jamtur01's picture
Upload folder using huggingface_hub
9c6594c verified
"""
This gives other modules access to the gritty details about characters and the
encodings that use them.
"""
from __future__ import annotations
import html
import itertools
import re
import unicodedata
# These are the encodings we will try to fix in ftfy, in the
# order that they should be tried.
CHARMAP_ENCODINGS = [
"latin-1",
"sloppy-windows-1252",
"sloppy-windows-1251",
"sloppy-windows-1250",
"sloppy-windows-1253",
"sloppy-windows-1254",
"sloppy-windows-1257",
"iso-8859-2",
"macroman",
"cp437",
]
SINGLE_QUOTE_RE = re.compile("[\u02bc\u2018-\u201b]")
DOUBLE_QUOTE_RE = re.compile("[\u201c-\u201f]")
def _build_regexes() -> dict[str, re.Pattern[str]]:
"""
ENCODING_REGEXES contain reasonably fast ways to detect if we
could represent a given string in a given encoding. The simplest one is
the 'ascii' detector, which of course just determines if all characters
are between U+0000 and U+007F.
"""
# Define a regex that matches ASCII text.
encoding_regexes = {"ascii": re.compile("^[\x00-\x7f]*$")}
for encoding in CHARMAP_ENCODINGS:
# Make a sequence of characters that bytes \x80 to \xFF decode to
# in each encoding, as well as byte \x1A, which is used to represent
# the replacement character οΏ½ in the sloppy-* encodings.
byte_range = bytes(list(range(0x80, 0x100)) + [0x1A])
charlist = byte_range.decode(encoding)
# The rest of the ASCII bytes -- bytes \x00 to \x19 and \x1B
# to \x7F -- will decode as those ASCII characters in any encoding we
# support, so we can just include them as ranges. This also lets us
# not worry about escaping regex special characters, because all of
# them are in the \x1B to \x7F range.
regex = f"^[\x00-\x19\x1b-\x7f{charlist}]*$"
encoding_regexes[encoding] = re.compile(regex)
return encoding_regexes
ENCODING_REGEXES = _build_regexes()
def _build_html_entities() -> dict[str, str]:
entities = {}
# Create a dictionary based on the built-in HTML5 entity dictionary.
# Add a limited set of HTML entities that we'll also decode if they've
# been case-folded to uppercase, such as decoding Ñ as "Γ‘".
for name, char in html.entities.html5.items(): # type: ignore
if name.endswith(";"):
entities["&" + name] = char
# Restrict the set of characters we can attempt to decode if their
# name has been uppercased. If we tried to handle all entity names,
# the results would be ambiguous.
if name == name.lower():
name_upper = name.upper()
entity_upper = "&" + name_upper
if html.unescape(entity_upper) == entity_upper:
entities[entity_upper] = char.upper()
return entities
HTML_ENTITY_RE = re.compile(r"&#?[0-9A-Za-z]{1,24};")
HTML_ENTITIES = _build_html_entities()
def possible_encoding(text: str, encoding: str) -> bool:
"""
Given text and a single-byte encoding, check whether that text could have
been decoded from that single-byte encoding.
In other words, check whether it can be encoded in that encoding, possibly
sloppily.
"""
return bool(ENCODING_REGEXES[encoding].match(text))
def _build_control_char_mapping() -> dict[int, None]:
"""
Build a translate mapping that strips likely-unintended control characters.
See :func:`ftfy.fixes.remove_control_chars` for a description of these
codepoint ranges and why they should be removed.
"""
control_chars: dict[int, None] = {}
for i in itertools.chain(
range(0x00, 0x09),
[0x0B],
range(0x0E, 0x20),
[0x7F],
range(0x206A, 0x2070),
[0xFEFF],
range(0xFFF9, 0xFFFD),
):
control_chars[i] = None
return control_chars
CONTROL_CHARS = _build_control_char_mapping()
