This PR and commit adds a new IK system for 3D with the Skeleton3D node
that adds several new IK solvers, as well as additional changes and functionality
for making bone manipulation in Godot easier.
This work was sponsored by GSoC 2020 and TwistedTwigleg
Full list of changes:
* Adds a SkeletonModification3D resource
* This resource is the base where all IK code is written and executed
* Adds a SkeletonModificationStack3D resource
* This node oversees the execution of the modifications and acts as a bridge of sorts for the modifications to the Skeleton3D node
* Adds SkeletonModification3D resources for LookAt, CCDIK, FABRIK, Jiggle, and TwoBoneIK
* Each modification is in it's own file
* Several changes to Skeletons, listed below:
* Added local_pose_override, which acts just like global_pose_override but keeps bone-child relationships intract
* So if you move a bone using local_pose_override, all of the bones that are children will also be moved. This is different than global_pose_override, which only affects the individual bone
* Internally bones keep track of their children. This removes the need of a processing list, makes it possible to update just a few select bones at a time, and makes it easier to traverse down the bone chain
* Additional functions added for converting from world transform to global poses, global poses to local poses, and all the same changes but backwards (local to global, global to world). This makes it much easier to work with bone transforms without needing to think too much about how to convert them.
* New signal added, bone_pose_changed, that can be used to tell if a specific bone changed its transform. Needed for BoneAttachment3D
* Added functions for getting the forward position of a bone
* BoneAttachment3D node refactored heavily
* BoneAttachment3D node is now completely standalone in its functionality.
* This makes the code easier and less interconnected, as well as allowing them to function properly without being direct children of Skeleton3D nodes
* BoneAttachment3D now can be set either using the index or the bone name.
* BoneAttachment3D nodes can now set the bone transform instead of just following it. This is disabled by default for compatibility
* BoneAttachment3D now shows a warning when not configured correctly
* Added rotate_to_align function in Basis
* Added class reference documentation for all changes
The Transform::xform and xform_inv are made safe for Planes when using non-uniform scaling.
Optimization of calling sites to prevent loss of performance from the changes to xform(Plane).
Found via `codespell -q 3 -S ./thirdparty,*.po,./DONORS.md -L ackward,ang,ans,ba,beng,cas,childs,childrens,dof,doubleclick,fave,findn,hist,inout,leapyear,lod,nd,numer,ois,ony,paket,seeked,sinc,switchs,te,uint`
* Clean-up of node_3d_editor_plugin.{h,cpp}: removed unused code, fixed some bugs.
* Moved node_3d_editor_gizmos.{h,cpp} to editor/plugins.
* Added support for multiple gizmos per node. This means custom gizmos will no longer override the built-in ones and that multiple gizmos can be used in more complex nodes.
* Added support for handle IDs. When adding handles to a gizmo, an ID can be specified for each one, making it easier to work with gizmos that have a variable number of handles.
* Added support for subgizmos, selectable elements that can be transformed without needing a node of their own. By overriding _subgizmo_intersect_frustum() and/or _subgizmo_intersect_ray() gizmos can define which subgizmos should be selected on a region or click selection. Subgizmo transformations are applied using get/set/commit virtual methods, similar to how handles work.
The `Math_INF` and `Math_NAN` defines were just aliases for those
constants, so we might as well use them directly.
Some portions of the code were already using `INFINITY` directly.
The order of numbers is not changed except for Transform2D. All logic is done inside of their structures (and not in Variant).
For the number of decimals printed, they now use String::num_real which works best with real_t, except for Color which is fixed at 4 decimals (this is a reliable number of float digits when converting from 16-bpc so it seems like a good choice)
This PR and commit adds a new IK system for 2D with the Skeleton2D node
that adds several new IK solvers, a way to control bones in a Skeleton2D
node similar to that in Skeleton3D. It also adds additional changes
and functionality.
This work was sponsored by GSoC 2020 and TwistedTwigleg.
