gstaichi 2.1.1rc3__cp310-cp310-macosx_11_0_arm64.whl

gstaichi/math/mathimpl.py

@@ -0,0 +1,886 @@
+# type: ignore
+
+# pylint: disable=W0622
+"""
+Math functions for glsl-like functions and other stuff.
+"""
+import math
+
+from gstaichi.lang import impl, ops
+from gstaichi.lang.impl import static, zero
+from gstaichi.lang.kernel_impl import func
+from gstaichi.lang.matrix import Matrix
+from gstaichi.lang.ops import (
+    acos,
+    asin,
+    atan2,
+    ceil,
+    cos,
+    exp,
+    floor,
+    log,
+    max,
+    min,
+    pow,
+    round,
+    sin,
+    sqrt,
+    tan,
+    tanh,
+)
+from gstaichi.types import matrix, template, vector
+from gstaichi.types.primitive_types import f64, u32, u64
+
+cfg = impl.default_cfg
+
+e = math.e
+"""The mathematical constant e = 2.718281….
+Directly imported from the Python standard library `math`.
+"""
+
+pi = math.pi
+"""The mathematical constant π = 3.141592….
+Directly imported from the Python standard library `math`.
+"""
+
+inf = math.inf
+"""A floating-point positive infinity. (For negative infinity, use `-inf`).
+Directly imported from the Python standard library `math`.
+"""
+
+nan = math.nan
+"""A floating-point "not a number" (NaN) value.
+Directly imported from the Python standard library `math`
+"""
+
+vec2 = vector(2, cfg().default_fp)
+"""2D floating vector type.
+"""
+
+vec3 = vector(3, cfg().default_fp)
+"""3D floating vector type.
+"""
+
+vec4 = vector(4, cfg().default_fp)
+"""4D floating vector type.
+"""
+
+ivec2 = vector(2, cfg().default_ip)
+"""2D signed int vector type.
+"""
+
+ivec3 = vector(3, cfg().default_ip)
+"""3D signed int vector type.
+"""
+
+ivec4 = vector(4, cfg().default_ip)
+"""3D signed int vector type.
+"""
+
+uvec2 = vector(2, cfg().default_up)
+"""2D unsigned int vector type.
+"""
+
+uvec3 = vector(3, cfg().default_up)
+"""3D unsigned int vector type.
+"""
+
+uvec4 = vector(4, cfg().default_up)
+"""4D unsigned int vector type.
+"""
+
+mat2 = matrix(2, 2, cfg().default_fp)
+"""2x2 floating matrix type.
+"""
+
+mat3 = matrix(3, 3, cfg().default_fp)
+"""3x3 floating matrix type.
+"""
+
+mat4 = matrix(4, 4, cfg().default_fp)
+"""4x4 floating matrix type.
+"""
+
+
+@func
+def mix(x, y, a):
+    """Performs a linear interpolation between `x` and `y` using
+    `a` to weight between them. The return value is computed as
+    `x * (1 - a) + a * y`.
+
+    The arguments can be scalars or :class:`~gstaichi.Matrix`,
+    as long as the operation can be performed.
+
+    This function is similar to the `mix` function in GLSL.
+
+    Args:
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specify
+            the start of the range in which to interpolate.
+        y (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specify
+            the end of the range in which to interpolate.
+        a (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specify
+            the weight to use to interpolate between x and y.
+
+    Returns:
+        (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): The linear
+            interpolation of `x` and `y` by weight `a`.
+
+    Example::
+
+        >>> x = ti.Vector([1, 1, 1])
+        >>> y = ti.Vector([2, 2, 2])
+        >>> a = ti.Vector([1, 0, 0])
+        >>> ti.math.mix(x, y, a)
+        [2.000000, 1.000000, 1.000000]
+        >>> x = ti.Matrix([[1, 2], [2, 3]], ti.f32)
+        >>> y = ti.Matrix([[3, 5], [4, 5]], ti.f32)
+        >>> a = 0.5
+        >>> ti.math.mix(x, y, a)
+        [[2.000000, 3.500000], [3.000000, 4.000000]]
+    """
+    return x * (1.0 - a) + y * a
+
+
+@func
+def clamp(x, xmin, xmax):
+    """Constrain a value to lie between two further values, element-wise.
+    The returned value is computed as `min(max(x, xmin), xmax)`.
