Module:Hani-sortkey/sandbox

---

The following documentation is located at Module:Hani-sortkey/sandbox/documentation. ^[edit] Categories were auto-generated by Module:module categorization. ^[edit]

Useful links: root page • root page’s subpages • links • transclusions • testcases • sandbox of (diff)

This module was used to develop the ideographic-description-sequence-detecting function in Module:Hani-sortkey.

---

• ⿰亻革 (⿰人00革00)
• ⿰亻革家 (⿰人00革00宀07)
• ⿺辶⿳穴⿲月⿱⿲幺言幺⿲長馬長刂心／⿺辶⿳穴⿲月⿱⿲幺言幺⿲长马长刂心 (⿺辵00⿳穴00⿲月00⿱⿲幺00言00幺00⿲長00馬00長00刀00心00)
• ⿺辶⿳穴⿲月⿱⿲幺言幺⿲長馬長刂心⿱苦⿲⿰⿹耳舌鼻⿱⿱平⿰惡意⿱眼⿰淨染⿰⿱女子身 (⿺辵00⿳穴00⿲月00⿱⿲幺00言00幺00⿲長00馬00長00刀00心00⿱艸05⿲⿰⿹耳00舌00鼻00⿱⿱干02⿰心08心09⿱目06⿰水08木05⿰女03身00)
• ⿺辶⿳穴⿲月⿱⿲幺言幺⿲長馬長刂心⿺辶⿳穴⿲月⿱⿲幺言幺⿲長馬長刂心麵／⿺辶⿳穴⿲月⿱⿲幺言幺⿲长马长刂心⿺辶⿳穴⿲月⿱⿲幺言幺⿲长马长刂心面 (⿺辵00⿳穴00⿲月00⿱⿲幺00言00幺00⿲長00馬00長00刀00心00⿺辵00⿳穴00⿲月00⿱⿲幺00言00幺00⿲長00馬00長00刀00心00麥09)

---

Hypothetical defective examples:

• ⿰石 (⿰石00)
• ⿰石⿳女子 (⿰石00⿳女00子00)
• ⿳女子 (⿳女00子00)

local export = {}

local namespace = mw.title.getCurrentTitle().nsText

local substring = mw.ustring.sub

local function log(...)
	if namespace == "Module" then
		mw.log(...)
	end
end

--[[
	The number of characters or ideographic sequences that must follow each
	ideographic description character.
]]
local IDchars = {
	["⿰"] = 2,
	["⿱"] = 2,
	["⿲"] = 3,
	["⿳"] = 3,
	["⿴"] = 2,
	["⿵"] = 2,
	["⿶"] = 2,
	["⿷"] = 2,
	["⿸"] = 2,
	["⿹"] = 2,
	["⿺"] = 2,
	["⿻"] = 2,
	
	--[[
	-- in future perhaps: https://www.unicode.org/L2/L2018/18012-irgn2273-four-new-idcs.pdf
	[mw.ustring.char(0x2FFC)] = 2,
	[mw.ustring.char(0x2FFD)] = 2,
	[mw.ustring.char(0x2FFE)] = 1,
	[mw.ustring.char(0x2FFF)] = 1,
	--]]
}

--[[
	Returns the index in the string where the ideographic description sequence
	(IDS) ends, or the index of the end of the string. Iterates whenever
	another ideographic description character (IDC) is found.
]]
local function findEndOfIDS(text, IDchar, i)
	if not ( text and IDchar and i) then
		return nil
	end
	
	local j = i
	local component = 1
	
	-- Number of components expected after current IDC.
	local components = IDchars[IDchar]
	
	while component <= components do
		j = j + 1
		
		local char = substring(text, j, j)
		
		if char == "" then
			break
		elseif IDchars[char] then
			j = findEndOfIDS(text, char, j)
		end
		
		component = component + 1
	end
	
	--[[
		If the expected number of components has been found,
		return the current index in the text.
	]]
	if component - components == 1 then
		return j
	else
		return nil
	end
end

local module_cache = {}

function export.getData(char, returnModule)
	if type(char) == "string" then
		char = mw.ustring.codepoint(char)
	elseif type(char) ~= number then
		error("getData must operate on a single character or codepoint.")
	end
	
	local data = mw.loadData("Module:Hani-sortkey/data/sandbox")
	
	local sectionStart = {
		0x3400,
		0xFA0E,
		0x20000,
		0x30000,
		math.huge
	}
	
	for k, v in ipairs(sectionStart) do
		if char < v then
			local start = 5 * (char - sectionStart[k - 1]) + 1
			return data[k - 1]:sub(start, start + 4)
		end
	end
end

local unsupported_data

function export.makeSortKey(text, lang, sc)
	local scripts = {
		Hani = true,
		Hans = true,
		Hant = true,
		Jpan = true,
		Kore = true
	}
	if sc and not scripts[sc] then
		return mw.ustring.upper(text)
	end
	
	local sort = {}
	
	local i = 1
	while i <= mw.ustring.len(text) do
		local character = substring(text, i, i)
		--[=[
			If we encounter an ideographic description character (IDC),
			find out if it begins a valid ideographic description sequence (IDS).
			
			If the IDS is valid and a sortkey for it is listed in
			[[Module:Hani-sortkey/data/unsupported]], then return
			the sortkey, and move to the next character after the
			IDS.
			
			Otherwise, insert the IDC into the sortkey and move to the next
			character after the IDC.
			
			If the IDS is valid and no sortkey for it is found, track it.
		]=]
		if IDchars[character] then
			local j = findEndOfIDS(text, character, i)
			local IDS, data
			if j then
				IDS = substring(text, i, j)
				unsupported_data = unsupported_data or mw.loadData("Module:Hani-sortkey/data/unsupported")
				data = unsupported_data[IDS]
			end
			if not data then
				if IDS then
					require("Module:debug").track("Hani-sortkey/IDS-without-sortkey")
					mw.log("ideographic description sequence without sortkey: '"
						.. IDS .. "'")
				else
					require("Module:debug").track("Hani-sortkey/invalid-IDS")
					mw.log("invalid ideographic description sequence at the beginning of '"
						.. substring(text, i) .. "'")
				end
			end
			if IDS and data then
				table.insert(sort, data)
				i = j
			else
				table.insert(sort, character)
			end
		else
			table.insert(sort, export.getData(character) or character)
		end
		i = i + 1
	end
	
	sort = table.concat(sort)

	return sort
end

return export