I believe the intention of this helper function is to have if_required1 be relvant for x1 and y1 and have if_required2 be relevant for x2 and y2.

That would mean that it is possible to control the required-ness of x1,y1 and x2,y2 independently.

As it is written now instead if_required1 is relevant for x1 and x2 and if_required2 is relevant for y1 and y2. If this functionality does match the intention then I would propose if_requiredX and if_requiredY as more descriptive variable names.

But I believe the x1,y1 and x2,y2 delineation was likely the original intention as it's plausible there could be a function that requires one coordinate point, but has the second one be optional. And seems less plausible to me that a function would have x1,x2 required and y1,y2 optional or vice versa for instance.

The only place where this helper function is called that the if_required arguments actually differ is here: https://github.com/adafruit/circuitpython/blob/main/shared-bindings/bitmaptools/__init__.c#L722

Using this statement which calls the affected function:

bitmaptools.arrayblit(bmp, array_2d, x2=32, y2=32)

under 10.0.0-alpha.2 results in this exception:

TypeError: 'y1' argument required

using this PR branch instead successfully runs as the docstrings indicate it should if x1 and y1 are omitted.

I do also believe there are other issues with the arrayblit() arg validation and/or docstrings, but it's separate from this issue with the swapped values in the helper function here, it just so happens to make use of this.