`is_extended` condition seems wrong

[mokurin000]

pyinterception/src/interception/_keycodes.py

Lines 394 to 396 in f956108

	res.is_extended = (
	bool(((scan_code >> 8) & 0xFF) & 0xE0) or res.vk_code in extended_keys
	)

In the expression, it ensures a single bit of 0xE0 matches to scan_code >> 8, and therefore the & 0xFF makes no sense. I think the later logical and sign should be equal to check

[kennyhml]

Hi,

yes, I believe you are correct. Truthfully I cannot remember whether I implemented it this way to also cover 0xE1.
It also doesnt seem like there exists many scan codes for which the high byte is not 0x00 but nevertheless, will definitely fix this.

Thanks!

[mokurin000]

yes, I believe you are correct. Truthfully I cannot remember whether I implemented it this way to also cover 0xE1.

I think for that the E1 flag should be used instead too? The E0 flag only works for E0
And a number contains 0xE100 must follow num & 0xE000 == 0xE000 too, but checking num & 0xE000 will make 0x8000 pass the test

(However, Pause/Break keys are only used by legacy software now)

It also doesnt seem like there exists many scan codes for which the high byte is not 0x00 but nevertheless, will definitely fix this.

I believe only a sort of VK code requires such manual rule based check, while in the Scan 1 Make table every extended keys comes with 0xE0 or 0xE1