writing_system.py

import unicodedata
from typing import Literal, NamedTuple

__all__ = [
	"WritingSystem",
	"getAllWritingSystemsFromText",
	"getWritingSystemFromText",
	# 'unicodeNextWord',
	"writingSystemByLowercaseName",
	"writingSystemByName",
	"writingSystemByUnicode",
	"writingSystemList",
]


class WritingSystem(NamedTuple):
	name: str
	iso: list[tuple[int, str]] | list[tuple[int, str, str]] = []
	unicode: list = []
	titleTag: str = "b"
	direction: Literal["ltr", "rtl", "ttb"] = "ltr"
	comma: str = ", "
	pop: int | float = 0  # population in millions


# digits and FULLWIDTH DIGITs are considered neutral/ignored, not Latin

# scripts are separated into multiple groups based on their popularity
# (usage in multiple live languages, and number of native speakers)


writingSystemList = [
	WritingSystem(
		name="Latin",
		iso=[(215, "Latn")],
		unicode=[
			"LATIN",
		],
		titleTag="b",
		comma=", ",
		pop=4900,
	),
	WritingSystem(
		name="Arabic",
		iso=[(160, "Arab")],
		unicode=["ARABIC"],
		titleTag="b",
		direction="rtl",
		comma="، ",
		pop=670,
	),
	WritingSystem(
		name="Cyrillic",
		iso=[(220, "Cyrl")],
		unicode=["CYRILLIC"],
		titleTag="b",
		comma=", ",
		pop=250,
	),
	WritingSystem(
		name="CJK",
		iso=[
			(285, "Bopo", "BOPOMOFO"),
			(286, "Hang", "HANGUL"),
			(410, "Hira", "HIRAGANA"),
			(411, "Kana", "KATAKANA"),
			(412, "Hrkt", "KATAKANA OR HIRAGANA"),
			(460, "Yiii", "Yi"),
			(499, "Nshu", "NUSHU"),
			(500, "Hani", "HAN"),  # aka Hanzi, Kanji, Hanja
			(501, "Hans", "SIMPLIFIED HAN"),
			(502, "Hant", "TRADITIONAL HAN"),
		],
		unicode=[
			"CJK",
			"HIRAGANA",
			"KATAKANA",
			"IDEOGRAPHIC",  # Ideographic Description Characters
			"DITTO",  # Ditto mark
			"HANGUL",  # Korean alphabet
			"HALFWIDTH KATAKANA",
			"HALFWIDTH HANGUL",
			"YI",  # https://en.wikipedia.org/wiki/Yi_script
			"FULLWIDTH LATIN",
			"BOPOMOFO",
			"NUSHU",
		],
		titleTag="big",
		comma="、",
		pop=1540,  # Chinese=1340, Kana=120, Hangul=78.7
	),
	WritingSystem(
		name="Devanagari",
		iso=[(315, "Deva")],
		unicode=["DEVANAGARI"],
		titleTag="big",
		comma=", ",
		pop=610,
	),
	# _____________________________________________________
	WritingSystem(
		name="Armenian",
		iso=[(230, "Armn")],
		unicode=["ARMENIAN"],
		titleTag="big",
		comma=", ",
		pop=12,
	),
	WritingSystem(
		name="Bengali-Assamese",
		iso=[(325, "Beng")],
		unicode=["BENGALI"],
		titleTag="big",
		comma=", ",
		pop=270,
	),
	WritingSystem(
		name="Burmese",
		iso=[(350, "Mymr")],
		unicode=["MYANMAR"],
		titleTag="big",
		comma=", ",  # almost not used except in English phrases
		pop=39,
	),
	WritingSystem(
		name="Ge'ez",
		iso=[(430, "Ethi")],
		unicode=["ETHIOPIC"],
		titleTag="big",
		comma=", ",
		pop=21,
	),
	WritingSystem(
		name="Greek",
		iso=[(200, "Grek")],
		unicode=["GREEK"],
		titleTag="b",
		comma=", ",
		pop=11,
	),
	WritingSystem(
		name="Gujarati",
		iso=[(320, "Gujr")],
		unicode=["GUJARATI"],
		titleTag="big",
		comma=", ",
		pop=48,
	),
	WritingSystem(
		name="Gurmukhi",
		iso=[(310, "Guru")],
		unicode=["GURMUKHI"],
		titleTag="big",
		comma=", ",
		pop=22,
	),
	WritingSystem(
		name="Hebrew",
		iso=[(125, "Hebr")],
		unicode=["HEBREW"],
		titleTag="big",
		direction="rtl",
		comma=", ",
		pop=14,
	),
	WritingSystem(
		name="Kannada",
		iso=[(345, "Knda")],
		unicode=["KANNADA"],
		titleTag="big",
		comma=", ",
		pop=45,
	),
	WritingSystem(
		name="Khmer",
		iso=[(355, "Khmr")],
		unicode=["KHMER"],
		titleTag="big",
		comma=", ",
		pop=11.4,
	),
	WritingSystem(
		name="Lao",
		iso=[(356, "Laoo")],
		unicode=["LAO"],
		titleTag="big",
		comma=", ",
		pop=22,
	),
	WritingSystem(
		name="Malayalam",
		iso=[(347, "Mlym")],
		unicode=["MALAYALAM"],
		titleTag="big",
		comma=", ",
		pop=38,
	),
