File: T01_basic_visual.py

package info (click to toggle)
python-vispy 0.15.2-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 8,868 kB
  • sloc: python: 59,799; javascript: 6,800; makefile: 69; sh: 6
file content (159 lines) | stat: -rw-r--r-- 6,494 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
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
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
# -*- coding: utf-8 -*-
# -----------------------------------------------------------------------------
# Copyright (c) Vispy Development Team. All Rights Reserved.
# Distributed under the (new) BSD License. See LICENSE.txt for more info.
# -----------------------------------------------------------------------------
"""
Tutorial: Creating Visuals
--------------------------

This tutorial is intended to guide developers who are interested in creating 
new subclasses of Visual. In most cases, this will not be necessary because
vispy's base library of visuals will be sufficient to create complex scenes as
needed. However, there are cases where a particular visual effect is desired 
that is not supported in the base library, or when a custom visual is needed to
optimize performance for a specific use case.

The purpose of a Visual is to encapsulate a single drawable object. This
drawable can be as simple or complex as desired. Some of the simplest visuals 
draw points, lines, or triangles, whereas more complex visuals invove multiple
drawing stages or make use of sub-visuals to construct larger objects.

In this example we will create a very simple Visual that draws a rectangle.
Visuals are defined by:

1. Creating a subclass of vispy.visuals.Visual that specifies the GLSL code
   and buffer objects to use.
2. Defining a _prepare_transforms() method that will be called whenever the
   user (or scenegraph) assigns a new set of transforms to the visual.


"""
from vispy import app, gloo, visuals, scene
import numpy as np

# Define a simple vertex shader. We use $template variables as placeholders for
# code that will be inserted later on. In this example, $position will become
# an attribute, and $transform will become a function. Important: using
# $transform in this way ensures that users of this visual will be able to 
# apply arbitrary transformations to it.
vertex_shader = """
void main() {
   gl_Position = $transform(vec4($position, 0, 1));
}
"""

# Very simple fragment shader. Again we use a template variable "$color", which
# allows us to decide later how the color should be defined (in this case, we
# will just use a uniform red color).
fragment_shader = """
void main() {
  gl_FragColor = $color;
}
"""


# Start the new Visual class. 
# By convention, all Visual subclass names end in 'Visual'.
# (Custom visuals may ignore this convention, but for visuals that are built 
# in to vispy, this is required to ensure that the VisualNode subclasses are 
# generated correctly.)
class MyRectVisual(visuals.Visual):
    """Visual that draws a red rectangle.

    Parameters
    ----------
    x : float
        x coordinate of rectangle origin
    y : float
        y coordinate of rectangle origin
    w : float
        width of rectangle
    h : float
        height of rectangle

    All parameters are specified in the local (arbitrary) coordinate system of
    the visual. How this coordinate system translates to the canvas will 
    depend on the transformation functions used during drawing.
    """

    # There are no constraints on the signature of the __init__ method; use
    # whatever makes the most sense for your visual.
    def __init__(self, x, y, w, h):
        # Initialize the visual with a vertex shader and fragment shader
        visuals.Visual.__init__(self, vertex_shader, fragment_shader)

        # vertices for two triangles forming a rectangle
        self.vbo = gloo.VertexBuffer(np.array([
            [x, y], [x+w, y], [x+w, y+h],
            [x, y], [x+w, y+h], [x, y+h]
        ], dtype=np.float32))

        # Assign values to the $position and $color template variables in 
        # the shaders. ModularProgram automatically handles generating the 
        # necessary attribute and uniform declarations with unique variable
        # names.
        self.shared_program.vert['position'] = self.vbo
        self.shared_program.frag['color'] = (1, 0, 0, 1)
        self._draw_mode = 'triangles'

    def _prepare_transforms(self, view):
        # This method is called when the user or the scenegraph has assigned
        # new transforms to this visual (ignore the *view* argument for now;
        # we'll get to that later). This method is thus responsible for
        # connecting the proper transform functions to the shader program.

        # The most common approach here is to simply take the complete
        # transformation from visual coordinates to render coordinates. Later
        # tutorials detail more complex transform handling.
        view.view_program.vert['transform'] = view.get_transform()


# At this point the visual is ready to use, but it takes some extra effort to
# set up a Canvas and TransformSystem for drawing (the examples in 
# examples/basics/visuals/ all follow this approach). 
# 
# An easier approach is to make the visual usable in a scenegraph, in which 
# case the canvas will take care of drawing the visual and setting up the 
# TransformSystem for us.
# 
# To be able to use our new Visual in a scenegraph, it needs to be
# a subclass of scene.Node. In vispy we achieve this by creating a parallel
# set of classes that inherit from both Node and each Visual subclass.
# This can be done automatically using scene.visuals.create_visual_node():
MyRect = scene.visuals.create_visual_node(MyRectVisual)

# By convention, these classes have the same name as the Visual they inherit 
# from, but without the 'Visual' suffix.

# The auto-generated class MyRect is basically equivalent to::
# 
#     class MyRect(MyRectVisual, scene.Node):
#        def __init__(self, *args, **kwds):
#            parent = kwds.pop('parent', None)
#            name = kwds.pop('name', None)
#            MyRectVisual.__init__(self, *args, **kwds)
#            Node.__init__(self, parent=parent, name=name)
#         


# Finally we will test the visual by displaying in a scene.

# Create a canvas to display our visual
canvas = scene.SceneCanvas(keys='interactive', show=True)

# Create two instances of MyRect, each using canvas.scene as their parent
rects = [MyRect(100, 100, 200, 300, parent=canvas.scene),
         MyRect(500, 100, 200, 300, parent=canvas.scene)]

# To test that the user-specified transforms work correctly, I'll rotate
# one rectangle slightly.
tr = visuals.transforms.MatrixTransform()
tr.rotate(5, (0, 0, 1))
rects[1].transform = tr

# ..and optionally start the event loop
if __name__ == '__main__':
    import sys
    if sys.flags.interactive != 1:
        app.run()