+
+    The arguments can be scalars or :class:`~gstaichi.Matrix`,
+    as long as they can be broadcasted to a common shape.
+
+    Args:
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specify
+            the value to constrain.
+        y (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specify
+            the lower end of the range into which to constrain `x`.
+        a (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specify
+            the upper end of the range into which to constrain `x`.
+
+    Returns:
+        The value of `x` constrained to the range `xmin` to `xmax`.
+
+    Example::
+
+        >>> v = ti.Vector([0, 0.5, 1.0, 1.5])
+        >>> ti.math.clamp(v, 0.5, 1.0)
+        [0.500000, 0.500000, 1.000000, 1.000000]
+        >>> x = ti.Matrix([[0, 1], [-2, 2]], ti.f32)
+        >>> y = ti.Matrix([[1, 2], [1, 2]], ti.f32)
+        >>> ti.math.clamp(x, 0.5, y)
+        [[0.500000, 1.000000], [0.500000, 2.000000]]
+    """
+    return ops.min(xmax, ops.max(xmin, x))
+
+
+@func
+def step(edge, x):
+    """Generate a step function by comparing two values, element-wise.
+
+    `step` generates a step function by comparing `x` to edge.
+    For element i of the return value, 0.0 is returned if x[i] < edge[i],
+    and 1.0 is returned otherwise.
+
+    The two arguments can be scalars or :class:`~gstaichi.Matrix`,
+    as long as they can be broadcasted to a common shape.
+
+    Args:
+        edge (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specify
+            the location of the edge of the step function.
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specify
+            the value to be used to generate the step function.
+
+    Returns:
+        The return value is computed as `x >= edge`, with type promoted.
+
+    Example::
+
+        >>> x = ti.Matrix([[0, 1], [2, 3]], ti.f32)
+        >>> y = 1
+        >>> ti.math.step(x, y)
+        [[1.000000, 1.000000], [0.000000, 0.000000]]
+    """
+    return ops.cast(x >= edge, float)
+
+
+@func
+def fract(x):
+    """Compute the fractional part of the argument, element-wise.
+    It's equivalent to `x - ti.floor(x)`.
+
+    Args:
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): The
+            input value.
+
+    Returns:
+        The fractional part of `x`.
+
+    Example::
+
+        >>> x = ti.Vector([-1.2, -0.7, 0.3, 1.2])
+        >>> ti.math.fract(x)
+        [0.800000, 0.300000, 0.300000, 0.200000]
+    """
+    return x - ops.floor(x)
+
+
+@func
+def smoothstep(edge0, edge1, x):
+    """Performs smooth Hermite interpolation between 0 and 1 when
+    `edge0 < x < edge1`, element-wise.
+
+    The arguments can be scalars or :class:`~gstaichi.Matrix`,
+    as long as they can be broadcasted to a common shape.
+
+    This function is equivalent to the `smoothstep` in GLSL.
+
+    Args:
+        edge0 (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specifies
+            the value of the lower edge of the Hermite function.
+        edge1 (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specifies
+            the value of the upper edge of the Hermite function.
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): Specifies
+            the source value for interpolation.
+
+    Returns:
+        The smoothly interpolated value.
+
+    Example::
+
+        >>> edge0 = ti.Vector([0, 1, 2])
+        >>> edge1 = 1
+        >>> x = ti.Vector([0.5, 1.5, 2.5])
+        >>> ti.math.smoothstep(edge0, edge1, x)
+        [0.500000, 1.000000, 0.000000]
+    """
+    t = clamp((x - edge0) / (edge1 - edge0), 0.0, 1.0)
+    return t * t * (3.0 - 2.0 * t)
+
+
+@func
+def sign(x):
+    """Extract the sign of the parameter, element-wise.
+
+    Args:
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): The
+            input value.
+
+    Returns:
+        -1.0 if `x` is less than 0.0, 0.0 if `x` is equal to 0.0,
+        and +1.0 if `x` is greater than 0.0.
+
+    Example::
+
+        >>> x = ti.Vector([-1.0, 0.0, 1.0])
+        >>> ti.math.sign(x)
+        [-1.000000, 0.000000, 1.000000]
+    """
+    return ops.cast((x >= 0.0), float) - ops.cast((x <= 0.0), float)
+
+
+@func
+def normalize(v):
+    """Calculates the unit vector in the same direction as the
+    original vector `v`.
+
+    It's equivalent to the `normalize` function is GLSL.
