如何在Matplotlib中为Artist直接绑定Pick Event回调函数

Matplotlib确实没有提供直接给单个Artist（比如线条）绑定独立pick回调的原生API，但不需要编写复杂的Pick_dispatcher类，有两种更简洁的方式实现需求：

方法1：利用闭包封装专属回调 #

通过闭包为每条线条生成专属处理逻辑，把线条的样式参数（比如竖线颜色）封装进去，避免全局判断：

import numpy as np
import matplotlib.pyplot as plt

def create_line_pick_handler(color):
    # 闭包封装竖线颜色，生成专属回调
    def handler(event):
        artist = event.artist
        i_sel = event.ind
        i_med = i_sel[len(i_sel) // 2]
        x = artist.get_xdata()[i_med]
        ax = artist.axes
        ax.axvline(x, ls='--', lw=1, color=color)
        artist.get_figure().canvas.draw_idle()
    return handler

if __name__ == "__main__":
    # 示例数据
    x = np.array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11])
    y1 = np.array([0, 0, 0, 1, 2, 3, 4, 5, 6, 6, 6, 6])
    y2 = np.array([0, 0, 1, 2, 3, 4, 5, 5, 5, 5, 5, 5])

    fig, ax = plt.subplots()
    tolerance = 5

    # 创建y1线条并绑定专属回调
    f, = ax.plot(x, y1, marker=".", ms=2, lw=1, label='y1', picker=True, pickradius=tolerance)
    fig.canvas.callbacks.connect('pick_event', create_line_pick_handler('blue'))

    # 创建y2线条并绑定专属回调
    g, = ax.plot(x, y2, marker=".", ms=2, lw=1, label='y2', picker=True, pickradius=tolerance)
    fig.canvas.callbacks.connect('pick_event', create_line_pick_handler('red'))

    plt.legend()
    plt.show()

方法2：给线条对象添加自定义回调属性 #

如果需要保留独立的on_pick_line1、on_pick_line2函数，可以给每条线条添加自定义属性存储对应的回调，再通过一个全局处理函数分发：

import numpy as np
import matplotlib.pyplot as plt

def on_pick_line1(event):
    artist = event.artist
    i_sel = event.ind
    i_med = i_sel[len(i_sel) // 2]
    x = artist.get_xdata()[i_med]
    ax = artist.axes
    ax.axvline(x, ls='--', lw=1, color='blue')
    artist.get_figure().canvas.draw_idle()

def on_pick_line2(event):
    artist = event.artist
    i_sel = event.ind
    i_med = i_sel[len(i_sel) // 2]
    x = artist.get_xdata()[i_med]
    ax = artist.axes
    ax.axvline(x, ls='--', lw=1, color='red')
    artist.get_figure().canvas.draw_idle()

def pick_handler(event):
    # 检查当前点击的线条是否有自定义pick回调
    if hasattr(event.artist, 'pick_callback'):
        event.artist.pick_callback(event)

if __name__ == "__main__":
    x = np.array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11])
    y1 = np.array([0, 0, 0, 1, 2, 3, 4, 5, 6, 6, 6, 6])
    y2 = np.array([0, 0, 1, 2, 3, 4, 5, 5, 5, 5, 5, 5])

    fig, ax = plt.subplots()
    tolerance = 5

    # 给y1线条绑定专属回调
    f, = ax.plot(x, y1, marker=".", ms=2, lw=1, label='y1', picker=True, pickradius=tolerance)
    f.pick_callback = on_pick_line1

    # 给y2线条绑定专属回调
    g, = ax.plot(x, y2, marker=".", ms=2, lw=1, label='y2', picker=True, pickradius=tolerance)
    g.pick_callback = on_pick_line2

    plt.legend()
    fig.canvas.callbacks.connect('pick_event', pick_handler)
    plt.show()

说明 #

两种方案都能替代复杂的调度类：

• 闭包方式适合回调逻辑相似、仅参数不同的场景，代码更紧凑
• 自定义属性方式适合保留独立回调函数的场景，逻辑更清晰

二者都能实现点击不同线条触发对应回调、绘制对应颜色竖线的需求，比原有的Pick_dispatcher更简洁直观。

内容的提问来源于stack exchange，提问作者user27243451