Spaces:
Sleeping
Sleeping
File size: 13,280 Bytes
69d9940 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 |
"""
Metadata about languages used by our model training code for our
SingleByteCharSetProbers. Could be used for other things in the future.
This code is based on the language metadata from the uchardet project.
"""
from string import ascii_letters
# TODO: Add Ukrainian (KOI8-U)
class Language:
"""Metadata about a language useful for training models
:ivar name: The human name for the language, in English.
:type name: str
:ivar iso_code: 2-letter ISO 639-1 if possible, 3-letter ISO code otherwise,
or use another catalog as a last resort.
:type iso_code: str
:ivar use_ascii: Whether or not ASCII letters should be included in trained
models.
:type use_ascii: bool
:ivar charsets: The charsets we want to support and create data for.
:type charsets: list of str
:ivar alphabet: The characters in the language's alphabet. If `use_ascii` is
`True`, you only need to add those not in the ASCII set.
:type alphabet: str
:ivar wiki_start_pages: The Wikipedia pages to start from if we're crawling
Wikipedia for training data.
:type wiki_start_pages: list of str
"""
def __init__(
self,
name=None,
iso_code=None,
use_ascii=True,
charsets=None,
alphabet=None,
wiki_start_pages=None,
):
super().__init__()
self.name = name
self.iso_code = iso_code
self.use_ascii = use_ascii
self.charsets = charsets
if self.use_ascii:
if alphabet:
alphabet += ascii_letters
else:
alphabet = ascii_letters
elif not alphabet:
raise ValueError("Must supply alphabet if use_ascii is False")
self.alphabet = "".join(sorted(set(alphabet))) if alphabet else None
self.wiki_start_pages = wiki_start_pages
def __repr__(self):
param_str = ", ".join(
f"{k}={v!r}" for k, v in self.__dict__.items() if not k.startswith("_")
)
return f"{self.__class__.__name__}({param_str})"
LANGUAGES = {
"Arabic": Language(
name="Arabic",
iso_code="ar",
use_ascii=False,
# We only support encodings that use isolated
# forms, because the current recommendation is
# that the rendering system handles presentation
# forms. This means we purposefully skip IBM864.
charsets=["ISO-8859-6", "WINDOWS-1256", "CP720", "CP864"],
alphabet="ءآأؤإئابةتثجحخدذرزسشصضطظعغػؼؽؾؿـفقكلمنهوىيًٌٍَُِّ",
wiki_start_pages=["الصفحة_الرئيسية"],
),
"Belarusian": Language(
name="Belarusian",
iso_code="be",
use_ascii=False,
charsets=["ISO-8859-5", "WINDOWS-1251", "IBM866", "MacCyrillic"],
alphabet="АБВГДЕЁЖЗІЙКЛМНОПРСТУЎФХЦЧШЫЬЭЮЯабвгдеёжзійклмнопрстуўфхцчшыьэюяʼ",
wiki_start_pages=["Галоўная_старонка"],
),
"Bulgarian": Language(
name="Bulgarian",
iso_code="bg",
use_ascii=False,
charsets=["ISO-8859-5", "WINDOWS-1251", "IBM855"],
alphabet="АБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЬЮЯабвгдежзийклмнопрстуфхцчшщъьюя",
wiki_start_pages=["Начална_страница"],
),
"Czech": Language(
name="Czech",
iso_code="cz",
use_ascii=True,
charsets=["ISO-8859-2", "WINDOWS-1250"],
alphabet="áčďéěíňóřšťúůýžÁČĎÉĚÍŇÓŘŠŤÚŮÝŽ",
wiki_start_pages=["Hlavní_strana"],
),
"Danish": Language(
name="Danish",
iso_code="da",
use_ascii=True,
charsets=["ISO-8859-1", "ISO-8859-15", "WINDOWS-1252"],
alphabet="æøåÆØÅ",
wiki_start_pages=["Forside"],
),
"German": Language(
name="German",
iso_code="de",
use_ascii=True,
charsets=["ISO-8859-1", "WINDOWS-1252"],
alphabet="äöüßÄÖÜ",
wiki_start_pages=["Wikipedia:Hauptseite"],
),
"Greek": Language(
name="Greek",
iso_code="el",
use_ascii=False,
charsets=["ISO-8859-7", "WINDOWS-1253"],
alphabet="αβγδεζηθικλμνξοπρσςτυφχψωάέήίόύώΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡΣΣΤΥΦΧΨΩΆΈΉΊΌΎΏ",
