Viewing File: /opt/alt/python38/lib/python3.8/site-packages/chardet/__pycache__/universaldetector.cpython-38.pyc
U
<_9Y0 @ s d Z ddlZddlZddlZddlmZ ddlmZmZm Z ddl
mZ ddlm
Z
ddlmZ dd lmZ G d
d deZdS )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)
InputStateLanguageFilterProbingState)EscCharSetProber)Latin1Prober)MBCSGroupProber)SBCSGroupProberc @ sn e Zd ZdZdZedZedZedZ dddd d
ddd
dZ
ejfddZ
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 [-]zWindows-1252zWindows-1250zWindows-1251zWindows-1256zWindows-1253zWindows-1255zWindows-1254zWindows-1257)z
iso-8859-1z
iso-8859-2z
iso-8859-5z
iso-8859-6z
iso-8859-7z
iso-8859-8z
iso-8859-9ziso-8859-13c C sN d | _ g | _d | _d | _d | _d | _d | _|| _t t
| _d | _|
d S )N)_esc_charset_prober_charset_probersresultdone _got_data_input_state
_last_charlang_filterloggingZ getLogger__name__logger_has_win_bytesreset)selfr r J/opt/alt/python38/lib/python3.8/site-packages/chardet/universaldetector.py__init__Q s zUniversalDetector.__init__c C sV dddd| _ d| _d| _d| _tj| _d| _| jr>| j | j
D ]}| qDdS )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
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|tjrJdddd| _nv|tj tj
frldddd| _nT|drdddd| _n:|d rd
ddd| _n |tjtjfrdddd| _d| _| jd
dk rd| _ dS | j
tjkr.| j|rtj| _
n*| j
tjkr.| j| j| r.tj| _
|dd | _| j
tjkr| js^t| j| _| j|tjkr:| jj| j | jjd| _d| _ n| j
tjkr:| jst | jg| _| jt!j"@ r| j#t$ | j#t% | jD ]:}||tjkr|j| |jd| _d| _ q&q| 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-SIG ? r zUTF-32s zX-ISO-10646-UCS-4-3412s zX-ISO-10646-UCS-4-2143zUTF-16Tr )(r len
isinstance bytearrayr
startswithcodecsBOM_UTF8r BOM_UTF32_LEBOM_UTF32_BEBOM_LEBOM_BEr r r# HIGH_BYTE_DETECTORsearch HIGH_BYTEESC_DETECTORr Z ESC_ASCIIr r r feedr ZFOUND_ITcharset_nameget_confidencer! r
r r ZNON_CJKappendr
r WIN_BYTE_DETECTORr )r Zbyte_strr$ r r r r6 o s
zUniversalDetector.feedc C st | j r| jS d| _ | js&| jd n| jtjkrBdddd| _n| jtjkrd}d}d}| j D ]"}|sjq`|
}||kr`|}|}q`|r|| jkr|j}|j
}|
}|d r| jr| j||}|||jd| _| j tjkrn| jd
dkrn| jd | j D ]`}|sqt|trP|jD ] }| jd|j|j|
q,n| jd|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&