Skip to content

Channels

This page contains classes defined in hakowan.channel module.

Channel dataclass

Channel base class.

Source code in hakowan/grammar/channel/channel.py 
10
11
12
13
14
@dataclass(kw_only=True, slots=True)
class Channel:
    """Channel base class."""

    pass

Position dataclass

Bases: Channel

Position channel

This class is used to specify the mapping from an attribute to the position channel. Note that, by default, the vertex coordinates of the data frame is used as the position channel. Thus, this class is mainly useful when we want to use non-vertex-coordinates as the position channel. For example, this method can be used for visualizing a deformed shape when the deformed position is stored as a vertex attribute in the data frame.

Attributes:

Name Type Description
data AttributeLike

The attribute used to encode the position field.
Source code in hakowan/grammar/channel/channel.py 
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
@dataclass(slots=True)
class Position(Channel):
    """Position channel

    This class is used to specify the mapping from an attribute to the position channel.
    Note that, by default, the vertex coordinates of the data frame is used as the position
    channel. Thus, this class is mainly useful when we want to use non-vertex-coordinates as the
    position channel. For example, this method can be used for visualizing a deformed shape when
    the deformed position is stored as a vertex attribute in the data frame.

    Attributes:
        data (AttributeLike): The attribute used to encode the position field.
    """

    data: AttributeLike

Normal dataclass

Bases: Channel

Normal channel

This class is used to specify the mapping from an attribute to the normal channel. By default, Hakowan will automatically compute the normal field from the geometry if normal channel is not specified. This class is useful for ensure the visualization uses a pre-defined normal field.

Attributes:

Name Type Description
data AttributeLike

The attribute used to encode the normal field.
Source code in hakowan/grammar/channel/channel.py 
34
35
36
37
38
39
40
41
42
43
44
45
46
47
@dataclass(slots=True)
class Normal(Channel):
    """Normal channel

    This class is used to specify the mapping from an attribute to the normal channel.
    By default, Hakowan will automatically compute the normal field from the geometry if normal
    channel is not specified. This class is useful for ensure the visualization uses a pre-defined
    normal field.

    Attributes:
        data (AttributeLike): The attribute used to encode the normal field.
    """

    data: AttributeLike

Size dataclass

Bases: Channel

Size channel

This class is used to specify the mapping from an attribute or value to the size channel. If a value is used, all elements will have the same size. Note that size is defined in the same unit as the input geometry.

Attributes:

Name Type Description
data AttributeLike | float

The attribute or value used to encode the size field.
Source code in hakowan/grammar/channel/channel.py 
50
51
52
53
54
55
56
57
58
59
60
61
62
@dataclass(slots=True)
class Size(Channel):
    """Size channel

    This class is used to specify the mapping from an attribute or value to the size channel. If a
    value is used, all elements will have the same size. Note that size is defined in the same unit
    as the input geometry.

    Attributes:
        data (AttributeLike | float): The attribute or value used to encode the size field.
    """

    data: AttributeLike | float

VectorField dataclass

Bases: Channel

Vector field channel

This class is used to specify the mapping from an attribute to the vector field channel.

A vector field can be define over the vertices or facets of the geometry. The vector field must have the same dimension as the geometry.

Attributes:

Name Type Description
data AttributeLike

The attribute used to encode the vector field.
refinement_level int

The refinement level of the vector field. This parameter is used to control the density of the vector field. The default value is 0.
style CurveStyle | None

The style of the vector field. If None, the default style will be used.
end_type Literal['point', 'arrow', 'flat']

The type of the vector field end. "point" tapers the tip to zero (cone/spike); "arrow" renders a flared arrowhead; "flat" keeps a constant radius at both ends (cylinder). The default value is "point".
normalize bool

If True, every vector is rescaled to unit length so that all arrows have the same length and only encode direction. The magnitude is freed up to be mapped to another channel (e.g. size or color via hakowan.norm()). Normalization is applied before any scale attached to data, so a uniform scale on data controls the common arrow length. By default (False), arrow length is proportional to the vector magnitude. The default value is False.
Source code in hakowan/grammar/channel/channel.py 
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
@dataclass(slots=True)
class VectorField(Channel):
    """Vector field channel

    This class is used to specify the mapping from an attribute to the vector field channel.

    A vector field can be define over the vertices or facets of the geometry. The vector field must
    have the same dimension as the geometry.

