mirror of
https://github.com/pre-commit/pre-commit-hooks.git
synced 2026-04-17 00:50:18 +00:00
[pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
This commit is contained in:
pre-commit-ci[bot]
parent
72ad6dc953
commit
f4cd1ba0d6
813 changed files with 66015 additions and 58839 deletions
|
|
@ -1,4 +1,3 @@
|
|||
# coding: utf-8
|
||||
"""
|
||||
|
||||
webencodings
|
||||
|
|
@ -11,8 +10,7 @@
|
|||
:license: BSD, see LICENSE for details.
|
||||
|
||||
"""
|
||||
|
||||
from __future__ import unicode_literals
|
||||
from __future__ import annotations
|
||||
|
||||
import codecs
|
||||
|
||||
|
|
@ -27,7 +25,8 @@ PYTHON_NAMES = {
|
|||
'iso-8859-8-i': 'iso-8859-8',
|
||||
'x-mac-cyrillic': 'mac-cyrillic',
|
||||
'macintosh': 'mac-roman',
|
||||
'windows-874': 'cp874'}
|
||||
'windows-874': 'cp874',
|
||||
}
|
||||
|
||||
CACHE = {}
|
||||
|
||||
|
|
@ -106,7 +105,7 @@ def _get_encoding(encoding_or_label):
|
|||
return encoding
|
||||
|
||||
|
||||
class Encoding(object):
|
||||
class Encoding:
|
||||
"""Reresents a character encoding such as UTF-8,
|
||||
that can be used for decoding or encoding.
|
||||
|
||||
|
|
@ -121,6 +120,7 @@ class Encoding(object):
|
|||
See :func:`codecs.register`.
|
||||
|
||||
"""
|
||||
|
||||
def __init__(self, name, codec_info):
|
||||
self.name = name
|
||||
self.codec_info = codec_info
|
||||
|
|
@ -269,7 +269,7 @@ def _iter_encode_generator(input, encode):
|
|||
yield output
|
||||
|
||||
|
||||
class IncrementalDecoder(object):
|
||||
class IncrementalDecoder:
|
||||
"""
|
||||
“Push”-based decoder.
|
||||
|
||||
|
|
@ -280,6 +280,7 @@ class IncrementalDecoder(object):
|
|||
:raises: :exc:`~exceptions.LookupError` for an unknown encoding label.
|
||||
|
||||
"""
|
||||
|
||||
def __init__(self, fallback_encoding, errors='replace'):
|
||||
# Fail early if `encoding` is an invalid label.
|
||||
self._fallback_encoding = _get_encoding(fallback_encoding)
|
||||
|
|
@ -320,7 +321,7 @@ class IncrementalDecoder(object):
|
|||
return decoder(input, final)
|
||||
|
||||
|
||||
class IncrementalEncoder(object):
|
||||
class IncrementalEncoder:
|
||||
"""
|
||||
“Push”-based encoder.
|
||||
|
||||
|
|
@ -337,6 +338,7 @@ class IncrementalEncoder(object):
|
|||
:returns: A byte string.
|
||||
|
||||
"""
|
||||
|
||||
def __init__(self, encoding=UTF8, errors='strict'):
|
||||
encoding = _get_encoding(encoding)
|
||||
self.encode = encoding.codec_info.incrementalencoder(errors).encode
|
||||
|
|
|
|||
|
|
@ -9,223 +9,223 @@
|
|||
:license: BSD, see LICENSE for details.
|
||||
|
||||
"""
