b637d693e4c347ee90a0455 | 58c3f24d96a052b3486ea74a47a7ced4e5c5c1f8 | "2023-06-17T21:51:38Z" | c++ | "2023-06-19T13:40:21Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,380 | ["modules/multiplayer/multiplayer_spawner.cpp"] | Parse error when attempting to export var of Callable type in a resource | ### Godot version
4.1.beta.a83eb16fb
### System information
Arch Linux
### Issue description
When attempting to have a scene load a resource that uses a script that exports a var of Callable type, a parse error will occur due to Callable not being serializable.
### Steps to reproduce
Create new project.
create a resource script that exports a Callable type:
```
class_name BreakStuff
extends Resource
@export var something: Callable
```
create a .tres file that uses the script.
create a scene that uses that resource, such as with a script like
```
extends Node
@export var stuff: Resource
```
set that resource to that .tres file.
close project and reopen.
Be greeted with error message about a parse error and be unable to open that scene.
Within plain text of .tscn file you can see a line that reads
something =
which causes a parse error as something has nothing to be set to.
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78380 | https://github.com/godotengine/godot/pull/78409 | fee49f4f9d9603a041cc30e85e33712a72fb8fc4 | 2d19cec17417f21408b16e0a9fe420ab500c32fc | "2023-06-17T21:44:27Z" | c++ | "2023-06-23T08:13:11Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,376 | ["modules/gdscript/gdscript.cpp"] | Static variables will raise an error if game will close. | ### Godot version
v4.1.beta2.official [a2575cba4]
### System information
Godot v4.1.beta2 - Windows 10.0.19045 - Vulkan (Forward+) - dedicated NVIDIA GeForce RTX 2070 SUPER (NVIDIA; 31.0.15.3141) - AMD Ryzen 9 3900X 12-Core Processor (24 Threads)
### Issue description
With the latest 4.1.beta2 static variables will now raise an error in console and log. No crash, only a error:
```
USER ERROR: Condition "p_elem->_root != this" is true.
at: remove (./core/templates/self_list.h:80)
```
I don't know if pull #78138 has anything to do with it?
Anyway, if `static` keyword will removed, the error is gone.
### Steps to reproduce
Pretty simple:
Make a new scene, attach a new script and copy/paste this:
```GDScript
extends Node2D
static var my_static_int: int = 0
func _ready() -> void:
pass
```
Run scene and close it.
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78376 | https://github.com/godotengine/godot/pull/78521 | 1e44b479702a577037c49600ce50f4b8a61b102c | f2ce0b6741049b69efb455394b8a161f68a468a6 | "2023-06-17T17:45:54Z" | c++ | "2023-06-21T19:52:15Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,374 | ["modules/text_server_adv/thorvg_svg_in_ot.cpp", "modules/text_server_fb/thorvg_svg_in_ot.cpp"] | Some glyphs from color OpenType SVG font are not rendered | ### Godot version
v4.1.beta2.official [a2575cba4]
### System information
Godot v4.1.beta2 - Windows 10.0.19045 - Vulkan (Compatibility) - NVIDIA GeForce RTX 2070 (NVIDIA; 31.0.15.3179) - AMD Ryzen 5 3600 6-Core Processor (12 Threads)
### Issue description
### Description:
Most Emoji glyphs from color OpenType SVG font are not displayed in Control nodes, such as Label or RichTextLabel.
When using https://github.com/adobe-fonts/noto-emoji-svg/blob/main/fonts/NotoColorEmoji-SVG.otf, some emoji are rendered correctly, such as βΊοΈ (U+263A), but others are not visible at all, e.g. β€οΈ (U+2764), π (U+1F3E0).
"Output" panel of Godot editor shows multiple error messages "_Failed to create SVG canvas._"
### Additional info:
All these characters are rendered correctly by Firefox browser (as seen in https://github.com/adobe-fonts/noto-emoji-svg/blob/main/test.html)
Regular English text is visible when using OT SVG fonts, e.g. https://www.fontspace.com/icecreamy-font-f82130 (although rendered with some kerning issues).
Monochrome OT font https://github.com/adobe-fonts/noto-emoji-svg/blob/main/fonts/NotoEmoji.otf appears to be working correctly.
This limits usage of color fonts, since CBDT/CBLC TTF fonts are blurry if the font size is larger than PNGs which are embedded into the font, and Godot doesn't support COLR vector fonts.
Note: None of the glyphs of colorful OpenType SVG fonts are rendered at all in Godot 4.0
### Steps to reproduce
- Add Label control node to a scene
- Select Label node
- Import https://github.com/adobe-fonts/noto-emoji-svg/blob/main/fonts/NotoColorEmoji-SVG.otf as a Font value using Theme or Theme Overrides property (make sure)
- Input π β€οΈβΊοΈ as a value of "text" property
- Check visible characters in 2D panel of Godot editor
Expected result:
- all colorful characters are rendered correctly
- no error messages
Actual result:
- Only βΊοΈ (U+263A) is visible, other two are invisible (but occupy some space)
- Multiple error messages "_Failed to create SVG canvas._" are present in the Output panel of Godot editor
### Minimal reproduction project
[svg_font_sandbox_4_1.zip](https://github.com/godotengine/godot/files/11779682/svg_font_sandbox_4_1.zip)
| https://github.com/godotengine/godot/issues/78374 | https://github.com/godotengine/godot/pull/78543 | de3d4e3cec4e45739fdc8fb7ceff84950aab5b15 | 18bc912bc0595a2d4ab2b5d42f9a6993c3b9a814 | "2023-06-17T17:42:50Z" | c++ | "2023-06-22T16:31:14Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,373 | ["servers/rendering/renderer_scene_cull.cpp"] | Vulkan: VoxelGI Indirect Lighting from Static Omni Light is broken until manually re-setting gi data in Rendering Server | ### Godot version
4.0.3 stable, 4.1 beta 2
### System information
Windows 10 Vulkan forward+ Radeon RX 6700xt
### Issue description
VoxelGI doesn't seem to show any bounce light from a static baked omni light on loading a scene. This happens both in the editor and when running the game.
Immediately after baking the voxelGI data the bounce light is visible and working properly, but it breaks as soon as you reload the scene, switch to another scene tab in the editor, and is also broken in the game window when run.
Once the baked data exists and is saved to a resource file, the lighting can be fixed by simply using the "load" option on the VoxelGIData and selecting the file it's already assigned to.
Calling DisplayServer.instance_set_base() or just using the set_base() method of the voxelGI node to re-set the GI data works to fix the issue in a running game (after a short period of starting the game)
This does not happen with bounce lighting from directional lights. EDIT SpotLight3D also seems to be unaffected by this issue.
### Steps to reproduce
Create new forward+ project
- In a 3d scene disable the preview sun and add 2 cube meshes positioned close enough to see bounce light.
- Add an omniLight3D set the bake mode to static and set the light energy high so that bounce light is easily visible, position it so that it doesn't directly illuminate the gap between the meshes.
- Add a VoxelGI node, create the voxelGIData resource and bake the GI
- bounce lighting is visible
- click Scene -> new scene, then click the tab to go back to the 3d scene with the GI
- bounce lighting is now gone
- bounce lighting is also not present when you run the scene
### Minimal reproduction project
[BrokenIL_MRP.zip](https://github.com/godotengine/godot/files/11779588/BrokenIL_MRP.zip)
The VoxelGI node in this scene has an exported checkbox that if selected does the fix of re-setting the gi data in the display server when the scene is run for comparison | https://github.com/godotengine/godot/issues/78373 | https://github.com/godotengine/godot/pull/81124 | d351d40e2902b05597123eaf6dd5cf15bda2caa5 | f2ba8ec6ca2e4ff493cba11f95621d10ab8bb363 | "2023-06-17T17:05:27Z" | c++ | "2023-10-05T22:24:29Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,338 | ["drivers/unix/file_access_unix.cpp"] | Incorrect permissions in macOS export bundle blocks upload to App Store | ### Godot version
4.1.beta2
### System information
macOS 13.3.1
### Issue description
Trying out macOS export in the new 4.1 beta I've run into a new issue. While the previous issues (see #74154 and #73876) now seems to be handles correctly, which makes the process much smoother for generating a macOS .pck for upload to the App Store, unfortunately a new error has cropped up in validation when uploading the package via Transporter:
```
Asset validation failed (90255)
The installer package includes files that are only readable by the root user. This will prevent verification of the application's code signature when your app is run. Ensure that non-root users can read the files in your app. (ID: 8a1239bb-7ef7-4350-8ba4-9beb5d9e12b2)
```
It seems this is caused by some files in the bundle not having read permission for everyone, and inspecting the bundle for example the Resources .pck, icon, PkgInfo and even info.plist are all set to no access for the everyone group, with only Read and Write access for my user. This seems to be a regression since app bundles exported by previous versions of Godot 4.x to not have this issue.
However, trying to change the permissions of these files in the bundle and then manually signing and creating the pkg doesn't seem to solve the problem, so it may be that the incorrect permissions on these files aren't the issue, just another symptom of some underlying issue that is setting incorrect permissions files in macOS exports. Or I'm for some reason not finding all the files with these issues.
### Steps to reproduce
Reproducing the entire process of signing and packaging up an application to upload it to the Apple App Store via Transporter is complex and requires a bit of work. However, if I'm correct that this is an issue at the core with the basic permissions of files in exported macOS app bundles all you need to do is export any project, like the simple MRP below with a basic macOS export prepared.
- Open the reproduction project and export it using the macOS preset defined.
- Inspect the exported test.app bundle and observe that multiple files in there have permissions `no access` set for everyone.
