CasperSecurity
o
�n�_�0 � @ s� d Z ddlZddlZddlZddlmZ ddlmZmZm Z ddl
m Z ddl m
Z
ddlmZ dd lmZ G d
d � d e�ZdS ) a
Module containing the UniversalDetector detector class, which is the primary
class a user of ``chardet`` should use.
:author: Mark Pilgrim (initial port to Python)
:author: Shy Shalom (original C code)
:author: Dan Blanchard (major refactoring for 3.0)
:author: Ian Cordasco
� N� )�CharSetGroupProber)�
InputState�LanguageFilter� ProbingState)�EscCharSetProber)� Latin1Prober)�MBCSGroupProber)�SBCSGroupProberc @ sn e Zd ZdZdZe�d�Ze�d�Ze�d�Z dddd d
d d d
d�Z
e j fdd�Z
dd� Zdd� Zdd� ZdS )�UniversalDetectoraq
The ``UniversalDetector`` class underlies the ``chardet.detect`` function
and coordinates all of the different charset probers.
To get a ``dict`` containing an encoding and its confidence, you can simply
run:
.. code::
u = UniversalDetector()
u.feed(some_bytes)
u.close()
detected = u.result
g�������?s [�-�]s (|~{)s [�-�]z Windows-1252z Windows-1250z Windows-1251z Windows-1256z Windows-1253z Windows-1255z Windows-1254z Windows-1257)z
iso-8859-1z
iso-8859-2z
iso-8859-5z
iso-8859-6z
iso-8859-7z
iso-8859-8z
iso-8859-9z iso-8859-13c C sN d | _ g | _d | _d | _d | _d | _d | _|| _t� t
�| _ d | _ | �
� d S )N)�_esc_charset_prober�_charset_probers�result�done� _got_data� _input_state�
_last_char� lang_filter�logging� getLogger�__name__�logger�_has_win_bytes�reset)�selfr � r �;/usr/lib/python3/dist-packages/chardet/universaldetector.py�__init__Q s zUniversalDetector.__init__c C sV dddd�| _ d| _d| _d| _tj| _d| _| jr | j� � | j
D ]}|� � q"dS )z�
Reset the UniversalDetector and all of its probers back to their
initial states. This is called by ``__init__``, so you only need to
call this directly in between analyses of different documents.
N� ��encoding�
confidence�languageF� ) r r r r r �
PURE_ASCIIr r r r r
)r �proberr r r r ^ s
�zUniversalDetector.resetc C s6 | j rdS t|�s dS t|t�st|�}| jso|�tj�r%dddd�| _n;|�tj tj
f�r6dddd�| _n*|�d�rCdddd�| _n |�d �rPd
ddd�| _n|�tj tj f�r`d ddd�| _d | _| jd
durod | _ dS | j
tjkr�| j�|�r�tj| _
n| j
tjkr�| j�| j| �r�tj| _
|dd� | _| j
tjkr�| js�t| j�| _| j�|�tjkr�| jj | j� � | jj d�| _d | _ dS dS | j
tjk�r| j s�t | j�g| _ | jt!j"@ r�| j �#t$� � | j �#t%� � | j D ]}|�|�tjk�r
|j |� � |j d�| _d | _ nq�| j&�|��rd | _'dS dS dS )a�
Takes a chunk of a document and feeds it through all of the relevant
charset probers.
After calling ``feed``, you can check the value of the ``done``
attribute to see if you need to continue feeding the
``UniversalDetector`` more data, or if it has made a prediction
(in the ``result`` attribute).
.. note::
You should always call ``close`` when you're done feeding in your
document if ``done`` is not already ``True``.
Nz UTF-8-SIG� �?� r zUTF-32s �� zX-ISO-10646-UCS-4-3412s ��zX-ISO-10646-UCS-4-2143zUTF-16Tr ���)(r �len�
isinstance� bytearrayr �
startswith�codecs�BOM_UTF8r � BOM_UTF32_LE� BOM_UTF32_BE�BOM_LE�BOM_BEr r r$ �HIGH_BYTE_DETECTOR�search� HIGH_BYTE� ESC_DETECTORr � ESC_ASCIIr r r �feedr �FOUND_IT� charset_name�get_confidencer" r
r r �NON_CJK�appendr
r �WIN_BYTE_DETECTORr )r �byte_strr% r r r r8 o s�
��
�
�
��
� �
�
��
�zUniversalDetector.feedc C sf | j r| jS d| _ | js| j�d� nY| jtjkr!dddd�| _nK| jtjkrld}d}d}| j D ]}|s5q0|�
� }||krA|}|}q0|rl|| j krl|j }|j �
� }|�
� }|�d �rd| jrd| j�||�}|||jd�| _| j�� tjkr�| jd
du r�| j�d � | j D ]+}|s�q�t|t�r�|jD ]}| j�d |j |j|�
� � q�q�| j�d |j |j|�
� � q�| jS )
z�
Stop analyzing the current document and come up with a final
prediction.
:returns: The ``result`` attribute, a ``dict`` with the keys
`encoding`, `confidence`, and `language`.
Tzno data received!�asciir& r' r Nr ziso-8859r z no probers hit minimum thresholdz%s %s confidence = %s)r r r r �debugr r r$ r5 r
r; �MINIMUM_THRESHOLDr: �lowerr, r � ISO_WIN_MAP�getr" �getEffectiveLevelr �DEBUGr* r �probers) r �prober_confidence�max_prober_confidence�
max_proberr% r: �lower_charset_namer! � group_proberr r r �close� sn
�
�
��
���zUniversalDetector.closeN)r �
__module__� __qualname__�__doc__rB �re�compiler3 r6 r> rD r �ALLr r r8 rN r r r r r 3 s&
�
mr )rQ r- r rR �charsetgroupproberr �enumsr r r � escproberr � latin1proberr �mbcsgroupproberr �sbcsgroupproberr
�objectr r r r r �<module> s