# Recognize UTF-8 sequences that would be valid if it weren't for a b'\xa0'
# that some Windows-1252 program converted to a plain space.
#
# The smaller values are included on a case-by-case basis, because we don't want
# to decode likely input sequences to unlikely characters. These are the ones
# that *do* form likely characters before 0xa0:
#
# 0xc2 -> U+A0 NO-BREAK SPACE
# 0xc3 -> U+E0 LATIN SMALL LETTER A WITH GRAVE
# 0xc5 -> U+160 LATIN CAPITAL LETTER S WITH CARON
# 0xce -> U+3A0 GREEK CAPITAL LETTER PI
# 0xd0 -> U+420 CYRILLIC CAPITAL LETTER ER
# 0xd9 -> U+660 ARABIC-INDIC DIGIT ZERO
#
# In three-character sequences, we exclude some lead bytes in some cases.
#
# When the lead byte is immediately followed by 0xA0, we shouldn't accept
# a space there, because it leads to some less-likely character ranges:
#
# 0xe0 -> Samaritan script
# 0xe1 -> Mongolian script (corresponds to Latin-1 'Γ‘' which is too common)
#
# We accept 0xe2 and 0xe3, which cover many scripts. Bytes 0xe4 and
# higher point mostly to CJK characters, which we generally don't want to
# decode near Latin lowercase letters.
#
# In four-character sequences, the lead byte must be F0, because that accounts
# for almost all of the usage of high-numbered codepoints (tag characters whose
# UTF-8 starts with the byte F3 are only used in some rare new emoji sequences).
#
# This is meant to be applied to encodings of text that tests true for `is_bad`.
# Any of these could represent characters that legitimately appear surrounded by
# spaces, particularly U+C5 (Γ…), which is a word in multiple languages!
#
# We should consider checking for b'\x85' being converted to ... in the future.
# I've seen it once, but the text still wasn't recoverable.
ALTERED_UTF8_RE = re.compile(
b"[\xc2\xc3\xc5\xce\xd0\xd9][ ]"
b"|[\xe2\xe3][ ][\x80-\x84\x86-\x9f\xa1-\xbf]"
b"|[\xe0-\xe3][\x80-\x84\x86-\x9f\xa1-\xbf][ ]"
b"|[\xf0][ ][\x80-\xbf][\x80-\xbf]"
b"|[\xf0][\x80-\xbf][ ][\x80-\xbf]"
b"|[\xf0][\x80-\xbf][\x80-\xbf][ ]"
)
# This expression matches UTF-8 and CESU-8 sequences where some of the
# continuation bytes have been lost. The byte 0x1a (sometimes written as ^Z) is
# used within ftfy to represent a byte that produced the replacement character
# \ufffd. We don't know which byte it was, but we can at least decode the UTF-8
# sequence as \ufffd instead of failing to re-decode it at all.
#
# In some cases, we allow the ASCII '?' in place of \ufffd, but at most once per
# sequence.
LOSSY_UTF8_RE = re.compile(
b"[\xc2-\xdf][\x1a]"
b"|[\xc2-\xc3][?]"
b"|\xed[\xa0-\xaf][\x1a?]\xed[\xb0-\xbf][\x1a?\x80-\xbf]"
b"|\xed[\xa0-\xaf][\x1a?\x80-\xbf]\xed[\xb0-\xbf][\x1a?]"
b"|[\xe0-\xef][\x1a?][\x1a\x80-\xbf]"
b"|[\xe0-\xef][\x1a\x80-\xbf][\x1a?]"
b"|[\xf0-\xf4][\x1a?][\x1a\x80-\xbf][\x1a\x80-\xbf]"
b"|[\xf0-\xf4][\x1a\x80-\xbf][\x1a?][\x1a\x80-\xbf]"
b"|[\xf0-\xf4][\x1a\x80-\xbf][\x1a\x80-\xbf][\x1a?]"
b"|\x1a"
)
# This regex matches C1 control characters, which occupy some of the positions
# in the Latin-1 character map that Windows assigns to other characters instead.
C1_CONTROL_RE = re.compile(r"[\x80-\x9f]")
# A translate mapping that breaks ligatures made of Latin letters. While
# ligatures may be important to the representation of other languages, in Latin
# letters they tend to represent a copy/paste error. It omits ligatures such
# as Γ¦ that are frequently used intentionally.
#
# This list additionally includes some Latin digraphs that represent two
# characters for legacy encoding reasons, not for typographical reasons.
#
# Ligatures and digraphs may also be separated by NFKC normalization, but that
# is sometimes more normalization than you want.
LIGATURES = {
ord("Δ²"): "IJ", # Dutch ligatures
ord("Δ³"): "ij",
ord("Ε‰"): "ΚΌn", # Afrikaans digraph meant to avoid auto-curled quote
ord("Η±"): "DZ", # Serbian/Croatian digraphs for Cyrillic conversion
ord("Η²"): "Dz",
ord("Η³"): "dz",
ord("Η„"): "DΕ½",
ord("Η…"): "DΕΎ",
ord("Η†"): "dΕΎ",
ord("Η‡"): "LJ",
ord("Lj"): "Lj",
ord("Η‰"): "lj",
ord("NJ"): "NJ",
ord("Η‹"): "Nj",
ord("nj"): "nj",
ord("ff"): "ff", # Latin typographical ligatures
ord("fi"): "fi",
ord("fl"): "fl",
ord("ffi"): "ffi",
ord("ffl"): "ffl",
ord("ο¬…"): "ΕΏt",
ord("st"): "st",
}
def _build_width_map() -> dict[int, str]:
"""
Build a translate mapping that replaces halfwidth and fullwidth forms
with their standard-width forms.
"""
# Though it's not listed as a fullwidth character, we'll want to convert
# U+3000 IDEOGRAPHIC SPACE to U+20 SPACE on the same principle, so start
# with that in the dictionary.
width_map = {0x3000: " "}
for i in range(0xFF01, 0xFFF0):
char = chr(i)
alternate = unicodedata.normalize("NFKC", char)
if alternate != char:
width_map[i] = alternate
return width_map
WIDTH_MAP = _build_width_map()