|
||||
|
||||
# XXX Do not edit!
|
||||
# This file is automatically generated by mklabels.py
|
||||
from __future__ import annotations
|
||||
|
||||
LABELS = {
|
||||
'unicode-1-1-utf-8': 'utf-8',
|
||||
'utf-8': 'utf-8',
|
||||
'utf8': 'utf-8',
|
||||
'866': 'ibm866',
|
||||
'cp866': 'ibm866',
|
||||
'csibm866': 'ibm866',
|
||||
'ibm866': 'ibm866',
|
||||
'csisolatin2': 'iso-8859-2',
|
||||
'iso-8859-2': 'iso-8859-2',
|
||||
'iso-ir-101': 'iso-8859-2',
|
||||
'iso8859-2': 'iso-8859-2',
|
||||
'iso88592': 'iso-8859-2',
|
||||
'iso_8859-2': 'iso-8859-2',
|
||||
'iso_8859-2:1987': 'iso-8859-2',
|
||||
'l2': 'iso-8859-2',
|
||||
'latin2': 'iso-8859-2',
|
||||
'csisolatin3': 'iso-8859-3',
|
||||
'iso-8859-3': 'iso-8859-3',
|
||||
'iso-ir-109': 'iso-8859-3',
|
||||
'iso8859-3': 'iso-8859-3',
|
||||
'iso88593': 'iso-8859-3',
|
||||
'iso_8859-3': 'iso-8859-3',
|
||||
'iso_8859-3:1988': 'iso-8859-3',
|
||||
'l3': 'iso-8859-3',
|
||||
'latin3': 'iso-8859-3',
|
||||
'csisolatin4': 'iso-8859-4',
|
||||
'iso-8859-4': 'iso-8859-4',
|
||||
'iso-ir-110': 'iso-8859-4',
|
||||
'iso8859-4': 'iso-8859-4',
|
||||
'iso88594': 'iso-8859-4',
|
||||
'iso_8859-4': 'iso-8859-4',
|
||||
'iso_8859-4:1988': 'iso-8859-4',
|
||||
'l4': 'iso-8859-4',
|
||||
'latin4': 'iso-8859-4',
|
||||
'csisolatincyrillic': 'iso-8859-5',
|
||||
'cyrillic': 'iso-8859-5',
|
||||
'iso-8859-5': 'iso-8859-5',
|
||||
'iso-ir-144': 'iso-8859-5',
|
||||
'iso8859-5': 'iso-8859-5',
|
||||
'iso88595': 'iso-8859-5',
|
||||
'iso_8859-5': 'iso-8859-5',
|
||||
'iso_8859-5:1988': 'iso-8859-5',
|
||||
'arabic': 'iso-8859-6',
|
||||
'asmo-708': 'iso-8859-6',
|
||||
'csiso88596e': 'iso-8859-6',
|
||||
'csiso88596i': 'iso-8859-6',
|
||||
'csisolatinarabic': 'iso-8859-6',
|
||||
'ecma-114': 'iso-8859-6',
|
||||
'iso-8859-6': 'iso-8859-6',
|
||||
'iso-8859-6-e': 'iso-8859-6',
|
||||
'iso-8859-6-i': 'iso-8859-6',
|
||||
'iso-ir-127': 'iso-8859-6',
|
||||
'iso8859-6': 'iso-8859-6',
|
||||
'iso88596': 'iso-8859-6',
|
||||
'iso_8859-6': 'iso-8859-6',
|
||||
'iso_8859-6:1987': 'iso-8859-6',
|
||||
'csisolatingreek': 'iso-8859-7',
|
||||
'ecma-118': 'iso-8859-7',
|
||||
'elot_928': 'iso-8859-7',
|
||||
'greek': 'iso-8859-7',
|
||||
'greek8': 'iso-8859-7',
|
||||
'iso-8859-7': 'iso-8859-7',
|
||||
'iso-ir-126': 'iso-8859-7',
|
||||
'iso8859-7': 'iso-8859-7',
|
||||
'iso88597': 'iso-8859-7',
|
||||
'iso_8859-7': 'iso-8859-7',
|
||||
'iso_8859-7:1987': 'iso-8859-7',
|
||||
'sun_eu_greek': 'iso-8859-7',
|
||||
'csiso88598e': 'iso-8859-8',
|
||||
'csisolatinhebrew': 'iso-8859-8',
|
||||
'hebrew': 'iso-8859-8',
|
||||
'iso-8859-8': 'iso-8859-8',
|
||||
'iso-8859-8-e': 'iso-8859-8',
|
||||
'iso-ir-138': 'iso-8859-8',
|
||||
'iso8859-8': 'iso-8859-8',
|
||||
'iso88598': 'iso-8859-8',
|
||||
'iso_8859-8': 'iso-8859-8',
|
||||
'iso_8859-8:1988': 'iso-8859-8',
|
||||
'visual': 'iso-8859-8',
|
||||
'csiso88598i': 'iso-8859-8-i',
|
||||
'iso-8859-8-i': 'iso-8859-8-i',
