3
\page Usd_Page_Datatypes Basic Datatypes for Scene Description Provided by Sdf
5
\section Usd_Datatypes Attribute value types
7
USD supports a variety of data types for attribute values.
8
These types are encoded in an SdfValueTypeName object that corresponds to an
9
underlying C++ type and is what a user would receive from a call to
10
UsdAttribute::GetTypeName. Methods for looking up these objects are provided
11
by \c SdfSchema and they're enumerated in the
12
object \c SdfValueTypeNames -- \sa SdfSchema
14
For example, to create a custom, Matrix4d-valued attribute on prim \em foo:
16
if (UsdAttribute mat = foo.CreateAttribute(TfToken("myMatrix"), SdfValueTypeNames->Matrix4d)){
17
mat.Set(GfMatrix4d(1)); // Assign identity matrix
19
// error creating the attribute
24
mat = foo.CreateAttribute("myMatrix", Sdf.ValueTypeNames.Matrix4d)
26
mat.Set(Gf.Matrix4d(1)) # Assign identity matrix
28
# error creating the attribute
31
\section Usd_Basic_Datatypes Basic data types
33
This table lists the fundamental data types supported by USD.
35
| Value type token | C++ type | Description |
36
|:-----------------|:--------------|:-------------------------------------------------|
38
| uchar | uint8_t | 8 bit unsigned integer |
39
| int | int32_t | 32 bit signed integer |
40
| uint | uint32_t | 32 bit unsigned integer |
41
| int64 | int64_t | 64 bit signed integer |
42
| uint64 | uint64_t | 64 bit unsigned integer |
43
| half | half | 16 bit floating point (imath half) |
44
| float | float | 32 bit floating point |
45
| double | double | 64 bit floating point |
46
| string | std::string | stl string |
47
| token | TfToken | interned string with fast comparison and hashing |
48
| asset | SdfAssetPath | represents a resolvable path to another asset |
49
| matrix2d | GfMatrix2d | 2x2 matrix of doubles |
50
| matrix3d | GfMatrix3d | 3x3 matrix of doubles |
51
| matrix4d | GfMatrix4d | 4x4 matrix of doubles |
52
| quatd | GfQuatd | double-precision quaternion |
53
| quatf | GfQuatf | single-precision quaternion |
54
| quath | GfQuath | half-precision quaternion |
55
| double2 | GfVec2d | vector of 2 doubles |
56
| float2 | GfVec2f | vector of 2 floats |
57
| half2 | GfVec2h | vector of 2 half's |
58
| int2 | GfVec2i | vector of 2 ints |
59
| double3 | GfVec3d | vector of 3 doubles |
60
| float3 | GfVec3f | vector of 3 floats |
61
| half3 | GfVec3h | vector of 3 half's |
62
| int3 | GfVec3i | vector of 3 ints |
63
| double4 | GfVec4d | vector of 4 doubles |
64
| float4 | GfVec4f | vector of 4 floats |
65
| half4 | GfVec4h | vector of 4 half's |
66
| int4 | GfVec4i | vector of 4 ints |
68
\section Usd_Roles Roles
70
Value types may also be specified by one of a set of special type names.
71
These names correspond to the basic data types listed above but provide
72
extra semantics about how the data should be interpreted. For instance,
73
a value of type "frame4d" is a GfMatrix4d, but represents a coordinate frame.
75
These type names are grouped by a "role" name, such that all types with the
76
same role have the same semantics. All of these "semantic types"
77
are also provided in SdfValueTypeNames
79
| Value type token | Underlying type | Role name | Role Meaning |
80
|:------------------|:----------------|:-----------|:----------------------------------------------------|
81
| point3d | Vec3d | Point | transform as a position |
82
| point3f | Vec3f | Point | transform as a position |
83
| point3h | Vec3h | Point | transform as a position |
84
| normal3d | Vec3d | Normal | transform as a normal |
85
| normal3f | Vec3f | Normal | transform as a normal |
86
| normal3h | Vec3h | Normal | transform as a normal |
87
| vector3d | Vec3d | Vector | transform as a direction |
88
| vector3f | Vec3f | Vector | transform as a direction |
89
| vector3h | Vec3h | Vector | transform as a direction |
90
| color3d | Vec3d | Color | energy-linear RGB |
91
| color3f | Vec3f | Color | energy-linear RGB |
92
| color3h | Vec3h | Color | energy-linear RGB |
93
| color4d | Vec4d | Color | energy-linear RGBA, <b>not</b> pre-alpha multiplied |
94
| color4f | Vec4f | Color | energy-linear RGBA, <b>not</b> pre-alpha multiplied |
95
| color4h | Vec4h | Color | energy-linear RGBA, <b>not</b> pre-alpha multiplied |
96
| frame4d | Matrix4d | Frame | defines a coordinate frame |
98
The following exist for historical/internal reasons,
99
but we expect to eliminate them prior to open-sourcing.
101
| Value type token | Underlying type | Role name |
102
|:------------------|:----------------|:-----------|
103
| Transform | Matrix4d | Transform |
104
| PointIndex | int | PointIndex |
105
| EdgeIndex | Vec2i | EdgeIndex |
106
| FaceIndex | int | FaceIndex |
108
\section Usd_Array_Datatypes Array data types
110
USD also supports arrays of the above data types. The value type
111
name for these arrays is simply the name of the underlying value type
112
and "[]", e.g. "bool[]", "float[]", etc., and are also provided by
113
SdfValueTypeNames with a "Array" suffix on the basic datatype
114
(e.g. SdfValueTypeNames->FloatArray provides "float[]").
115
The corresponding C++ type is VtArray, e.g. VtArray<bool>, VtArray<float>, etc.
117
\section Usd_Dictionary_Type Dictionary-valued Metadata
119
Metadata in USD (See \ref Usd_OM_Metadata) can take on several other datatypes.
120
Most of these are highly-specialized, pertaining to composition,
121
like SdfListOp<SdfReference>, but we also provide one very versatile
122
datatype <b>dictionary</b> <em>that can only be assumed by metadata</em>,
123
such as \ref UsdObject::GetCustomData() "customData". A dictionary's entries
124
are keyed by string, and can take on any other scene description value type,
125
\em including dictionary, allowing for nested dictionaries.
127
Dictionaries have special value-resolution semantics, in that each field
128
resolves independently of all others, allowing dictionaries to be sparsely
129
described and overridden in many layers. This behavior, which we have
130
found highly desirable, also makes dictionaries more expensive to fully
131
compose than all other value types (except those that
132
are \ref Usd_OM_ListOps "list-edited"). It is for this reason that we
133
disallow dictionary as a value type for attributes: it would eliminate
134
the extremely important performance characteristic of attribute
135
value resolution that "strongest opinion wins".
137
You can use the builtin customData for your truly ad-hoc user-data needs,
138
but you can also add new dictionary-valued metadata by defining it in a
139
module plugInfo.json file -
140
see \ref Usd_Extending_Metadata for details on how to do so.