Sumerian Cuneiform Number 1 (U+12000) accepted as start of identifier

[Mohammed-600]

Bug report

Bug description:

Description
I found a bug where Python's parser accepts the Sumerian Cuneiform Number One (𒁹) at the beginning of a function name. According to Python's rules, identifiers should not start with a digit. Even though this is an ancient digit, it is still a numeric value and should trigger a SyntaxError.

Repetition
I tested this on 3 different installations of Python on my machine, and they all allowed it.

Code Example
def 𒁹():
return "Success"
Python Verison 3.14.2
Reporter: Mohammed Al-Ghazali/ Mohammed-600

CPython versions tested on:

3.14

Operating systems tested on:

Windows

[aisk]

Thank you for reporting this. I'm not an expert on Unicode, but according to the Unicode database: https://www.compart.com/en/unicode/U+12079, the category of "𒁹" is Other Letter rather than Decimal Number, so we shouldn't treat it as a digit.

Other numerals from different languages follow a similar pattern. For example, the CJK character "一" is also marked as Other Letter in Unicode, and it can be used as the first character of an identifier:

>>> def 一(): return 1
...
>>> 一()
1

[StanFromIreland]

This is not a bug, U+12000 has the category Lo (Letter, other). It is not a number, and it is in the xid_start set making it a valid identifier start in Python:

>>> unicodedata.isxidstart('𒁹')
True

You can read more about this in our docs.

[ZeroIntensity]

More importantly, why are you trying to write Python code in Sumerian Cuneiform?