### Minimal reproduction project
[MacExportPermissionsTest.zip](https://www.lostminds.com/godot/MacExportPermissionsTest.zip) | https://github.com/godotengine/godot/issues/78338 | https://github.com/godotengine/godot/pull/78347 | d63794d466626fd0d176a1ce9ee558fdaf8b3fe5 | b7976f4ad65bbda075f400843a40d6c75c385b49 | "2023-06-16T16:03:29Z" | c++ | "2023-06-18T14:31:05Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,315 | ["editor/gui/editor_spin_slider.cpp"] | Unwarranted value incremented in text boxes on focus | ### Godot version
4.1 beta1, 4.1.beta2
### System information
Windows 11
### Issue description
Whenever you try to select an editable field like Scale.X and Scale.Y. The value is changed as soon as the focus is on.
This is random behavior but consistently reproducible throughout the editor.
https://github.com/godotengine/godot/assets/84369217/2773c297-bf4b-4dd7-86f7-0c837b8d4a12
**It is not reproducible in 4.0.3**
### Steps to reproduce
Reproducible in most of the editable text fields with floating point values.
It is random but consistent
### Minimal reproduction project
N.A | https://github.com/godotengine/godot/issues/78315 | https://github.com/godotengine/godot/pull/78318 | 6d1ca527924f2c868ac31a4d732a5dfc1fd09712 | ada10b57f3b7f4afb280622044df0573637a8e39 | "2023-06-16T06:44:27Z" | c++ | "2023-06-16T08:24:14Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,314 | ["platform/macos/export/lipo.cpp"] | MacOS Export Error: "Failed to create fat binary" (LipO can't open cache file) | ### Godot version
v4.1.beta2.mono.official [a2575cba4]
### System information
Godot v4.1.beta2.mono - Windows 10.0.22000 - Vulkan (Forward+) - dedicated AMD Radeon(TM) Graphics () - Intel(R) Core(TM) i5-4690K CPU @ 3.50GHz (4 Threads)
### Issue description
Console output as follows:
```
LipO: Can't open file: "<path>/editor_data/cache/_lipo/_exe_0."
CodeSign: Failed to create fat binary.
editor/export/editor_export_platform.h:175 - Code Signing: Built-in CodeSign failed with error "Failed to create fat binary.".
```
### Steps to reproduce
Open attached project then: Project->Export->Export All...->Release
(Attached project is just simple scene with basic/default export options)
### Minimal reproduction project
[TestProj.zip](https://github.com/godotengine/godot/files/11766951/TestProj.zip)
| https://github.com/godotengine/godot/issues/78314 | https://github.com/godotengine/godot/pull/78319 | ada10b57f3b7f4afb280622044df0573637a8e39 | 598378513b256e69e9b824c36136774c41cc763c | "2023-06-16T06:33:34Z" | c++ | "2023-06-16T08:24:37Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,299 | ["editor/plugins/gpu_particles_2d_editor_plugin.cpp", "editor/plugins/gpu_particles_3d_editor_plugin.cpp", "editor/plugins/sprite_2d_editor_plugin.cpp", "editor/scene_tree_dock.cpp", "editor/scene_tree_dock.h"] | changing type of a node removes reference from `Node` properties | ### Godot version
v4.1.beta.custom_build [dbd482faa]
### System information
macOS 13.3.0 - Vulkan (Forward+) - integrated Apple M1 Pro - Apple M1 Pro (8 Threads)
### Issue description
Some editor action like changing type will replace the node by another node. Right now Node properties will still point to the old Node.
### Steps to reproduce
1. create a scene with a root node
2. attach a script that exports a node property
3. set the property to point to the root node
4. change the type of the root node
5. look at the property it will error out and the editor will be blank
### Minimal reproduction project
[Test.zip](https://github.com/godotengine/godot/files/11764206/Test.zip)
with MRP you can start at step 4 | https://github.com/godotengine/godot/issues/78299 | https://github.com/godotengine/godot/pull/78300 | 54c0d0ff320e5bb7bfc8254451a7000b28581100 | e4b8dc81b8d06d2e263b9865017c48fe5d98c287 | "2023-06-16T00:22:14Z" | c++ | "2023-08-03T20:37:40Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,296 | ["core/object/script_language.h", "scene/gui/code_edit.cpp"] | Script Editor auto completions sorting is broken | ### Godot version
c0d8d91
### System information
Windows 11
### Issue description
There's weird bug going on with the latest master branch. The auto completion in script editor is sorted in a wacky ways, putting sort of constants ahead of anything because it's ahead alphabetically
Now compare it to my build about two weeks ago
Both are using the same content of 'editor_data' settings folder. And loading up the same project.
### Steps to reproduce
- Open a new blank project
- Make a new script
- Triggers any auto completion in Script Editor
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78296 | https://github.com/godotengine/godot/pull/78321 | ce60395b4a462c6cc5e8b2c115fc28d2d57909ae | c65aba7aff1d5251001a48cfbe797ffe91a77d6d | "2023-06-15T23:09:19Z" | c++ | "2023-06-16T14:52:27Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,295 | ["editor/editor_data.cpp", "editor/editor_node.cpp"] | Very noticeable performance issues when adding nodes to a scene that contains many scripted nodes (editor) | ### Godot version
Godot 4.1-beta2
### System information
Manjaro Gnome, Intel Core i7-11800H
### Issue description
On a relatively modern laptop (Intel Core i7-11800H), in a scene with many scripted nodes (total of 900+ with roughly half of them scripted), adding or duplicating a node freezes the editor for several seconds. Godot 4.0.2 handles such scenes much better!
Here is a flame graph captured while repeatedly duplicating simple Marker2D nodes with Ctrl+D in a complicated scene:
Performance data file to use with hotspot v1.4.1 (relevant data starts at the 75-second mark and lasts about 20+ seconds):
[perf.data.perfparser.zip](https://github.com/godotengine/godot/files/11763700/perf.data.perfparser.zip)
### Steps to reproduce
1) Add 1000 nodes with a script attached.
2) Duplicate a node and witness lags.
Also affects scrolling the scene tree dock.
### Minimal reproduction project
[add_node_lags_repro.zip](https://github.com/godotengine/godot/files/11763731/add_node_lags_repro.zip)
| https://github.com/godotengine/godot/issues/78295 | https://github.com/godotengine/godot/pull/78670 | 1b1e2403f9ee1f0c8152e414c124d42491473e8e | d08ba5f3dca8fb01e60bd369b9d5a77138659b50 | "2023-06-15T22:27:45Z" | c++ | "2023-06-26T12:56:47Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,291 | ["editor/gui/scene_tree_editor.cpp"] | Batch rename won't check for duplicate unique name | ### Godot version
v4.1.beta.custom_build [c0d8d91b1]
### System information
macOS 13.3.0 - Vulkan (Forward+) - integrated Apple M1 Pro - Apple M1 Pro (8 Threads)
### Issue description
When batch renaming nodes it won't check for already used unique name. This will lead to the unique name being revoked with those warning.
<img width="870" alt="Capture dβeΜcran 2023-06-15 aΜ 23 56 06" src="https://github.com/godotengine/godot/assets/66184050/8dbd861f-ca06-4217-9942-7f660dcbc76d">
Expected behavior would be a popup similar to this one being fired when doing the same rename with basic rename and preventing node rename.
<img width="439" alt="Capture dβeΜcran 2023-06-15 aΜ 23 56 21" src="https://github.com/godotengine/godot/assets/66184050/db77881b-fa6c-46b1-8746-e715b3846d9f">
### Steps to reproduce
1. create a scene with a root two child and one grand child
2. make unique name the scond child and the grand child
3. batch rename the two child to "Node"
4. see the warning and the unique name being revoked
### Minimal reproduction project
[Test.zip](https://github.com/godotengine/godot/files/11763614/Test.zip)
with MRP start at step 3 | https://github.com/godotengine/godot/issues/78291 | https://github.com/godotengine/godot/pull/78292 | 725beaa7afce294f38d4aca1fa287249021b6bf8 | 179e3d6ab94bc546721e321802fac32d7016ab4f | "2023-06-15T22:01:55Z" | c++ | "2023-08-03T16:33:37Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,290 | ["editor/gui/scene_tree_editor.cpp"] | Batch rename to empty string will behave strangely. | ### Godot version
v4.1.beta.custom_build [01453ab7f]
### System information
macOS 13.3.0 - Vulkan (Forward+) - integrated Apple M1 Pro - Apple M1 Pro (8 Threads)
### Issue description
When Batch renaming multiple nodes to empty nodes, this will error out instead of using the class name. Moreover item name will be set to empty instead of reverting them back.
https://github.com/godotengine/godot/assets/66184050/ca8dcd75-a306-428e-87ff-7c39c231d101
### Steps to reproduce
1. create a scene with one root and multiple child nodes
2. batch rename them by replacing N by P
3. batch rename them using regular expression to set them all to empty string (using regular expression `.*`)
4. see errors and bugged out items
5. unselect nodes their name will reappear you can see nothing has been set
### Minimal reproduction project
with MRP you can start at step 3
[Test.zip](https://github.com/godotengine/godot/files/11763532/Test.zip)
| https://github.com/godotengine/godot/issues/78290 | https://github.com/godotengine/godot/pull/78292 | 725beaa7afce294f38d4aca1fa287249021b6bf8 | 179e3d6ab94bc546721e321802fac32d7016ab4f | "2023-06-15T21:51:28Z" | c++ | "2023-08-03T16:33:37Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,281 | ["editor/editor_audio_buses.cpp", "editor/editor_node.cpp", "editor/editor_node.h", "editor/plugins/animation_player_editor_plugin.cpp", "editor/plugins/debugger_editor_plugin.cpp", "editor/plugins/shader_editor_plugin.cpp"] | Theme editor starts opened and crashes upon interaction. | ### Godot version
4.1 beta2
### System information
Windows 10 Home
### Issue description
I had a control with a custom theme which i removed. When loading the project again the Theme Editor is opened and some actions, like Show Default or Override All, cause the editor to crash.
