I have a CP932 encoded file which contains CP932 specific Kanji character (such as U+9AD9). When I try to detect the encoding by using chardet.detect() function, I get the following result:
{'encoding': None, 'confidence': 0.0, 'language': None}

If I use chardet.detect_all() function, I get different result.
[{'encoding': 'SHIFT_JIS', 'confidence': 0.9472926785393057, 'language': 'Japanese'}]

Both are different from the expected result, 'CP932'.