Mercurial Repositories > eric / file comparison
comparison: ThirdParty/CharDet/chardet/charsetprober.py
ThirdParty/CharDet/chardet/charsetprober.py
- changeset 12
- 1d8dd9706f46
- parent 0
- de9c2efb9d02
- child 3537
- 7662053c3906
equal
deleted
inserted
replaced
| 11:b0996e4a289e | 12:1d8dd9706f46 |
|---|---|
| 24 # License along with this library; if not, write to the Free Software | 24 # License along with this library; if not, write to the Free Software |
| 25 # Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA | 25 # Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA |
| 26 # 02110-1301 USA | 26 # 02110-1301 USA |
| 27 ######################### END LICENSE BLOCK ######################### | 27 ######################### END LICENSE BLOCK ######################### |
| 28 | 28 |
| 29 import constants, re | 29 from . import constants |
| 30 import re | |
| 30 | 31 |
| 31 class CharSetProber: | 32 class CharSetProber: |
| 32 def __init__(self): | 33 def __init__(self): |
| 33 pass | 34 pass |
| 34 | 35 |
| 46 | 47 |
| 47 def get_confidence(self): | 48 def get_confidence(self): |
| 48 return 0.0 | 49 return 0.0 |
| 49 | 50 |
| 50 def filter_high_bit_only(self, aBuf): | 51 def filter_high_bit_only(self, aBuf): |
| 51 aBuf = re.sub(r'([\x00-\x7F])+', ' ', aBuf) | 52 aBuf = re.sub(b'([\x00-\x7F])+', b' ', aBuf) |
| 52 return aBuf | 53 return aBuf |
| 53 | 54 |
| 54 def filter_without_english_letters(self, aBuf): | 55 def filter_without_english_letters(self, aBuf): |
| 55 aBuf = re.sub(r'([A-Za-z])+', ' ', aBuf) | 56 aBuf = re.sub(b'([A-Za-z])+', b' ', aBuf) |
| 56 return aBuf | 57 return aBuf |
| 57 | 58 |
| 58 def filter_with_english_letters(self, aBuf): | 59 def filter_with_english_letters(self, aBuf): |
| 59 # TODO | 60 # TODO |
| 60 return aBuf | 61 return aBuf |