|
||||
'logical': 'iso-8859-8-i',
|
||||
'csisolatin6': 'iso-8859-10',
|
||||
'iso-8859-10': 'iso-8859-10',
|
||||
'iso-ir-157': 'iso-8859-10',
|
||||
'iso8859-10': 'iso-8859-10',
|
||||
'iso885910': 'iso-8859-10',
|
||||
'l6': 'iso-8859-10',
|
||||
'latin6': 'iso-8859-10',
|
||||
'iso-8859-13': 'iso-8859-13',
|
||||
'iso8859-13': 'iso-8859-13',
|
||||
'iso885913': 'iso-8859-13',
|
||||
'iso-8859-14': 'iso-8859-14',
|
||||
'iso8859-14': 'iso-8859-14',
|
||||
'iso885914': 'iso-8859-14',
|
||||
'csisolatin9': 'iso-8859-15',
|
||||
'iso-8859-15': 'iso-8859-15',
|
||||
'iso8859-15': 'iso-8859-15',
|
||||
'iso885915': 'iso-8859-15',
|
||||
'iso_8859-15': 'iso-8859-15',
|
||||
'l9': 'iso-8859-15',
|
||||
'iso-8859-16': 'iso-8859-16',
|
||||
'cskoi8r': 'koi8-r',
|
||||
'koi': 'koi8-r',
|
||||
'koi8': 'koi8-r',
|
||||
'koi8-r': 'koi8-r',
|
||||
'koi8_r': 'koi8-r',
|
||||
'koi8-u': 'koi8-u',
|
||||
'csmacintosh': 'macintosh',
|
||||
'mac': 'macintosh',
|
||||
'macintosh': 'macintosh',
|
||||
'x-mac-roman': 'macintosh',
|
||||
'dos-874': 'windows-874',
|
||||
'iso-8859-11': 'windows-874',
|
||||
'iso8859-11': 'windows-874',
|
||||
'iso885911': 'windows-874',
|
||||
'tis-620': 'windows-874',
|
||||
'windows-874': 'windows-874',
|
||||
'cp1250': 'windows-1250',
|
||||
'windows-1250': 'windows-1250',
|
||||
'x-cp1250': 'windows-1250',
|
||||
'cp1251': 'windows-1251',
|
||||
'windows-1251': 'windows-1251',
|
||||
'x-cp1251': 'windows-1251',
|
||||
'ansi_x3.4-1968': 'windows-1252',
|
||||
'ascii': 'windows-1252',
|
||||
'cp1252': 'windows-1252',
|
||||
'cp819': 'windows-1252',
|
||||
'csisolatin1': 'windows-1252',
|
||||
'ibm819': 'windows-1252',
|
||||
'iso-8859-1': 'windows-1252',
|
||||
'iso-ir-100': 'windows-1252',
|
||||
'iso8859-1': 'windows-1252',
|
||||
'iso88591': 'windows-1252',
|
||||
'iso_8859-1': 'windows-1252',
|
||||
'iso_8859-1:1987': 'windows-1252',
|
||||
'l1': 'windows-1252',
|
||||
'latin1': 'windows-1252',
|
||||
'us-ascii': 'windows-1252',
|
||||
'windows-1252': 'windows-1252',
|
||||
'x-cp1252': 'windows-1252',
|
||||
'cp1253': 'windows-1253',
|
||||
'windows-1253': 'windows-1253',
|
||||
'x-cp1253': 'windows-1253',
|
||||
'cp1254': 'windows-1254',
|
||||
'csisolatin5': 'windows-1254',
|
||||
'iso-8859-9': 'windows-1254',
|
||||
'iso-ir-148': 'windows-1254',
|
||||
'iso8859-9': 'windows-1254',
|
||||
'iso88599': 'windows-1254',
|
||||
'iso_8859-9': 'windows-1254',
|
||||
'iso_8859-9:1989': 'windows-1254',
|
||||
'l5': 'windows-1254',
|
||||
'latin5': 'windows-1254',
|
||||
'windows-1254': 'windows-1254',
|
||||
'x-cp1254': 'windows-1254',
|
||||
'cp1255': 'windows-1255',
|
||||
'windows-1255': 'windows-1255',
|
||||
'x-cp1255': 'windows-1255',
|
||||
'cp1256': 'windows-1256',
|
||||
'windows-1256': 'windows-1256',
|
||||
'x-cp1256': 'windows-1256',
|
||||
'cp1257': 'windows-1257',
|
||||
'windows-1257': 'windows-1257',
|
||||
'x-cp1257': 'windows-1257',
|
||||
