如何以防止xdata变为None的方式更新matplotlib图形？

Question

我正在编写一个应用程序，该应用程序定期（大约每秒一次）从硬件设备收集一些数据，并使用matplotlib将其显示在图形中。 用户将定期单击图形，当发生这种情况时，我需要知道图形的坐标系中测量的单击的x和y坐标。

这很容易做到：如果canvas是FigureCanvas实例，则可以说

canvas.mpl_connect('button_release_event', mouse_up_handler)

并定义

def mouse_up_handler(event):
    print('The user clicked at ({}, {})'.format(event.xdata, event.ydata))

以在图形的坐标系中显示点击坐标。 但是，有时用户会在图形更新时单击鼠标右键，然后event.xdata和.ydata为None 。 现在，我只是舍弃这些点击，但是当应用程序忽略他们的点击时，这显然让用户感到沮丧。

我已经附上了完整的示例程序，以便您可以看到如何用新的图形替换当前的图形。 （如果运行它并重复单击图形，您有时会看到报告的x坐标有时为None 。）我可以用永远不会给出None的方式替换图形的数据吗？ 我只希望一次显示一组数据。

范例程序

from __future__ import print_function
from math import exp
from matplotlib.backends.backend_wxagg import FigureCanvasWxAgg \
    as FigureCanvas
from matplotlib.figure import Figure
from random import normalvariate
from threading import Timer
from time import sleep
import wx

class GraphView(wx.Panel):
    def __init__(self, parent):
        wx.Panel.__init__(self, parent)

        self.figure = Figure()
        canvas = FigureCanvas(self, -1, self.figure)
        canvas.mpl_connect('button_release_event', self.mouse_up_handler)

        self.axes = None

        sizer = wx.BoxSizer(wx.VERTICAL)
        sizer.Add(canvas, 1, wx.GROW)
        self.SetSizer(sizer)

        self.generate_new_data()

    def generate_new_data(self):
        x_values = range(16384)
        y_values = [exp(normalvariate(0, 10)) for _ in range(16384)]

        if self.axes is not None:
            self.figure.delaxes(self.axes)

        sleep(0.4)  # simulate some additional work being done with the data

        self.axes = self.figure.add_subplot(1, 1, 1)
        self.axes.set_yscale('log')
        self.axes.autoscale(enable=True)
        self.axes.set_xlim(x_values[0], x_values[-1])
        self.axes.plot(x_values, y_values, antialiased=False)

        self.figure.canvas.draw()

        Timer(1.0, self.generate_new_data).start()

    def mouse_up_handler(self, event):
        print('Click at x = {}'.format(event.xdata))


class GraphWindow(wx.Frame):
    def __init__(self, parent):
        wx.Frame.__init__(self, parent)
        graph_view = GraphView(self)


if __name__ == '__main__':
    app = wx.App(False)
    GraphWindow(None).Show()
    app.MainLoop()

Answer 1

就像是：

def __init__(self, parent):
    wx.Panel.__init__(self, parent)

    self.figure = Figure()
    canvas = FigureCanvas(self, -1, self.figure)
    canvas.mpl_connect('button_release_event', self.mouse_up_handler)

    # make axes
    self.ax = self.figure.add_subplot(1, 1, 1)
    # set up details
    self.ax.set_yscale('log')
    self.ax.autoscale(enable=True)
    self.ax.set_xlim(x_values[0], x_values[-1]) 
    # make the line
    self.ln, = self.ax.plot([], []) # add what ever parameters you want

    sizer = wx.BoxSizer(wx.VERTICAL)
    sizer.Add(canvas, 1, wx.GROW)
    self.SetSizer(sizer)

    self.generate_new_data()



def generate_new_data(self):
    x_values = range(16384)
    y_values = [exp(normalvariate(0, 10)) for _ in range(16384)]

    sleep(0.4)  # simulate some additional work being done with the data

    # update your existing line
    self.ln.set_data(x_values, y_values, antialiased=False)

    self.figure.canvas.draw()

    Timer(1.0, self.generate_new_data).start()

重用轴和Line2D对象。