The point is that `RayCast`s are checked against objects' `collision_layer`(s), but they themselves are considered no to _belong_ to any layer. Therefore, the correct name for their property is `collision_mask`, rather than `collision_layer`.
Only renaming is needed since the behavior was already the right one, only that it wasn't matching what users would expect from the name and description of the property.
Fixes#7589, where it's also discussed.
Rename user facing methods and variables as well as the corresponding
C++ methods according to the folloming changes:
* pos -> position
* rot -> rotation
* loc -> location
C++ variables are left as is.
After discussing with @reduz and @akien-mga it was decided that we do
not allow assignments or declarations in if statements. This PR removes
the instances of this I could find by automated means.
Fix 3D joint handling of collision exceptions, so that they lose effect when the joint is not valid in every case; also some redundant code removed.
Also avoid trying to create the joint when not neither body A nor body B are set.
Make 2D joints be handled like their 3D counterparts, which adds the fixes to them while also removing duplicated code.
(adapted from commit 4a98a59aa6)
- Makes vararg methods automatically use PROPERTY_USAGE_NIL_IS_VARIANT on return types
- Completely removes the ":type" suffix for method names. Virtual methods must use the MethodInfo constructors that takes Variant::Type or PropertyHint as the first parameter for the return type (with CLASS_INFO as a helper to get the PropertyInfo). Parameters must use PROPERTY_HINT_RESOURCE_TYPE and hint string.
- PROPERTY_USAGE_NIL_IS_VARIANT is no longer needed for parameters, because parameters cannot be void.
- Adds missing PROPERTY_USAGE_NIL_IS_VARIANT to virtual and built-in methods that return Variant.
Currently the shadow gradient can't be disabled and always has a minimum value of 1.
I'm not sure why though, the default value seems to be 0 and the rasterizer checks for 0 [here](8fc6bb8f77/drivers/gles3/rasterizer_canvas_gles3.cpp (L1284)) before enabling it.
Feels like a bug to me, but if there's a reason for that let me know.
As a side effect, this edit effectively changes the default value from 1 to 0 for new projects.
Currently we rely on some undefined behavior when Object->cast_to() gets
called with a Null pointer. This used to work fine with GCC < 6 but
newer versions of GCC remove all codepaths in which the this pointer is
Null. However, the non-static cast_to() was supposed to be null safe.
This patch makes cast_to() Null safe and removes the now redundant Null
checks where they existed.
It is explained in this article: https://www.viva64.com/en/b/0226/
The change in #10524 subtly changes the behavior of set_motion_scale()
and set_motion_offset() if the ParallaxLayer does not have a parent
node. Previously it would still set the corresponding property, but
after this change the property change would be discarded.
I'm not entirely sure if this actually matters as there doesn't appear
to be any code that picks up this change if the ParallaxLayer gets
re-parented later, but it's better to not change behavior regardless.
