<_9Y� � @ s� d d l m Z m Z d d l m Z d d l m Z d d l m Z d d l m
Z
d d l m Z Gd d � d e � Z
d S)
� )� ProbingState� MachineState)�MultiByteCharSetProber)�CodingStateMachine)�EUCJPDistributionAnalysis)�EUCJPContextAnalysis)�EUCJP_SM_MODELc sp e Z d Z � f d d � Z � f d d � Z e d d � � Z e d d � � Z d d
� Z d d � Z � S)
� EUCJPProberc sH t t | � j � t t � | _ t � | _ t � | _ | j
� d S)N) �superr �__init__r r � coding_smr �distribution_analyzerr �context_analyzer�reset)�self)� __class__� �/eucjpprober.pyr % s
zEUCJPProber.__init__c s$ t t | � j � | j j � d S)N)r
r r r )r )r r r r , s zEUCJPProber.resetc C s d S)NzEUC-JPr )r r r r � charset_name0 s zEUCJPProber.charset_namec C s d S)NZJapaneser )r r r r �language4 s zEUCJPProber.languagec C s� xAt 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 | � | j j | j | � q | j j | | d | d � | � | 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 %s� r ���)�range�lenr Z
next_stater �ERROR�logger�debugr r r ZNOT_ME�_stateZITS_MEZFOUND_ITZSTARTZget_current_charlenZ
_last_charr �feedr
�stateZ DETECTINGZgot_enough_data�get_confidenceZSHORTCUT_THRESHOLD)r Zbyte_str�iZ coding_stateZchar_lenr r r r 8 s4 zEUCJPProber.feedc C s+ | j j � } | j j � } t | | � S)N)r r r
�max)r Z context_confZ distrib_confr r r r Y s zEUCJPProber.get_confidence)
�__name__�
__module__� __qualname__r r �propertyr r r r r r )r r r $ s !r N)Zenumsr r Zmbcharsetproberr Zcodingstatemachiner Zchardistributionr Zjpcntxr Zmbcssmr r r r r r �<module> s