morpheus65535/bazarr
1
1
Fork 0
mirror of https://github.com/morpheus65535/bazarr.git synced 2025-01-12 09:47:40 +08:00
Code Issues Projects Releases Packages Wiki Activity

Switched to UnicodeDammit instead of chardet which give me better result #37

Browse source
This commit is contained in:
morpheus65535 2018-01-13 23:20:20 -05:00
parent 47261c0c91
commit 98de479389
1 changed files with 3 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
list_subtitles.py
View file

@ -6,7 +6,7 @@ import pycountry
import sqlite3
import ast
import langdetect
import chardet
from bs4 import UnicodeDammit
from itertools import islice
from get_general_settings import *
@ -38,9 +38,9 @@ def store_subtitles(file):
with open(path_replace(os.path.join(os.path.dirname(file), subtitle)), 'r') as f:
text = list(islice(f, 20))
text = ' '.join(text)
encoding = chardet.detect(text)['encoding']
encoding = UnicodeDammit(text)
try:
text = text.decode(encoding)
text = text.decode(encoding.original_encoding)
except Exception as e:
logging.exception('Error trying to detect character encoding for this subtitles file: ' + path_replace(os.path.join(os.path.dirname(file), subtitle)))
else: