Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Use python stdlib html 5 entity list

Browse Source
This commit is contained in:
Kovid Goyal 2024-09-12 13:45:51 +05:30
parent b4713b3020
commit 35740a2c48
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 7 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
src/calibre/ebooks/html_entities.py
View File

@ -91,7 +91,7 @@ html5_entities = {
'DifferentialD': 'ⅆ', 'DifferentialD': 'ⅆ',
'Dopf': '𝔻', 'Dopf': '𝔻',
'Dot': '¨', 'Dot': '¨',
'DotDot': '⃜\u20dc', 'DotDot': '⃜',
'DotEqual': '≐', 'DotEqual': '≐',
'DoubleContourIntegral': '∯', 'DoubleContourIntegral': '∯',
'DoubleDot': '¨', 'DoubleDot': '¨',
@ -502,7 +502,7 @@ html5_entities = {
'TRADE': '™', 'TRADE': '™',
'TSHcy': 'Ћ', 'TSHcy': 'Ћ',
'TScy': 'Ц', 'TScy': 'Ц',
'Tab': ' ', 'Tab': '\t',
'Tau': 'Τ', 'Tau': 'Τ',
'Tcaron': 'Ť', 'Tcaron': 'Ť',
'Tcedil': 'Ţ', 'Tcedil': 'Ţ',
@ -2133,11 +2133,10 @@ html5_entities = {
} }
if __name__ == '__main__': def generate_entity_lists():
import re import re
from html import entities
from html5lib.constants import entities entities = {k.rstrip(';'): entities.html5[k] for k in entities.html5}
entities = {k.replace(';', ''): entities[k] for k in entities}
lines = [] lines = []
for k in sorted(entities): for k in sorted(entities):
@ -2145,8 +2144,7 @@ if __name__ == '__main__':
with open(__file__, 'r+b') as f: with open(__file__, 'r+b') as f:
raw = f.read().decode('utf-8') raw = f.read().decode('utf-8')
raw = re.sub(r'^# ENTITY_DATA {{{.+^# }}}', pat = re.compile(r'^# ENTITY_DATA {{{.+^# }}}', flags=re.M | re.DOTALL)
'# ENTITY_DATA {{{\n' + '\n'.join(lines) + '\n# }}}', raw = pat.sub(lambda m: '# ENTITY_DATA {{{\n' + '\n'.join(lines) + '\n# }}}', raw)
raw, flags=re.M | re.DOTALL)
f.seek(0), f.truncate() f.seek(0), f.truncate()
f.write(raw.encode('utf-8')) f.write(raw.encode('utf-8'))