    Attributes:
        data (AttributeLike): The attribute used to encode the vector field.
        refinement_level (int): The refinement level of the vector field. This parameter is used to
            control the density of the vector field. The default value is 0.
        style (CurveStyle | None): The style of the vector field. If None, the default style will
            be used.
        end_type (Literal["point", "arrow", "flat"]): The type of the vector field end.
            ``"point"`` tapers the tip to zero (cone/spike); ``"arrow"`` renders a
            flared arrowhead; ``"flat"`` keeps a constant radius at both ends
            (cylinder). The default value is ``"point"``.
        normalize (bool): If True, every vector is rescaled to unit length so
            that all arrows have the same length and only encode direction.
            The magnitude is freed up to be mapped to another channel (e.g.
            ``size`` or color via ``hakowan.norm()``). Normalization is
            applied *before* any scale attached to ``data``, so a uniform scale
            on ``data`` controls the common arrow length. By default (False),
            arrow length is proportional to the vector magnitude. The default
            value is ``False``.
    """

    data: AttributeLike
    refinement_level: int = 0
    style: CurveStyle | None = None
    end_type: Literal["point", "arrow", "flat"] = "point"
    normalize: bool = False

Covariance dataclass

Bases: Channel

Covariance channel

This class is used to specify the mapping from an attribute to the covariance matrix channel. The covariance channel only applies to point mark. It is represented as a per-vertex 3x3 symmetric matrix, which defines the stretch and rotation of the point marks.

Attributes:

Name Type Description
data AttributeLike

The attribute used to encode the covariance matrix.
full bool

(bool): If True, the full covariance matrix is stored in the attribute. If False, its "square root", M, is stored. The full covariance matrix is ∑ := M @ M^T. The matrix M represenst the stretch and rotation transform applied on each mark.
Source code in hakowan/grammar/channel/channel.py 
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
@dataclass(slots=True)
class Covariance(Channel):
    """Covariance channel

    This class is used to specify the mapping from an attribute to the covariance matrix channel.
    The covariance channel only applies to point mark. It is represented as a per-vertex 3x3
    symmetric matrix, which defines the stretch and rotation of the point marks.

    Attributes:
        data (AttributeLike): The attribute used to encode the covariance matrix.
        full: (bool): If True, the full covariance matrix is stored in the attribute.
            If False, its "square root", M, is stored. The full covariance matrix is ∑ := M @ M^T.
            The matrix M represenst the stretch and rotation transform applied on each mark.
    """

    data: AttributeLike
    full: bool = False

Shape dataclass

Bases: Channel

Shape channel

This class is used to specify the mapping from an attribute to the shape channel. This channel is only used for point mark.

Attributes:

Name Type Description
base_shape Literal['sphere', 'disk', 'cube']

The base shape used to represent a point. The default value is "sphere".
orientation AttributeLike | None

The attribute used to encode the normal orientation of the shape. If None, orientation will be identity (i.e. normal along z-axis).
Source code in hakowan/grammar/channel/channel.py 
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
@dataclass(slots=True)
class Shape(Channel):
    """Shape channel

    This class is used to specify the mapping from an attribute to the shape channel.
    This channel is only used for point mark.

    Attributes:
        base_shape (Literal["sphere", "disk", "cube"]): The base shape used to represent a point.
            The default value is ``"sphere"``.
        orientation (AttributeLike | None): The attribute used to encode the normal orientation
            of the shape. If None, orientation will be identity (i.e. normal along z-axis).
    """

    base_shape: Literal["sphere", "disk", "cube"] = "sphere"
    orientation: Optional[AttributeLike] = None

BumpMap dataclass

Bases: Channel

Bump map channel

This class specifies the bump map channel.

Attributes:

Name Type Description
texture TextureLike | None

The texture used to encode the bump map.
scale float

The scale of the bump map. The default value is 1.0.
Source code in hakowan/grammar/channel/channel.py 
138
139
140
141
142
143
144
145
146
147
148
149
150
@dataclass(slots=True)
class BumpMap(Channel):
    """Bump map channel

    This class specifies the bump map channel.

    Attributes:
        texture (TextureLike | None): The texture used to encode the bump map.
        scale (float): The scale of the bump map. The default value is 1.0.
    """

    texture: TextureLike
    scale: float = 1.0

NormalMap dataclass

Bases: Channel

Normal map channel

This class specifies the normal map channel.

Attributes:

Name Type Description
texture TextureLike | None

The texture used to encode the normal map.
Source code in hakowan/grammar/channel/channel.py 
153
154
155
156
157
158
159
160
161
162
163
@dataclass(slots=True)
class NormalMap(Channel):
    """Normal map channel

    This class specifies the normal map channel.

    Attributes:
        texture (TextureLike | None): The texture used to encode the normal map.
    """

    texture: TextureLike