set xtick label colors based on label string-CodePudding

I have a plot with a list of labels. What I'd like to do is color code each label based on the string. So, for instance, all labels with the string ':Q_D' would refer to quadrupoles and should get a certain color. The other substrings to color by are ':DCH_D' and ':DCV_D' for horizontal and vertical corrector magnets.

I see that you can change the color of all xtick labels using this: plt.xticks(color='r') but I'd like to have them individually colored. Is this possible?

CodePudding user response：

You could loop through the tick labels, check whether they contain the given string, and set the color:

from matplotlib import pyplot as plt
import numpy as np
from random import choice, randint

# first create some test data
labels = [f"REA_BTS{randint(25, 35)}:{choice(['DCV_D', 'DCH_D', 'Q_D'])}{randint(1100, 1500)}:I_CSET"
          for _ in range(20)]

# create the plot
fig, ax = plt.subplots(figsize=(12, 6))
for _ in range(7):
    ax.plot(labels, np.random.randn(20).cumsum())
ax.tick_params(axis='x', labelrotation=90)
ax.margins(x=0.01)  # less whitespace inside the plot left and right
plt.tight_layout()

# set the colors of the tick labels; note that the tick labels are 
# only filled in after plt.tight_layout() or after a draw
for t in ax.get_xticklabels():
    txt = t.get_text()
    print(t.get_text())
    if 'Q_D' in txt:
        t.set_color('tomato')
    elif 'DCV_D' in txt:
        t.set_color('cornflowerblue')
    elif 'DCH_D' in txt:
        t.set_color('lime')
plt.show()