3
OXS) @ s d Z ddlmZ ddlZddlmZ dZddd d
d Zi Zd d
Z dd Z
dd Z G dd de Z
e
dZe
dZe
dZd+ddZdd Zed fd d Zd,d d Zd!d" Zed fd#d$Zd%d& ZG d'd( d(e ZG d)d* d*e ZdS )-a
webencodings
~~~~~~~~~~~~
This is a Python implementation of the `WHATWG Encoding standard
`. See README for details.
:copyright: Copyright 2012 by Simon Sapin
:license: BSD, see LICENSE for details.
)unicode_literalsN )LABELSz0.5.1z
iso-8859-8z mac-cyrillicz mac-romanZcp874)z iso-8859-8-izx-mac-cyrillic macintoshz windows-874c C s | j dj jdS )a9 Transform (only) ASCII letters to lower case: A-Z is mapped to a-z.
:param string: An Unicode string.
:returns: A new Unicode string.
This is used for `ASCII case-insensitive
`_
matching of encoding labels.
The same matching is also used, among other things,
for `CSS keywords `_.
This is different from the :meth:`~py:str.lower` method of Unicode strings
which also affect non-ASCII characters,
sometimes mapping them into the ASCII range:
>>> keyword = u'Bac\N{KELVIN SIGN}ground'
>>> assert keyword.lower() == u'background'
>>> assert ascii_lower(keyword) != keyword.lower()
>>> assert ascii_lower(keyword) == u'bac\N{KELVIN SIGN}ground'
utf8)encodelowerdecode)string r /usr/lib/python3.6/__init__.py ascii_lower# s r
c C sx t | jd} tj| }|dkr$dS tj|}|dkrt|dkrLddlm} ntj||}tj |}t
||}|t|< |S )u<
Look for an encoding by its label.
This is the spec’s `get an encoding
`_ algorithm.
Supported labels are listed there.
:param label: A string.
:returns:
An :class:`Encoding` object, or :obj:`None` for an unknown label.
z
Nzx-user-definedr )
codec_info) r
stripr getCACHEZx_user_definedr PYTHON_NAMEScodecslookupEncoding)Zlabelnameencodingr Z python_namer r r r = s
r c C s. t | dr| S t| }|dkr*td| |S )z
Accept either an encoding object or label.
:param encoding: An :class:`Encoding` object or a label string.
:returns: An :class:`Encoding` object.
:raises: :exc:`~exceptions.LookupError` for an unknown label.
r NzUnknown encoding label: %r)hasattrr LookupError)Zencoding_or_labelr r r r
_get_encoding[ s
r c @ s e Zd ZdZdd Zdd ZdS )r aO Reresents a character encoding such as UTF-8,
that can be used for decoding or encoding.
.. attribute:: name
Canonical name of the encoding
.. attribute:: codec_info
The actual implementation of the encoding,
a stdlib :class:`~codecs.CodecInfo` object.
See :func:`codecs.register`.
c C s || _ || _d S )N)r r )selfr r r r r __init__| s zEncoding.__init__c C s
d| j S )Nz
)r )r r r r __repr__ s zEncoding.__repr__N)__name__
__module__ __qualname____doc__r r r r r r r m s r zutf-8zutf-16lezutf-16bereplacec C s2 t |}t| \}} |p|}|jj| |d |fS )a
Decode a single string.
:param input: A byte string
:param fallback_encoding:
An :class:`Encoding` object or a label string.
The encoding to use if :obj:`input` does note have a BOM.
:param errors: Type of error handling. See :func:`codecs.register`.
:raises: :exc:`~exceptions.LookupError` for an unknown encoding label.
:return:
A ``(output, encoding)`` tuple of an Unicode string
and an :obj:`Encoding`.
r )r _detect_bomr r )inputfallback_encodingerrorsZ bom_encodingr r r r r s r c C sV | j drt| dd fS | j dr4t| dd fS | j drNt| dd fS d| fS )zBReturn (bom_encoding, input), with any BOM removed from the input.s Ns s )
startswith_UTF16LE_UTF16BEUTF8)r$ r r r r# s
r# strictc C s t |jj| |d S )a;
Encode a single string.