'cp1258': 'windows-1258',
|
||||
'windows-1258': 'windows-1258',
|
||||
'x-cp1258': 'windows-1258',
|
||||
'x-mac-cyrillic': 'x-mac-cyrillic',
|
||||
'x-mac-ukrainian': 'x-mac-cyrillic',
|
||||
'chinese': 'gbk',
|
||||
'csgb2312': 'gbk',
|
||||
'csiso58gb231280': 'gbk',
|
||||
'gb2312': 'gbk',
|
||||
'gb_2312': 'gbk',
|
||||
'gb_2312-80': 'gbk',
|
||||
'gbk': 'gbk',
|
||||
'iso-ir-58': 'gbk',
|
||||
'x-gbk': 'gbk',
|
||||
'gb18030': 'gb18030',
|
||||
'hz-gb-2312': 'hz-gb-2312',
|
||||
'big5': 'big5',
|
||||
'big5-hkscs': 'big5',
|
||||
'cn-big5': 'big5',
|
||||
'csbig5': 'big5',
|
||||
'x-x-big5': 'big5',
|
||||
'unicode-1-1-utf-8': 'utf-8',
|
||||
'utf-8': 'utf-8',
|
||||
'utf8': 'utf-8',
|
||||
'866': 'ibm866',
|
||||
'cp866': 'ibm866',
|
||||
'csibm866': 'ibm866',
|
||||
'ibm866': 'ibm866',
|
||||
'csisolatin2': 'iso-8859-2',
|
||||
'iso-8859-2': 'iso-8859-2',
|
||||
'iso-ir-101': 'iso-8859-2',
|
||||
'iso8859-2': 'iso-8859-2',
|
||||
'iso88592': 'iso-8859-2',
|
||||
'iso_8859-2': 'iso-8859-2',
|
||||
'iso_8859-2:1987': 'iso-8859-2',
|
||||
'l2': 'iso-8859-2',
|
||||
'latin2': 'iso-8859-2',
|
||||
'csisolatin3': 'iso-8859-3',
|
||||
'iso-8859-3': 'iso-8859-3',
|
||||
'iso-ir-109': 'iso-8859-3',
|
||||
'iso8859-3': 'iso-8859-3',
|
||||
'iso88593': 'iso-8859-3',
|
||||
'iso_8859-3': 'iso-8859-3',
|
||||
'iso_8859-3:1988': 'iso-8859-3',
|
||||
'l3': 'iso-8859-3',
|
||||
'latin3': 'iso-8859-3',
|
||||
'csisolatin4': 'iso-8859-4',
|
||||
'iso-8859-4': 'iso-8859-4',
|
||||
'iso-ir-110': 'iso-8859-4',
|
||||
'iso8859-4': 'iso-8859-4',
|
||||
'iso88594': 'iso-8859-4',
|
||||
'iso_8859-4': 'iso-8859-4',
|
||||
'iso_8859-4:1988': 'iso-8859-4',
|
||||
'l4': 'iso-8859-4',
|
||||
'latin4': 'iso-8859-4',
|
||||
'csisolatincyrillic': 'iso-8859-5',
|
||||
'cyrillic': 'iso-8859-5',
|
||||
'iso-8859-5': 'iso-8859-5',
|
||||
'iso-ir-144': 'iso-8859-5',
|
||||
'iso8859-5': 'iso-8859-5',
|
||||
'iso88595': 'iso-8859-5',
|
||||
'iso_8859-5': 'iso-8859-5',
|
||||
'iso_8859-5:1988': 'iso-8859-5',
|
||||
'arabic': 'iso-8859-6',
|
||||
'asmo-708': 'iso-8859-6',
|
||||
'csiso88596e': 'iso-8859-6',
|
||||
'csiso88596i': 'iso-8859-6',
|
||||
'csisolatinarabic': 'iso-8859-6',
|
||||
'ecma-114': 'iso-8859-6',
|
||||
'iso-8859-6': 'iso-8859-6',
|
||||
'iso-8859-6-e': 'iso-8859-6',
|
||||
'iso-8859-6-i': 'iso-8859-6',
|
||||
'iso-ir-127': 'iso-8859-6',
|
||||
'iso8859-6': 'iso-8859-6',
|
||||
'iso88596': 'iso-8859-6',
|
||||
'iso_8859-6': 'iso-8859-6',
|
||||
'iso_8859-6:1987': 'iso-8859-6',
|
||||
'csisolatingreek': 'iso-8859-7',
|
||||
'ecma-118': 'iso-8859-7',
|
||||
'elot_928': 'iso-8859-7',
|
||||
'greek': 'iso-8859-7',
|
||||
'greek8': 'iso-8859-7',
|
||||
'iso-8859-7': 'iso-8859-7',
|
||||
'iso-ir-126': 'iso-8859-7',
|
||||
'iso8859-7': 'iso-8859-7',
|
||||
'iso88597': 'iso-8859-7',
|
||||
'iso_8859-7': 'iso-8859-7',
|
||||
'iso_8859-7:1987': 'iso-8859-7',
|
||||
'sun_eu_greek': 'iso-8859-7',
|
||||