# Character classes that help us pinpoint embedded mojibake. These can
# include common characters, because we'll also check them for 'badness'.
#
# Though they go on for many lines, the members of this dictionary are
# single concatenated strings.
#
# This code is generated using scripts/char_data_table.py.
UTF8_CLUES: dict[str, str] = {
# Letters that decode to 0xC2 - 0xDF in a Latin-1-like encoding
"utf8_first_of_2": (
"\N{LATIN CAPITAL LETTER A WITH BREVE}" # windows-1250:C3
"\N{LATIN CAPITAL LETTER A WITH CIRCUMFLEX}" # latin-1:C2
"\N{LATIN CAPITAL LETTER A WITH DIAERESIS}" # latin-1:C4
"\N{LATIN CAPITAL LETTER A WITH MACRON}" # windows-1257:C2
"\N{LATIN CAPITAL LETTER A WITH RING ABOVE}" # latin-1:C5
"\N{LATIN CAPITAL LETTER A WITH TILDE}" # latin-1:C3
"\N{LATIN CAPITAL LETTER AE}" # latin-1:C6
"\N{LATIN CAPITAL LETTER C WITH ACUTE}" # windows-1250:C6
"\N{LATIN CAPITAL LETTER C WITH CARON}" # windows-1250:C8
"\N{LATIN CAPITAL LETTER C WITH CEDILLA}" # latin-1:C7
"\N{LATIN CAPITAL LETTER D WITH CARON}" # windows-1250:CF
"\N{LATIN CAPITAL LETTER D WITH STROKE}" # windows-1250:D0
"\N{LATIN CAPITAL LETTER E WITH ACUTE}" # latin-1:C9
"\N{LATIN CAPITAL LETTER E WITH CARON}" # windows-1250:CC
"\N{LATIN CAPITAL LETTER E WITH CIRCUMFLEX}" # latin-1:CA
"\N{LATIN CAPITAL LETTER E WITH DIAERESIS}" # latin-1:CB
"\N{LATIN CAPITAL LETTER E WITH DOT ABOVE}" # windows-1257:CB
"\N{LATIN CAPITAL LETTER E WITH GRAVE}" # latin-1:C8
"\N{LATIN CAPITAL LETTER E WITH MACRON}" # windows-1257:C7
"\N{LATIN CAPITAL LETTER E WITH OGONEK}" # windows-1250:CA
"\N{LATIN CAPITAL LETTER ETH}" # latin-1:D0
"\N{LATIN CAPITAL LETTER G WITH BREVE}" # windows-1254:D0
"\N{LATIN CAPITAL LETTER G WITH CEDILLA}" # windows-1257:CC
"\N{LATIN CAPITAL LETTER I WITH ACUTE}" # latin-1:CD
"\N{LATIN CAPITAL LETTER I WITH CIRCUMFLEX}" # latin-1:CE
"\N{LATIN CAPITAL LETTER I WITH DIAERESIS}" # latin-1:CF
"\N{LATIN CAPITAL LETTER I WITH DOT ABOVE}" # windows-1254:DD
"\N{LATIN CAPITAL LETTER I WITH GRAVE}" # latin-1:CC
"\N{LATIN CAPITAL LETTER I WITH MACRON}" # windows-1257:CE
"\N{LATIN CAPITAL LETTER K WITH CEDILLA}" # windows-1257:CD
"\N{LATIN CAPITAL LETTER L WITH ACUTE}" # windows-1250:C5
"\N{LATIN CAPITAL LETTER L WITH CEDILLA}" # windows-1257:CF
"\N{LATIN CAPITAL LETTER L WITH STROKE}" # windows-1257:D9
"\N{LATIN CAPITAL LETTER N WITH ACUTE}" # windows-1250:D1
"\N{LATIN CAPITAL LETTER N WITH CARON}" # windows-1250:D2
"\N{LATIN CAPITAL LETTER N WITH CEDILLA}" # windows-1257:D2
"\N{LATIN CAPITAL LETTER N WITH TILDE}" # latin-1:D1
"\N{LATIN CAPITAL LETTER O WITH ACUTE}" # latin-1:D3
"\N{LATIN CAPITAL LETTER O WITH CIRCUMFLEX}" # latin-1:D4
"\N{LATIN CAPITAL LETTER O WITH DIAERESIS}" # latin-1:D6
"\N{LATIN CAPITAL LETTER O WITH DOUBLE ACUTE}" # windows-1250:D5
"\N{LATIN CAPITAL LETTER O WITH GRAVE}" # latin-1:D2
"\N{LATIN CAPITAL LETTER O WITH MACRON}" # windows-1257:D4
"\N{LATIN CAPITAL LETTER O WITH STROKE}" # latin-1:D8
"\N{LATIN CAPITAL LETTER O WITH TILDE}" # latin-1:D5
"\N{LATIN CAPITAL LETTER R WITH CARON}" # windows-1250:D8
"\N{LATIN CAPITAL LETTER S WITH ACUTE}" # windows-1257:DA
"\N{LATIN CAPITAL LETTER S WITH CARON}" # windows-1257:D0
"\N{LATIN CAPITAL LETTER S WITH CEDILLA}" # windows-1254:DE
"\N{LATIN CAPITAL LETTER T WITH CEDILLA}" # windows-1250:DE
"\N{LATIN CAPITAL LETTER THORN}" # latin-1:DE
"\N{LATIN CAPITAL LETTER U WITH ACUTE}" # latin-1:DA
"\N{LATIN CAPITAL LETTER U WITH CIRCUMFLEX}" # latin-1:DB
"\N{LATIN CAPITAL LETTER U WITH DIAERESIS}" # latin-1:DC
"\N{LATIN CAPITAL LETTER U WITH DOUBLE ACUTE}" # windows-1250:DB
"\N{LATIN CAPITAL LETTER U WITH GRAVE}" # latin-1:D9
"\N{LATIN CAPITAL LETTER U WITH MACRON}" # windows-1257:DB
"\N{LATIN CAPITAL LETTER U WITH OGONEK}" # windows-1257:D8
"\N{LATIN CAPITAL LETTER U WITH RING ABOVE}" # windows-1250:D9
"\N{LATIN CAPITAL LETTER Y WITH ACUTE}" # latin-1:DD
"\N{LATIN CAPITAL LETTER Z WITH ACUTE}" # windows-1257:CA
"\N{LATIN CAPITAL LETTER Z WITH CARON}" # windows-1257:DE
"\N{LATIN CAPITAL LETTER Z WITH DOT ABOVE}" # windows-1257:DD
"\N{LATIN SMALL LETTER SHARP S}" # latin-1:DF
"\N{MULTIPLICATION SIGN}" # latin-1:D7
"\N{GREEK CAPITAL LETTER BETA}" # windows-1253:C2
"\N{GREEK CAPITAL LETTER GAMMA}" # windows-1253:C3
"\N{GREEK CAPITAL LETTER DELTA}" # windows-1253:C4
"\N{GREEK CAPITAL LETTER EPSILON}" # windows-1253:C5
"\N{GREEK CAPITAL LETTER ZETA}" # windows-1253:C6
"\N{GREEK CAPITAL LETTER ETA}" # windows-1253:C7
"\N{GREEK CAPITAL LETTER THETA}" # windows-1253:C8
"\N{GREEK CAPITAL LETTER IOTA}" # windows-1253:C9
"\N{GREEK CAPITAL LETTER KAPPA}" # windows-1253:CA
"\N{GREEK CAPITAL LETTER LAMDA}" # windows-1253:CB
"\N{GREEK CAPITAL LETTER MU}" # windows-1253:CC
"\N{GREEK CAPITAL LETTER NU}" # windows-1253:CD
"\N{GREEK CAPITAL LETTER XI}" # windows-1253:CE
"\N{GREEK CAPITAL LETTER OMICRON}" # windows-1253:CF
"\N{GREEK CAPITAL LETTER PI}" # windows-1253:D0
"\N{GREEK CAPITAL LETTER RHO}" # windows-1253:D1
"\N{GREEK CAPITAL LETTER SIGMA}" # windows-1253:D3
"\N{GREEK CAPITAL LETTER TAU}" # windows-1253:D4
"\N{GREEK CAPITAL LETTER UPSILON}" # windows-1253:D5
"\N{GREEK CAPITAL LETTER PHI}" # windows-1253:D6
"\N{GREEK CAPITAL LETTER CHI}" # windows-1253:D7
"\N{GREEK CAPITAL LETTER PSI}" # windows-1253:D8
"\N{GREEK CAPITAL LETTER OMEGA}" # windows-1253:D9
"\N{GREEK CAPITAL LETTER IOTA WITH DIALYTIKA}" # windows-1253:DA
"\N{GREEK CAPITAL LETTER UPSILON WITH DIALYTIKA}" # windows-1253:DB
"\N{GREEK SMALL LETTER ALPHA WITH TONOS}" # windows-1253:DC
"\N{GREEK SMALL LETTER EPSILON WITH TONOS}" # windows-1253:DD
"\N{GREEK SMALL LETTER ETA WITH TONOS}" # windows-1253:DE
"\N{GREEK SMALL LETTER IOTA WITH TONOS}" # windows-1253:DF
"\N{CYRILLIC CAPITAL LETTER VE}" # windows-1251:C2
"\N{CYRILLIC CAPITAL LETTER GHE}" # windows-1251:C3
"\N{CYRILLIC CAPITAL LETTER DE}" # windows-1251:C4
"\N{CYRILLIC CAPITAL LETTER IE}" # windows-1251:C5
"\N{CYRILLIC CAPITAL LETTER ZHE}" # windows-1251:C6
"\N{CYRILLIC CAPITAL LETTER ZE}" # windows-1251:C7
"\N{CYRILLIC CAPITAL LETTER I}" # windows-1251:C8
"\N{CYRILLIC CAPITAL LETTER SHORT I}" # windows-1251:C9
"\N{CYRILLIC CAPITAL LETTER KA}" # windows-1251:CA
"\N{CYRILLIC CAPITAL LETTER EL}" # windows-1251:CB
"\N{CYRILLIC CAPITAL LETTER EM}" # windows-1251:CC
"\N{CYRILLIC CAPITAL LETTER EN}" # windows-1251:CD
"\N{CYRILLIC CAPITAL LETTER O}" # windows-1251:CE
"\N{CYRILLIC CAPITAL LETTER PE}" # windows-1251:CF
"\N{CYRILLIC CAPITAL LETTER ER}" # windows-1251:D0
"\N{CYRILLIC CAPITAL LETTER ES}" # windows-1251:D1
"\N{CYRILLIC CAPITAL LETTER TE}" # windows-1251:D2
"\N{CYRILLIC CAPITAL LETTER U}" # windows-1251:D3
"\N{CYRILLIC CAPITAL LETTER EF}" # windows-1251:D4
"\N{CYRILLIC CAPITAL LETTER HA}" # windows-1251:D5
"\N{CYRILLIC CAPITAL LETTER TSE}" # windows-1251:D6
"\N{CYRILLIC CAPITAL LETTER CHE}" # windows-1251:D7
"\N{CYRILLIC CAPITAL LETTER SHA}" # windows-1251:D8
"\N{CYRILLIC CAPITAL LETTER SHCHA}" # windows-1251:D9
"\N{CYRILLIC CAPITAL LETTER HARD SIGN}" # windows-1251:DA
"\N{CYRILLIC CAPITAL LETTER YERU}" # windows-1251:DB
"\N{CYRILLIC CAPITAL LETTER SOFT SIGN}" # windows-1251:DC
"\N{CYRILLIC CAPITAL LETTER E}" # windows-1251:DD
"\N{CYRILLIC CAPITAL LETTER YU}" # windows-1251:DE
"\N{CYRILLIC CAPITAL LETTER YA}" # windows-1251:DF
),
# Letters that decode to 0xE0 - 0xEF in a Latin-1-like encoding
"utf8_first_of_3": (
"\N{LATIN SMALL LETTER A WITH ACUTE}" # latin-1:E1
"\N{LATIN SMALL LETTER A WITH BREVE}" # windows-1250:E3
"\N{LATIN SMALL LETTER A WITH CIRCUMFLEX}" # latin-1:E2
"\N{LATIN SMALL LETTER A WITH DIAERESIS}" # latin-1:E4
"\N{LATIN SMALL LETTER A WITH GRAVE}" # latin-1:E0
"\N{LATIN SMALL LETTER A WITH MACRON}" # windows-1257:E2
"\N{LATIN SMALL LETTER A WITH OGONEK}" # windows-1257:E0
"\N{LATIN SMALL LETTER A WITH RING ABOVE}" # latin-1:E5
"\N{LATIN SMALL LETTER A WITH TILDE}" # latin-1:E3
"\N{LATIN SMALL LETTER AE}" # latin-1:E6
"\N{LATIN SMALL LETTER C WITH ACUTE}" # windows-1250:E6
"\N{LATIN SMALL LETTER C WITH CARON}" # windows-1250:E8
"\N{LATIN SMALL LETTER C WITH CEDILLA}" # latin-1:E7
"\N{LATIN SMALL LETTER D WITH CARON}" # windows-1250:EF
"\N{LATIN SMALL LETTER E WITH ACUTE}" # latin-1:E9
"\N{LATIN SMALL LETTER E WITH CARON}" # windows-1250:EC
"\N{LATIN SMALL LETTER E WITH CIRCUMFLEX}" # latin-1:EA
"\N{LATIN SMALL LETTER E WITH DIAERESIS}" # latin-1:EB
"\N{LATIN SMALL LETTER E WITH DOT ABOVE}" # windows-1257:EB
"\N{LATIN SMALL LETTER E WITH GRAVE}" # latin-1:E8
"\N{LATIN SMALL LETTER E WITH MACRON}" # windows-1257:E7
"\N{LATIN SMALL LETTER E WITH OGONEK}" # windows-1250:EA
"\N{LATIN SMALL LETTER E WITH OGONEK}" # windows-1250:EA
"\N{LATIN SMALL LETTER G WITH CEDILLA}" # windows-1257:EC
"\N{LATIN SMALL LETTER I WITH ACUTE}" # latin-1:ED
"\N{LATIN SMALL LETTER I WITH CIRCUMFLEX}" # latin-1:EE
"\N{LATIN SMALL LETTER I WITH DIAERESIS}" # latin-1:EF
"\N{LATIN SMALL LETTER I WITH GRAVE}" # latin-1:EC
"\N{LATIN SMALL LETTER I WITH MACRON}" # windows-1257:EE
"\N{LATIN SMALL LETTER I WITH OGONEK}" # windows-1257:E1
"\N{LATIN SMALL LETTER K WITH CEDILLA}" # windows-1257:ED
"\N{LATIN SMALL LETTER L WITH ACUTE}" # windows-1250:E5
"\N{LATIN SMALL LETTER L WITH CEDILLA}" # windows-1257:EF
"\N{LATIN SMALL LETTER R WITH ACUTE}" # windows-1250:E0
"\N{LATIN SMALL LETTER Z WITH ACUTE}" # windows-1257:EA
"\N{GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND TONOS}" # windows-1253:E0