### Steps to reproduce
1. create a scene with a Control node
2. add a custom theme to it
3. click the theme so the Theme Editor opens
4. remove the theme
5. save the scene
6. close the project and open it again (Theme Editor needs to be opened when closing the project)
7. click Show Default or Override All
### Minimal reproduction project
[Godot41_themeEditorCrash.zip](https://github.com/godotengine/godot/files/11760454/Godot41_themeEditorCrash.zip)
| https://github.com/godotengine/godot/issues/78281 | https://github.com/godotengine/godot/pull/78611 | 46b2619223d5928381b1792e9736fd7e89b3bb53 | f44b2d7898402a2a337bdc6a321f093a762dc18d | "2023-06-15T16:25:57Z" | c++ | "2023-06-23T15:52:35Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,252 | ["modules/gdscript/gdscript_compiler.cpp", "modules/gdscript/gdscript_parser.h", "modules/gdscript/tests/scripts/runtime/features/export_group_no_name_conflict_with_properties.gd", "modules/gdscript/tests/scripts/runtime/features/export_group_no_name_conflict_with_properties.out"] | Using @export_category using the same name as a variable will overwrite the first variable in the category. | ### Godot version
4.0.2.stable.mono
### System information
Windows 10 - 4.0.2.stable.mono - Any Renderer
### Issue description
In a script with an `@export` variable (with a non-default value), Godot will replace the first variable in the category with the same name. For example, if you're exporting a variable called "Vector", and the value is set to a non-default value, the first variable in the category called "Vector" will also be changed to the same value.
### Steps to reproduce
Attach said script to a node:
```
extends Node
@export var Vector: Vector2
@export_category("Vector")
@export var Var2: float = 1
func _process(delta):
print(Var2)
```
When changing the value of the `Vector` export variable, the variable Var2 will contain the same value. This does not happen when the `@export_category` has a different name.
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78252 | https://github.com/godotengine/godot/pull/78254 | 62b4643d854755b36d7c7ace0047d7a40e812f1a | efbff1369a9eb366667c1d641ed6c853bdd8c830 | "2023-06-15T02:37:15Z" | c++ | "2023-07-31T19:01:03Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,238 | ["scene/resources/packed_scene.cpp"] | Missing script that has exported `Array[Node]` make godot crash when opening the scene | ### Godot version
4.1.dev (4d5f10fc3a)
### System information
macOS 13.3.0 - Vulkan (Forward+) - integrated Apple M1 Pro - Apple M1 Pro (8 Threads)
### Issue description
When a script was attached to a node in a saved scene and this script is exporting a member typed with `Array[Node]` . Reopening the scene when the script is missing because it was deleted or moved will cause a crash
### Steps to reproduce
1. Create a scene with two nodes
2. attach a script to the root with just one exported member typed as `Array[Node]`
3. Add both nodes to the exported member
4. Save the scene
5. close the scene
6. Delete the script
7. Try to open the scene again godot will crash
### Minimal reproduction project
with the MRP you can start reproduction steps at the sixth step.
[Test.zip](https://github.com/godotengine/godot/files/11749000/Test.zip) | https://github.com/godotengine/godot/issues/78238 | https://github.com/godotengine/godot/pull/78240 | fe1aa263ca3b5cc77c3c9f8aec4a29f3b107f087 | 41b34e3e56e9e63a19834d8e47443fcf70d18583 | "2023-06-14T18:59:42Z" | c++ | "2023-06-15T08:50:45Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,230 | ["modules/gdscript/gdscript.cpp"] | Autoload scripts compile error are not reported (except confusing "Script does not inherit from Node") | ### Godot version
v4.0.3.stable.official [5222a99f5]
### System information
Win 10
### Issue description
When Autoload scripts have compile errors, the error is not printed to Output. Instead, they fail when Running with an error completely unrelated to what's wrong:
```
USER ERROR: Script does not inherit from Node: res://thetrouble.gd.
at: start (main/main.cpp:2776)
```
There's no other information. I assume this message is because their failure to compile means they don't get associated as a Node child (related to #75388?)? But I don't understand why they don't have any errors reported. I think it only occurs if they were added to Autoload before they got compiled/used?
I am using an external editor (gvim) and godot's LSP support. That correctly shows the error, but only when I'm editing the broken file which makes it unclear what's wrong. (When I first encountered this problem, I looked at the files, saw they were extending Node2D and didn't understand the problem. It's possible the lsp was failing to show any errors, but it shows errors now.)
### Steps to reproduce
1. New project, add scene
2. Add script "thetrouble.gd"
3. Add thetrouble.gd as an Autoload
4. Add invalid code to thetrouble.gd (two variables named `seq`)
* **Expected**: compile error reported
* **Actual**: No output
5. Run project
* **Expected**: compile error reported
* **Actual**: Script does not inherit from Node: res://thetrouble.gd.
6. Add valid code to thetrouble.gd (assign a value to `seq`)
* **Expected**: compile error reported
* **Actual**: No output
Note: If you mix up steps 2 and 3, the issue will not occur.
Workaround: Edit Autoloads and disable and re-enable thetrouble's autoload state and error messages appear. New errors even appear after this point.
### Minimal reproduction project
[repro.zip](https://github.com/godotengine/godot/files/11747339/repro.zip)
| https://github.com/godotengine/godot/issues/78230 | https://github.com/godotengine/godot/pull/78540 | 4ed08409b044114a0744aff9ef3fa3f0c6ae5726 | 5e1671afd96fd3f8f02188d8e238affac5b82a97 | "2023-06-14T14:57:13Z" | c++ | "2023-08-03T12:46:33Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,228 | ["drivers/gles3/storage/texture_storage.cpp"] | Shader: Incorrect texture read in Editor (Compatibility Renderer) | ### Godot version
v4.0.3.stable.arch_linux
### System information
Arch linux - OpenGL API 4.2 (Core Profile) Mesa 21.3.9 Amber - Compatibility - Using Device: Intel Open Source Technology Center - Mesa DRI Intel(R) HD Graphics 2500 (IVB GT1)
### Issue description
I wrote a simple colorize shader:
```
uniform sampler2D color_map;
void fragment()
{
vec2 uv = COLOR.rg;
COLOR.rgb = texture(color_map, uv).rgb;
}
```
But when I change the uniform parameter in the editor, it turned into this
However, this issue does not appear on Vulkan renderer (Foward+ and Mobile)
### Steps to reproduce
Change value in the gradient in the shader parameter.
### Minimal reproduction project
[shader-bug.zip](https://github.com/godotengine/godot/files/11746761/shader-bug.zip)
| https://github.com/godotengine/godot/issues/78228 | https://github.com/godotengine/godot/pull/78287 | 1710a7d30313728e90ea35243b0230e411e17483 | 9e81aeda335a1fb0006965e7b44b099b2e2a55c1 | "2023-06-14T14:27:36Z" | c++ | "2023-06-16T08:23:03Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,224 | ["scene/gui/base_button.cpp"] | Button's shorcut ignore button_group "allow_unpress" attribute | ### Godot version
4.1.beta.custom_build.824c139f6
### System information
EndeavourOs - Linux kernel 6.3.7-arch1-1
### Issue description
Pressing a button through an associated shortcut allow to ignore the button group attribute "allow_unpress".
Button can then be unpressed
The button properly stay pressed when you click on it, the issue seem to only come from "clicking it" with the shortcut
### Steps to reproduce
- Create multiple buttons linked with the same button group
- Associate a shortcut to each button
- Use the shortcut to "press" the already selected button
### Minimal reproduction project
[button_group_shortcut.zip](https://github.com/godotengine/godot/files/11745281/button_group_shortcut.zip)
| https://github.com/godotengine/godot/issues/78224 | https://github.com/godotengine/godot/pull/78293 | 598378513b256e69e9b824c36136774c41cc763c | ce60395b4a462c6cc5e8b2c115fc28d2d57909ae | "2023-06-14T11:48:17Z" | c++ | "2023-06-16T14:52:01Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,213 | ["servers/rendering/shader_language.cpp"] | Cannot set filter/repeat mode for uniform arrays of Sampler2Ds | ### Godot version
v4.0.3.stable.official [5222a99f5]
### System information
Godot v4.0.3.stable - Windows 10.0.19045 - Vulkan (Forward+) - dedicated NVIDIA GeForce GTX 1080 (NVIDIA; 31.0.15.1694) - Intel(R) Core(TM) i5-10600KF CPU @ 4.10GHz (12 Threads)
### Issue description
There is no way to set the filter or repeat mode of a sampler2D array, trying to do so using shader hints `filter_nearest`/`filter_linear` or `repeat_enable`/`repeat_disable` results in an error:
`error(3): This hint is not supported for uniform arrays.`
Note: not a duplicate of #62472, which applies to `sampler2DArray`s, but not arrays of `sampler2D`s
### Steps to reproduce
```
shader_type spatial;
uniform sampler2D[8] textures : source_color, filter_nearest;
```
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/78213 | https://github.com/godotengine/godot/pull/79100 | 64689c146ca7905836b5c30bffe2d34fa2cafce0 | 65cc4db0ae7a80a9ec2b365f0cf6cbefcf2926f4 | "2023-06-14T04:57:07Z" | c++ | "2023-07-12T12:18:34Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,210 | ["scene/gui/nine_patch_rect.cpp", "scene/gui/nine_patch_rect.h"] | Nine patch rect doesn't update when texture atlas region is updated | ### Godot version
v4.1.beta1.mono.official [828ec2c5d]
### System information
Kubuntu 22.04 Compatability
### Issue description
As funky as it sounds I tried using a atlas texture as the texture for a nice patch and discovered that the nine patch is not updated in the scene when the atlas texture region is changed
### Steps to reproduce
- Add a altas texture to the nice patch node
- Mess with the atlas texture's region property (not the nice slices region property)
- See it doesn't update in the scene
### Minimal reproduction project
I will upload one if needed as it stand it seems like a very simple thing to reproduce | https://github.com/godotengine/godot/issues/78210 | https://github.com/godotengine/godot/pull/78236 | 0653f7a5b05861288d6ffe519de2a2cf5ee21d8e | 94a4da9de60170e8b634fb7cc793cd4be43b9496 | "2023-06-14T03:14:54Z" | c++ | "2023-06-20T07:21:15Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,208 | ["scene/resources/packed_scene.cpp", "scene/resources/packed_scene.h"] | Exported array gets emptied when saving scene script | ### Godot version
v4.1.beta1.official [828ec2c5d]
### System information
Godot v4.1.beta1 - Windows 10.0.23475 - Vulkan (Mobile) - dedicated NVIDIA GeForce RTX 3060 (NVIDIA; 31.0.15.3550) - AMD Ryzen 5 5500 (12 Threads)
### Issue description
I have an exported array for a given type I created with class_name
```
extends Node3D
class_name Enemy
```
It is exported as follow
```
@export var rollable_hazards: Array[Enemy]
```
I populate it using the editor and it works fine, but sometimes it gets emptied after saving the exporting scene. Not sure if it relates with #62916 and/or #77735.
