如何在Matplotlib pick_event中一次仅选择一个点

Question

I have a Python script that plots lots (n) of lines, each of 10 points, and I am trying to make it so that I can click on a line and it will print the id of the line and the id of the point in the line. So far I have got this:

def onpick(event):
    ind = event.ind
    s = event.artist.get_gid()
    print s, ind

#x and y are n x 10 arrays
#s is the id of the line
for s in range(n):
    ax.plot(x[s,:],y[s,:],'^',color=colors(s),picker=2,gid=str(s))

That works fine and gives me a plot a bit like this (I have previously put the coloured boxes and colorbar in place for reference):

I can click on a point and it prints something like

1 [1]

**The problem is this - ** if I click between two points that are very close it prints

0 [2 3]

or similar. I can't reduce the "picker" distance any further because that makes it very hard to get the mouse in the exact right position to pick a point.

What I would like is a way of limiting the pick to be ONLY the closest point. Any ideas?

Answer 1

If you want to print only the index of the closest point, you need to find out which one of those is the closest to the mouseevent.

The location of the mouseevent in data coordinates is obtained via event.mouseevent.xdata (or ydata ). Then the distance needs to be calculated and the index of the point which is closest can be returned.

import numpy as np; np.random.seed(1)
import matplotlib.pyplot as plt

x = np.logspace(1,10,base=1.8)
y = np.random.rayleigh(size=(2,len(x)))

def onpick(event):
    ind = event.ind
    if len(ind) > 1:
        datax,datay = event.artist.get_data()
        datax,datay = [datax[i] for i in ind],[datay[i] for i in ind]
        msx, msy = event.mouseevent.xdata, event.mouseevent.ydata
        dist = np.sqrt((np.array(datax)-msx)**2+(np.array(datay)-msy)**2)
        ind = [ind[np.argmin(dist)]]
    s = event.artist.get_gid()
    print s, ind

colors=["crimson","darkblue"]
fig,ax = plt.subplots()
for s in range(2):
    ax.plot(x,y[s,:],'^',color=colors[s],picker=2,gid=str(s))

fig.canvas.mpl_connect("pick_event", onpick)

plt.show()