morpheus65535/bazarr
1
1
Fork 0
mirror of https://github.com/morpheus65535/bazarr.git synced 2025-01-11 01:07:36 +08:00
Code Issues Projects Releases Packages Wiki Activity

improve character encoding detection

Browse source
This commit is contained in:
josdion 2020-03-26 20:10:58 +02:00
parent 92497a8822
commit be0411d50a
1 changed files with 5 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
bazarr/list_subtitles.py
View file

@ -379,10 +379,12 @@ def guess_external_subtitles(dest_folder, subtitles):
try: try:
# to improve performance, use only the first 32K to detect encoding # to improve performance, use only the first 32K to detect encoding
if len(text) > 32768: guess = chardet.detect(text[:32768]) guess = chardet.detect(text[:32768])
else: guess = chardet.detect(text) logging.debug('BAZARR detected encoding %r', guess)
if guess["confidence"] < 0.8: if guess["confidence"] < 0.6:
raise UnicodeError raise UnicodeError
if guess["confidence"] < 0.8 or guess["encoding"] == "ascii":
guess["encoding"] = "utf-8"
text = text.decode(guess["encoding"]) text = text.decode(guess["encoding"])
detected_language = guess_language(text) detected_language = guess_language(text)
except UnicodeError: except UnicodeError: