[Bug]: `set_text` does not work when accessed from `axis.get_xticklabels`

[hoxbro]

Bug summary

I want to update the x-ticks label in a for loop with set_text, but it does not update the text.

Code for reproduction

import matplotlib.pyplot as plt

fig, axis = plt.subplots()
for t in axis.get_xticklabels():
    t.set_text('a')

plt.show()

Actual outcome

Nothing happens with the xticklabels.

Expected outcome

I would expect all ticks on x-axis to be labeled 'a'. It works if I use t.set_y with the offset.

Additional information

No response

Operating system

No response

Matplotlib Version

3.8.4

Matplotlib Backend

No response

Python version

No response

Jupyter version

No response

Installation

None

[tacaswell]

See https://matplotlib.org/devdocs/users/explain/axes/axes_ticks.html for details.

To get what you want use ax.set_xticks(locs, labels=['a']*len(locs))

The very short version is that the ticking system is made up of a locator (which decides where the tick should go) and the formatter (which decides what the tick label should be). We do expose Text objects for the tick labels, however they are controlled by the ticking system so at draw time we will over-write the location and strings. This is clearest when panning / zooming as the ticks must both move (on the screne) and get new labels.

It is an implementation details that set_y works because for efficiency reasons we re-use Text objects from a pool (rather than making new ones every time).

[hoxbro]

Thank you for the explanation.