Here's a screen recording I did showing the issue
https://github.com/godotengine/godot/assets/8123916/09d5df4e-83af-4459-b46b-1c62f78c2bb5
Please let me know if you need any additional information.
### Steps to reproduce
1. Export an array from a scene script
2. Populate it with other nodes
3. Edit the exporting script and save it
### Minimal reproduction project
If you check this branch out, commit [327d03d](https://github.com/lazaromenezes/dungeon-explorer-godot/commit/327d03da6c995ed586b8bc8764bdb051bb42bf7b) was the base for the changes I had into the screen recording. | https://github.com/godotengine/godot/issues/78208 | https://github.com/godotengine/godot/pull/77735 | b06b00eb447526739e854cc30c4edb1445ba9325 | 75ef6e486c7f0e1e50463c54b65f454810608a21 | "2023-06-14T01:33:14Z" | c++ | "2023-06-19T22:04:01Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,205 | ["editor/plugins/text_shader_editor.cpp"] | Ctrl+K to comment lines in shader editor isn't working if the shader is a .gdshaderinc | ### Godot version
Godot 4.0.3
### System information
Windows 10 64 bits NVIDIA GeForce GTX 1060
### Issue description
Ctrl+K is a shortcut normally used to comment out the current line. It works in the shader editor, but for some reason, it doesn't if your shader has the `.gdshaderinc` extension (aka it's a shader include). Shader Includes are the same language, we should be able to use that feature.
### Steps to reproduce
Create a `.gdshaderinc` file. Write some code in it. Observe you can't use Ctrl+K to comment out lines.
### Minimal reproduction project
[CommentLinesShaderInc.zip](https://github.com/godotengine/godot/files/11739881/CommentLinesShaderInc.zip)
| https://github.com/godotengine/godot/issues/78205 | https://github.com/godotengine/godot/pull/79158 | 79b31a8357e26a510ed68edebb3c9cf905f807e9 | bdcbfd71a87f446b9742d690d92838aad16f2bcb | "2023-06-13T22:09:40Z" | c++ | "2023-07-08T16:22:42Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,202 | ["editor/editor_plugin_settings.cpp", "editor/gui/scene_tree_editor.cpp"] | Plugin description tooltip has no limits | ### Godot version
4.1 beta1
### System information
Windows 10.0.19045 - Vulkan (Forward+) - dedicated NVIDIA GeForce GTX 1060 (NVIDIA; 30.0.15.1403) - Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz (8 Threads)
### Issue description
This is a plugin description that spans 2 screens and goes beyond the right screen border.
I just wanted to test if there is any limit to plugin description size. There is none. You can potentially put a text that will cover your whole screen (don't try it π).
Not sure how we should solve that. We could autowrap the label, but to gracefully display extremely long texts they either need to be scrolled (in a tooltip??) or truncated.
### Steps to reproduce
1. Open Project Settings
2. Go to Plugins -> Create New Plugin
3. Write a book in the Description field
### Minimal reproduction project
https://www.lipsum.com/ | https://github.com/godotengine/godot/issues/78202 | https://github.com/godotengine/godot/pull/79090 | a2604ff7123174fd95c22a4c7f3e2972bb4fb514 | 5205ff69ca1cf971cadbf4c418fb99179f2bf6e8 | "2023-06-13T21:35:46Z" | c++ | "2023-07-07T14:04:41Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,196 | ["doc/classes/RichTextLabel.xml"] | Appending to RichTextLabel.text restores any cleared text | ### Godot version
v4.0.3.stable.official [5222a99f5]
### System information
Windows 10 - Godot 4.0.3
### Issue description
Calling `RichTextLabel.clear()` and then appending to its `text` field restores all of the cleared text.
This seems related to #54951, but this issue has nothing to do with `bbcode_text` field -- it occurs simply by writing text and calling the `clear` function.
This bug can be trivially worked around by calling `RichTextLabel.text = ""` in addition to `RichTextLabel.clear()`. However this is nonsensical and contradicts the documentation, which explicitly states that the clear() function "Clears the tag stack and sets text to an empty string"
### Steps to reproduce
1. Click the "Add Text" button. A line of text is added to the RichTextLabel.
2. Click the "Clear" button. The RichTextLabel is cleared.
3. Click the "Add Text" button. The cleared lines are restored, and additional text is added beneath them.
### Minimal reproduction project
[RichTextLabel-clear.zip](https://github.com/godotengine/godot/files/11737734/RichTextLabel-clear.zip)
| https://github.com/godotengine/godot/issues/78196 | https://github.com/godotengine/godot/pull/78203 | 838b19a1ccd641c2567d625ee49b1d7458bf4c43 | 72ee877b22669ddef4d64df81aba30a313da6a11 | "2023-06-13T17:53:59Z" | c++ | "2023-06-14T09:13:44Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,185 | ["core/object/undo_redo.h"] | _set method for a String property triggered too much times, when in Resource property's embedded inspector, and text changes fast | ### Godot version
v4.0.3.stable.official [5222a99f5]
### System information
Windows 11
### Issue description
See screen recording below.
~~I type in a String property which is within a PROPERTY_USAGE_ARRAY / EditorInspectorArray.~~
I type in a String property which is within a Resource property's (DataflowGraph.Function: DataflowFunction) embedded inspector
When I type slowly in a String property. 1 char change -> 1 trigger of _set method
When I type faster, 1 char change -> Triggers _set method for all char changes since the last typing
When I again type slow, 1 char change -y 1 trigger again
~~End of recording: Not observed for String property, which is not in array property.~~
End of recording: Not observed for String property, which is the property of the outer Resource (DataflowGraph)
### Steps to reproduce
Get project from attached ZIP.
Open res://addons/dataflow_engine/test_dataflow_graph.tres in Inspector
Do what you read/see in issue description.
### Minimal reproduction project
[Project.zip](https://github.com/godotengine/godot/files/11735719/Project.zip)
| https://github.com/godotengine/godot/issues/78185 | https://github.com/godotengine/godot/pull/78809 | 16dd4e5725a2576ba5765f9f15bccb519b83482c | ab430b0b4a2d4434eabf54f8704459b37cd19dee | "2023-06-13T14:05:03Z" | c++ | "2023-06-29T13:42:36Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,150 | ["scene/resources/packed_scene.cpp", "scene/resources/packed_scene.h"] | Exported typed array in instanced child scene gets empty elements added and crashes editor | ### Godot version
4.1.beta1
### System information
macOS 13.3.1
### Issue description
If you have a scene (ChildScene) with a node that has an exported typed array, for example
`@export var export_array:Array[Node]`
use the inspector to add a number of child nodes as the elements of this array and then instance this scene in another scene (ParentScene) the editor becomes prone to crashing, seemingly every time it tries to load or save ParentScene. This reliably crashes the editor, and also when opening the project after the crash, the export_array array now has added empty elements, same as the number of elements it originally had.
### Steps to reproduce
Set up a child scene with a node with a typed export array, and set up a few child nodes to set as the items in the array in the inspector.
Instanciate the childScene in another scene parentScene
Now, go back to the childScene and save it again (with or without modification) open the parent scene and the editor will crash. When opening the editor again after the crash you can see that the array in both the instance in parentScene and the original in childScene has added empty elements (but not in the same places!).
If you change the child scene, removing the empty rows and try to go to the parentScene you'll crash again. Open the project and the empty elements will be back. If you don't remove them and crash again, additional empty elements will have been added.
https://github.com/godotengine/godot/assets/17763524/373ef395-178a-4206-9a73-fec55e2b5a68
You can also get a warning this is happening if when you save a scene you get a bunch of errors:
`scene/main/node.cpp:1988 - Condition "!common_parent" is true. Returning: NodePath()`
Since having typed export arrays with empty items apparently causes this error (maybe this is also a bug?) so you can see that something has gone wrong and empty items have been added to your exported array.
### Minimal reproduction project
[ExportArrayTest.zip](https://www.lostminds.com/godot/ExportArrayTest.zip) | https://github.com/godotengine/godot/issues/78150 | https://github.com/godotengine/godot/pull/77735 | b06b00eb447526739e854cc30c4edb1445ba9325 | 75ef6e486c7f0e1e50463c54b65f454810608a21 | "2023-06-12T17:40:55Z" | c++ | "2023-06-19T22:04:01Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,146 | ["modules/gdscript/gdscript.cpp", "modules/gdscript/gdscript_compiler.cpp"] | Unexpected "Parser Error : Identifier not found" when using parent class member | ### Godot version
4.0.2
### System information
Linux Mint 21
### Issue description
On the minimal reproduction project, with a 3 level hierarchy and 2 autoloads, when running the game we get a parse error in daughter.gd where we use a parent member.
**Additional remarks :**
- We have a circular dependency involving one of the autoloads : Autoload2 references Daughter which references Mother which references GrandMother which references Autoload2
- Autoload1 references Mother which references GrandMother which references Autoload2
- We don't have any error if we switch the 2 autoloads order in Project Settings
- We get the same error if _member is put in GrandMother class instead of Mother class
- We still get the issue if we disable Autoload1 in Project Settings
**A quick look in Godot source code :**
- GDScriptCache::full_gdscript_cache contains invalid scripts (ie script->is_valid() is false)
- I think the engine tries to compile GrandMother, then Daughter (but it fails), then Mother
(This issue might be related to https://github.com/godotengine/godot/issues/75388.)