'csiso88598e': 'iso-8859-8',
|
||||
'csisolatinhebrew': 'iso-8859-8',
|
||||
'hebrew': 'iso-8859-8',
|
||||
'iso-8859-8': 'iso-8859-8',
|
||||
'iso-8859-8-e': 'iso-8859-8',
|
||||
'iso-ir-138': 'iso-8859-8',
|
||||
'iso8859-8': 'iso-8859-8',
|
||||
'iso88598': 'iso-8859-8',
|
||||
'iso_8859-8': 'iso-8859-8',
|
||||
'iso_8859-8:1988': 'iso-8859-8',
|
||||
'visual': 'iso-8859-8',
|
||||
'csiso88598i': 'iso-8859-8-i',
|
||||
'iso-8859-8-i': 'iso-8859-8-i',
|
||||
'logical': 'iso-8859-8-i',
|
||||
'csisolatin6': 'iso-8859-10',
|
||||
'iso-8859-10': 'iso-8859-10',
|
||||
'iso-ir-157': 'iso-8859-10',
|
||||
'iso8859-10': 'iso-8859-10',
|
||||
'iso885910': 'iso-8859-10',
|
||||
'l6': 'iso-8859-10',
|
||||
'latin6': 'iso-8859-10',
|
||||
'iso-8859-13': 'iso-8859-13',
|
||||
'iso8859-13': 'iso-8859-13',
|
||||
'iso885913': 'iso-8859-13',
|
||||
'iso-8859-14': 'iso-8859-14',
|
||||
'iso8859-14': 'iso-8859-14',
|
||||
'iso885914': 'iso-8859-14',
|
||||
'csisolatin9': 'iso-8859-15',
|
||||
'iso-8859-15': 'iso-8859-15',
|
||||
'iso8859-15': 'iso-8859-15',
|
||||
'iso885915': 'iso-8859-15',
|
||||
'iso_8859-15': 'iso-8859-15',
|
||||
'l9': 'iso-8859-15',
|
||||
'iso-8859-16': 'iso-8859-16',
|
||||
'cskoi8r': 'koi8-r',
|
||||
'koi': 'koi8-r',
|
||||
'koi8': 'koi8-r',
|
||||
'koi8-r': 'koi8-r',
|
||||
'koi8_r': 'koi8-r',
|
||||
'koi8-u': 'koi8-u',
|
||||
'csmacintosh': 'macintosh',
|
||||
'mac': 'macintosh',
|
||||
'macintosh': 'macintosh',
|
||||
'x-mac-roman': 'macintosh',
|
||||
'dos-874': 'windows-874',
|
||||
'iso-8859-11': 'windows-874',
|
||||
'iso8859-11': 'windows-874',
|
||||
'iso885911': 'windows-874',
|
||||
'tis-620': 'windows-874',
|
||||
'windows-874': 'windows-874',
|
||||
'cp1250': 'windows-1250',
|
||||
'windows-1250': 'windows-1250',
|
||||
'x-cp1250': 'windows-1250',
|
||||
'cp1251': 'windows-1251',
|
||||
'windows-1251': 'windows-1251',
|
||||
'x-cp1251': 'windows-1251',
|
||||
'ansi_x3.4-1968': 'windows-1252',
|
||||
'ascii': 'windows-1252',
|
||||
'cp1252': 'windows-1252',
|
||||
'cp819': 'windows-1252',
|
||||
'csisolatin1': 'windows-1252',
|
||||
'ibm819': 'windows-1252',
|
||||
'iso-8859-1': 'windows-1252',
|
||||
'iso-ir-100': 'windows-1252',
|
||||
'iso8859-1': 'windows-1252',
|
||||
'iso88591': 'windows-1252',
|
||||
'iso_8859-1': 'windows-1252',
|
||||
'iso_8859-1:1987': 'windows-1252',
|
||||
'l1': 'windows-1252',
|
||||
'latin1': 'windows-1252',
|
||||
'us-ascii': 'windows-1252',
|
||||
'windows-1252': 'windows-1252',
|
||||
'x-cp1252': 'windows-1252',
|
||||
'cp1253': 'windows-1253',
|
||||
'windows-1253': 'windows-1253',
|
||||
'x-cp1253': 'windows-1253',
|
||||
'cp1254': 'windows-1254',
|
||||
'csisolatin5': 'windows-1254',
|
||||
'iso-8859-9': 'windows-1254',
|
||||
'iso-ir-148': 'windows-1254',
|
||||
'iso8859-9': 'windows-1254',
|
||||
'iso88599': 'windows-1254',
|
||||
'iso_8859-9': 'windows-1254',
|
||||
'iso_8859-9:1989': 'windows-1254',
|
||||