"\N{GREEK SMALL LETTER ALPHA}" # windows-1253:E1
"\N{GREEK SMALL LETTER BETA}" # windows-1253:E2
"\N{GREEK SMALL LETTER GAMMA}" # windows-1253:E3
"\N{GREEK SMALL LETTER DELTA}" # windows-1253:E4
"\N{GREEK SMALL LETTER EPSILON}" # windows-1253:E5
"\N{GREEK SMALL LETTER ZETA}" # windows-1253:E6
"\N{GREEK SMALL LETTER ETA}" # windows-1253:E7
"\N{GREEK SMALL LETTER THETA}" # windows-1253:E8
"\N{GREEK SMALL LETTER IOTA}" # windows-1253:E9
"\N{GREEK SMALL LETTER KAPPA}" # windows-1253:EA
"\N{GREEK SMALL LETTER LAMDA}" # windows-1253:EB
"\N{GREEK SMALL LETTER MU}" # windows-1253:EC
"\N{GREEK SMALL LETTER NU}" # windows-1253:ED
"\N{GREEK SMALL LETTER XI}" # windows-1253:EE
"\N{GREEK SMALL LETTER OMICRON}" # windows-1253:EF
"\N{CYRILLIC SMALL LETTER A}" # windows-1251:E0
"\N{CYRILLIC SMALL LETTER BE}" # windows-1251:E1
"\N{CYRILLIC SMALL LETTER VE}" # windows-1251:E2
"\N{CYRILLIC SMALL LETTER GHE}" # windows-1251:E3
"\N{CYRILLIC SMALL LETTER DE}" # windows-1251:E4
"\N{CYRILLIC SMALL LETTER IE}" # windows-1251:E5
"\N{CYRILLIC SMALL LETTER ZHE}" # windows-1251:E6
"\N{CYRILLIC SMALL LETTER ZE}" # windows-1251:E7
"\N{CYRILLIC SMALL LETTER I}" # windows-1251:E8
"\N{CYRILLIC SMALL LETTER SHORT I}" # windows-1251:E9
"\N{CYRILLIC SMALL LETTER KA}" # windows-1251:EA
"\N{CYRILLIC SMALL LETTER EL}" # windows-1251:EB
"\N{CYRILLIC SMALL LETTER EM}" # windows-1251:EC
"\N{CYRILLIC SMALL LETTER EN}" # windows-1251:ED
"\N{CYRILLIC SMALL LETTER O}" # windows-1251:EE
"\N{CYRILLIC SMALL LETTER PE}" # windows-1251:EF
),
# Letters that decode to 0xF0 or 0xF3 in a Latin-1-like encoding.
# (Other leading bytes correspond only to unassigned codepoints)
"utf8_first_of_4": (
"\N{LATIN SMALL LETTER D WITH STROKE}" # windows-1250:F0
"\N{LATIN SMALL LETTER ETH}" # latin-1:F0
"\N{LATIN SMALL LETTER G WITH BREVE}" # windows-1254:F0
"\N{LATIN SMALL LETTER O WITH ACUTE}" # latin-1:F3
"\N{LATIN SMALL LETTER S WITH CARON}" # windows-1257:F0
"\N{GREEK SMALL LETTER PI}" # windows-1253:F0
"\N{GREEK SMALL LETTER SIGMA}" # windows-1253:F3
"\N{CYRILLIC SMALL LETTER ER}" # windows-1251:F0
"\N{CYRILLIC SMALL LETTER U}" # windows-1251:F3
),
# Letters that decode to 0x80 - 0xBF in a Latin-1-like encoding,
# including a space standing in for 0xA0
"utf8_continuation": (
"\x80-\xbf"
"\N{SPACE}" # modification of latin-1:A0, NO-BREAK SPACE
"\N{LATIN CAPITAL LETTER A WITH OGONEK}" # windows-1250:A5
"\N{LATIN CAPITAL LETTER AE}" # windows-1257:AF
"\N{LATIN CAPITAL LETTER L WITH CARON}" # windows-1250:BC
"\N{LATIN CAPITAL LETTER L WITH STROKE}" # windows-1250:A3
"\N{LATIN CAPITAL LETTER O WITH STROKE}" # windows-1257:A8
"\N{LATIN CAPITAL LETTER R WITH CEDILLA}" # windows-1257:AA
"\N{LATIN CAPITAL LETTER S WITH ACUTE}" # windows-1250:8C
"\N{LATIN CAPITAL LETTER S WITH CARON}" # windows-1252:8A
"\N{LATIN CAPITAL LETTER S WITH CEDILLA}" # windows-1250:AA
"\N{LATIN CAPITAL LETTER T WITH CARON}" # windows-1250:8D
"\N{LATIN CAPITAL LETTER Y WITH DIAERESIS}" # windows-1252:9F
"\N{LATIN CAPITAL LETTER Z WITH ACUTE}" # windows-1250:8F
"\N{LATIN CAPITAL LETTER Z WITH CARON}" # windows-1252:8E
"\N{LATIN CAPITAL LETTER Z WITH DOT ABOVE}" # windows-1250:AF
"\N{LATIN CAPITAL LIGATURE OE}" # windows-1252:8C
"\N{LATIN SMALL LETTER A WITH OGONEK}" # windows-1250:B9
"\N{LATIN SMALL LETTER AE}" # windows-1257:BF
"\N{LATIN SMALL LETTER F WITH HOOK}" # windows-1252:83
"\N{LATIN SMALL LETTER L WITH CARON}" # windows-1250:BE
"\N{LATIN SMALL LETTER L WITH STROKE}" # windows-1250:B3
"\N{LATIN SMALL LETTER O WITH STROKE}" # windows-1257:B8
"\N{LATIN SMALL LETTER R WITH CEDILLA}" # windows-1257:BA
"\N{LATIN SMALL LETTER S WITH ACUTE}" # windows-1250:9C
"\N{LATIN SMALL LETTER S WITH CARON}" # windows-1252:9A
"\N{LATIN SMALL LETTER S WITH CEDILLA}" # windows-1250:BA
"\N{LATIN SMALL LETTER T WITH CARON}" # windows-1250:9D
"\N{LATIN SMALL LETTER Z WITH ACUTE}" # windows-1250:9F
"\N{LATIN SMALL LETTER Z WITH CARON}" # windows-1252:9E
"\N{LATIN SMALL LETTER Z WITH DOT ABOVE}" # windows-1250:BF
"\N{LATIN SMALL LIGATURE OE}" # windows-1252:9C
"\N{MODIFIER LETTER CIRCUMFLEX ACCENT}" # windows-1252:88
"\N{CARON}" # windows-1250:A1
"\N{BREVE}" # windows-1250:A2
"\N{OGONEK}" # windows-1250:B2
"\N{SMALL TILDE}" # windows-1252:98
"\N{DOUBLE ACUTE ACCENT}" # windows-1250:BD
"\N{GREEK TONOS}" # windows-1253:B4
"\N{GREEK DIALYTIKA TONOS}" # windows-1253:A1
"\N{GREEK CAPITAL LETTER ALPHA WITH TONOS}" # windows-1253:A2
