Python re matching fails to work for extened unicode range

Question

import re

pat = re.compile(r"[\u20000-\u2A6D6]+")

pat.match("Hello World!")

This will give us a result

<re.Match object; span=(0, 5), match='Hello'>

But in fact, the input string here is fully ASCII which is not from the unicode range.

Is this expected? If so, how to compile those unicode range in practice?

Answer 1

The pattern currently describes a character class consisting of either \u2000, any character in the range 0-\u2A6D or 6.

For python character literals that are wider than 2 bytes, you need to use the escape sequence \U with 8 hex digits:

pat = re.compile(r"[\U00020000-\U0002A6D6]+")