local data = {}
local U = require("Module:string/char")
local anusvAra = U(0x902)
local visarga = U(0x903)
local virAma = U(0x94D)
local avagraha = "ऽ"
local consonants = "कखगघङचछजझञटठडढणतथदधनपफबभमयरलळवशषसह"
local consonant = "[" .. consonants .. "]"
local Lconsonants = "kgṅcjñṭḍṇtdnpbmyrlvśṣs" -- excludes h and ḷ
local Lcons1 = "[" .. Lconsonants .. "h]"
local Lcons2 = "[" .. Lconsonants .. "]"
local Lvowels = "āeĕēiïīoŏōuŭuṛṝl̥̄l̥ḹ" -- excludes a
local Lvowel1 = "[" .. Lvowels .. "]"
local Lvowel2 = "[" .. Lvowels .. "a]"
local accents = U(0x301, 0x306, 0x308) -- combining acute, breve and diaeresis
local accent = "[" .. accents .. "]"
local acute = U(0x301) -- combining acute
data["sa"] = {
-- Resolve ḷ where possible
{"("..Lcons1..")(ḷ)", "%1l̥"}, -- It's a vowel next to a consonant
{"(ḷ)("..Lcons2..")", "l̥%2"},
{"("..Lvowel1..accent.."?)(ḷ)", "%1L"}, -- It's a consonant next to a vowel.
{"(ḷ)(h?"..Lvowel2..")", "L%2"},
-- Vowels and modifiers. Do the diphthongs and diaereses first.
{"ai", "ऐ"},
{"au", "औ"},
{".", { ["ä"] = "अ", ["ö"] = "ओ", ["ï"] = "इ", ["ü"] = "उ",
["a"] = "अ", ["ā"] = "आ", ["i"] = "इ", ["ī"] = "ई",
["u"] = "उ", ["ū"] = "ऊ", ["e"] = "ए", ["o"] = "ओ",
["ṝ"] = "ॠ", ["ṛ"] = "ऋ", ["ḹ"] = "ॡ", ["ḷ"] = "ऌ", }},
{"r̥", "ऋ"},
{"l̥", "ऌ"},
{"l̥̄", "ॡ"},
{"(अ)[%-/]([इउ])", "%1%2"}, -- a-i, a-u for अइ, अउ; must follow rules for "ai", "au"
-- Two-letter consonants must go before h.
{".h", {["kh"] = "ख", ["gh"] = "घ", ["ch"] = "छ", ["jh"] = "झ",
["ṭh"] = "ठ", ["ḍh"] = "ढ", ["th"] = "थ", ["dh"] = "ध",
["ph"] = "फ", ["bh"] = "भ", }},
-- Other letters
{".", {h = "ह",
-- Other stops.
k = "क", g = "ग", c = "च", j = "ज",
["ṭ"] = "ट", ["ḍ"] = "ड", t = "त", d = "द",
p = "प", b = "ब",
-- Nasals.
["ṅ"] = "ङ", ["ñ"] = "ञ", ["ṇ"] = "ण", ["n"] = "न",
n = "न", m = "म",
-- Remaining consonants.
y = "य", r = "र", l = "ल", L = "ळ",
v = "व", ["ś"] = "श", ["ṣ"] = "ष", s = "स",
["ẏ"] = "य़", ["ṃ"] = anusvAra, ["ḥ"] = visarga,
["'"] = avagraha,
}},
-- 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. Must follow all consonant conversions.
{"(" .. consonant .. ")(" .. consonant .. ")", "%1" .. virAma .. "%2"},
{"(" .. consonant .. ")(" .. consonant .. ")", "%1" .. virAma .. "%2"},
{"(" .. consonant .. ")$", "%1" .. virAma},
{acute, ""},
}
local vowels = {
["इ"] = U(0x93F),
["उ"] = U(0x941),
["ऋ"] = U(0x943),
["ऌ"] = U(0x962),
["ए"] = U(0x947),
["ओ"] = U(0x94B),
["आ"] = U(0x93E),
["ई"] = U(0x940),
["ऊ"] = U(0x942),
["ॠ"] = U(0x944),
["ॡ"] = U(0x963),
["ऐ"] = U(0x948),
["औ"] = U(0x94C),
}
-- Convert independent vowels to diacritics after consonants. Must go after all consonant conversions.
local independentForms = {}
for independentForm in pairs(vowels) do
-- assert(mw.ustring.len(independentForm) == 1)
table.insert(independentForms, independentForm)
end
table.insert(data["sa"], {"%f[^" .. consonants .. "]([" .. table.concat(independentForms) .. "])", vowels})
-- This must go last, after independent vowels are converted to diacritics, or "aï", "aü" won't work.
table.insert(data["sa"], {"(" .. consonant .. ")अ", "%1"})
-- [[w:Harvard-Kyoto]] to [[w:International Alphabet of Sanskrit Transliteration]]
data["sa-tr"] = {
[1] = {
["A"] = "ā",
["ĕ"] = "e",
["I"] = "ī",
["U"] = "ū",
["ŏ"] = "o",
["J"] = "ñ",
["T"] = "ṭ",
["D"] = "ḍ",
["N"] = "ṇ",
["G"] = "ṅ",
["z"] = "ś",
["S"] = "ṣ",
["M"] = "ṃ",
["H"] = "ḥ",
["lRR"] = "ḹ",
["/"] = acute,
},
[2] = {
["lR"] = "l̥", -- was "ḷ",
["RR"] = "ṝ",
},
[3] = {
["R"] = "ṛ",
},
}
return data