wiki_start_pages=["Πύλη:Κύρια"],
),
"English": Language(
name="English",
iso_code="en",
use_ascii=True,
charsets=["ISO-8859-1", "WINDOWS-1252"],
wiki_start_pages=["Main_Page"],
),
"Esperanto": Language(
name="Esperanto",
iso_code="eo",
# Q, W, X, and Y not used at all
use_ascii=False,
charsets=["ISO-8859-3"],
alphabet="abcĉdefgĝhĥijĵklmnoprsŝtuŭvzABCĈDEFGĜHĤIJĴKLMNOPRSŜTUŬVZ",
wiki_start_pages=["Vikipedio:Ĉefpaĝo"],
),
"Spanish": Language(
name="Spanish",
iso_code="es",
use_ascii=True,
charsets=["ISO-8859-1", "ISO-8859-15", "WINDOWS-1252"],
alphabet="ñáéíóúüÑÁÉÍÓÚÜ",
wiki_start_pages=["Wikipedia:Portada"],
),
"Estonian": Language(
name="Estonian",
iso_code="et",
use_ascii=False,
charsets=["ISO-8859-4", "ISO-8859-13", "WINDOWS-1257"],
# C, F, Š, Q, W, X, Y, Z, Ž are only for
# loanwords
alphabet="ABDEGHIJKLMNOPRSTUVÕÄÖÜabdeghijklmnoprstuvõäöü",
wiki_start_pages=["Esileht"],
),
"Finnish": Language(
name="Finnish",
iso_code="fi",
use_ascii=True,
charsets=["ISO-8859-1", "ISO-8859-15", "WINDOWS-1252"],
alphabet="ÅÄÖŠŽåäöšž",
wiki_start_pages=["Wikipedia:Etusivu"],
),
"French": Language(
name="French",
iso_code="fr",
use_ascii=True,
charsets=["ISO-8859-1", "ISO-8859-15", "WINDOWS-1252"],
alphabet="œàâçèéîïùûêŒÀÂÇÈÉÎÏÙÛÊ",
wiki_start_pages=["Wikipédia:Accueil_principal", "Bœuf (animal)"],
),
"Hebrew": Language(
name="Hebrew",
iso_code="he",
use_ascii=False,
charsets=["ISO-8859-8", "WINDOWS-1255"],
alphabet="אבגדהוזחטיךכלםמןנסעףפץצקרשתװױײ",
wiki_start_pages=["עמוד_ראשי"],
),
"Croatian": Language(
name="Croatian",
iso_code="hr",
# Q, W, X, Y are only used for foreign words.
use_ascii=False,
charsets=["ISO-8859-2", "WINDOWS-1250"],
alphabet="abcčćdđefghijklmnoprsštuvzžABCČĆDĐEFGHIJKLMNOPRSŠTUVZŽ",
wiki_start_pages=["Glavna_stranica"],
),
"Hungarian": Language(
name="Hungarian",
iso_code="hu",
# Q, W, X, Y are only used for foreign words.
use_ascii=False,
charsets=["ISO-8859-2", "WINDOWS-1250"],
alphabet="abcdefghijklmnoprstuvzáéíóöőúüűABCDEFGHIJKLMNOPRSTUVZÁÉÍÓÖŐÚÜŰ",
wiki_start_pages=["Kezdőlap"],
),
"Italian": Language(
name="Italian",
iso_code="it",
use_ascii=True,
charsets=["ISO-8859-1", "ISO-8859-15", "WINDOWS-1252"],
alphabet="ÀÈÉÌÒÓÙàèéìòóù",
wiki_start_pages=["Pagina_principale"],
),
"Lithuanian": Language(
name="Lithuanian",
iso_code="lt",
use_ascii=False,
charsets=["ISO-8859-13", "WINDOWS-1257", "ISO-8859-4"],
# Q, W, and X not used at all
alphabet="AĄBCČDEĘĖFGHIĮYJKLMNOPRSŠTUŲŪVZŽaąbcčdeęėfghiįyjklmnoprsštuųūvzž",
wiki_start_pages=["Pagrindinis_puslapis"],
),
"Latvian": Language(
name="Latvian",
iso_code="lv",
use_ascii=False,
charsets=["ISO-8859-13", "WINDOWS-1257", "ISO-8859-4"],
# Q, W, X, Y are only for loanwords
alphabet="AĀBCČDEĒFGĢHIĪJKĶLĻMNŅOPRSŠTUŪVZŽaābcčdeēfgģhiījkķlļmnņoprsštuūvzž",
wiki_start_pages=["Sākumlapa"],
),
"Macedonian": Language(
name="Macedonian",
iso_code="mk",
use_ascii=False,
charsets=["ISO-8859-5", "WINDOWS-1251", "MacCyrillic", "IBM855"],
alphabet="АБВГДЃЕЖЗЅИЈКЛЉМНЊОПРСТЌУФХЦЧЏШабвгдѓежзѕијклљмнњопрстќуфхцчџш",
wiki_start_pages=["Главна_страница"],
),
"Dutch": Language(
name="Dutch",
iso_code="nl",
use_ascii=True,
charsets=["ISO-8859-1", "WINDOWS-1252"],
wiki_start_pages=["Hoofdpagina"],
),
