<_9YU
� @ s@ d d l m Z d d l m Z m Z Gd d � d e � Z d S)� )�
CharSetProber)� ProbingState� MachineStatec sy e Z d Z d Z d � f d d � Z � f d d � Z e d d � � Z e d d
� � Z d d � Z d
d � Z
� S)�MultiByteCharSetProberz
MultiByteCharSetProber
Nc s> t t | � j d | � d | _ d | _ d d g | _ d S)N� lang_filter� )�superr �__init__�distribution_analyzer� coding_sm�
_last_char)�selfr )� __class__� �/mbcharsetprober.pyr ' s z MultiByteCharSetProber.__init__c sR t t | � j � | j r) | j j � | j r? | j j � d d g | _ d S)Nr )r r �resetr r
r )r
)r r r r - s
z MultiByteCharSetProber.resetc C s
t � d S)N)�NotImplementedError)r
r r r � charset_name5 s z#MultiByteCharSetProber.charset_namec C s
t � d S)N)r )r
r r r �language9 s z MultiByteCharSetProber.languagec C sc xt t | � � D]� } | j j | | � } | t j k rm | j j d | j | j | � t
j | _ Pq | t j
k r� t
j | _ Pq | t j k r | j j � } | d k r� | d | j d <| j j | j | � q | j j | | d | d � | � q W| d | j d <| j t
j k r\| j j � r\| j � | j k r\t
j | _ | j S)Nz!%s %s prober hit error at byte %sr r ���)�range�lenr Z
next_stater �ERROR�logger�debugr r r ZNOT_ME�_stateZITS_MEZFOUND_ITZSTARTZget_current_charlenr r
�feed�stateZ DETECTINGZgot_enough_data�get_confidenceZSHORTCUT_THRESHOLD)r
Zbyte_str�iZ coding_stateZchar_lenr r r r = s. zMultiByteCharSetProber.feedc C s
| j j � S)N)r
r )r
r r r r Z s z%MultiByteCharSetProber.get_confidence) �__name__�
__module__� __qualname__�__doc__r r �propertyr r r r r r )r r r " s r N)Z
charsetproberr Zenumsr r r r r r r �<module> s