"\N{GREEK CAPITAL LETTER EPSILON WITH TONOS}" # windows-1253:B8
"\N{GREEK CAPITAL LETTER ETA WITH TONOS}" # windows-1253:B9
"\N{GREEK CAPITAL LETTER IOTA WITH TONOS}" # windows-1253:BA
"\N{GREEK CAPITAL LETTER OMICRON WITH TONOS}" # windows-1253:BC
"\N{GREEK CAPITAL LETTER UPSILON WITH TONOS}" # windows-1253:BE
"\N{GREEK CAPITAL LETTER OMEGA WITH TONOS}" # windows-1253:BF
"\N{CYRILLIC CAPITAL LETTER IO}" # windows-1251:A8
"\N{CYRILLIC CAPITAL LETTER DJE}" # windows-1251:80
"\N{CYRILLIC CAPITAL LETTER GJE}" # windows-1251:81
"\N{CYRILLIC CAPITAL LETTER UKRAINIAN IE}" # windows-1251:AA
"\N{CYRILLIC CAPITAL LETTER DZE}" # windows-1251:BD
"\N{CYRILLIC CAPITAL LETTER BYELORUSSIAN-UKRAINIAN I}" # windows-1251:B2
"\N{CYRILLIC CAPITAL LETTER YI}" # windows-1251:AF
"\N{CYRILLIC CAPITAL LETTER JE}" # windows-1251:A3
"\N{CYRILLIC CAPITAL LETTER LJE}" # windows-1251:8A
"\N{CYRILLIC CAPITAL LETTER NJE}" # windows-1251:8C
"\N{CYRILLIC CAPITAL LETTER TSHE}" # windows-1251:8E
"\N{CYRILLIC CAPITAL LETTER KJE}" # windows-1251:8D
"\N{CYRILLIC CAPITAL LETTER SHORT U}" # windows-1251:A1
"\N{CYRILLIC CAPITAL LETTER DZHE}" # windows-1251:8F
"\N{CYRILLIC SMALL LETTER IO}" # windows-1251:B8
"\N{CYRILLIC SMALL LETTER DJE}" # windows-1251:90
"\N{CYRILLIC SMALL LETTER GJE}" # windows-1251:83
"\N{CYRILLIC SMALL LETTER UKRAINIAN IE}" # windows-1251:BA
"\N{CYRILLIC SMALL LETTER DZE}" # windows-1251:BE
"\N{CYRILLIC SMALL LETTER BYELORUSSIAN-UKRAINIAN I}" # windows-1251:B3
"\N{CYRILLIC SMALL LETTER YI}" # windows-1251:BF
"\N{CYRILLIC SMALL LETTER JE}" # windows-1251:BC
"\N{CYRILLIC SMALL LETTER LJE}" # windows-1251:9A
"\N{CYRILLIC SMALL LETTER NJE}" # windows-1251:9C
"\N{CYRILLIC SMALL LETTER TSHE}" # windows-1251:9E
"\N{CYRILLIC SMALL LETTER KJE}" # windows-1251:9D
"\N{CYRILLIC SMALL LETTER SHORT U}" # windows-1251:A2
"\N{CYRILLIC SMALL LETTER DZHE}" # windows-1251:9F
"\N{CYRILLIC CAPITAL LETTER GHE WITH UPTURN}" # windows-1251:A5
"\N{CYRILLIC SMALL LETTER GHE WITH UPTURN}" # windows-1251:B4
"\N{EN DASH}" # windows-1252:96
"\N{EM DASH}" # windows-1252:97
"\N{HORIZONTAL BAR}" # windows-1253:AF
"\N{LEFT SINGLE QUOTATION MARK}" # windows-1252:91
"\N{RIGHT SINGLE QUOTATION MARK}" # windows-1252:92
"\N{SINGLE LOW-9 QUOTATION MARK}" # windows-1252:82
"\N{LEFT DOUBLE QUOTATION MARK}" # windows-1252:93
"\N{RIGHT DOUBLE QUOTATION MARK}" # windows-1252:94
"\N{DOUBLE LOW-9 QUOTATION MARK}" # windows-1252:84
"\N{DAGGER}" # windows-1252:86
"\N{DOUBLE DAGGER}" # windows-1252:87
"\N{BULLET}" # windows-1252:95
"\N{HORIZONTAL ELLIPSIS}" # windows-1252:85
"\N{PER MILLE SIGN}" # windows-1252:89
"\N{SINGLE LEFT-POINTING ANGLE QUOTATION MARK}" # windows-1252:8B
"\N{SINGLE RIGHT-POINTING ANGLE QUOTATION MARK}" # windows-1252:9B
"\N{EURO SIGN}" # windows-1252:80
"\N{NUMERO SIGN}" # windows-1251:B9
"\N{TRADE MARK SIGN}" # windows-1252:99
),
# Letters that decode to 0x80 - 0xBF in a Latin-1-like encoding,
# and don't usually stand for themselves when adjacent to mojibake.
# This excludes spaces, dashes, 'bullet', quotation marks, and ellipses.
"utf8_continuation_strict": (
"\x80-\xbf"
"\N{LATIN CAPITAL LETTER A WITH OGONEK}" # windows-1250:A5
"\N{LATIN CAPITAL LETTER AE}" # windows-1257:AF
"\N{LATIN CAPITAL LETTER L WITH CARON}" # windows-1250:BC
"\N{LATIN CAPITAL LETTER L WITH STROKE}" # windows-1250:A3
"\N{LATIN CAPITAL LETTER O WITH STROKE}" # windows-1257:A8
"\N{LATIN CAPITAL LETTER R WITH CEDILLA}" # windows-1257:AA
"\N{LATIN CAPITAL LETTER S WITH ACUTE}" # windows-1250:8C
"\N{LATIN CAPITAL LETTER S WITH CARON}" # windows-1252:8A
"\N{LATIN CAPITAL LETTER S WITH CEDILLA}" # windows-1250:AA
"\N{LATIN CAPITAL LETTER T WITH CARON}" # windows-1250:8D
"\N{LATIN CAPITAL LETTER Y WITH DIAERESIS}" # windows-1252:9F
"\N{LATIN CAPITAL LETTER Z WITH ACUTE}" # windows-1250:8F
"\N{LATIN CAPITAL LETTER Z WITH CARON}" # windows-1252:8E
"\N{LATIN CAPITAL LETTER Z WITH DOT ABOVE}" # windows-1250:AF
"\N{LATIN CAPITAL LIGATURE OE}" # windows-1252:8C
"\N{LATIN SMALL LETTER A WITH OGONEK}" # windows-1250:B9
"\N{LATIN SMALL LETTER AE}" # windows-1257:BF
"\N{LATIN SMALL LETTER F WITH HOOK}" # windows-1252:83
"\N{LATIN SMALL LETTER L WITH CARON}" # windows-1250:BE
"\N{LATIN SMALL LETTER L WITH STROKE}" # windows-1250:B3
"\N{LATIN SMALL LETTER O WITH STROKE}" # windows-1257:B8
"\N{LATIN SMALL LETTER R WITH CEDILLA}" # windows-1257:BA