	WritingSystem(
		name="Odia",
		iso=[(327, "Orya")],
		unicode=["ORIYA"],
		titleTag="big",
		comma=", ",
		pop=21,
	),
	WritingSystem(
		name="Sinhala",
		iso=[(348, "Sinh")],
		unicode=["SINHALA"],
		titleTag="big",
		comma=", ",
		pop=14.4,
	),
	WritingSystem(
		name="Sundanese",
		iso=[(362, "Sund")],
		unicode=["SUNDANESE"],
		titleTag="big",
		comma=", ",
		pop=38,
	),
	WritingSystem(
		name="Brahmi",
		iso=[
			(300, "Brah", "Brahmi"),
		],
		unicode=["BRAHMI"],
		titleTag="big",
		comma=", ",
	),
	WritingSystem(
		name="Tamil",
		iso=[
			(346, "Taml", "Tamil"),
		],
		unicode=["TAMIL"],
		titleTag="big",
		# Parent scripts: Brahmi, Tamil-Brahmi, Pallava
		comma=", ",
		pop=70,
	),
	WritingSystem(
		name="Telugu",
		iso=[(340, "Telu")],
		unicode=["TELUGU"],
		titleTag="big",
		comma=", ",
		pop=74,
	),
	WritingSystem(
		name="Thai",
		iso=[(352, "Thai")],
		unicode=["THAI"],
		titleTag="big",
		comma=", ",
		pop=38,
	),
	# _____________________________________________________
	WritingSystem(
		name="Syriac",
		iso=[(135, "Syrc")],
		unicode=["SYRIAC"],
		titleTag="b",
		direction="rtl",
		comma="، ",
		pop=8,  # Syriac=0.4, Lontara=7.6
		# Lontara is a separate script according to Wikipedia
		# but not according to Unicode
	),
	WritingSystem(
		name="Tibetan",
		iso=[(330, "Tibt")],
		unicode=["TIBETAN"],
		titleTag="big",
		comma=", ",  # almost not used expect in numbers!
		pop=5,
	),
	WritingSystem(
		name="Georgian",
		iso=[(240, "Geor")],
		unicode=["GEORGIAN"],
		titleTag="big",
		comma=", ",
		pop=4.5,
	),
	WritingSystem(
		name="Mongolian",
		iso=[(145, "Mong")],
		unicode=["MONGOLIAN"],
		titleTag="big",
		direction="ltr",  # historically ttb?
		comma=", ",
		pop=2,
	),
	WritingSystem(
		name="Thaana",
		iso=[(170, "Thaa")],
		unicode=["THAANA"],
		titleTag="big",
		direction="rtl",
		comma="، ",
		pop=0.35,
	),
	# _____________________________________________________
	WritingSystem(
		name="Javanese",
		iso=[(361, "Java")],
		unicode=["JAVANESE"],
		titleTag="big",
		# Since around 1945 Javanese script has largely been
		# supplanted by Latin script to write Javanese.
	),
	WritingSystem(
		# aka CANADIAN ABORIGINAL or UCAS
		name="Canadian syllabic",
		iso=[(440, "Cans")],
		unicode=["CANADIAN SYLLABICS"],
		titleTag="big",
		comma=", ",
	),
	WritingSystem(
		name="Takri",
		iso=[(321, "Takr")],
		unicode=["TAKRI"],
		titleTag="b",
		# comma="", FIXME
	),
	# _____________________________________________________
	WritingSystem(
		name="SignWriting",
		iso=[(95, "Sgnw")],
		unicode=["SIGNWRITING"],
		titleTag="big",
		direction="ttb",
		comma="𝪇",
	),
	# _____________________________________________________
	WritingSystem(
		name="Adlam",
		iso=[(166, "Adlm")],
		unicode=["ADLAM"],
		titleTag="big",
		direction="rtl",
	),
	WritingSystem(
		name="Avestan",
		iso=[(134, "Avst")],
		unicode=["AVESTAN"],
		titleTag="b",
		direction="rtl",
	),
	WritingSystem(
		name="Glagolitic",
		iso=[(225, "Glag")],
		unicode=["GLAGOLITIC"],  # Unicode 4.1
		titleTag="b",
	),
	WritingSystem(
		name="Khojki",
		iso=[(322, "Khoj")],
		unicode=["KHOJKI"],
		titleTag="big",
	),
	WritingSystem(
		name="Khudabadi",  # aka: "Khudawadi", "Sindhi"
		iso=[(318, "Sind")],
		unicode=["KHUDAWADI"],
		titleTag="big",
	),
	WritingSystem(
		name="N'Ko",
		iso=[(165, "Nkoo")],
		unicode=["NKO"],
		titleTag="big",
	),
	# _____________________________________________________
	# WritingSystem(
	# 	name="Baybayin",
	# 	iso=[(370, "Tglg")],
	# 	unicode=["TAGALOG"],  # added in Unicode 3.2
	# ),
	# WritingSystem(
	# 	name="Rejang",
	# 	iso=[(363, "Rjng")],
	# 	unicode=["REJANG"],
	# ),
	# WritingSystem(
	# 	name="Mandombe",
	# 	unicode=[],
	# ),
	# WritingSystem(
	# 	name="Mwangwego",
	# 	unicode=[],
	# ),
]