### Steps to reproduce
To get the parse error :
- Open the minimal reproduction project
- Run the game
You'll get `Parser Error: Identifier not found: _member`.
### Minimal reproduction project
[ParseError_Project.zip](https://github.com/godotengine/godot/files/11725756/ParseError_Project.zip)
| https://github.com/godotengine/godot/issues/78146 | https://github.com/godotengine/godot/pull/79205 | 5954c58ee1912d6ddceb165edfc27cc6084a4fa2 | 2c552140686b5366e48c61a1a15001f15359cccc | "2023-06-12T16:39:20Z" | c++ | "2023-07-17T17:15:16Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,145 | ["modules/navigation/nav_region.cpp"] | NavigationAgent3D takes incorrect paths when NavigationRegion is rotated | ### Godot version
4.1.beta1
### System information
Godot v4.1.beta1 - Windows 10.0.22621 - Vulkan (Forward+) - dedicated NVIDIA GeForce RTX 3060 Ti (NVIDIA; 31.0.15.2802) - 11th Gen Intel(R) Core(TM) i5-11400F @ 2.60GHz (12 Threads)
### Issue description
The agent finds the correct path between a and b.
Then the scene is rotated 180 degress.
Now the agent is taking a longer path than it needs to.
https://github.com/godotengine/godot/assets/6430599/3fa04c4f-49a0-468d-9d4c-4a12dee3cb5b
### Steps to reproduce
Download project. Press "a" and "d" to rotate the scene around the z axis.
### Minimal reproduction project
[movenav.zip](https://github.com/godotengine/godot/files/11725682/movenav.zip)
| https://github.com/godotengine/godot/issues/78145 | https://github.com/godotengine/godot/pull/78173 | ad935c4d93f6aa90b2c6a1f9de6911500e785012 | 2b373f795c42f330c02998c3a889d17752f1b3ba | "2023-06-12T16:35:52Z" | c++ | "2023-06-13T11:53:09Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,142 | ["editor/editor_node.cpp"] | `set_edited_scene` out of bounds errors on empty project | ### Godot version
4.1.beta (7b1387ff2)
### System information
Mageia 9, Linux
### Issue description
When opening an empty project in latest `master`, I get these errors:
```
ERROR: Index p_idx = -1 is out of bounds (edited_scene.size() = 1).
at: get_scene_path (editor/scu/../editor_data.cpp:880)
ERROR: Index p_idx = -1 is out of bounds (edited_scene.size() = 1).
at: check_and_update_scene (editor/scu/../editor_data.cpp:683)
ERROR: Index p_idx = -1 is out of bounds (edited_scene.size() = 1).
at: set_edited_scene (editor/scu/../editor_data.cpp:745)
```
Possibly related to recent changes to restore past sessions, but there's nothing to restore in this case?
### Steps to reproduce
- Create a new project, edit it
- Quit to project manager
- Re-open the newly created project -> errors
### Minimal reproduction project
Empty project. | https://github.com/godotengine/godot/issues/78142 | https://github.com/godotengine/godot/pull/78152 | c43b6b5ac8d6c464d1e24a691d7901190400a32f | a5206e449ead208b6cb79a3731ef58bf871d5f05 | "2023-06-12T15:38:42Z" | c++ | "2023-06-12T20:56:46Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,140 | ["editor/import/scene_import_settings.cpp"] | Can't reset root node type to Node3D in advanced import settings | ### Godot version
4.0.3 and latest 4.1 (2d6b880987bc600cda586b281fcbe26791e92e09)
### System information
Manjaro Linux
### Issue description
When you change the root type of an imported scene (in my case it's a glb file) you can't change it back to a Node3D. Changing it to any other 3D node works though.
### Steps to reproduce
1. Import a glb file
2. Double click it to open the advanced import settings and change the root type to StaticBody3D or any other type that inherits from Node3D. Click reimport.
3. Now try to change it from StaticBody3D to Node3D. This will fail silently.
### Minimal reproduction project
[RootTypeBug.zip](https://github.com/godotengine/godot/files/11725039/RootTypeBug.zip) | https://github.com/godotengine/godot/issues/78140 | https://github.com/godotengine/godot/pull/79907 | 2992ffd25523cb8b3e0ef1b873c40115b8940178 | ba3fb66d71cef59c1152a13eab9491ea25953ff3 | "2023-06-12T15:14:42Z" | c++ | "2023-08-01T20:30:07Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,136 | ["modules/gdscript/gdscript.cpp"] | Game crashes on exit if closed while awaiting a signal and there are unfreed objects | ### Godot version
4.0.3
### System information
Ubuntu 22.04.2 LTS
### Issue description
If awaiting a signal, and there are unfreed objects in memory, and the user closes the game, then the executable crashes with this error:
```
ERROR: Condition "_first != nullptr" is true.
at: ~List (./core/templates/self_list.h:106)
```
In a build (rather than running from the editor), this is followed by this error:
```
WARNING: ObjectDB instances leaked at exit (run with --verbose for details).
at: cleanup (core/object/object.cpp:1982)
ERROR: Resources still in use at exit (run with --verbose for details).
at: clear (core/io/resource.cpp:489)
```
This happens whether the object in question is an `Object` subclass that has not been freed or a `RefCounted`.
What I expect to happen is that the application should close without throwing these errors. I expect any such memory to be automatically freed on application exit, or at least for the application to close gracefully, without throwing an error. The application's giving an error code produces a bad user experience that I cannot avoid if using signals and non-Node scripts.
I have seen the same behavior on Windows 10, although I am writing this report from Linux.
Note that these errors show up in the terminal that launched Godot (the engine or the built game), not within Godot's "Output" window.
### Steps to reproduce
Please see the attached minimal reproduction project. Run `main.tscn` and close the running application: it will produce an error. Run the application again and click the button, whose `pressed` signal is being awaited in the script. Now close the application, and no error will be produced.
The error is easiest to see if godot itself was started from the command-line: this error is reported in the terminal that launched Godot, not within Godot's built-in terminal.
Whether the allocated object is an unfreed `Object` or a `RefCounted` does not seem to matter. The example project uses `item.gd` as a minimal `RefCounted`. Once can change it to an `Object` and stick an `_item.free()` at the end of the `_ready()` function, and the same behavior will occur.
To make your own reproduction:
- Make a custom `Object` or `RefCounted` script
- In the main scene, e.g. in `_ready`, create an instance, then await a signal
- Run the application and close it before that signal is received.
Here is a sample. Given a script like this:
```
class_name Item
extends RefCounted
```
Then the following scene will generate the problem if closed before clicking a button in it:
```
extends Control
func _ready():
var _item := Item.new()
await $Button.pressed
```
### Minimal reproduction project
[repro_await.zip](https://github.com/godotengine/godot/files/11722690/repro_await.zip)
| https://github.com/godotengine/godot/issues/78136 | https://github.com/godotengine/godot/pull/78138 | 375156a637ad2a997f3cda7bbfb8ed51f4b7beab | 5145098e6e3407c309f20101df4e06457516978e | "2023-06-12T12:09:09Z" | c++ | "2023-06-14T07:25:34Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,118 | ["scene/gui/container.cpp", "scene/gui/control.cpp"] | 4.1.beta1, mouse-over slider shows offset handle highlight | ### Godot version
v4.1.beta1.official [828ec2c5d]
### System information
Godot v4.1.beta1 - Windows 10.0.19045 - Vulkan (Compatibility) - GeForce GTX 550 Ti () - Intel(R) Core(TM) i3-2100 CPU @ 3.10GHz (4 Threads)
### Issue description
Issue observed in:
- v4.1.beta1.official [828ec2c5d]
Expected behavior in:
- v4.1.dev4.official [5c2295ff5]
- v4.0.stable.official [92bee43ad]
- v3.5.2.stable.official [170ba337a]
### Description
A sliders mouse-over handle highlight (a circle with a ring around it) appears offset (down and to the right), the first time the mouse is moved over a slider.
### Expected
The slider handle highlight, should appear on the sliders handle on the first mouse-over.
As it does in the Godot versions before: 4.1.beta1
### Steps to reproduce
- open the attached project: [issue, mouse over slider, handle highlight, offset.zip](https://github.com/godotengine/godot/files/11714943/issue.mouse.over.slider.handle.highlight.offset.zip)
Node2D is selected, in the Scene panel, and the Transform group/section is unfolded in the Inspector panel.
- move the mouse over the Transform > Rotation, slider
### Observed
The sliders mouse-over handle highlight (a circle with a ring around it) appears offset (down and to the right) from the sliders handle.
The mouse-over handle highlight moves to the slider handle when:
- the mouse is held still over the slider until the tooltip appears
- or if the mouse is moved away from the slider and back over the slider
Now the second and subsequent slider mouse-overs keep showing the handle highlight correctly.
The issue can be reproduced by deselecting and selecting Node2D in the Scene panel.
https://github.com/godotengine/godot/assets/13420573/c817788c-6032-44e0-8693-35801f406c51
### Expected
The sliders mouse-over handle highlight should appear on the sliders handle, the first time the mouse is moved over the slider.
### Minimal reproduction project
[issue, mouse over slider, handle highlight, offset.zip](https://github.com/godotengine/godot/files/11714943/issue.mouse.over.slider.handle.highlight.offset.zip)
| https://github.com/godotengine/godot/issues/78118 | https://github.com/godotengine/godot/pull/78009 | 35ff936b93969f3e6d316a485cee3722bb17bba0 | 1b5620d2a9d30598f68a828b3dd4a07296e2c546 | "2023-06-11T15:39:39Z" | c++ | "2023-06-12T12:14:20Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,116 | ["core/object/script_language_extension.h"] | ScriptInstanceExtension GDExtensionObjectPtr is no longer readable on Godot Side | ### Godot version
v4.1.beta
### System information
Windows 10
### Issue description
```
virtual Ref<Script> get_script() const override {
if (native_info->get_script_func) {
GDExtensionObjectPtr script = native_info->get_script_func(instance);
return Ref<Script>(reinterpret_cast<Script *>(script));
}
return Ref<Script>();
```
Is not creating a functional reference.
