Hi @Arikad0 - can you address the current check failures? (check out above the Linting, Mypy and Tests failures)

Also, if this PR closes the issue you mentioned in the description, you can use one of the magic phrases such as "Closes #" so that GitHub understands to link the issue and PR. Thanks!

As it stands, the implementation here is identical to that for the vertical lines, so we have

import matplotlib.pyplot as plt

fig, ax = plt.subplots()
ax.fill_between([0, 1], [0, 1], hatch="_")
plt.show()

I believe the reason the tests fail is because we now have two vertical lines on top of each other for the "+" hatching. E.g here are the result and diff images for the failing legend test:

@Arikad0 do you have thoughts on how to actually create the dashed pattern?

I'm a bit lost on how to proceed on this issue. Do you have any idea that may help? Thanks in advance

Gentle ping - any reviewers available to take a look? Thanks!

I wondered if we could use the Shapes type of hatch with a very simple straight path for this and actually it seems to work well:

Defining the class as

class Dashes(Shapes):
    size = 0.35
    def __init__(self, hatch, density):
        self.num_rows = (hatch.count('_')) * density
        path = Path([[0, 0], [1, 0]], [Path.MOVETO, Path.LINETO])
        self.shape_vertices = path.vertices
        self.shape_codes = path.codes
        super().__init__(hatch, density)

and adapting the example from #27146

import matplotlib.pyplot as plt
import numpy as np

# Data
cols = 10
rows = 4

data = (
    np.reshape(np.arange(0, cols, 1), (1, -1)) ** 2
    + np.reshape(np.arange(0, rows), (-1, 1))
    + np.random.random((rows, cols))*5
)

# Plot
fig, ax = plt.subplots()

x = range(data.shape[1])
ax.stackplot(
    x, data,
    hatch=["_", "__", "___", "____"]
)

ax.set_xlim(0, 9)
ax.set_ylim(0, 350)

plt.show()

I have

Possibly there are better ways.

Is there any way to bypass the MatplotlibDeprecationWarning, in case of adding custom class (e.g. dashed).

def _validate_hatch_pattern(hatch):
    valid_hatch_patterns = set(r'_-+|/\xXoO.*')    # Adding '_' Dashed pattern

After implementing above and menetioning it in hatch.pyi , still getting

MatplotlibDeprecationWarning: hatch must consist of a string of "*+-./OX\ox|" or None, but found the following invalid values "_". Passing invalid values is deprecated since 3.4 and will become an error in 3.9.

Did you search for that warning in the source code?

Did you search for that warning in the source code?

Thanks, it is working after a restart.

I am going to close this in favour of #28072.