"\N{LATIN SMALL LETTER S WITH ACUTE}" # windows-1250:9C
"\N{LATIN SMALL LETTER S WITH CARON}" # windows-1252:9A
"\N{LATIN SMALL LETTER S WITH CEDILLA}" # windows-1250:BA
"\N{LATIN SMALL LETTER T WITH CARON}" # windows-1250:9D
"\N{LATIN SMALL LETTER Z WITH ACUTE}" # windows-1250:9F
"\N{LATIN SMALL LETTER Z WITH CARON}" # windows-1252:9E
"\N{LATIN SMALL LETTER Z WITH DOT ABOVE}" # windows-1250:BF
"\N{LATIN SMALL LIGATURE OE}" # windows-1252:9C
"\N{MODIFIER LETTER CIRCUMFLEX ACCENT}" # windows-1252:88
"\N{CARON}" # windows-1250:A1
"\N{BREVE}" # windows-1250:A2
"\N{OGONEK}" # windows-1250:B2
"\N{SMALL TILDE}" # windows-1252:98
"\N{DOUBLE ACUTE ACCENT}" # windows-1250:BD
"\N{GREEK TONOS}" # windows-1253:B4
"\N{GREEK DIALYTIKA TONOS}" # windows-1253:A1
"\N{GREEK CAPITAL LETTER ALPHA WITH TONOS}" # windows-1253:A2
"\N{GREEK CAPITAL LETTER EPSILON WITH TONOS}" # windows-1253:B8
"\N{GREEK CAPITAL LETTER ETA WITH TONOS}" # windows-1253:B9
"\N{GREEK CAPITAL LETTER IOTA WITH TONOS}" # windows-1253:BA
"\N{GREEK CAPITAL LETTER OMICRON WITH TONOS}" # windows-1253:BC
"\N{GREEK CAPITAL LETTER UPSILON WITH TONOS}" # windows-1253:BE
"\N{GREEK CAPITAL LETTER OMEGA WITH TONOS}" # windows-1253:BF
"\N{CYRILLIC CAPITAL LETTER IO}" # windows-1251:A8
"\N{CYRILLIC CAPITAL LETTER DJE}" # windows-1251:80
"\N{CYRILLIC CAPITAL LETTER GJE}" # windows-1251:81
"\N{CYRILLIC CAPITAL LETTER UKRAINIAN IE}" # windows-1251:AA
"\N{CYRILLIC CAPITAL LETTER DZE}" # windows-1251:BD
"\N{CYRILLIC CAPITAL LETTER BYELORUSSIAN-UKRAINIAN I}" # windows-1251:B2
"\N{CYRILLIC CAPITAL LETTER YI}" # windows-1251:AF
"\N{CYRILLIC CAPITAL LETTER JE}" # windows-1251:A3
"\N{CYRILLIC CAPITAL LETTER LJE}" # windows-1251:8A
"\N{CYRILLIC CAPITAL LETTER NJE}" # windows-1251:8C
"\N{CYRILLIC CAPITAL LETTER TSHE}" # windows-1251:8E
"\N{CYRILLIC CAPITAL LETTER KJE}" # windows-1251:8D
"\N{CYRILLIC CAPITAL LETTER SHORT U}" # windows-1251:A1
"\N{CYRILLIC CAPITAL LETTER DZHE}" # windows-1251:8F
"\N{CYRILLIC SMALL LETTER IO}" # windows-1251:B8
"\N{CYRILLIC SMALL LETTER DJE}" # windows-1251:90
"\N{CYRILLIC SMALL LETTER GJE}" # windows-1251:83
"\N{CYRILLIC SMALL LETTER UKRAINIAN IE}" # windows-1251:BA
"\N{CYRILLIC SMALL LETTER DZE}" # windows-1251:BE
"\N{CYRILLIC SMALL LETTER BYELORUSSIAN-UKRAINIAN I}" # windows-1251:B3
"\N{CYRILLIC SMALL LETTER YI}" # windows-1251:BF
"\N{CYRILLIC SMALL LETTER JE}" # windows-1251:BC
"\N{CYRILLIC SMALL LETTER LJE}" # windows-1251:9A
"\N{CYRILLIC SMALL LETTER NJE}" # windows-1251:9C
"\N{CYRILLIC SMALL LETTER TSHE}" # windows-1251:9E
"\N{CYRILLIC SMALL LETTER KJE}" # windows-1251:9D
"\N{CYRILLIC SMALL LETTER SHORT U}" # windows-1251:A2
"\N{CYRILLIC SMALL LETTER DZHE}" # windows-1251:9F
"\N{CYRILLIC CAPITAL LETTER GHE WITH UPTURN}" # windows-1251:A5
"\N{CYRILLIC SMALL LETTER GHE WITH UPTURN}" # windows-1251:B4
"\N{DAGGER}" # windows-1252:86
"\N{DOUBLE DAGGER}" # windows-1252:87
"\N{PER MILLE SIGN}" # windows-1252:89
"\N{SINGLE LEFT-POINTING ANGLE QUOTATION MARK}" # windows-1252:8B
"\N{SINGLE RIGHT-POINTING ANGLE QUOTATION MARK}" # windows-1252:9B
"\N{EURO SIGN}" # windows-1252:80
"\N{NUMERO SIGN}" # windows-1251:B9
"\N{TRADE MARK SIGN}" # windows-1252:99
),
}
# This regex uses UTF8_CLUES to find sequences of likely mojibake.
# It matches them with + so that several adjacent UTF-8-looking sequences
# get coalesced into one, allowing them to be fixed more efficiently
# and not requiring every individual subsequence to be detected as 'badness'.
#
# We accept spaces in place of "utf8_continuation", because spaces might have
# been intended to be U+A0 NO-BREAK SPACE.
#
# We do a lookbehind to make sure the previous character isn't a
# "utf8_continuation_strict" character, so that we don't fix just a few
# characters in a huge garble and make the situation worse.
#
# Unfortunately, the matches to this regular expression won't show their
# surrounding context, and including context would make the expression much
# less efficient. The 'badness' rules that require context, such as a preceding
# lowercase letter, will prevent some cases of inconsistent UTF-8 from being
# fixed when they don't see it.
UTF8_DETECTOR_RE = re.compile(
"""
(?<! [{utf8_continuation_strict}])
(
[{utf8_first_of_2}] [{utf8_continuation}]
|
[{utf8_first_of_3}] [{utf8_continuation}]{{2}}
|
[{utf8_first_of_4}] [{utf8_continuation}]{{3}}
)+
""".format(**UTF8_CLUES),
re.VERBOSE,
)