GDExtensionObjectPtr is a part of gdextension_interface and is a void pointer to pas data pointers between GDExtesion and Godot Core.
In the images below you are able to see that the extension `reinterpret_cast `works fine.
And that the pointer is passed to core.
However at core `Ref<Script> script_test` stays a NULL value
I was hoping this would be solved by solving [cpp 1119](https://github.com/godotengine/godot-cpp/issues/1119)
### Steps to reproduce
Create a custom script instance and create it using `internal::gdextension_interface_script_instance_create` and a `GDExtensionScriptInstanceInfo` that only implements `get_property_list_func` and `get_script_func`.
### Minimal reproduction project
Difficult to create a minimal project.
Mostly looking for advice on what to test to find the actual source problem | https://github.com/godotengine/godot/issues/78116 | https://github.com/godotengine/godot/pull/78392 | 58fae90ff39c30d19ecd902d3ae9a4de136ca501 | 329652b6feac2114c4e30062f18ece36d4e3addc | "2023-06-11T14:22:27Z" | c++ | "2023-06-20T19:19:20Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,091 | ["scene/gui/container.cpp", "scene/gui/control.cpp"] | Atlas preview jumps after first scroll event | ### Godot version
4.1 beta1
### System information
Windows 10.0.19045 - Vulkan (Forward+) - dedicated NVIDIA GeForce GTX 1060 (NVIDIA; 30.0.15.1403) - Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz (8 Threads)
### Issue description
When I change tile source, the view is initialized at a wrong position. After scrolling it for the first time, the position gets re-adjusted.
https://github.com/godotengine/godot/assets/2223172/830cef4a-10f0-4fa5-9172-e7f8015330eb
It wasn't happening before.
### Steps to reproduce
1. Create a TileMap with TileSet
2. Add a sufficiently big tile atlas source
3. Add another source, so that you can switch between the 2
4. Switch to the other source and then to the big one
5. Scroll
### Minimal reproduction project
The texture from the video: https://github.com/godotengine/godot/assets/2223172/7ac78271-84d0-48ab-ae28-45841e2e0880 | https://github.com/godotengine/godot/issues/78091 | https://github.com/godotengine/godot/pull/78009 | 35ff936b93969f3e6d316a485cee3722bb17bba0 | 1b5620d2a9d30598f68a828b3dd4a07296e2c546 | "2023-06-10T18:46:42Z" | c++ | "2023-06-12T12:14:20Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,088 | ["modules/gltf/editor/editor_scene_exporter_gltf_plugin.cpp"] | Exporting a GLTF file does not automatically update the filesystem tree | ### Godot version
Master, commit 37d1dfef9d81aade2
### System information
PopOs 22.04, any renderer, Gnome, X11
### Issue description
When exporting a scene to GLTF, the new file does not show up in the filesystem tree, until a new file or folder is created.
https://github.com/godotengine/godot/assets/3624853/ede88d50-f93d-4487-913f-1d87bbd8a7bb
Granted, this is not a very common case. But still, the behavior might seem strange and confusing. It gives the impression that the export did not work.
### Steps to reproduce
- Open any 3D scene and export it to GLTF by going to `Scene > Export to ... > gLTF2.0 Scene`
- Set the file path to your `res://` folder
- The file does not show up
- Create a new folder in the FileSystem tree. This will update the tree and your GLTF file will show up then.
### Minimal reproduction project
No need.
Just export any 3D scene to GLTF. | https://github.com/godotengine/godot/issues/78088 | https://github.com/godotengine/godot/pull/78096 | d4951ead3ae64a76e7a985827e228fa71d64f25e | f016852c9fe9c617b5f73fe6b9f7ee5b74c38330 | "2023-06-10T17:11:18Z" | c++ | "2023-06-12T09:41:04Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,087 | ["scene/main/viewport.cpp"] | In the Advanced Import settings editor, tooltips only flash for a split second | ### Godot version
Master, commit 37d1dfef9d81aade2
### System information
PopOs 22.04, Any renderer, NVIDIA Corporation GP107GLM, Driver Version: 525.89.02
### Issue description
On the GLTF import screen, when you hover the mouse cursor over any property, the tooltip only flashes for a split second, making it very hard to read.
This does not happen in other areas of the editor, like when hovering over a property in the node inspector.
https://github.com/godotengine/godot/assets/3624853/532e36ba-1737-449d-a586-22d86df5cc30
### Steps to reproduce
- Double click a GLTF File to open its `Advanced Import Settings` screen
- Hover over any property
- The tooltip will quickly flash then vanish
### Minimal reproduction project
You only need to have GLTF file in any project.
Here is one :
[box-skeleton-blender.gltf.zip](https://github.com/godotengine/godot/files/11712890/box-skeleton-blender.gltf.zip)
| https://github.com/godotengine/godot/issues/78087 | https://github.com/godotengine/godot/pull/78110 | f016852c9fe9c617b5f73fe6b9f7ee5b74c38330 | 9fca49864776c06ba92a2738d9661be217faeffa | "2023-06-10T16:54:59Z" | c++ | "2023-06-12T09:55:58Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,080 | ["editor/plugins/curve_editor_plugin.cpp", "editor/plugins/gradient_texture_2d_editor_plugin.cpp"] | Crash opening visual shader editor for specific shader | ### Godot version
4.1.beta1
### System information
macOS 13.3.1 mobile renderer
### Issue description
Trying to open and edit one of my visual shaders now consistently crashes Godot 4.1beta1.
The shader was made and edited fine in earlier versions of Godot 4 and I'm not certain what is special with this shader (attached below for investigation). Opening and editing other visual shaders works fine for me. The only thing I can think of that this shader uses that other do not is the gradient node, though as I can't open it I'm not sure what else might be going on in it.
### Steps to reproduce
Open the attached visual shader (in any project)
### Minimal reproduction project
[ConnectBlendShader.tres.zip](https://www.lostminds.com/godot/ConnectBlendShader.tres.zip) | https://github.com/godotengine/godot/issues/78080 | https://github.com/godotengine/godot/pull/78171 | 52493767fcb793e27a7caea30c4eef1c2d0b5ecc | 7aa530c4cad478f5555f240d63e2e31c9766fbd5 | "2023-06-10T11:09:52Z" | c++ | "2023-06-13T11:53:02Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,079 | ["scene/gui/line_edit.cpp", "scene/gui/line_edit.h", "scene/gui/spin_box.cpp"] | Impossible to start text selection on unfocused SpinBox | ### Godot version
v4.1.beta.custom_build [d3651ea45]
### System information
Windows 11
### Issue description
It's not possible to start text selection on a SpinBox when it's unfocused.
Note: After a bit of digging in the code, I found that clicking on a SpintBox doesn't trigger `SpinBox::gui_input` no matter if it's focused or not. However, releasing mouse left button does trigger the function.
### Steps to reproduce
See the gif.
### Minimal reproduction project
It's simple to reproduce. | https://github.com/godotengine/godot/issues/78079 | https://github.com/godotengine/godot/pull/78092 | 1b5620d2a9d30598f68a828b3dd4a07296e2c546 | b02dd1c2a30fee8df7264e14780030bb617b7602 | "2023-06-10T09:38:25Z" | c++ | "2023-06-12T12:14:44Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,052 | ["editor/plugins/animation_tree_editor_plugin.cpp"] | AnimationTree(AnimationNodeStateMachine) error reported in the editor | ### Godot version
4.1b
### System information
win11 Vulkan API 1.3.236
### Issue description
Selecting AnimationTree will result in error message
`This function in this node can only be accessed from either the main thread or a thread group. Use call_deferred() instead.`
### Steps to reproduce
AnimationTree -> ROOT -> AnimationNodeStateMachine ->Add Animation.
There are several animations, clicking AnimationTree will result in several errors
### Minimal reproduction project
[test.zip](https://github.com/godotengine/godot/files/11708080/test.zip)
| https://github.com/godotengine/godot/issues/78052 | https://github.com/godotengine/godot/pull/79588 | 0a9006a5aa3f9fac41cb1bd9f4c06fdbc96dc87f | 1886dee28fc7560def60f92d1a94478bf4a701c0 | "2023-06-09T15:59:38Z" | c++ | "2023-08-02T19:36:56Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,049 | ["editor/editor_plugin.cpp"] | Can't use `EditorPlugin.add_autoload_singleton()` with a relative path | ### Godot version
4.0.3.stable
### System information
Linux/X11
### Issue description
when creating addons, i want to use relative paths so that things don't break if i (or another user) decides to rename and/or move the add-on's folder.
using relative paths with ``add_autoload_singleton()`` doesn't work because of an error when enabling the add-on.
## ``res://addons/my_plugin/plugin.gd``
```GDScript
@tool
extends EditorPlugin
func _enter_tree():
# this line doesn't work:
add_autoload_singleton('MyAutoload', 'my_autoload.gd')
# these lines work:
#add_autoload_singleton( 'MyAutoload',
# 'res://addons/my_plugin/my_autoload.gd'
#)
func _exit_tree():
remove_autoload_singleton('MyAutoload')
```
## error message window when enabling the add-on
```
βββββββββββββββββββββββββββββββββββββββββββββββββββ
β Warning! _ β‘ X β
βββββββββββββββββββββββββββββββββββββββββββββββββββ€
β Can't add Autoload: β
β res:// is an invalid path. File does not exist. β
β β
β [OK] [Copy Text] β
βββββββββββββββββββββββββββββββββββββββββββββββββββ
```
the add-on does not get disabled because of the error.
the error-free parts of the add-on still get activated, just the autoload doesn't get created.
