local data = {}
local U = require("Module:string/char")
local candrabindu = U(0xC81)
local anusvAra = U(0xC82)
local visarga = U(0xC83)
local virAma = U(0xCCD)
local nuktA = U(0xCBC)
local avagraha = "ಽ"
local consonants = "ಕಖಗಘಙಚಛಜಝಞಟಠಡಢಣತಥದಧನಪಫಬಭಮಯರಱಲವಶಷಸಹಳೞ"
local consonant = "[" .. consonants .. "]" .. nuktA .. "?"
local acute = U(0x301) -- combining acute
data["pra-Knda"] = {
-- Two-element vowels or vowel seqeunces.
{"a[iuïüö]", {["ai"] = "ಐ", ["au"] = "ಔ", ["aï"] = "ಅಇ", ["aü"] = "ಅಉ",
["aö"] = "ಅಓ"}},
{"l̥̄", "ೡ"},
{"l̥", "ಌ"},
-- Digraphs with 'h':
{".h", {["kh"] = "ಖ", ["gh"] = "ಘ", ["ch"] = "ಛ", ["jh"] = "ಝ",
["ṭh"] = "ಠ", ["ḍh"] = "ಢ", ["th"] = "ಥ", ["dh"] = "ಧ",
["ph"] = "ಫ", ["bh"] = "ಭ",}},
-- Non-ASCII single characters
{".", {["ḹ"] = "ೡ", ["ṃ"] = anusvAra, ["ḥ"] = visarga,
["ṅ"] = "ಙ", ["ñ"] = "ಞ", ["ṇ"] = "ಣ", ["ś"] = "ಶ",
["ṣ"] = "ಷ", ["ā"] = "ಆ", ["ī"] = "ಈ", ["ū"] = "ಊ",
["ĕ"] = "ಎ", ["ŏ"] = "ಒ", ["ṝ"] = "ೠ", ["ḷ"] = "ಳ",
["ṭ"] = "ಟ", ["ḍ"] = "ಡ", ['̈'] = "", ["ṛ"] = "ಋ",}},
-- ASCII letters
{"[a-z]", {n = "ನ", m = "ಮ", y = "ಯ", r = "ರ",
l = "ಲ", v = "ವ", s = "ಸ", a = "ಅ",
i = "ಇ", u = "ಉ", e = "ಏ", o = "ಓ",
k = "ಕ", g = "ಗ", c = "ಚ", j = "ಜ",
t = "ತ", d = "ದ", p = "ಪ", b = "ಬ",
h = "ಹ"}},
{"(ಅ)[%-/]([ಇಉ])", "%1%2"}, -- a-i, a-u for 𑀅𑀇, 𑀅𑀉; must follow rules for "ai", "au"
{"(" .. consonant .. ")$", "%1" .. virAma},
{acute, ""},
-- this rule must be applied twice because a consonant may only be in one capture per operation, so "CCC" will only recognize the first two consonants
{"(" .. consonant .. ")" .. "(" .. consonant .. ")", "%1" .. virAma .. "%2"},
{"(" .. consonant .. ")" .. "(" .. consonant .. ")", "%1" .. virAma .. "%2"},
{"i", "ಇ"},
{"u", "ಉ"},
}
local vowels = {
["ಇ"] = U(0xCBF),
["ಉ"] = U(0xCC1),
["ಋ"] = U(0xCC3),
["ಌ"] = U(0xCE2),
["ಎ"] = U(0xCC6),
["ಏ"] = U(0xCC7),
["ಒ"] = U(0xCCA),
["ಓ"] = U(0xCCB),
["ಆ"] = U(0xCBE),
["ಈ"] = U(0xCC0),
["ಊ"] = U(0xCC2),
["ೠ"] = U(0xCC4),
["ೡ"] = U(0xCE3),
["ಐ"] = U(0xCC8),
["ಔ"] = U(0xCCC),
}
for independentForm, diacriticalForm in pairs(vowels) do
table.insert(data["pra-Knda"], {"(" .. consonant .. ")" .. independentForm, "%1" .. diacriticalForm})
end
-- This must go last, after independent vowels are converted to diacritics, or "aï", "aü" won't work.
table.insert(data["pra-Knda"], {"(" .. consonant .. ")ಅ", "%1"})
-- [[w:Harvard-Kyoto]] to [[w:International Alphabet of Sanskrit Transliteration]]
data["pra-Knda-tr"] = {
[1] = {
["A"] = "ā",
["I"] = "ī",
["U"] = "ū",
["J"] = "ñ",
["T"] = "ṭ",
["D"] = "ḍ",
["N"] = "ṇ",
["G"] = "ṅ",
["z"] = "ś",
["S"] = "ṣ",
["M"] = "ṃ",
["H"] = "ḥ",
["lRR"] = "ḹ",
["/"] = acute,
},
[2] = {
["lR"] = "ḷ",
["RR"] = "ṝ",
},
[3] = {
["R"] = "ṛ",
},
}
return data