Fixed translation here, need someone to approve.

Also edited the description from 40+ = back to 41-60 = back and some edge tests in sample test, to make it more clear.

• More two years later, the bug is still there.
• The 40 expected Middle is right, as the description says 21-40 = middle, 40+ = back.

fixed

fixed for other languages

Because 0 was added as a test case.

fixed

fixed for remaining languages

Wouldn't this fail in a case where the unicode value was 2 or more bytes and the first or last byte happened to be the same?
Examples being:

U+00E9 (LATIN SMALL LETTER E WITH ACUTE, “é”)
• UTF‑8 encoding: 0xC3 0xA9
U+00EA (LATIN SMALL LETTER E WITH CIRCUMFLEX, “ê”)
• UTF‑8 encoding: 0xC3 0xAA

First byte is 0xC3 for both, but they are different runes

Failing Test case example:

Expect(Feast("ér", "êr")).To(BeFalse(), "Testing 'ér' vs 'êr'")

Could have a similar situation at the end, last bytes are equal but the runes are different

fixed

there is no longer such statement in the description

fixed

Approved fork, so this should be fixed.

Haskell fork that fixes the sort order to match the description.

Would it be maybe better to use a .expect("Array is empty") to handle that case for general use?

Added across all languages :-)