### Steps to reproduce
1. create a plugin with the subfolder "my_plugin" and script name "plugin.gd"
2. create an empty script in ``res://addons/my_plugin/my_autoload.gd``
3. edit ``res://addons/my_plugin/plugin.gd``:
```GDScript
@tool
extends EditorPlugin
func _enter_tree():
add_autoload_singleton('MyAutoload', 'my_autoload.gd')
func _exit_tree():
remove_autoload_singleton('MyAutoload')
```
4. enable the plugin
5. see the error message window
### Minimal reproduction project
[add_autoload_singleton.zip](https://github.com/godotengine/godot/files/11706746/add_autoload_singleton.zip)
(13.9 KiByte download, 11.4 KiByte unzipped) | https://github.com/godotengine/godot/issues/78049 | https://github.com/godotengine/godot/pull/78109 | 8c5c29f2a3d519693f153118e6495dab7335a350 | 3dca469f8c1bbeaca867dee2fe8954c7b7a1f413 | "2023-06-09T13:41:48Z" | c++ | "2023-06-15T08:50:33Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,035 | ["servers/rendering/renderer_canvas_cull.cpp"] | 2D Sprites rendered black with enabled Y-sorting | ### Godot version
v4.1.beta1.official [828ec2c5d]
### System information
Windows 11 Home, v4.1.beta1.official [828ec2c5d], All Renderers, NVidia GeForce RTX 3060 Ti, Drivers 535.98
### Issue description
When enabling the "Y Sort Enabled" ordering method for a Sprite2D, the sprite will be rendered entirely black.
### Steps to reproduce
Simply create a Sprite2D Node, assign any texture to it and activate Y-Sorting
### Minimal reproduction project
[GDB_4.1_ysort.zip](https://github.com/godotengine/godot/files/11704361/GDB_4.1_ysort.zip)
| https://github.com/godotengine/godot/issues/78035 | https://github.com/godotengine/godot/pull/78134 | b2335c0df78e1585dd6de92e7a3a8082a25e463d | 66423d440e53109627717edbada77856c514bb6f | "2023-06-09T08:59:24Z" | c++ | "2023-06-12T12:16:41Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,020 | ["scene/main/scene_tree.cpp"] | Node process priority has no effect in 4.1-beta1 | ### Godot version
v4.1.beta1.mono.official [828ec2c5d]
### System information
Godot v4.1.beta1.mono - Windows 10.0.22621 - Vulkan (Forward+) - dedicated NVIDIA GeForce RTX 4070 Ti (NVIDIA; 31.0.15.3598) - 13th Gen Intel(R) Core(TM) i5-13600K (20 Threads)
### Issue description
After the recent refactoring (https://github.com/godotengine/godot/pull/75901), Node process priority seems to be unimplemented. `ComparatorWithPriority` is no longer used anywhere in the engine code.
Nodes behave as if they all had the same priority, and running with `--single-threaded-scene` makes no difference.
### Steps to reproduce
Run the MRP, look at the order in which the nodes print their names.
In Godot 4.0.3: First, Default, Last
In Godot 4.1-beta1: Default, First, Last
### Minimal reproduction project
[ProcessPriorityBug.zip](https://github.com/godotengine/godot/files/11693028/ProcessPriorityBug.zip)
| https://github.com/godotengine/godot/issues/78020 | https://github.com/godotengine/godot/pull/78044 | b6bb244aad0e2520757f7e9db6e9606c8cd3f2f0 | 300748e52c03fd1761b716fc7eea2b9fb97b86f9 | "2023-06-08T21:10:08Z" | c++ | "2023-06-09T11:45:15Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 78,008 | ["servers/rendering/renderer_rd/renderer_canvas_render_rd.cpp"] | Mac rendering warnings such as invalid format | ### Godot version
On top of 4.0 branch. https://github.com/godotengine/godot/commit/fd5dc25240c1537d90b80507683f78bcf574f547
### System information
Mac Mini, Apple Silicon on Firefox
### Issue description
Error message in log.
```
e[1;31mERROR:e[0;91m vkCreateGraphicsPipelines failed with error -3 for shader 'CanvasOcclusionShaderRD:0'.
e[0;90m at: render_pipeline_create (drivers/vulkan/rendering_device_vulkan.cpp:6394)e[0m
[mvk-error] VK_ERROR_FORMAT_NOT_SUPPORTED: VkFormat VK_FORMAT_R64G64_SFLOAT is not supported for vertex buffers on this device.
```
### Steps to reproduce
Launching my project.
### Minimal reproduction project
I think it happens on all mac projects, but checking now. | https://github.com/godotengine/godot/issues/78008 | https://github.com/godotengine/godot/pull/85822 | 9dce1a45a067776f7cd620061fc5e973427f8287 | 74b6fad3c8a884b6c628365b6bddcb0b739a1e51 | "2023-06-08T16:02:28Z" | c++ | "2023-12-08T15:58:59Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 77,997 | ["scene/resources/packed_scene.cpp", "scene/resources/packed_scene.h"] | Instances of scenes with a Material flagged Local to Scene see their material instances become independent and change ID every time you reopen the scene containing them | ### Godot version
v4.0.2.stable.official [7a0977ce2]
### System information
Linux Ubuntu 22.04 with Unity desktop
### Issue description
Local to Scene is convenient for Materials because when changing properties / shader parameters by code, we often want to affect a single entity, not all of them.
However, when the material is placed on a scene to be instantiated (aka "prefab", "blueprint" or "archetype" in other engines; we'll call it "model scene" below), and flagged Local to Scene, each instance has the two following behaviors:
1. Independent: They create their own embedded material instance. While it seems to follow Local to Scene's design, this means such instances will stop being updated with new changes on the original material on the model scene. In addition, the Local to Scene tooltip says that it should affect *run-time*, not edit time. Although there is a point in making some changes in edit time (namely ungreying the fields to allow user to change some instance fields in the inspector to override defaults), this seems to go too far by breaking the original resource relationship.
2. ID change: Every time the scene containing the instances is closed and reopen, the material ID will change in both the SubResource declaration and reference, polluting VCS diff.
Saving the material as an asset (Resource) instead of being embed in the model scene doesn't fix the issue.
For now, my workaround is to manually duplicate the resource to make it a unique instance at runtime, on game start (or node instantiation), as I do in other engines:
`sprite.material = sprite.material.duplicate()`
This doesn't allow overriding property instances in the inspector at edit time, but at east it avoids sharing material shader parameter updates with all instances (whether due to scene instantiation or material resource asset reuse), while avoiding the repeated ID change.
### Steps to reproduce
The best is to download the MRP and read the Git history which explains each step and shows the diff. It also has additional steps to demonstrate cases where the bug does _not_ occur. Below, I'll sum-up the core steps:
**ID change**
1. Create a scene with a sprite with a CanvasItemMaterial flagged Local to Scene
2. Create another scene where you instantiate the model scene
3. Save the new scene and commit in Git.
4. Close and reopen the scene and observe the diff
Observed diff:
CanvasItemMaterial ID changes in both sub-resource declaration, and reference usage.
Since the ID change can be repeated indefinitely, you can also open the MRP and repeat step 4 as many times as you want. Note that the MRP has 2 scene model instances so you will see 2 IDs changing each time.
**Independence**
5. Change some property on the model scene material such as `particles_animation = true`
6. Save model scene, open scene with instance and save: particles_animation is not updated on the instance
**Bonus: Material as Resource**
Save material as a resource asset and try again: the bug still occurs.
### Minimal reproduction project
[v4.0.2 - Material of instance of scene with Local to Scene material keeps changing ID.zip](https://github.com/godotengine/godot/files/11686919/v4.0.2.-.Material.of.instance.of.scene.with.Local.to.Scene.material.keeps.changing.ID.zip)
| https://github.com/godotengine/godot/issues/77997 | https://github.com/godotengine/godot/pull/65011 | 31a7fd1951f6e9f68faa6edd2aaff87cb922a31c | 4330a94b1382937783f67f4f05173d025639300a | "2023-06-08T10:29:44Z" | c++ | "2023-09-08T09:50:30Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 77,995 | ["scene/3d/skeleton_3d.cpp"] | BoneMap Overwrite Axis breaks mesh bind poses in converted FBX | ### Godot version
4.1.dev 0a0132c - also present in 4.0 but broken slightly differently
### System information
Godot v4.1.dev (0a0132ccf) - Windows 10.0.19044 - Vulkan (Forward+) - dedicated NVIDIA GeForce RTX 3090 () - Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz (16 Threads)
### Issue description
Selecting a BoneMap during import should have no visual impact on the imported mesh. However, the Overwrite Axis option (selected by default, and required for effective cross-model retargeting) seems to be visually breaking these models.
The attached project has converted FBX files (90 degree rotation, 100 scale) from KenneyNL
Diasbling Overwrite Axis makes the imported model work.
### Steps to reproduce
1. open the attached project
2. note that the two models look different
3. the model not at origin is the same gltf but imported with BoneMap default settings.
### Minimal reproduction project
[fbx_conversion_retarget_broken.zip](https://github.com/godotengine/godot/files/11685705/fbx_conversion_retarget_broken.zip)
| https://github.com/godotengine/godot/issues/77995 | https://github.com/godotengine/godot/pull/78025 | 082cfd501a9f7140ec521cec6d875dfcb57c7f2e | da68c707f2daaf93f484f0c9de3f01430b064f57 | "2023-06-08T08:23:30Z" | c++ | "2023-06-09T09:07:42Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 77,994 | ["modules/mbedtls/crypto_mbedtls.cpp"] | Loading X509CertificateMbedTLS failing at editor/project startup | ### Godot version
72b59325cf7beba7e6e9170cf6023a079fd58672
### System information
Windows 11, 22H2
### Issue description
There is an error emitting at editor and project startup:
Looks like a regression after https://github.com/godotengine/godot/pull/76836, (it's happening on Windows after instalment by default). There should be a way to fix it, either by disable that error or providing a link where that certificate can be downloaded and installed.