Full list of changes:
* Adds a SkeletonModifier2D resource
* This resource is the base where all IK code is written and executed
* Has a function for clamping angles, since it is so commonly used
* Modifiers are unique when duplicated so it works with instancing
* Adds a SkeletonModifierStack2D resource
* This resource manages a series of SkeletonModification2Ds
* This is what the Skeleton2D directly interfaces with to make IK possible
* Adds SkeletonModifier2D resources for LookAt, CCDIK, FABRIK, Jiggle, and TwoBoneIK
* Each modification is in its own file
* There is also a SkeletonModifier2D resource that acts as a stack for using multiple stacks together
* Adds a PhysicalBone2D node
* Works similar to the PhysicalBone3D node, but uses a RigidBody2D node
* Changes to Skeleton2D listed below:
* Skeleton2D now holds a single SkeletonModificationStack2D for IK
* Skeleton2D now has a local_pose_override, which overrides the Bone2D position similar to how the overrides work in Skeleton3D
* Changes to Bone2D listed below:
* The default_length property has been changed to length. Length is the length of the bone to its child bone node
* New bone_angle property, which is the angle the bone has to its first child bone node
* Bone2D caches its transform when not modified by IK for IK interpolation purposes
* Bone2D draws its own editor gizmo, though this is stated to change in the future
* Changes to CanvasItemEditor listed below:
* Bone2D gizmo drawing code removed
* The 2D IK code is removed. Now Bone2D is the only bone system for 2D
* Transform2D now has a looking_at function for rotating to face a position
* Two new node notifications: NOTIFICATION_EDITOR_PRE_SAVE and NOTIFICATION_EDITOR_POST_SAVE
* These notifications only are called in the editor right before and after saving a scene
* Needed for not saving the IK position when executing IK in the editor
* Documentation for all the changes listed above.
Changes passing of current_tree from a member variable to a function argument, making bugs due to stale state less likely.
Fix a bug in deactivate where current_tree variable was stale. This may have resulted in visual anomalies.
The code is based on the current version of thirdparty/vhacd and modified to use Godot's types and code style.
Additional changes:
- extended PagedAllocator to allow leaked objects
- applied patch from https://github.com/bulletphysics/bullet3/pull/3037
Port lawnjelly's dynamic BVH implementation from 3.x to be used in
both 2D and 3D broadphases.
Removed alternative broadphase implementations which are not meant to be
used anymore since they are much slower.
Includes changes in Rect2, Vector2, Vector3 that help with the template
implementation of the dynamic BVH by uniformizing the interface between
2D and 3D math.
Co-authored-by: lawnjelly <lawnjelly@gmail.com>
- Move most properties from TileMap to TileSet,
- Make TileSet more flexible, supporting more feature (several
collision layers, etc...),
- Fusion both the TileMap and TileSet editor,
- Implement TileSetSources, and thus a new way to index tiles in the TileSet,
- Rework the TileSet and TileMap editors completely,
- Implement an editor zoom widget (and use it in several places)
Fixes#48420, fixes#48421.
The binding was missed when moving GDScript built-in to Global Scope it seems.
Co-authored-by: kleonc <9283098+kleonc@users.noreply.github.com>
This returns a double while the other values are all uint64_t. The
clang compiler gives a warning since converting the constant to double
loses precision.
We've been using standard C library functions `memcpy`/`memset` for these since
2016 with 67f65f6639.
There was still the possibility for third-party platform ports to override the
definitions with a custom header, but this doesn't seem useful anymore.
Added an occlusion culling system with support for static occluder meshes.
It can be enabled via `Project Settings > Rendering > Occlusion Culling > Use Occlusion Culling`.
Occluders are defined via the new `Occluder3D` resource and instanced using the new
`OccluderInstance3D` node. The occluders can also be automatically baked from a
scene using the built-in editor plugin.
Cylinder collision detection uses a mix of SAT and GJKEPA.
