mirror of
https://gh.wpcy.net/https://github.com/WeblateOrg/weblate.git
synced 2026-04-27 08:52:27 +08:00
35 lines
688 B
Python
Executable file
35 lines
688 B
Python
Executable file
#!/usr/bin/env python
|
|
|
|
# Copyright © Michal Čihař <michal@weblate.org>
|
|
#
|
|
# SPDX-License-Identifier: GPL-3.0-or-later
|
|
|
|
"""
|
|
Generates list of non-word chars.
|
|
|
|
Used in weblate/checks/data.py
|
|
"""
|
|
|
|
import pprint
|
|
import sys
|
|
import unicodedata
|
|
|
|
# Unicode categories to consider non word chars
|
|
CATEGORIES = {"Po", "Ps", "Zs", "Cc", "Sk"}
|
|
# Excluded chars
|
|
EXCLUDES = {
|
|
# Removed to avoid breaking regexp syntax
|
|
"]",
|
|
# We intentionally skip following
|
|
"-",
|
|
# Used in Catalan ŀ
|
|
"·",
|
|
"•",
|
|
}
|
|
pprint.pprint(
|
|
[
|
|
char
|
|
for char in map(chr, range(sys.maxunicode + 1))
|
|
if char not in EXCLUDES and unicodedata.category(char) in CATEGORIES
|
|
]
|
|
)
|
