a
    Oc7                     @   sN   d dl mZ ddlmZ ddlmZmZmZ edg dZG dd deZ	d	S )
    )
namedtuple   )CharSetProber)CharacterCategoryProbingStateSequenceLikelihoodSingleByteCharSetModel)charset_namelanguagechar_to_order_maplanguage_modeltypical_positive_ratiokeep_ascii_lettersalphabetc                       sb   e Zd ZdZd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  ZS )SingleByteCharSetProber@   i   gffffff?g?FNc                    sL   t    || _|| _|| _d | _d | _d | _d | _d | _	d | _
|   d S N)super__init___model	_reversed_name_prober_last_order_seq_counters_total_seqs_total_char_control_char
_freq_charreset)selfZmodelis_reversedname_prober	__class__ [/var/www/html/megastudio/lib/python3.9/site-packages/pip/_vendor/chardet/sbcharsetprober.pyr   6   s    
z SingleByteCharSetProber.__init__c                    s<   t    d| _dgt  | _d| _d| _d| _d| _	d S )N   r   )
r   r   r   r   get_num_categoriesr   r   r   r   r   r   r"   r$   r%   r   E   s    
zSingleByteCharSetProber.resetc                 C   s   | j r| j jS | jjS r   )r   r	   r   r(   r$   r$   r%   r	   P   s    z$SingleByteCharSetProber.charset_namec                 C   s   | j r| j jS | jjS r   )r   r
   r   r(   r$   r$   r%   r
   V   s    z SingleByteCharSetProber.languagec           	      C   sV  | j js| |}n
| |}|s(| jS | j j}| j j}|D ]}||tj	}|tj
k rf|  jd7  _|| jk r|  jd7  _| j| jk r|  jd7  _| js|| j | }n|| | j }| j|  d7  < || _q<| j j}| jtjkrP| j| jkrP|  }|| jkr(| jd|| tj| _n(|| jk rP| jd||| j tj| _| jS )Nr   z$%s confidence = %s, we have a winnerz8%s confidence = %s, below negative shortcut threshold %s)r   r   filter_international_wordsremove_xml_tagsstater   r   getr   	UNDEFINEDCONTROLr   SAMPLE_SIZEr   r   r   r   r   r	   r   	DETECTINGSB_ENOUGH_REL_THRESHOLDget_confidencePOSITIVE_SHORTCUT_THRESHOLDloggerdebugFOUND_IT_stateNEGATIVE_SHORTCUT_THRESHOLDNOT_ME)	r   byte_strr   r   charorderZlm_catr	   
confidencer$   r$   r%   feed\   sL    



zSingleByteCharSetProber.feedc                 C   sn   d}| j dkrj| jtj d| jtj   | j  | jj }|| j| j  | j }|| j	 | j }|dkrjd}|S )Ng{Gz?r   g      ?g      ?gGz?)
r   r   r   POSITIVELIKELYr   r   r   r   r   )r   rr$   r$   r%   r2      s    

z&SingleByteCharSetProber.get_confidence)FN)__name__
__module____qualname__r/   r1   r3   r8   r   r   propertyr	   r
   r>   r2   __classcell__r$   r$   r"   r%   r   0   s   

2r   N)
collectionsr   charsetproberr   enumsr   r   r   r   r   r$   r$   r$   r%   <module>   s   