### Steps to reproduce
Simply run any project or editor.
### Minimal reproduction project
No need, too easy to reproduce. | https://github.com/godotengine/godot/issues/77994 | https://github.com/godotengine/godot/pull/78716 | 1f9e540f14edbf2d496a1421f8d37e5b483c4c53 | 950f4c3b41acef0403618dd6169be90530dea5da | "2023-06-08T07:58:38Z" | c++ | "2023-06-27T04:58:34Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 77,991 | ["editor/gui/scene_tree_editor.cpp"] | Unselectable nodes when switching filter and no children match the new filter | ### Godot version
4.1.dev (4d5f10fc3a)
### System information
macOS 13.3.0 - Vulkan (Forward+) - integrated Apple M1 Pro - Apple M1 Pro (8 Threads)
### Issue description
When switching filter without coming back to an empty search those that were not matching the search but are included in the new search and none of their child is included will stay greyed out.
### Steps to reproduce
https://github.com/godotengine/godot/assets/66184050/2901bcfe-7834-4f1a-b71f-3b7ad08c3754
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/77991 | https://github.com/godotengine/godot/pull/77992 | 0da20d019eac3bfea5733bab2de14e9bc779a52a | 2c22e56e37e26ec82d771209c23f2adeb9873335 | "2023-06-08T03:33:03Z" | c++ | "2023-06-15T08:50:22Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 77,981 | ["core/io/resource.cpp", "core/io/resource.h", "scene/main/viewport.cpp", "scene/main/viewport.h"] | Indefinitely repeating errors when editing a preloaded scene containing TextureRect and ViewportTexture | ### Godot version
v4.0.3.stable.official [5222a99f5]
### System information
Debian 11 | Vulkan API 1.2.175 - Forward+ - Using Vulkan Device #0: NVIDIA - NVIDIA GeForce GTX 1080 | nvidia-driver 470.182.03-1 from Debian 11 | Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz
### Issue description
Re-opening a scene in the editor that contains a `TextureRect` with a `ViewportTexture` and is `preload`ed in an unopened script produces the following error repeatedly and rapidly (about 200 times per second) until either the scene is closed or the script is also opened.
>ERROR: Texture (binding: 0) should provide one ID referencing a texture (IDs provided: 0).
> at: uniform_set_create (drivers/vulkan/rendering_device_vulkan.cpp:5501)
>ERROR: Condition "!uniform_set" is true.
> at: draw_list_bind_uniform_set (drivers/vulkan/rendering_device_vulkan.cpp:7191)
>ERROR: Uniforms were never supplied for set (3) at the time of drawing, which are required by the pipeline
> at: draw_list_draw (drivers/vulkan/rendering_device_vulkan.cpp:7346)
### Steps to reproduce
1. Open the attached reproduction project
2. Open demo.tscn
3. Close demo.tscn
4. Open demo.tscn again
Do **not** open other.gd beforehand. Doing so will stop the problem from occurring.
### Minimal reproduction project
[reproduce_uniform_not_set.zip](https://github.com/godotengine/godot/files/11681977/reproduce_uniform_not_set.zip)
| https://github.com/godotengine/godot/issues/77981 | https://github.com/godotengine/godot/pull/79201 | 1537452aa94e3fdb19386eac9f8394b391998e8a | e73a4a382ee57c9eb006d7e8f11038643081b12b | "2023-06-07T21:09:47Z" | c++ | "2023-08-17T13:43:38Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 77,978 | ["editor/editor_file_system.cpp"] | Binary mesh resource disappears from MeshInstance3D upon project restart | ### Godot version
v4.1.beta.custom_build [72b59325c]
### System information
Windows 10, Vulkan
### Issue description
This has been plaguing me for a while in my main project. I can only partially reproduce it in a MRP:
When placing a binary mesh resource imported from a .glb file into a MeshInstance3D, it will be gone upon restart. It doesn't seem to happen to every resource, but I've had MIs that were fine and then start doing it out of nowhere.
I thought it was going to be related to materials due to things like https://github.com/godotengine/godot/issues/77007 but I can reliably trigger it with the steps listed below.
I have done limited testing on .tres files because I usually use .res for size/performance, but in the MRP .tres doesn't seem to trigger it, so perhaps it doesn't affect those.
Note on the Steps to reproduce: you have to delete the test_cube.res and redo the steps to trigger the bug again, but in my main project the affected resources always disappear.
### Steps to reproduce
1. Double click test_cube.glb
2. Click the mesh (test_cube_Cube_097), set save to file path as: res://test_cube.res
3. Make sure to click the little folder icon and then Save (this seems to make it consistent to proc)
4. Press Reimport
5. Drag the test_cube.res to the Mesh slot of the MeshInstance3D
6. Ctrl+s to save
7. Project -> Reload current project
8. Mesh is no longer in the MeshInstance3D
### Minimal reproduction project
[binary res bug.zip](https://github.com/godotengine/godot/files/11681009/binary.res.bug.zip) | https://github.com/godotengine/godot/issues/77978 | https://github.com/godotengine/godot/pull/85922 | e1d4b3cc07605a28f182e23d1924b6432fda2186 | e551672c2d7e0b4a484a647d5bf07acf08f59766 | "2023-06-07T19:03:35Z" | c++ | "2023-12-11T23:33:25Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 77,970 | ["drivers/gles3/storage/material_storage.cpp", "editor/shader_globals_editor.cpp", "scene/main/shader_globals_override.cpp", "scene/resources/material.cpp", "servers/rendering/renderer_rd/storage_rd/material_storage.cpp", "servers/rendering/shader_language.cpp", "servers/rendering/storage/variant_converters.h"] | sky.gdshader does not accept vec4 array from gdscript | ### Godot version
4.0.3
### System information
Windows10
### Issue description
sky shader accepts uniform vec3, vec4, vec3 array, but not vec4 array
### Steps to reproduce
simply pass Vector4 to uniform vec4[] from gdscript
### Minimal reproduction project
gdscript
sky_material.set_shader_parameter("v3",Vector3(1,0.8,0.5)) - v
sky_material.set_shader_parameter("v4",Vector4(1,0.8,0.5,0.2)) - v
sky_material.set_shader_parameter("v3a",[Vector3(1,0.8,0.5)]) - v
sky_material.set_shader_parameter("v4a",[Vector4(1,0.8,0.5,0.2)]) - x
sky gdshader
uniform vec3 v3;
uniform vec4 v4;
uniform vec3 v3a[1];
uniform vec4 v4a[1];
void sky(){
COLOR.r = v4a[0].w
} | https://github.com/godotengine/godot/issues/77970 | https://github.com/godotengine/godot/pull/74937 | e188d619227990001667821dac8bc8940076d4a9 | 166643df326724ab439721dcf1eb2d367ae5b744 | "2023-06-07T16:48:44Z" | c++ | "2023-06-09T09:04:16Z" |
closed | godotengine/godot | https://github.com/godotengine/godot | 77,968 | ["platform/windows/detect.py"] | Subsequent builds are slower on Windows + MSVC due to linking | ### Godot version
4.1.dev 72b59325cf7beba7e6e9170cf6023a079fd58672
### System information
Windows 11 22H2 - MSVC 2022 - Intel Core i9-13900K, OS and Godot repository on a fast M.2 SSD
### Issue description
Subsequent builds are slower on Windows + MSVC due to the linking step performing incremental linking, but it's counterproductive for Godot.
This is on MSVC 17.4.3 (version from early 2023):
```
1st build after cleaning all build files
[Time elapsed: 00:00:57.752]
2nd build after modifying scene/main/node.h
[Time elapsed: 00:01:30.433]
3rd build after modifying scene/main/node.h again
[Time elapsed: 00:01:27.513]
4th build after modifying scene/main/node.h yet again
[Time elapsed: 00:01:31.877]
5th build after removing bin/ folder and modifying scene/main/node.h another time
[Time elapsed: 00:00:26.683]
6th build after removing bin/, but not changing any other file
[Time elapsed: 00:00:05.874]
7th build after removing bin, not changing any other file and using progress=no (Windows console I/O is slow :))
[Time elapsed: 00:00:05.559]
```
Updating to MSVC 17.6.2 (latest version available as of June 2023) improves the situation somewhat, but it's still much slower until you remove `bin/`:
```
1st clean build
[Time elapsed: 00:00:51.550]
2nd build after modifying scene/main/node.h
[Time elapsed: 00:00:54.424]
3rd build after modifying scene/main/node.h again
[Time elapsed: 00:00:54.931]
4th build after removing bin and modifying scene/main/node.h again
[Time elapsed: 00:00:23.995]
5th build after removing bin, but not changing any other file
[Time elapsed: 00:00:05.793]
```
This behavior does not occur on Linux with GCC + mold. Link times are consistent regardless of whether the `bin/` folder was previously erased. Removing all `**/*.a` files also does not impact total build time.
### Steps to reproduce
Use `git clean -dfxi` to remove all build files before compiling (only the first time).
- Using MSVC 2022 on Windows, perform a clean build of Godot using it: `scons dev_build=yes scu_build=dev`.
- Modify `scene/main/node.h` to add a line break at the beginning of the file.
- Build again and compare the build time.
- Modify `scene/main/node.h` again to add a line break at the beginning of the file.
- Build again and compare the build time for a second time
- Remove the `bin/` folder, modify `scene/main/node.h` yet another time then build again. Compare the build time.
### Minimal reproduction project
N/A | https://github.com/godotengine/godot/issues/77968 | https://github.com/godotengine/godot/pull/80482 | a22cadfee92c6ee0c70a6168b643b581fb61e07f | 478b80310ccbaaaa028d09d567d504ce8aace5bb | "2023-06-07T16:22:38Z" | c++ | "2023-08-11T08:33:00Z" |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.