+
+    Args:
+        x (:class:`~gstaichi.Matrix`): The vector to normalize.
+
+    Returns:
+        The normalized vector :math:`v/|v|`.
+
+    Example::
+
+        >>> v = ti.Vector([1, 2, 3])
+        >>> ti.math.normalize(v)
+        [0.267261, 0.534522, 0.801784]
+    """
+    return v / v.norm()
+
+
+@func
+def log2(x):
+    """Return the base 2 logarithm of `x`, so that if :math:`2^y=x`,
+    then :math:`y=\\log2(x)`.
+
+    This is equivalent to the `log2` function is GLSL.
+
+    Args:
+        x (:class:`~gstaichi.Matrix`): The input value.
+
+    Returns:
+        The base 2 logarithm of `x`.
+
+    Example::
+
+        >>> x = ti.Vector([1., 2., 3.])
+        >>> ti.math.log2(x)
+        [0.000000, 1.000000, 1.584962]
+    """
+    return ops.log(x) / static(ops.log(2.0))
+
+
+@func
+def reflect(x, n):
+    """Calculate the reflection direction for an incident vector.
+
+    For a given incident vector `x` and surface normal `n` this
+    function returns the reflection direction calculated as
+    :math:`x - 2.0 * dot(x, n) * n`.
+
+    This is equivalent to the `reflect` function is GLSL.
+
+    `n` should be normalized in order to achieve the desired result.
+
+    Args:
+        x (:class:`~gstaichi.Matrix`): The incident vector.
+        n (:class:`~gstaichi.Matrix`): The normal vector.
+
+    Returns:
+        The reflected vector.
+
+    Example::
+
+        >>> x = ti.Vector([1., 2., 3.])
+        >>> n = ti.Vector([0., 1., 0.])
+        >>> ti.math.reflect(x, n)
+        [1.000000, -2.000000, 3.000000]
+    """
+    k = x.dot(n)
+    return x - 2.0 * k * n
+
+
+@func
+def degrees(x):
+    """Convert `x` in radians to degrees, element-wise.
+
+    Args:
+        x (:class:`~gstaichi.Matrix`): The input angle in radians.
+
+    Returns:
+        angle in degrees.
+
+    Example::
+
+        >>> x = ti.Vector([-pi/2, pi/2])
+        >>> ti.math.degrees(x)
+        [-90.000000, 90.000000]
+    """
+    return x * static(180.0 / pi)
+
+
+@func
+def radians(x):
+    """Convert `x` in degrees to radians, element-wise.
+
+    Args:
+        x (:class:`~gstaichi.Matrix`): The input angle in degrees.
+
+    Returns:
+        angle in radians.
+
+    Example::
+
+        >>> x = ti.Vector([-90., 45., 90.])
+        >>> ti.math.radians(x) / pi
+        [-0.500000, 0.250000, 0.500000]
+    """
+    return x * static(pi / 180.0)
+
+
+@func
+def distance(x, y):
+    """Calculate the distance between two points.
+
+    This function is equivalent to the `distance` function is GLSL.
+
+    Args:
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): The first input point.
+        y (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): The second input point.
+
+    Returns:
+        The distance between the two points.
+
+    Example::
+
+        >>> x = ti.Vector([0, 0, 0])
+        >>> y = ti.Vector([1, 1, 1])
+        >>> ti.math.distance(x, y)
+        1.732051
+    """
+    return (x - y).norm()
+
+
+@func
+def refract(x, n, eta):
+    """Calculate the refraction direction for an incident vector.
+
+    This function is equivalent to the `refract` function in GLSL.
+
+    Args:
+        x (:class:`~gstaichi.Matrix`): The incident vector.
+        n (:class:`~gstaichi.Matrix`): The normal vector.
+        eta (float): The ratio of indices of refraction.
+
+    Returns:
+        :class:`~gstaichi.Matrix`: The refraction direction vector.
+
+    Example::
+
+        >>> x = ti.Vector([1., 1., 1.])
+        >>> y = ti.Vector([0, 1., 0])
+        >>> ti.math.refract(x, y, 2.0)
+        [2.000000, -1.000000, 2.000000]
+    """
+    dxn = x.dot(n)
+    result = zero(x)
+    k = 1.0 - eta * eta * (1.0 - dxn * dxn)
+    if k >= 0.0:
+        result = eta * x - (eta * dxn + ops.sqrt(k)) * n
+    return result
+
+
+@func
+def dot(x, y):
+    """Calculate the dot product of two vectors.
