Working with the Backend Objects

Convert a backend object to whitecanvas canvas

The wrap_canvas converts a backend object to a whitecanvas canvas.

from whitecanvas import wrap_canvas
import matplotlib.pyplot as plt

fig, axes = plt.subplots()
axes.set_title("Title")  # operations in the backend side

canvas = wrap_canvas(axes)  # axes --> canvas
canvas.add_line([0, 1, 2, 3, 4])

plt.show()  # backend methods still work

Retrieve the backend object from a whitecanvas canvas

The backend object is available at the native property.

from whitecanvas import new_canvas

canvas = new_canvas("matplotlib")
canvas.native

Output

<Axes: >

Combine whitecanvas with Applications

1. Control matplotlib Qt application with whitecanvas

#!skip
# use `%gui qt` in IPython
from whitecanvas import new_canvas
import matplotlib.pyplot as plt
from qtpy import QtWidgets as QtW

canvas = new_canvas("matplotlib:qt")
qt_canvas = canvas.native.get_figure().canvas

main = QtW.QMainWindow()
main.setWindowTitle("myapp")
main.setCentralWidget(qt_canvas)
main.show()

Since the canvas points to the same canvas as in the Qt application, you can control the application with whitecanvas API.

canvas.add_line([0, 1, 2, 3, 4])