| Dir : /proc/self/root/opt/saltstack/salt/extras-3.10/chardet/ |
| Server: Linux ngx353.inmotionhosting.com 4.18.0-553.22.1.lve.1.el8.x86_64 #1 SMP Tue Oct 8 15:52:54 UTC 2024 x86_64 IP: 209.182.202.254 |
| Dir : //proc/self/root/opt/saltstack/salt/extras-3.10/chardet/enums.py |
"""
All of the Enums that are used throughout the chardet package.
:author: Dan Blanchard (dan.blanchard@gmail.com)
"""
from enum import Enum, Flag
class InputState:
"""
This enum represents the different states a universal detector can be in.
"""
PURE_ASCII = 0
ESC_ASCII = 1
HIGH_BYTE = 2
class LanguageFilter(Flag):
"""
This enum represents the different language filters we can apply to a
``UniversalDetector``.
"""
NONE = 0x00
CHINESE_SIMPLIFIED = 0x01
CHINESE_TRADITIONAL = 0x02
JAPANESE = 0x04
KOREAN = 0x08
NON_CJK = 0x10
ALL = 0x1F
CHINESE = CHINESE_SIMPLIFIED | CHINESE_TRADITIONAL
CJK = CHINESE | JAPANESE | KOREAN
class ProbingState(Enum):
"""
This enum represents the different states a prober can be in.
"""
DETECTING = 0
FOUND_IT = 1
NOT_ME = 2
class MachineState:
"""
This enum represents the different states a state machine can be in.
"""
START = 0
ERROR = 1
ITS_ME = 2
class SequenceLikelihood:
"""
This enum represents the likelihood of a character following the previous one.
"""
NEGATIVE = 0
UNLIKELY = 1
LIKELY = 2
POSITIVE = 3
@classmethod
def get_num_categories(cls) -> int:
""":returns: The number of likelihood categories in the enum."""
return 4
class CharacterCategory:
"""
This enum represents the different categories language models for
``SingleByteCharsetProber`` put characters into.
Anything less than CONTROL is considered a letter.
"""
UNDEFINED = 255
LINE_BREAK = 254
SYMBOL = 253
DIGIT = 252
CONTROL = 251