+
+    Args:
+        x (:class:`~gstaichi.Matrix`): The first input vector.
+        y (:class:`~gstaichi.Matrix`): The second input vector.
+
+    Returns:
+        The dot product of two vectors.
+
+    Example::
+
+        >>> x = ti.Vector([1., 1., 0.])
+        >>> y = ti.Vector([0., 1., 1.])
+        >>> ti.math.dot(x, y)
+        1.000000
+    """
+    return x.dot(y)
+
+
+@func
+def cross(x, y):
+    """Calculate the cross product of two vectors.
+
+    The two input vectors must have the same dimension :math:`d <= 3`.
+
+    This function calls the `cross` method of :class:`~gstaichi.Vector`.
+
+    Args:
+        x (:class:`~gstaichi.Matrix`): The first input vector.
+        y (:class:`~gstaichi.Matrix`): The second input vector.
+
+    Returns:
+        The cross product of two vectors.
+
+    Example::
+
+        >>> x = ti.Vector([1., 0., 0.])
+        >>> y = ti.Vector([0., 1., 0.])
+        >>> ti.math.cross(x, y)
+        [0.000000, 0.000000, 1.000000]
+    """
+    return x.cross(y)
+
+
+@func
+def mod(x, y):
+    """Compute value of one parameter modulo another, element-wise.
+
+    Args:
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): The first input.
+        y (:mod:`~gstaichi.types.primitive_types`, :class:`~gstaichi.Matrix`): The second input.
+
+    Returns:
+        the value of `x` modulo `y`. This is computed as `x - y * floor(x/y)`.
+
+    Example::
+
+        >>> x = ti.Vector([-0.5, 0.5, 1.])
+        >>> y = 1.0
+        >>> ti.math.mod(x, y)
+        [0.500000, 0.500000, 0.000000]
+    """
+    return x - y * ops.floor(x / y)
+
+
+@func
+def translate(dx, dy, dz):
+    """Constructs a translation Matrix with shape (4, 4).
+
+    Args:
+        dx (float): delta x.
+        dy (float): delta y.
+        dz (float): delta z.
+
+    Returns:
+        :class:`~gstaichi.math.mat4`: translation matrix.
+
+    Example::
+
+        >>> ti.math.translate(1, 2, 3)
+        [[ 1. 0. 0. 1.]
+         [ 0. 1. 0. 2.]
+         [ 0. 0. 1. 3.]
+         [ 0. 0. 0. 1.]]
+    """
+    return mat4(
+        [
+            [1.0, 0.0, 0.0, dx],
+            [0.0, 1.0, 0.0, dy],
+            [0.0, 0.0, 1.0, dz],
+            [0.0, 0.0, 0.0, 1.0],
+        ]
+    )
+
+
+@func
+def scale(sx, sy, sz):
+    """Constructs a scale Matrix with shape (4, 4).
+
+    Args:
+        sx (float): scale x.
+        sy (float): scale y.
+        sz (float): scale z.
+
+    Returns:
+        :class:`~gstaichi.math.mat4`: scale matrix.
+
+    Example::
+
+        >>> ti.math.scale(1, 2, 3)
+        [[ 1. 0. 0. 0.]
+         [ 0. 2. 0. 0.]
+         [ 0. 0. 3. 0.]
+         [ 0. 0. 0. 1.]]
+    """
+    return mat4(
+        [
+            [sx, 0.0, 0.0, 0.0],
+            [0.0, sy, 0.0, 0.0],
+            [0.0, 0.0, sz, 0.0],
+            [0.0, 0.0, 0.0, 1.0],
+        ]
+    )
+
+
+@func
+def rot_by_axis(axis, ang):
+    """Returns the 4x4 matrix representation of a 3d rotation with given axis `axis` and angle `ang`.
