This function's docstrings don't match its behavior in two ways.

1. Currently if you pass -1 for x2 or y2 as specified in the docstring it will raise an exception ValueError: x2 must be 0-80

We could update the validation to accept -1 but this same validation helper in used in multiple other places and I'm not sure if all of them share the same support for and meaning of -1 value.

In any case the current code already provides None as a value that does make it get taken as the width and/or height of the image. I think there is no need to have two different argument values that lead to that same behavior. So I've just updated the docstring to indicate None instead of -1.

2. if x2 and y2 are omitted as the docstrings say they can be the code raises an exception TypeError: 'x2' argument required. This is resolved by removing the ARG_REQUIRED flag from for if_required2 argument to the validation helper call.

Tested on a pyportal with this code:

import math

import bitmaptools
import supervisor
import displayio
import ulab.numpy as np

# Make the display context. Change size if you want
display = supervisor.runtime.display
display.auto_refresh = False

# Make the display context
main_group = displayio.Group(scale=4)
display.root_group = main_group

bmp = displayio.Bitmap(320//4, 240//4, 16)
palette = displayio.Palette(16)
palette[0] = 0x000000
palette[1] = 0xffffff
palette[2] = 0xff0000
palette[3] = 0x00ff00
palette[4] = 0x0000ff
palette[5] = 0xffff00
palette[6] = 0x00ffff
palette[7] = 0xff00ff

bmp.fill(3)

tg = displayio.TileGrid(bmp, pixel_shader=palette)
main_group.append(tg)
rows = []
for y in range(bmp.height):
    for x in range(bmp.width):
        rows.extend([(y*bmp.width + x + y) % 8])
array_2d = np.array(rows, dtype=np.uint8)
bitmaptools.arrayblit(bmp, array_2d)
while True:
    pass