"Polish": Language(
name="Polish",
iso_code="pl",
# Q and X are only used for foreign words.
use_ascii=False,
charsets=["ISO-8859-2", "WINDOWS-1250"],
alphabet="AĄBCĆDEĘFGHIJKLŁMNŃOÓPRSŚTUWYZŹŻaąbcćdeęfghijklłmnńoóprsśtuwyzźż",
wiki_start_pages=["Wikipedia:Strona_główna"],
),
"Portuguese": Language(
name="Portuguese",
iso_code="pt",
use_ascii=True,
charsets=["ISO-8859-1", "ISO-8859-15", "WINDOWS-1252"],
alphabet="ÁÂÃÀÇÉÊÍÓÔÕÚáâãàçéêíóôõú",
wiki_start_pages=["Wikipédia:Página_principal"],
),
"Romanian": Language(
name="Romanian",
iso_code="ro",
use_ascii=True,
charsets=["ISO-8859-2", "WINDOWS-1250"],
alphabet="ăâîșțĂÂÎȘȚ",
wiki_start_pages=["Pagina_principală"],
),
"Russian": Language(
name="Russian",
iso_code="ru",
use_ascii=False,
charsets=[
"ISO-8859-5",
"WINDOWS-1251",
"KOI8-R",
"MacCyrillic",
"IBM866",
"IBM855",
],
alphabet="абвгдеёжзийклмнопрстуфхцчшщъыьэюяАБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ",
wiki_start_pages=["Заглавная_страница"],
),
"Slovak": Language(
name="Slovak",
iso_code="sk",
use_ascii=True,
charsets=["ISO-8859-2", "WINDOWS-1250"],
alphabet="áäčďéíĺľňóôŕšťúýžÁÄČĎÉÍĹĽŇÓÔŔŠŤÚÝŽ",
wiki_start_pages=["Hlavná_stránka"],
),
"Slovene": Language(
name="Slovene",
iso_code="sl",
# Q, W, X, Y are only used for foreign words.
use_ascii=False,
charsets=["ISO-8859-2", "WINDOWS-1250"],
alphabet="abcčdefghijklmnoprsštuvzžABCČDEFGHIJKLMNOPRSŠTUVZŽ",
wiki_start_pages=["Glavna_stran"],
),
# Serbian can be written in both Latin and Cyrillic, but there's no
# simple way to get the Latin alphabet pages from Wikipedia through
# the API, so for now we just support Cyrillic.
"Serbian": Language(
name="Serbian",
iso_code="sr",
alphabet="АБВГДЂЕЖЗИЈКЛЉМНЊОПРСТЋУФХЦЧЏШабвгдђежзијклљмнњопрстћуфхцчџш",
charsets=["ISO-8859-5", "WINDOWS-1251", "MacCyrillic", "IBM855"],
wiki_start_pages=["Главна_страна"],
),
"Thai": Language(
name="Thai",
iso_code="th",
use_ascii=False,
charsets=["ISO-8859-11", "TIS-620", "CP874"],
alphabet="กขฃคฅฆงจฉชซฌญฎฏฐฑฒณดตถทธนบปผฝพฟภมยรฤลฦวศษสหฬอฮฯะัาำิีึืฺุู฿เแโใไๅๆ็่้๊๋์ํ๎๏๐๑๒๓๔๕๖๗๘๙๚๛",
wiki_start_pages=["หน้าหลัก"],
),
"Turkish": Language(
name="Turkish",
iso_code="tr",
# Q, W, and X are not used by Turkish
use_ascii=False,
charsets=["ISO-8859-3", "ISO-8859-9", "WINDOWS-1254"],
alphabet="abcçdefgğhıijklmnoöprsştuüvyzâîûABCÇDEFGĞHIİJKLMNOÖPRSŞTUÜVYZÂÎÛ",
wiki_start_pages=["Ana_Sayfa"],
),
"Vietnamese": Language(
name="Vietnamese",
iso_code="vi",
use_ascii=False,
# Windows-1258 is the only common 8-bit
# Vietnamese encoding supported by Python.
# From Wikipedia:
# For systems that lack support for Unicode,
# dozens of 8-bit Vietnamese code pages are
# available.[1] The most common are VISCII
# (TCVN 5712:1993), VPS, and Windows-1258.[3]
# Where ASCII is required, such as when
# ensuring readability in plain text e-mail,
# Vietnamese letters are often encoded
# according to Vietnamese Quoted-Readable
# (VIQR) or VSCII Mnemonic (VSCII-MNEM),[4]
# though usage of either variable-width
# scheme has declined dramatically following
# the adoption of Unicode on the World Wide
# Web.
charsets=["WINDOWS-1258"],
alphabet="aăâbcdđeêghiklmnoôơpqrstuưvxyAĂÂBCDĐEÊGHIKLMNOÔƠPQRSTUƯVXY",
wiki_start_pages=["Chữ_Quốc_ngữ"],
),
}
|