()
| 1030 | # implementation. |
| 1031 | @lru_cache() |
| 1032 | def find_new_unicode_codepoints(): |
| 1033 | new = set() |
| 1034 | characters = [chr(c) for c in range(0x80, 0x11000) |
| 1035 | if not (0xD800 <= c < 0xE000)] |
| 1036 | is_printable = pc.utf8_is_printable(pa.array(characters)).to_pylist() |
| 1037 | for i, c in enumerate(characters): |
| 1038 | if is_printable[i] != c.isprintable(): |
| 1039 | new.add(ord(c)) |
| 1040 | return new |
| 1041 | |
| 1042 | |
| 1043 | # Python claims there are not alpha, not sure why, they are in |
no test coverage detected