GJKEPA is used to find the best separation axis in cases where finding
it analytically is too complex.
Changes in SAT solver:
Added support for generating separation axes for cylinder shape.
Added support for generating contact points with circle feature.
Changes in GJKEPA solver:
Updated from latest Bullet version which includes EPA fixes in some
scenarios.
Setting a lower EPA_ACCURACY to fix accuracy problems with cylinder vs.
cylinder in some cases.
Channels that are inactive -or when playback has not started yet- will report -200 dB as their peak value (which is also the lowest value possible during playback).
Named color constants renamed to UPPERCASE. Unlike #41019, this PR
is complete and implements these changes in the simplest way possible.
Co-authored-by: Shivam Mukherjee <mshivam98@gmail.com>
Happy new year to the wonderful Godot community!
2020 has been a tough year for most of us personally, but a good year for
Godot development nonetheless with a huge amount of work done towards Godot
4.0 and great improvements backported to the long-lived 3.2 branch.
We've had close to 400 contributors to engine code this year, authoring near
7,000 commit! (And that's only for the `master` branch and for the engine code,
there's a lot more when counting docs, demos and other first-party repos.)
Here's to a great year 2021 for all Godot users 🎆
The calls to the BVH need to use the world space AABB, rather than local space for it to work. Also, update was not being called which is required to update the AABB as objects move.
-Much greater pairing/unpairing performance
-For now, using it for culling too, but this will change in a couple of days.
-Added a paged allocator, to efficiently alloc/free some types of objects.
-Based on Bullet Dbvh, has style and functional changes.
-Provides efficient pairing
-Needed to optimize rendering
-Needed to optimize physics
This PR is up for others to review the implementation.
-Happens on import by default for all models
-Just works (tm)
-Biasing can be later adjusted per node or per viewport (as well as globally)
-Disabled AABB.get_support test because its broken
- added `state` as a property to restore internal state of RNG;
- `get_seed()` returns last seed used to initialize the state rather than the current state.
Co-authored-by: MidZik <matt.idzik1@gmail.com>
* Using C-style function pointers now, InternalMethod is gone.
* This ensures much better performance in typed code.
* Renamed builtin_funcs to utility_funcs, to avoid naming confusion
-Moved Expression to use this, removed its own.
-Eventually GDScript/VisualScript/GDNative need to be moved to this.
-Given the JSON functions were hacked-in, removed them and created a new JSONParser class
-Made sure these functions appear properly in documentation, since they will be removed from GDScript
-Discern between named, indexed and keyed
-Get direct access to functions for typed GDScript and GDNative bindings
-Small changes to some classes in order to work with the new setget binder
It's non-deterministic so it's better to show a fixed value like 0 instead of
having it potentially change whenever `randomize()` is called.
Fixes#43317.
Also separated Light2D in PointLight2D and DirectionalLight2D.
Used PointLight2D because its more of a point, and it does not work
the same as OmniLight (as shape depends on texture).
Added a few utility methods to Rect2D I needed.
Moved to a system using variadic templates, shared with CallableBind.
New code is cleaner, faster and allows for much better optimization of core
type functions from GDScript and GDNative.
Added Variant::InternalMethod function for direct call access.
is_equal_approx is able to handle values of any size, and is_equal_approx_ratio is no longer used in any exposed APIs, so we don't need is_equal_approx_ratio anymore. Also, add #ifdef MATH_CHECKS for a method that is only used when MATH_CHECKS is defined.
Add __NetBSD__ to `platform_config.h` so that it can find `alloca`
and use the proper `pthread_setname_np` format.
Rename RANDOM_MAX to avoid conflict with NetBSD stdlib.
Fixes#42145.
The input to smoothstep is not actually a weight, and the decscription
of smoothstep was pretty hard to understand and easy to misinterpret.
