3
`9Y0 @ s d Z ddlZddlZddlZddlmZ ddlmZmZm Z ddl
m Z ddl m
Z
ddlmZ dd lmZ G d
d d eZdS ) 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)
InputStateLanguageFilter ProbingState)EscCharSetProber) Latin1Prober)MBCSGroupProber)SBCSGroupProberc @ sn e Zd ZdZdZejdZejdZejdZ dddd d
d d d
dZ
e j fddZ
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 | _|| _tj t
| _ d | _ | j
d S )N)_esc_charset_prober_charset_probersresultdone _got_data _input_state
_last_char lang_filterloggingZ getLogger__name__logger_has_win_bytesreset)selfr r '/usr/lib/python3.6/universaldetector.py__init__Q s zUniversalDetector.__init__c C sZ dddd| _ d| _d| _d| _tj| _d| _| jr>| jj x| j
D ] }|j qFW 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.
Ng )encoding
confidencelanguageF ) r r r r r
PURE_ASCIIr r r r r
)r proberr r r r ^ s
zUniversalDetector.resetc C s> | j r
dS t|sdS t|ts(t|}| js|jtjrJdddd| _nv|jtj tj
frldddd| _nT|jdrdddd| _n:|jd rd
ddd| _n |jtj tj frd ddd| _d | _| jd
dk rd | _ dS | j
tjkr.| jj|rtj| _
n*| j
tjkr.| jj| j| r.tj| _
|dd | _| j
tjkr| js^t| j| _| jj|tjkr:| jj | jj | jj d| _d | _ n| j
tjkr:| j st | jg| _ | jt!j"@ r| j j#t$ | j j#t% x@| j D ]6}|j|tjkr|j |j |j d| _d | _ P qW | j&j|r:d | _'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-SIGg ? )r r r zUTF-32s zX-ISO-10646-UCS-4-3412s zX-ISO-10646-UCS-4-2143zUTF-16Tr r )(r len
isinstance bytearrayr
startswithcodecsBOM_UTF8r BOM_UTF32_LE BOM_UTF32_BEBOM_LEBOM_BEr r r! HIGH_BYTE_DETECTORsearch HIGH_BYTE ESC_DETECTORr Z ESC_ASCIIr r r feedr ZFOUND_IT charset_nameget_confidencer r
r r ZNON_CJKappendr
r WIN_BYTE_DETECTORr )r Zbyte_strr" r r r r3 o s|
zUniversalDetector.feedc C s | j r | jS d| _ | js&| jjd n| jtjkrBdddd| _n| jtjkrd}d}d}x,| j D ]"}|slqb|j
}||krb|}|}qbW |r|| j kr|j }|j j
}|j
}|jd r| jr| jj||}|||jd| _| jj tjkrz| jd
dkrz| jjd xn| j D ]d}|s qt|trZxF|jD ] }| jjd |j |j|j
q4W n| jjd |j |j|j
qW | 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!asciig ?r# )r r r Ng ziso-8859r z no probers hit minimum thresholdz%s %s confidence = %s)r r r r debugr r r! r1 r
r5 MINIMUM_THRESHOLDr4 lowerr( r ISO_WIN_MAPgetr ZgetEffectiveLevelr DEBUGr& r Zprobers) r Zprober_confidenceZmax_prober_confidenceZ
max_proberr" r4 Zlower_charset_namer Z group_proberr r r close s`
zUniversalDetector.closeN)r
__module__ __qualname____doc__r: recompiler/ r2 r7 r< r ZALLr r r3 r? r r r r r 3 s"
mr )rB r) r rC Zcharsetgroupproberr Zenumsr r r Z escproberr Z latin1proberr Zmbcsgroupproberr Zsbcsgroupproberr
objectr r r r r $ s