Matplotlib：Draggable 不适用于使用 twinx() 生成的多个图例

Question

我有两个图例，如下图，我发现我无法拖动第一个图例，这是什么问题？如何处理？谢谢！

    import matplotlib.pyplot as plt

    fig1, ax1 = plt.subplots()
    ax2 = ax1.twinx()

    ax1.plot([1,2,3],[0.1,0.82,0.3],'y*', label="one")
    ax2.plot([1,2,3],[5,6,7],'ro', label="two")

    leg1 = ax1.legend()
    leg2 = ax2.legend()

    leg1.draggable(state=True)
    leg2.draggable(state=True)
    plt.show()

Answer 1

这是孪生轴的一般限制。选择事件仅限于 "top" 轴。

这有几个原因，但基本上可以归结为 "it takes a good deal of refactoring to change, might break backwards compatibility in a subtle way, and it only impacts a few things"。

有一个解决方法，但有点笨拙。我还将把第二个图例从第一个图例上移开。否则，我们将同时拖动它们。这是一个例子：

import matplotlib.pyplot as plt

class Workaround(object):
    def __init__(self, artists):
        self.artists = artists
        artists[0].figure.canvas.mpl_connect('button_press_event', self)

    def __call__(self, event):
        for artist in self.artists:
            artist.pick(event)

fig1, ax1 = plt.subplots()
ax2 = ax1.twinx()

ax1.plot([1,2,3],[0.1,0.82,0.3],'y*', label="one")
ax2.plot([1,2,3],[5,6,7],'ro', label="two")

leg1 = ax1.legend()
leg2 = ax2.legend(loc='upper left')

leg1.draggable(True)
leg2.draggable(True)

draggable_workaround = Workaround([leg1, leg2])

plt.show()

Matplotlib：Draggable 不适用于使用 twinx() 生成的多个图例

Matplotlib: Draggable is not working on multiple legends of plots which are generated with twinx()

python

matplotlib

plot

legend