for _ws in writingSystemList:
	if not _ws.name:
		raise ValueError(f"empty name in {_ws}")

writingSystemByUnicode = {uni: ws for ws in writingSystemList for uni in ws.unicode}

writingSystemByName = {ws.name: ws for ws in writingSystemList}

writingSystemByLowercaseName = {ws.name.lower(): ws for ws in writingSystemList}

unicodeNextWord = {
	"HALFWIDTH",
	"FULLWIDTH",
	"CANADIAN",
}


def _getWritingSystemFromChar(char: str) -> "WritingSystem | None":
	try:
		unicodeWords = unicodedata.name(char).split(" ")
	except ValueError:
		# if c not in string.whitespace:
		# 	print(f"{c=}, {e}")
		return None
	alias = unicodeWords[0]
	ws = writingSystemByUnicode.get(alias)
	if ws:
		return ws
	if alias in unicodeNextWord:
		return writingSystemByUnicode.get(" ".join(unicodeWords[:2]))
	return None


def _getWritingSystemFromText(
	st: str,
	start: int,
	end: int,
) -> "WritingSystem | None":
	for char in st[start:end]:
		ws = _getWritingSystemFromChar(char)
		if ws:
			return ws
	return None


def getWritingSystemFromText(
	st: str,
	beginning: bool = False,
) -> "WritingSystem | None":
	st = st.strip()
	if not st:
		return None
	# some special first words in unicodedata.name(c):
	# "RIGHT", "ASTERISK", "MODIFIER"
	k = 0 if beginning else (len(st) + 1) // 2 - 1
	ws = _getWritingSystemFromText(st, k, len(st))
	if ws:
		return ws
	return _getWritingSystemFromText(st, 0, k)


def getAllWritingSystemsFromText(
	st: str,
) -> "set[str]":
	st = st.strip()
	if not st:
		return set()
	wsSet = set()
	for char in st:
		ws = _getWritingSystemFromChar(char)
		if ws:
			wsSet.add(ws.name)
	return wsSet