Clarified what it means to be approximately equal.
nearest_po2 does not do what the descriptions says it does. For one,
it returns the same power if the input is a power of 2. Second, it
returns 0 if the input is negative or 0, while the smallest possible
integral power of 2 actually is 1 (2^0 = 1). Due to the implementation
and how it is used in a lot of places, it does not seem wise to change
such a core function however, and I decided it is better to alter the
description of the built-in.
Added a few examples/clarifications/edge-cases.
- Fixed floating point issue on the old one.
- Fixed the equation on the get_euler_yxz function.
- Added unit tests.
This work has been kindly sponsored by IMVU.
Semicolons are not necessary after function definitions or control flow
blocks, and having some code use them makes things inconsistent (and
occasionally can mess up `clang-format`'s formatting).
Removing them is tedious work though, I had to do this manually (regex
+ manual review) as I couldn't find a tool for that. All other code
folders would need to get the same treatment.
I couldn't find a tool that enforces it, so I went the manual route:
```
find -name "thirdparty" -prune \
-o -name "*.cpp" -o -name "*.h" -o -name "*.m" -o -name "*.mm" \
-o -name "*.glsl" > files
perl -0777 -pi -e 's/\n}\n([^#])/\n}\n\n\1/g' $(cat files)
misc/scripts/fix_style.sh -c
```
This adds a newline after all `}` on the first column, unless they
are followed by `#` (typically `#endif`). This leads to having lots
of places with two lines between function/class definitions, but
clang-format then fixes it as we enforce max one line of separation.
This doesn't fix potential occurrences of function definitions which
are indented (e.g. for a helper class defined in a .cpp), but it's
better than nothing. Also can't be made to run easily on CI/hooks so
we'll have to be careful with new code.
Part of #33027.
Which means that reduz' beloved style which we all became used to
will now be changed automatically to remove the first empty line.
This makes us lean closer to 1TBS (the one true brace style) instead
of hybridating it with some Allman-inspired spacing.
There's still the case of braces around single-statement blocks that
needs to be addressed (but clang-format can't help with that, but
clang-tidy may if we agree about it).
Part of #33027.
Using `clang-tidy`'s `modernize-use-default-member-init` check and
manual review of the changes, and some extra manual changes that
`clang-tidy` failed to do.
Also went manually through all of `core` to find occurrences that
`clang-tidy` couldn't handle, especially all initializations done
in a constructor without using initializer lists.
-Added LocalVector (needed it)
-Added stb_rect_pack (It's pretty cool, we could probably use it for other stuff too)
-Fixes and changes all around the place
-Added library for 128 bits fixed point (required for Delaunay3D)
This reverts commit ec7b481170.
This was wrong, `d` is not a distance but the `d` constant in the
parametric equation `ax + by + cz = d` describing the plane.
Part of #33027, also discussed in #29848.
Enforcing the use of brackets even on single line statements would be
preferred, but `clang-format` doesn't have this functionality yet.
Skew is x-axis only, because it must be bidirectionally convertible to a 2x3 matrix, but you can subtract it to the rotation to get the effect on y-axis
This fixes numerous false positives coming out of the culling system.
AABB checks are now a full separating-axis check against the frustum, with the points of the frustum being compared to the planes of the box just as the points of the box were being compared to the planes of the frustum. This fixes large objects behind the camera not being culled correctly.
Some systems that used frustums that were (sometimes mistakenly?) unbounded on one or more side have been modified to be fully enclosed.
Configured for a max line length of 120 characters.
psf/black is very opinionated and purposely doesn't leave much room for
configuration. The output is mostly OK so that should be fine for us,
but some things worth noting:
- Manually wrapped strings will be reflowed, so by using a line length
of 120 for the sake of preserving readability for our long command
calls, it also means that some manually wrapped strings are back on
the same line and should be manually merged again.
- Code generators using string concatenation extensively look awful,
since black puts each operand on a single line. We need to refactor
these generators to use more pythonic string formatting, for which
many options are available (`%`, `format` or f-strings).
