local p = {}
local sparse_concat = require("Module:table").sparseConcat
local ustring = mw.ustring
local U = ustring.char
local get_codepoint = ustring.codepoint
local ufind = ustring.find
local ugsub = ustring.gsub
local decompose = ustring.toNFD
local str_gsub = string.gsub
local UTF8_char = "[%z\1-\127\194-\244][\128-\191]*" -- roughly equivalent to "." in Ustring patterns
local one_UTF8_char_or_none = "[%z\1-\127\194-\244]?[\128-\191]*" -- roughly equivalent to ".?" in Ustring patterns
local subscript = U(0x345) -- iota subscript (ypogegrammeni)
local macron = U(0x304) -- macron
local spacing_macron = U(0xAF)
local modifier_macron = U(0x2C9) -- modifier letter macron
local breve = U(0x306) -- breve
local spacing_breve = "˘" -- spacing breve
local diaeresis = U(0x308) -- diaeresis
local rough = U(0x314) -- rough breathing (reversed comma)
local smooth = U(0x313) -- smooth breathing (comma)
local acute = U(0x301) -- acute
local grave = U(0x300) -- grave
local circumflex = U(0x342) -- Greek circumflex (perispomeni)
local question_mark = U(0x37E) -- Greek question mark
local spacing_rough = "῾" -- spacing rough breathing
local spacing_smooth = "᾿" -- spacing smooth breathing
local combining_diacritic = table.concat{
"[",
macron, breve,
rough, smooth, diaeresis,
acute, grave, circumflex,
subscript,
"]",
}
-- The numbers are used to sort series of diacritics.
local diacritic_position = {
[macron] = 1,
[breve] = 2,
[rough] = 3,
[smooth] = 3,
[diaeresis] = 3,
[acute] = 4,
[grave] = 4,
[circumflex] = 4,
[subscript] = 5,
}
-- Perform a function on each Unicode character in a string.
local function for_each(str, func)
for char in string.gmatch(str, UTF8_char) do
func(char)
end
end
--[=[ This function arranges diacritics in the following order:
1. macron or breve
2. breathings or diaeresis
3. acute, circumflex, or grave
4. iota subscript
Used by [[Module:typing-aids]].
Returns an error if a sequence of diacritics contains more than one
of each category.
]=]
local function get_relative_position(diacritic1, diacritic2)
return diacritic_position[diacritic1] < diacritic_position[diacritic2]
end
local function chars_to_table(chars)
local t = {}
local i = 0
for char in string.gmatch(chars, "[%z\1-\127\194-\244][\128-\191]*") do
i = i + 1
t[i] = char
end
return t
end
local function reorder_diacritic_sequence(diacritics)
diacritics = chars_to_table(diacritics)
table.sort(diacritics, get_relative_position)
return table.concat(diacritics)
end
function p.reorder_diacritics(text)
return (ugsub(decompose(text),
combining_diacritic .. combining_diacritic .. "+",
reorder_diacritic_sequence))
end
local multiple = {
["_i"] = subscript,
}
local single = {
["a"] = "α", ["A"] = "Α",
["b"] = "β", ["B"] = "Β",
["c"] = "ξ", ["C"] = "Ξ",
["d"] = "δ", ["D"] = "Δ",
["e"] = "ε", ["E"] = "Ε",
["f"] = "φ", ["F"] = "Φ",
["g"] = "γ", ["G"] = "Γ",
["h"] = "η", ["H"] = "Η",
["i"] = "ι", ["I"] = "Ι",
["k"] = "κ", ["K"] = "Κ",
["l"] = "λ", ["L"] = "Λ",
["m"] = "μ", ["M"] = "Μ",
["n"] = "ν", ["N"] = "Ν",
["o"] = "ο", ["O"] = "Ο",
["p"] = "π", ["P"] = "Π",
["q"] = "θ", ["Q"] = "Θ",
["r"] = "ρ", ["R"] = "Ρ",
["s"] = "σ", ["S"] = "Σ",
["t"] = "τ", ["T"] = "Τ",
["u"] = "υ", ["U"] = "Υ",
["v"] = "ϝ", ["V"] = "Ϝ",
["w"] = "ω", ["W"] = "Ω",
["x"] = "χ", ["X"] = "Χ",
["y"] = "ψ", ["Y"] = "Ψ",
["z"] = "ζ", ["Z"] = "Ζ",
-- vowel length
["_"] = macron, [spacing_macron] = macron, [modifier_macron] = macron,
["^"] = breve, [spacing_breve] = breve,
-- diaeresis and breathings
["+"] = diaeresis, ["("] = rough, [")"] = smooth,
-- accents
["/"] = acute, ["\\"] = grave,
["="] = circumflex, ["{{=}}"] = circumflex, ["~"] = circumflex,
-- punctuation
["'"] = "’",
["?"] = question_mark,
[";"] = "·",
["*"] = "", -- place after s to prevent it from turning into final sigma
-- pipe
["!"] = "|",
}
local function convert_s_to_sigma(text)
text = str_gsub(text,
"s(" .. one_UTF8_char_or_none .. ")",
function (following)
return ((following == ""
or following ~= "*" and following ~= "-" and ufind(following, "[%s%p]"))
and "ς"
or "σ") .. following
end)
return text
end
local function combining_to_spacing(text)
for _, accents in ipairs{ { rough, spacing_rough }, { smooth, spacing_smooth } } do
local combining, spacing = unpack(accents)
text = str_gsub(text,
"(" .. one_UTF8_char_or_none .. ")" .. combining,
function (preceding)
if preceding == "" then
return spacing
else
return preceding .. combining
end
end)
end
return text
end
function p.to_Greek(text)
if type(text) ~= "string" then
error("first argument to to_greek should be string, not " .. type(text))
end
text = convert_s_to_sigma(text)
for k, v in pairs(multiple) do
text = str_gsub(text, k, v)
end
text = str_gsub(text, UTF8_char, single)
text = combining_to_spacing(text)
return p.reorder_diacritics(text)
end
function p.to_Greek_t(frame)
local args = {}
for k, v in pairs(frame:getParent().args) do
if k == 1 then
v = mw.text.trim(v)
if v == "" then
v = nil
end
args[k] = v
end
end
if not args[1] then
if mw.title.getCurrentTitle().nsText == "Template" then
args[1] = "le/cis"
else
error("Parameter 1 is required.")
end
end
return p.to_Greek(args[1])
end
local function process(char)
if char == "" then
return char
end
local entity = ("&#x%X;"):format(get_codepoint(char))
if diacritic_position[char] then
return "◌" .. entity
else
return entity
end
end
function p.show_shortcuts(frame)
local output = { '{| class="wikitable"' }
local function comp(item1, item2)
-- non-letters after letters
if item1:find("^%a$") ~= item2:find("^%a$") then
return item1:find("^%a$")
end
local lower1, lower2 = item1:lower(), item2:lower()
-- capitals before lowercase
if lower1 == lower2 then
return item1 < item2
-- otherwise case-insensitive sorting
else
return lower1 < lower2
end
end
local i = 1
for k, v in require("Module:Table").sortedPairs(single, comp) do
i = i + 1
output[i] = '| <code>' .. k .. '</code> || <span lang="grc">' .. process(v) .. '</span>'
if i % 3 == 0 then -- 3 because each row consists of row syntax |- and two pairs of cells
i = i + 1
output[i] = '|-'
end
end
table.insert(output, '|}')
return table.concat(output, '\n')
end
return p
