Drawing hatch filled rectangles with differing angles in python

drawing, python

Solution

I have a basic idea although I suspect you will need to do quite a bit more depending how flexible you want the result to be.

from numpy import cos, sin
import numpy as np
import matplotlib.pyplot as plt

def plot_hatches(ax, angle, offset=.1):
    angle_radians = np.radians(angle)
    x = np.linspace(-2, 2, 10)
    for c in np.arange(-2, 2, offset):
        yprime = cos(angle_radians) * c + sin(angle_radians) * x
        xprime = sin(angle_radians) * c - cos(angle_radians) * x
        ax.plot(xprime, yprime, color="k")
    ax.set_ylim(0, 1)
    ax.set_xlim(0, 1)
    return ax


fig, axes = plt.subplots(nrows=4, ncols=4, figsize=(8,8), sharex=True, sharey=True)

for i in range(len(axes.flat)):
    plot_hatches(axes.flat[i], np.random.uniform(0, 90))

fig.subplots_adjust(hspace=0, wspace=0)

There is two things parts here: firstly a function `plot_hatches` which draws hatches over the unit square on the axes `ax`. This is done by taking a single line `x, y=c` and rotating it using rotation matrix to get a `xprime` and `yprime` which are the coordinates of lines at an angle to the x axis with offset `c`. Iterating over several values of `c` covers the unit square, the lines can be made denser by making the `offset` argument smaller.

Secondly we need a method to draw the axes next to each other. This I have done using `subplots`. This returns `fig, axes` the `axes` is an array of axes instances, so we iteratate through them passing them into the function to plot hatches and give it a random angle each time.

EDIT I have changed the plot_hatches code to rotate in an anticlockwise manner (it was clockwise before this edit). This will now produce exactly the image given in the question with the array `[0, -45, 0, 45, 0]`:

Problem

I'm trying to draw a 2D rectangular plate in python. This plate will be split into a variable number of sections, each of these sections will be filled with a hatched pattern. This hatched pattern will have a specified angle. As an example for a rectangle with 5 sections and an array of its section hatch orientation (in degrees) being [0, 45, 0, -45, 0] is shown below. It will need to be able to display any orientation, not just the usual 90, 45, 0, i.e. 33, 74.5 etc. Any idea how I could do this? Essentially I just want to show the orientation in each section, any other methods of expressing the same results would be much appreciated e.g. a single line instead of hatched. Edit (after question was answered): Edited script provided by Greg is shown below. ``` from numpy import cos, sin import numpy as np import matplotlib.pyplot as plt angles = [0,10,20,30,40,50] numberOfSections = len(angles) def plot_hatches(ax, angle, offset=.1): angle_radians = np.radians(angle) x = np.linspace(-1, 1, 10) for c in np.arange(-2, 2, offset): yprime = cos(angle_radians) * c - sin(angle_radians) * x xprime = sin(angle_radians) * c + cos(angle_radians) * x ax.plot(xprime, yprime, color="b", linewidth=2) ax.set_ylim(0, 1) ax.set_xlim(0, 1) return ax fig, axes = plt.subplots(nrows=1, ncols=numberOfSections, figsize=(16,(16/numberOfSections)), sharex=True, sharey=True) for i in range(len(axes.flat)): plot_hatches(axes.flat[i], angles[i]) fig.subplots_adjust(hspace=0, wspace=0) plt.show() ``` Produces an figure shown below. But on inspection the angles do not match the input angles.

Original source