- CI checks and a pre-commit hook will be added to ensure that future
buildsystem changes are well-formatted.
For us, it practically only changes the fact that `A<A<int>>` is now
used instead of the C++03 compatible `A<A<int> >`.
Note: clang-format 10+ changed the `Standard` arguments to fully
specified `c++11`, `c++14`, etc. versions, but we can't use `c++17`
now if we want to preserve compatibility with clang-format 8 and 9.
`Cpp11` is still supported as deprecated alias for `Latest`.
We now require a compiler with C++17 support, so we don't need to
check for features added to GCC 5 or Clang 3.2.
Clang builtin availability checks were unused anyway as Clang defines
`__GNUC__` as it's also a GNU C implementation.
Fixes#36986.
- Renames PackedIntArray to PackedInt32Array.
- Renames PackedFloatArray to PackedFloat32Array.
- Adds PackedInt64Array and PackedFloat64Array.
- Renames Variant::REAL to Variant::FLOAT for consistency.
Packed arrays are for storing large amount of data and creating stuff like
meshes, buffers. textures, etc. Forcing them to be 64 is a huge waste of
memory. That said, many users requested the ability to have 64 bits packed
arrays for their games, so this is just an optional added type.
For Variant, the float datatype is always 64 bits, and exposed as `float`.
We still have `real_t` which is the datatype that can change from 32 to 64
bits depending on a compile flag (not entirely working right now, but that's
the idea). It affects math related datatypes and code only.
Neither Variant nor PackedArray make use of real_t, which is only intended
for math precision, so the term is removed from there to keep only float.
WARNING: Requires C++17 'guaranteed copy elision' to fix ambiguous
operator problems in Variant.
This was added for this commit (and future C++17 uses) in #36457.
Fixes#26637.
Fixes#19900.
The viewport_size returned by get_viewport_size was previously incorrect, being half the correct value. The function is renamed to get_viewport_half_extents, and now returns a Vector2.
Code which called this function has also been modified accordingly.
This PR also fixes shadow culling when using ortho cameras, because the correct input for CameraMatrix::set_orthogonal should be the full HEIGHT from get_viewport_half_extents, and not half the width.
It also fixes state.ubo_data.viewport_size in rasterizer_scene_gles3.cpp to be the width and the height of the viewport in pixels as stated in the documentation, rather than the current value which is half the viewport extents in worldspace, presumed to be a bug.
Happy new year to the wonderful Godot community!
We're starting a new decade with a well-established, non-profit, free
and open source game engine, and tons of further improvements in the
pipeline from hundreds of contributors.
Godot will keep getting better, and we're looking forward to all the
games that the community will keep developing and releasing with it.
Underscaled arc tolerance produced very small values so that changes
to this parameter were negligible when scaled internally, hence significant
performance drop (lots of intermediate points inserted in an arc). Now the
performance is mostly the same compared to other types of offsetting
(SQUARE, MITER).
s * edge0 = -d / a * edge0 = -edge0⋅v0 / (edge0⋅edge0) * edge0 = vector projection of -v0 onto edge0
By incorrectly using -e/c instead of -d/a, Face3::get_closest_point_to was returning the wrong point in certain cases. Specifically, I noticed it returning vertex[0] when it should have been returning vertex[1].
Previously, disabled points will not be considered when performing
get_closest_point. This commit changes that by introducing an additional
flag for this behavior. Related issue: #31814
Similarly to `Vector2` and `Rect2` transforms in 2D and Vector3, Plane,
and AABB in 3D. PoolVector2Array and PoolVector3Array were the only
missing Variant types in both Transform2D and Transform respectively.
Condensed some if and ERR statements. Added dots to end of error messages
Couldn't figure out EXPLAINC. These files gave me trouble: core/error_macros.h, core/io/file_access_buffered_fa.h (where is it?),
core/os/memory.cpp,
drivers/png/png_driver_common.cpp,
drivers/xaudio2/audio_driver_xaudio2.cpp (where is it?)