:param input: An Unicode string.
:param encoding: An :class:`Encoding` object or a label string.
:param errors: Type of error handling. See :func:`codecs.register`.
:raises: :exc:`~exceptions.LookupError` for an unknown encoding label.
:return: A byte string.
r )r r r )r$ r r& r r r r s r c C s$ t ||}t| |}t|}||fS )a
"Pull"-based decoder.
:param input:
An iterable of byte strings.
The input is first consumed just enough to determine the encoding
based on the precense of a BOM,
then consumed on demand when the return value is.
:param fallback_encoding:
An :class:`Encoding` object or a label string.
The encoding to use if :obj:`input` does note have a BOM.
:param errors: Type of error handling. See :func:`codecs.register`.
:raises: :exc:`~exceptions.LookupError` for an unknown encoding label.
:returns:
An ``(output, encoding)`` tuple.
:obj:`output` is an iterable of Unicode strings,
:obj:`encoding` is the :obj:`Encoding` that is being used.
)IncrementalDecoder_iter_decode_generatornext)r$ r% r& decoder generatorr r r r iter_decode s
r3 c c s |j }t| } xf| D ].}||}|r|jdk s2t|jV |V P qW |ddd}|jdk s`t|jV |rr|V dS x | D ]}||}|r||V q|W |ddd}|r|V dS )zqReturn a generator that first yields the :obj:`Encoding`,
then yields output chukns as Unicode strings.
N T)final)r iterr AssertionError)r$ r1 r chunckoutputr r r r/ s,
r/ c C s t ||j}t| |S )uY
“Pull”-based encoder.
:param input: An iterable of Unicode strings.
:param encoding: An :class:`Encoding` object or a label string.
:param errors: Type of error handling. See :func:`codecs.register`.
:raises: :exc:`~exceptions.LookupError` for an unknown encoding label.
:returns: An iterable of byte strings.
)IncrementalEncoderr _iter_encode_generator)r$ r r& r r r r iter_encode s r< c c s: x | D ]}||}|r|V qW |ddd}|r6|V d S )N T)r5 r )r$ r r8 r9 r r r r; s
r; c @ s$ e Zd ZdZd ddZd
ddZdS ) r. uO
“Push”-based decoder.
:param fallback_encoding:
An :class:`Encoding` object or a label string.
The encoding to use if :obj:`input` does note have a BOM.
:param errors: Type of error handling. See :func:`codecs.register`.
:raises: :exc:`~exceptions.LookupError` for an unknown encoding label.
r" c C s& t || _|| _d| _d | _d | _d S )Nr4 )r _fallback_encoding_errors_buffer_decoderr )r r% r& r r r r s
zIncrementalDecoder.__init__Fc C s~ | j }|dk r|||S | j| }t|\}}|dkrXt|dk rR| rR|| _dS | j}|jj| jj}|| _ || _ |||S )zDecode one chunk of the input.
:param input: A byte string.
:param final:
Indicate that no more input is available.
Must be :obj:`True` if this is the last call.
:returns: An Unicode string.
Nr( r= )
rA r@ r# lenr> r incrementaldecoderr? r r )r r$ r5 r1 r r r r r ' s
zIncrementalDecoder.decodeN)r" )F)r r r r! r r r r r r r. s
r. c @ s e Zd ZdZedfddZdS )r: u
“Push”-based encoder.
:param encoding: An :class:`Encoding` object or a label string.
:param errors: Type of error handling. See :func:`codecs.register`.
:raises: :exc:`~exceptions.LookupError` for an unknown encoding label.
.. method:: encode(input, final=False)
:param input: An Unicode string.
:param final:
Indicate that no more input is available.
Must be :obj:`True` if this is the last call.
:returns: A byte string.
r- c C s t |}|jj|j| _d S )N)r r incrementalencoderr )r r r& r r r r T s zIncrementalEncoder.__init__N)r r r r! r, r r r r r r: C s r: )r" )r" )r! Z
__future__r r Zlabelsr VERSIONr r r
r r objectr r, r* r+ r r# r r3 r/ r< r; r. r: r r r r
s2
3