'l5': 'windows-1254',
|
||||
'latin5': 'windows-1254',
|
||||
'windows-1254': 'windows-1254',
|
||||
'x-cp1254': 'windows-1254',
|
||||
'cp1255': 'windows-1255',
|
||||
'windows-1255': 'windows-1255',
|
||||
'x-cp1255': 'windows-1255',
|
||||
'cp1256': 'windows-1256',
|
||||
'windows-1256': 'windows-1256',
|
||||
'x-cp1256': 'windows-1256',
|
||||
'cp1257': 'windows-1257',
|
||||
'windows-1257': 'windows-1257',
|
||||
'x-cp1257': 'windows-1257',
|
||||
'cp1258': 'windows-1258',
|
||||
'windows-1258': 'windows-1258',
|
||||
'x-cp1258': 'windows-1258',
|
||||
'x-mac-cyrillic': 'x-mac-cyrillic',
|
||||
'x-mac-ukrainian': 'x-mac-cyrillic',
|
||||
'chinese': 'gbk',
|
||||
'csgb2312': 'gbk',
|
||||
'csiso58gb231280': 'gbk',
|
||||
'gb2312': 'gbk',
|
||||
'gb_2312': 'gbk',
|
||||
'gb_2312-80': 'gbk',
|
||||
'gbk': 'gbk',
|
||||
'iso-ir-58': 'gbk',
|
||||
'x-gbk': 'gbk',
|
||||
'gb18030': 'gb18030',
|
||||
'hz-gb-2312': 'hz-gb-2312',
|
||||
'big5': 'big5',
|
||||
'big5-hkscs': 'big5',
|
||||
'cn-big5': 'big5',
|
||||
'csbig5': 'big5',
|
||||
'x-x-big5': 'big5',
|
||||
'cseucpkdfmtjapanese': 'euc-jp',
|
||||
'euc-jp': 'euc-jp',
|
||||
'x-euc-jp': 'euc-jp',
|
||||
'csiso2022jp': 'iso-2022-jp',
|
||||
'iso-2022-jp': 'iso-2022-jp',
|
||||
'csshiftjis': 'shift_jis',
|
||||
'ms_kanji': 'shift_jis',
|
||||
'shift-jis': 'shift_jis',
|
||||
'shift_jis': 'shift_jis',
|
||||
'sjis': 'shift_jis',
|
||||
'windows-31j': 'shift_jis',
|
||||
'x-sjis': 'shift_jis',
|
||||
'cseuckr': 'euc-kr',
|
||||
'csksc56011987': 'euc-kr',
|
||||
'euc-kr': 'euc-kr',
|
||||
'iso-ir-149': 'euc-kr',
|
||||
'korean': 'euc-kr',
|
||||
'ks_c_5601-1987': 'euc-kr',
|
||||
'ks_c_5601-1989': 'euc-kr',
|
||||
'ksc5601': 'euc-kr',
|
||||
'ksc_5601': 'euc-kr',
|
||||
'windows-949': 'euc-kr',
|
||||
'csiso2022kr': 'iso-2022-kr',
|
||||
'iso-2022-kr': 'iso-2022-kr',
|
||||
'utf-16be': 'utf-16be',
|
||||
'utf-16': 'utf-16le',
|
||||
'utf-16le': 'utf-16le',
|
||||
'x-user-defined': 'x-user-defined',
|
||||
'euc-jp': 'euc-jp',
|
||||
'x-euc-jp': 'euc-jp',
|
||||
'csiso2022jp': 'iso-2022-jp',
|
||||
'iso-2022-jp': 'iso-2022-jp',
|
||||
'csshiftjis': 'shift_jis',
|
||||
'ms_kanji': 'shift_jis',
|
||||
'shift-jis': 'shift_jis',
|
||||
'shift_jis': 'shift_jis',
|
||||
'sjis': 'shift_jis',
|
||||
'windows-31j': 'shift_jis',
|
||||
'x-sjis': 'shift_jis',
|
||||
'cseuckr': 'euc-kr',
|
||||
'csksc56011987': 'euc-kr',
|
||||
'euc-kr': 'euc-kr',
|
||||
'iso-ir-149': 'euc-kr',
|
||||
'korean': 'euc-kr',
|
||||
'ks_c_5601-1987': 'euc-kr',
|
||||
'ks_c_5601-1989': 'euc-kr',
|
||||
'ksc5601': 'euc-kr',
|
||||
'ksc_5601': 'euc-kr',
|
||||
'windows-949': 'euc-kr',
|
||||
'csiso2022kr': 'iso-2022-kr',
|
||||
'iso-2022-kr': 'iso-2022-kr',
|
||||
'utf-16be': 'utf-16be',
|
||||
'utf-16': 'utf-16le',
|
||||
'utf-16le': 'utf-16le',
|
||||
'x-user-defined': 'x-user-defined',
|
||||
}
|
||||
|
|
|
|||
|
|
@ -9,6 +9,7 @@
|
|||
:license: BSD, see LICENSE for details.