Also allow lifting the decimal step formatting with a hint range step
of 0. A new `range_step_decimals()` is added for this to avoid breaking
compatibility on the general purpose `step_decimals()` (which still
returns 0 for an input step of 0).
Supersedes #25470.
Partial fix for #18251.
"posmod" is the integer version of "fposmod". We do not need a "mod" because of the % operator.
I changed the default arg names from "x" and "y" to "a" and "b" because they are not coordinates. I also changed pow's arg names to "base" and "exp". Also, I reorganized the code in the VS built-in funcs switch statement.
By combining all scalar factors we can get rid of a scalar * vector
multiplication and a square root operation, since the resulting formula
only uses the squared length.
For clarity, assign-to-release idiom for PoolVector::Read/Write
replaced with a function call.
Existing uses replaced (or removed if already handled by scope)
It's the recommended way to set those, and is more portable
(automatically prepends -D for GCC/Clang and /D for MSVC).
We still use CPPFLAGS for some pre-processor flags which are not
defines.
Godot core needs MD5/SHA256/AES/Base64 which used to be provided by
separate libraries.
Since we bundle mbedtls in most cases, and we can easily only include
the needed sources if we so desire, let's use it.
To simplify library changes in the future, and better isolate header
dependencies all functions have been wrapped around inside a class in
`core/math/crypto_base.h`.
If the mbedtls module is disabled, we only bundle the needed source
files independently of the `builtin_mbedtls` option.
If the module is enabled, the `builtin_mbedtls` option works as usual.
Also remove some unused headers from StreamPeerMbedTLS which were
causing build issues.
Can be used via scripting as `Geometry.triangulate_delaunay_2d(points)`
The interface is the same as in `Triangulate` library, returning indices
into triangulated points.
Clipper 6.4.2 is used internally to perform polypaths clipping, as well
as inflating/deflating polypaths. The following methods were added:
```
Geometry.merge_polygons_2d(poly_a, poly_b) # union
Geometry.clip_polygons_2d(poly_a, poly_b) # difference
Geometry.intersect_polygons_2d(poly_a, poly_b) # intersection
Geometry.exclude_polygons_2d(poly_a, poly_b) # xor
Geometry.clip_polyline_with_polygon_2d(poly_a, poly_b)
Geometry.intersect_polyline_with_polygon_2d(poly_a, poly_b)
Geometry.offset_polygon_2d(polygon, delta) # inflate/deflate
Geometry.offset_polyline_2d(polyline, delta) # returns polygons
// This one helps to implement CSG-like behaviour:
Geometry.transform_points_2d(points, transform)
```
All the methods return an array of polygons/polylines. The resulting
polygons could possibly be holes which could be checked with
`Geometry.is_polygon_clockwise()` which was exposed to scripting as well.
Made AudioFrame and Vector2 equivalent for casting.
Added ability to obtain the playback object from stream players.
Added ability to obtain effect instance from audio server.
It seems to stay compatible with formatting done by clang-format 6.0 and 7.0,
so contributors can keep using those versions for now (they will not undo those
changes).
Network peers get_var/put_var
File get_var/store_var
GDScript/Mono/VisualScript bytes2var/var2bytes
Add MultiplayerAPI.allow_object_decoding member which deprecates PacketPeer.allow_object_decoding.
Break ABI compatibaility (API compatibility for GDNative).
Replace inline with _FORCE_INLINE_ in short methods.
Remove unused and redundant method no_area() as we already have has_no_area().
Add grow_individual() grow_margin() and expand() to Rect2i.
When generating single precision floats, Godot casts a uint32_t to float,
causing uniformity loss.
This new randf, inspired by T. R. Campbell's random_real, samples the output
of rand as the fraction part of an infinite binary number, with some tricks
to reduce ops and branching. This method provides "good enough" uniformity at
decent speed, for floats greater than 2^-64. Smaller numbers are floored to 0.