+
+    Args:
+        axis (vec3): rotation axis
+        ang (float): angle in radians unit
+
+    Returns:
+        :class:`~gstaichi.math.mat4`: rotation matrix
+    """
+    c = ops.cos(ang)
+    s = ops.sin(ang)
+
+    axis = normalize(axis)
+    temp = (1 - c) * axis
+    return mat4(
+        [
+            [
+                c + temp[0] * axis[0],
+                temp[0] * axis[1] + s * axis[2],
+                temp[0] * axis[2] - s * axis[1],
+                0.0,
+            ],
+            [
+                temp[1] * axis[0] - s * axis[2],
+                c + temp[1] * axis[1],
+                temp[1] * axis[2] + s * axis[0],
+                0.0,
+            ],
+            [
+                temp[2] * axis[0] + s * axis[1],
+                temp[2] * axis[1] - s * axis[0],
+                c + temp[2] * axis[2],
+                0.0,
+            ],
+            [0.0, 0.0, 0.0, 1.0],
+        ]
+    )
+
+
+@func
+def rot_yaw_pitch_roll(yaw, pitch, roll):
+    """Returns a 4x4 homogeneous rotation matrix representing the 3d rotation with Euler angles (rotate with Y axis first, X axis second, Z axis third).
+
+    Args:
+        yaw   (float): yaw angle in radians unit
+        pitch (float): pitch angle in radians unit
+        roll  (float): roll angle in radians unit
+
+    Returns:
+        :class:`~gstaichi.math.mat4`: rotation matrix
+    """
+    ch = ops.cos(yaw)
+    sh = ops.sin(yaw)
+    cp = ops.cos(pitch)
+    sp = ops.sin(pitch)
+    cb = ops.cos(roll)
+    sb = ops.sin(roll)
+
+    return mat4(
+        [
+            [ch * cb + sh * sp * sb, sb * cp, -sh * cb + ch * sp * sb, 0.0],
+            [-ch * sb + sh * sp * cb, cb * cp, sb * sh + ch * sp * cb, 0.0],
+            [sh * cp, -sp, ch * cp, 0.0],
+            [0.0, 0.0, 0.0, 1.0],
+        ]
+    )
+
+
+@func
+def rotation2d(ang):
+    """Returns the matrix representation of a 2d counter-clockwise rotation,
+    given the angle of rotation.
+
+    Args:
+        ang (float): Angle of rotation in radians.
+
+    Returns:
+        :class:`~gstaichi.math.mat2`: 2x2 rotation matrix.
+
+    Example::
+
+        >>>ti.math.rotation2d(ti.math.radians(30))
+        [[0.866025, -0.500000], [0.500000, 0.866025]]
+    """
+    ca, sa = ops.cos(ang), ops.sin(ang)
+    return mat2([[ca, -sa], [sa, ca]])
+
+
+@func
+def rotation3d(ang_x, ang_y, ang_z):
+    """Returns a 4x4 homogeneous rotation matrix representing the 3d rotation with Euler angles (rotate with Y axis first, X axis second, Z axis third).
+
+    Args:
+        ang_x (float): angle in radians unit around X axis
+        ang_y (float): angle in radians unit around Y axis
+        ang_z (float): angle in radians unit around Z axis
+    Returns:
+        :class:`~gstaichi.math.mat4`: rotation matrix
+    Example:
+
+        >>> ti.math.rotation3d(0.52, -0.785, 1.046)
+        [[ 0.05048351 -0.61339645 -0.78816002  0.        ]
+        [ 0.65833154  0.61388511 -0.4355969   0.        ]
+        [ 0.75103329 -0.49688014  0.4348093   0.        ]
+        [ 0.          0.          0.          1.        ]]
+    """
+    return rot_yaw_pitch_roll(ang_z, ang_x, ang_y)
+
+
+@func
+def eye(n: template()):
+    """Returns the nxn identity matrix.
+
+    Alias for :func:`~gstaichi.Matrix.identity`.
+    """
+    return Matrix.identity(float, n)
+
+
+@func
+def length(x):
+    """Calculate the length of a vector.
+
+    This function is equivalent to the `length` function in GLSL.
+    Args:
+        x (:class:`~gstaichi.Matrix`): The vector of which to calculate the length.
+
+    Returns:
+        The Euclidean norm of the vector.
+
+    Example::
+
+        >>> x = ti.Vector([1, 1, 1])
+        >>> ti.math.length(x)
+        1.732051
+    """
+    return x.norm()
+
+
+@func
+def determinant(m):
+    """Alias for :func:`gstaichi.Matrix.determinant`."""
+    return m.determinant()
+
+
+@func
+def inverse(mat):  # pylint: disable=R1710
+    """Calculate the inverse of a matrix.
+
+    This function is equivalent to the `inverse` function in GLSL.
+
+    Args:
+        mat (:class:`gstaichi.Matrix`): The matrix of which to take the inverse. \
+            Supports only 2x2, 3x3 and 4x4 matrices.
