no log: added failsafe to encoding detection

bazarr/subtitles/indexer/utils.py

@@ -76,25 +76,26 @@ def guess_external_subtitles(dest_folder, subtitles, media_type, previously_inde
                 with open(subtitle_path, 'rb') as f:
                     text = f.read()
 
+                encoding = detect(text)
+                if encoding and 'encoding' in encoding:
                     encoding = detect(text)['encoding']
-                if not encoding:
+                else:
                     logging.debug("BAZARR skipping this subtitles because we can't guess the encoding. "
                                   "It's probably a binary file: " + subtitle_path)
                     continue
-                if 'UTF' in encoding:
-                    text = text.decode('utf-8')
+                text = text.decode(encoding)
+
                 detected_language = guess_language(text)
+
                 # add simplified and traditional chinese detection
                 if detected_language == 'zh':
                     traditional_chinese_fuzzy = [u"繁", u"雙語"]
                     traditional_chinese = [".cht", ".tc", ".zh-tw", ".zht", ".zh-hant", ".zhhant", ".zh_hant",
                                            ".hant", ".big5", ".traditional"]
-                        if str(os.path.splitext(subtitle)[0]).lower().endswith(tuple(traditional_chinese)) or (str(subtitle_path).lower())[:-5] in traditional_chinese_fuzzy:
-                            detected_language == 'zt'
-                else:
-                    text = text.decode(encoding)
+                    if str(os.path.splitext(subtitle)[0]).lower().endswith(tuple(traditional_chinese)) or \
+                            (str(subtitle_path).lower())[:-5] in traditional_chinese_fuzzy:
+                        detected_language = 'zt'
 
-                detected_language = guess_language(text)
                 if detected_language:
                     logging.debug("BAZARR external subtitles detected and guessed this language: " + str(
                         detected_language))
@@ -127,14 +128,13 @@ def guess_external_subtitles(dest_folder, subtitles, media_type, previously_inde
                 with open(subtitle_path, 'rb') as f:
                     text = f.read()
 
+                encoding = detect(text)
+                if encoding and 'encoding' in encoding:
                     encoding = detect(text)['encoding']
-                if not encoding:
+                else:
                     logging.debug("BAZARR skipping this subtitles because we can't guess the encoding. "
                                   "It's probably a binary file: " + subtitle_path)
                     continue
-                if 'UTF' in encoding:
-                    text = text.decode('utf-8')
-                else:
                 text = text.decode(encoding)
 
                 if bool(re.search(hi_regex, text)):