Module:Sandbox/DePiep/uchar-helper

-- FROM [module:Unicode data] 28-4-2022

-- Creates a function to look up data in a module that contains "singles" (a
-- code point-to-data map) and "ranges" (an array containing arrays that contain
-- the low and high code points of a range and the data associated with that
-- range).
-- "loader" loads and returns the "singles" and "ranges" tables.
-- "match_func" is passed the code point and either the data or the "dots", and
-- generates the final result of the function.
-- The varargs ("dots") describes the default data to be returned if there wasn't
-- a match.
-- In case the function is used more than once, "cache" saves ranges that have
-- already been found to match, or a range whose data is the default if there
-- was no match.
local function memo_lookup(data_module_subpage, match_func, ...)
	local dots = { ... }
	local cache = {}
	local singles, ranges

	return function (codepoint)
		if not singles then
			local data_module = loader[data_module_subpage]
			singles, ranges = data_module.singles, data_module.ranges
		end

		if singles[codepoint] then
			return match_func(codepoint, singles[codepoint])
		end

		local range = binary_range_search(codepoint, cache)
		if range then
			return match_func(codepoint, manual_unpack(range, 3))
		end
		
		local range, index = binary_range_search(codepoint, ranges)
		if range then
			table.insert(cache, range)
			table.sort(cache, compare_ranges)
			return match_func(codepoint, manual_unpack(range, 3))
		end
		
		if ranges[index] then
			local dots_range
			if codepoint > ranges[index][2] then
				dots_range = {
					ranges[index][2] + 1,
					ranges[index + 1] and ranges[index + 1][1] - 1 or 0x10FFFF,
					unpack(dots)
				}
			else -- codepoint < range[index][1]
				dots_range = {
					ranges[index - 1] and ranges[index - 1][2] + 1 or 0,
					ranges[index][1] - 1,
					unpack(dots)
				}
			end
			table.sort(cache, compare_ranges)
		end
		
		return match_func(codepoint)
	end
end
Usage