+
+    Returns:
+        Inverse of the input matrix.
+
+    Example::
+
+        >>> m = ti.math.mat3([(1, 1, 0), (0, 1, 1), (0, 0, 1)])
+        >>> ti.math.inverse(m)
+        [[1.000000, -1.000000, 1.000000],
+         [0.000000, 1.000000, -1.000000],
+         [0.000000, 0.000000, 1.000000]]
+    """
+    return mat.inverse()
+
+
+@func
+def isinf(x):
+    """Determines whether the parameter is positive or negative infinity, element-wise.
+
+    Args:
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`gstaichi.Matrix`): The input.
+
+    Example:
+
+       >>> x = ti.math.vec4(inf, -inf, nan, 1)
+       >>> ti.math.isinf(x)
+       [1, 1, 0, 0]
+
+    Returns:
+        For each element i of the result, returns 1 if x[i] is posititve or negative floating point infinity and 0 otherwise.
+    """
+    ftype = impl.get_runtime().default_fp
+    fx = ops.cast(x, ftype)
+    if static(ftype == f64):
+        y = ops.bit_cast(fx, u64)
+        return (ops.cast(y >> 32, u32) & 0x7FFFFFFF) == 0x7FF00000 and (ops.cast(y, u32) == 0)
+
+    y = ops.bit_cast(fx, u32)
+    return (y & 0x7FFFFFFF) == 0x7F800000
+
+
+@func
+def isnan(x):
+    """Determines whether the parameter is a number, element-wise.
+
+    Args:
+        x (:mod:`~gstaichi.types.primitive_types`, :class:`gstaichi.Matrix`): The input.
+
+    Example:
+
+       >>> x = ti.math.vec4(nan, -nan, inf, 1)
+       >>> ti.math.isnan(x)
+       [1, 1, 0, 0]
+
+    Returns:
+        For each element i of the result, returns 1 if x[i] is posititve or negative floating point NaN (Not a Number) and 0 otherwise.
+    """
+    ftype = impl.get_runtime().default_fp
+    fx = ops.cast(x, ftype)
+    if static(ftype == f64):
+        y = ops.bit_cast(fx, u64)
+        return (ops.cast(y >> 32, u32) & 0x7FFFFFFF) + (ops.cast(y, u32) != 0) > 0x7FF00000
+
+    y = ops.bit_cast(fx, u32)
+    return (y & 0x7FFFFFFF) > 0x7F800000
+
+
+@func
+def vdir(ang):
+    """Returns the 2d unit vector with argument equals `ang`.
+
+    x (:mod:`~gstaichi.types.primitive_types`): The input angle in radians.
+
+    Example:
+
+        >>> x = pi / 2
+        >>> ti.math.vdir(x)
+        [0, 1]
+
+    Returns:
+        a 2d vector with argument equals `ang`.
+    """
+    return vec2(cos(ang), sin(ang))
+
+
+@func
+def popcnt(x):
+    return ops.popcnt(x)
+
+
+@func
+def clz(x):
+    return ops.clz(x)
+
+
+__all__ = [
+    "acos",
+    "asin",
+    "atan2",
+    "ceil",
+    "clamp",
+    "clz",
+    "cos",
+    "cross",
+    "degrees",
+    "determinant",
+    "distance",
+    "dot",
+    "e",
+    "exp",
+    "eye",
+    "floor",
+    "fract",
+    "inf",
+    "inverse",
+    "isinf",
+    "isnan",
+    "ivec2",
+    "ivec3",
+    "ivec4",
+    "length",
+    "log",
+    "log2",
+    "mat2",
+    "mat3",
+    "mat4",
+    "max",
+    "min",
+    "mix",
+    "mod",
+    "translate",
+    "scale",
+    "nan",
+    "normalize",
+    "pi",
+    "pow",
+    "popcnt",
+    "radians",
+    "reflect",
+    "refract",
+    "rot_by_axis",
+    "rot_yaw_pitch_roll",
+    "rotation2d",
+    "rotation3d",
+    "round",
+    "sign",
+    "sin",
+    "smoothstep",
+    "sqrt",
+    "step",
+    "tan",
+    "tanh",
+    "uvec2",
+    "uvec3",
+    "uvec4",
+    "vdir",
+    "vec2",
+    "vec3",
+    "vec4",
+]