|
||||
|
||||
"""
|
||||
from __future__ import annotations
|
||||
|
||||
import json
|
||||
try:
|
||||
|
|
@ -42,15 +43,19 @@ def generate(url):
|
|||
LABELS = {
|
||||
''']
|
||||
labels = [
|
||||
(repr(assert_lower(label)).lstrip('u'),
|
||||
repr(encoding['name']).lstrip('u'))
|
||||
(
|
||||
repr(assert_lower(label)).lstrip('u'),
|
||||
repr(encoding['name']).lstrip('u'),
|
||||
)
|
||||
for category in json.loads(urlopen(url).read().decode('ascii'))
|
||||
for encoding in category['encodings']
|
||||
for label in encoding['labels']]
|
||||
for label in encoding['labels']
|
||||
]
|
||||
max_len = max(len(label) for label, name in labels)
|
||||
parts.extend(
|
||||
' %s:%s %s,\n' % (label, ' ' * (max_len - len(label)), name)
|
||||
for label, name in labels)
|
||||
' {}:{} {},\n'.format(label, ' ' * (max_len - len(label)), name)
|
||||
for label, name in labels
|
||||
)
|
||||
parts.append('}')
|
||||
return ''.join(parts)
|
||||
|
||||
|
|
|
|||
|
|
@ -1,4 +1,3 @@
|
|||
# coding: utf-8
|
||||
"""
|
||||
|
||||
webencodings.tests
|
||||
|
|
@ -10,11 +9,17 @@
|
|||
:license: BSD, see LICENSE for details.
|
||||
|
||||
"""
|
||||
from __future__ import annotations
|
||||
|
||||
from __future__ import unicode_literals
|
||||
|
||||
from . import (lookup, LABELS, decode, encode, iter_decode, iter_encode,
|
||||
IncrementalDecoder, IncrementalEncoder, UTF8)
|
||||
from . import decode
|
||||
from . import encode
|
||||
from . import IncrementalDecoder
|
||||
from . import IncrementalEncoder
|
||||
from . import iter_decode
|
||||
from . import iter_encode
|
||||
from . import LABELS
|
||||
from . import lookup
|
||||
from . import UTF8
|
||||
|
||||
|
||||
def assert_raises(exception, function, *args, **kwargs):
|
||||
|
|
@ -116,20 +121,35 @@ def test_iter_decode():
|
|||
assert iter_decode_to_string([b'hell', b'o'], 'latin1') == 'hello'
|
||||
assert iter_decode_to_string([b'\xc3\xa9'], 'latin1') == 'é'
|
||||
assert iter_decode_to_string([b'\xEF\xBB\xBF\xc3\xa9'], 'latin1') == 'é'
|
||||
assert iter_decode_to_string([
|
||||
b'\xEF\xBB\xBF', b'\xc3', b'\xa9'], 'latin1') == 'é'
|
||||
assert iter_decode_to_string([
|
||||
b'\xEF\xBB\xBF', b'a', b'\xc3'], 'latin1') == 'a\uFFFD'
|
||||
assert iter_decode_to_string([
|
||||
b'', b'\xEF', b'', b'', b'\xBB\xBF\xc3', b'\xa9'], 'latin1') == 'é'
|
||||
assert iter_decode_to_string(
|
||||
[
|
||||
b'\xEF\xBB\xBF', b'\xc3', b'\xa9',
|
||||
], 'latin1',
|
||||
) == 'é'
|
||||
assert iter_decode_to_string(
|
||||
[
|
||||
b'\xEF\xBB\xBF', b'a', b'\xc3',
|
||||
], 'latin1',
|
||||
) == 'a\uFFFD'
|
||||
assert iter_decode_to_string(
|
||||
[
|
||||
