mirrors/kitty
1
0
Fork 0
mirror of https://github.com/kovidgoyal/kitty synced 2026-07-18 22:44:50 +02:00
Code Issues Packages Projects Releases Wiki Activity

Ignore the non-characters from the unicode standard in addition to ignoring the control characters

Browse Source
This commit is contained in:
Kovid Goyal
2018-06-14 10:20:13 +05:30
parent a157f3c5e3
commit 61dd52b50f
7 changed files with 28623 additions and 27403 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
gen-wcwidth.py
View File

@@ -15,6 +15,12 @@ from urllib.request import urlopen
os.chdir(os.path.dirname(os.path.abspath(__file__)))
non_characters = frozenset(range(0xfffe, 0x10ffff, 0x10000))
non_characters |= frozenset(range(0xffff, 0x10ffff + 1, 0x10000))
non_characters |= frozenset(range(0xfdd0, 0xfdf0))
if len(non_characters) != 66:
raise SystemExit('non_characters table incorrect')
def get_data(fname, folder='UCD'):
url = f'https://www.unicode.org/Public/{folder}/latest/{fname}'
@@ -189,11 +195,12 @@ def gen_emoji():
p('\treturn false;\n}')
def category_test(name, p, classes, comment, static=False):
def category_test(name, p, classes, comment, static=False, extra_chars=frozenset()):
static = 'static inline ' if static else ''
chars = set()
for c in classes:
chars |= class_maps[c]
chars |= extra_chars
p(f'{static}bool\n{name}(char_type code) {{')
p(f'\t// {comment} ({len(chars)} codepoints)' + ' {{' '{')
p('\tswitch(code) {')
@@ -245,7 +252,7 @@ def gen_ucd():
with create_header('kitty/unicode-data.c') as p:
p('#include "unicode-data.h"')
category_test('is_combining_char', p, {c for c in class_maps if c.startswith('M')}, 'M category (marks)')
category_test('is_ignored_char', p, 'Cc Cf Cs'.split(), 'Control characters (Cc Cf Cs)')
category_test('is_ignored_char', p, 'Cc Cf Cs'.split(), 'Control characters and non-characters', extra_chars=non_characters)
category_test('is_word_char', p, {c for c in class_maps if c[0] in 'LN'}, 'L and N categories')
category_test('is_CZ_category', p, cz, 'C and Z categories')
category_test('is_P_category', p, {c for c in class_maps if c[0] == 'P'}, 'P category (punctuation)')
@@ -258,7 +265,7 @@ def gen_ucd():
p('combining_type mark_for_codepoint(char_type c) {')
rmap = codepoint_to_mark_map(p, mark_map)
p('}\n')
if rmap[0xfe0e] != 1275:
if rmap[0xfe0e] != 1280:
raise ValueError('The mark for 0xfe0e has changed, you have to update VS15 to {} and VS16 to {} in unicode-data.h'.format(
rmap[0xfe0e], rmap[0xfe0f]
))