BitSequence comparison issue

[addinh]

It seems the comparison functions for BitSequence may be erroneous? There are nots in the functions for <=, <, >, >= that shouldn't be there, unless I'm understanding the purpose of comparison wrong

pyftdi/pyftdi/bits.py

Lines 234 to 250 in 2ad8a1a

	def __eq__(self, other):
	return self._cmp(other) == 0
	def __ne__(self, other):
	return not self == other
	def __le__(self, other):
	return not self._cmp(other) <= 0
	def __lt__(self, other):
	return not self._cmp(other) < 0
	def __ge__(self, other):
	return not self._cmp(other) >= 0
	def __gt__(self, other):
	return not self._cmp(other) > 0

Adding to this, the base function _cmp(self, other) also seems odd. I'm assuming the comparison is based on int value, but the function compares length first (which would fail in an example case: 101 vs 0010), then compares bit by bit. However, n from enumerate should only have values from 0 to len(self)-1. This makes the function less than always return True if both BitSequences have the same length.

pyftdi/pyftdi/bits.py

Lines 252 to 260 in 2ad8a1a

	def _cmp(self, other):
	# the bit sequence should be of the same length
	ld = len(self) - len(other)
	if ld:
	return ld
	for n, (x, y) in enumerate(zip(self._seq, other.sequence()), start=1):
	if xor(x, y):
	return n
	return 0

A simplier way to compare would be to convert to int and compare directly, for example int(self) and int(other). Again, I may be understanding the purpose wrong, but if comparison deals with the int values of BitSequences then using the conversion function directly should make it simple.

[eblot]

I think you are very right.
This code has not been maintained for over 10 years, and is not part of the CI/CD.

Patches are very welcome - there are even better if you can include some tests.