b'', b'\xEF', b'', b'', b'\xBB\xBF\xc3', b'\xa9',
|
||||
], 'latin1',
|
||||
) == 'é'
|
||||
assert iter_decode_to_string([b'\xEF\xBB\xBF'], 'latin1') == ''
|
||||
assert iter_decode_to_string([b'\xEF\xBB'], 'latin1') == 'ï»'
|
||||
assert iter_decode_to_string([b'\xFE\xFF\x00\xe9'], 'latin1') == 'é'
|
||||
assert iter_decode_to_string([b'\xFF\xFE\xe9\x00'], 'latin1') == 'é'
|
||||
assert iter_decode_to_string([
|
||||
b'', b'\xFF', b'', b'', b'\xFE\xe9', b'\x00'], 'latin1') == 'é'
|
||||
assert iter_decode_to_string([
|
||||
b'', b'h\xe9', b'llo'], 'x-user-defined') == 'h\uF7E9llo'
|
||||
assert iter_decode_to_string(
|
||||
[
|
||||
b'', b'\xFF', b'', b'', b'\xFE\xe9', b'\x00',
|
||||
], 'latin1',
|
||||
) == 'é'
|
||||
assert iter_decode_to_string(
|
||||
[
|
||||
b'', b'h\xe9', b'llo',
|
||||
], 'x-user-defined',
|
||||
) == 'h\uF7E9llo'
|
||||
|
||||
|
||||
def test_iter_encode():
|
||||
|
|
@ -140,8 +160,13 @@ def test_iter_encode():
|
|||
assert b''.join(iter_encode(['', 'é', '', ''], 'utf-16')) == b'\xe9\x00'
|
||||
assert b''.join(iter_encode(['', 'é', '', ''], 'utf-16le')) == b'\xe9\x00'
|
||||
assert b''.join(iter_encode(['', 'é', '', ''], 'utf-16be')) == b'\x00\xe9'
|
||||
assert b''.join(iter_encode([
|
||||
'', 'h\uF7E9', '', 'llo'], 'x-user-defined')) == b'h\xe9llo'
|
||||
assert b''.join(
|
||||
iter_encode(
|
||||
[
|
||||
'', 'h\uF7E9', '', 'llo',
|
||||
], 'x-user-defined',
|
||||
),
|
||||
) == b'h\xe9llo'
|
||||
|
||||
|
||||
def test_x_user_defined():
|
||||
|
|
|
|||
|
|
@ -1,4 +1,3 @@
|
|||
# coding: utf-8
|
||||
"""
|
||||
|
||||
webencodings.x_user_defined
|
||||
|
|
@ -10,13 +9,12 @@
|
|||
:license: BSD, see LICENSE for details.
|
||||
|
||||
"""
|
||||
|
||||
from __future__ import unicode_literals
|
||||
from __future__ import annotations
|
||||
|
||||
import codecs
|
||||
|
||||
|
||||
### Codec APIs
|
||||
# Codec APIs
|
||||
|
||||
class Codec(codecs.Codec):
|
||||
|
||||
|
|
@ -45,7 +43,7 @@ class StreamReader(Codec, codecs.StreamReader):
|
|||
pass
|
||||
|
||||
|
||||
### encodings module API
|
||||
# encodings module API
|
||||
|
||||
codec_info = codecs.CodecInfo(
|
||||
name='x-user-defined',
|
||||
|
|
@ -58,7 +56,7 @@ codec_info = codecs.CodecInfo(
|
|||
)
|
||||
|
||||
|
||||
### Decoding Table
|
||||
# Decoding Table
|
||||
|
||||
# Python 3:
|
||||
# for c in range(256): print(' %r' % chr(c if c < 128 else c + 0xF700))
|
||||
|
|
@ -321,5 +319,5 @@ decoding_table = (
|
|||
'\uf7ff'
|
||||
)
|
||||
|
||||
### Encoding table
|
||||
# Encoding table
|
||||
encoding_table = codecs.charmap_build(decoding_table)
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue