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

Remove all punctuation when checking tokens

Browse Source
This commit is contained in:
Kovid Goyal 2022-07-28 22:40:03 +05:30
parent 17b65bbc49
commit 8b01843531
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/calibre/ebooks/metadata/sources/google.py
View File

@ -6,6 +6,7 @@ from __future__ import absolute_import, division, print_function, unicode_litera
import hashlib import hashlib
import re import re
import time import time
import regex
try: try:
from queue import Empty, Queue from queue import Empty, Queue
except ImportError: except ImportError:
@ -185,7 +186,7 @@ def to_metadata(browser, log, entry_, timeout): # {{{
class GoogleBooks(Source): class GoogleBooks(Source):
name = 'Google' name = 'Google'
version = (1, 0, 5) version = (1, 0, 6)
minimum_calibre_version = (2, 80, 0) minimum_calibre_version = (2, 80, 0)
description = _('Downloads metadata and covers from Google Books') description = _('Downloads metadata and covers from Google Books')
@ -376,6 +377,7 @@ class GoogleBooks(Source):
): ):
isbn = check_isbn(identifiers.get('isbn', None)) isbn = check_isbn(identifiers.get('isbn', None))
q = [] q = []
strip_punc_pat = regex.compile(r'[\p{C}|\p{M}|\p{P}|\p{S}|\p{Z}]+', regex.UNICODE)
def to_check_tokens(*tokens): def to_check_tokens(*tokens):
for t in tokens: for t in tokens:
@ -384,7 +386,7 @@ class GoogleBooks(Source):
t = t.lower() t = t.lower()
if t in ('and', 'not', 'the'): if t in ('and', 'not', 'the'):
continue continue
yield t.strip(':') yield strip_punc_pat.sub('', t)
check_tokens = set() check_tokens = set()
if isbn is not None: if isbn is not None: