Module:Unichar

MyWikiBiz, Author Your Legacy — Friday November 15, 2024
Revision as of 07:44, 16 July 2021 by Zoran (talk | contribs) (Pywikibot 6.4.0)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

Tests

  • U+00A9 © COPYRIGHT SIGN
  • U+0378 <reserved-0378>
  • U+AC00HANGUL SYLLABLE GA
  • U+0020 SPACE



local p = {}

local Unicode_data = require "Module:Unicode data"

local function errorf(level, ...)
	if type(level) == "number" then
		return error(string.format(...), level + 1)
	else -- level is actually the format string.
		return error(string.format(level, ...), 2)
	end
end

 -- from [[Template:Unichar]]
local styles = {
	smallcaps = 'class="smallcaps" style="font-variant: small-caps; font-size: smaller;"',
	monospace = 'style="font-family: monospace, monospace;"', -- why repeat it?
	background_color = 'style="background: lightblue;"', -- for space characters
}
local function style(text, type)
	if not styles[type] then
		errorf("Style %s not recognized", type)
	end
	return ('<span %s>%s</span>'):format(styles[type], text)
end

local U = mw.ustring.char
local function show(codepoint)
	-- Format characters that at least are visible to me.
	-- The parentheses will short-circuit the evaluation of some of the conditions.
	-- Arabic number sign, Arabic sign sanah, Arabic footnote marker, Arabic sign safha, Arabic sign samvat
	if 0x600 <= codepoint and (codepoint <= 0x604
			-- Arabic end of ayah, Syriac abbreviation mark, Arabic disputed end of ayah
			or codepoint == 0x6DD or codepoint == 0x70F or codepoint == 0x8E2
			-- interlinear annotation anchor, separator, terminator
			or 0xFFF9 <= codepoint and (codepoint <= 0xFFFB
			-- shorthand format letter overlap, continuing overlap, downstep
			or 0x1BCA0 <= codepoint and codepoint <= 0x1BCA3))
			or Unicode_data.is_printable(codepoint) then
		local printed_codepoint = U(codepoint)
		if mw.ustring.toNFC(printed_codepoint) ~= printed_codepoint then
			-- Prevent MediaWiki software from normalizing the character.
			printed_codepoint = ("&#x%X;"):format(codepoint)
		end
		if Unicode_data.is_combining(codepoint) then
			printed_codepoint = "◌" .. printed_codepoint
		end
		if Unicode_data.is_whitespace(codepoint) then
			printed_codepoint = style(printed_codepoint, "background_color")
		end
		return printed_codepoint
	else
		return ""
	end
end

local function u_plus(codepoint)
	return ("U+%04X"):format(codepoint)
end

local function get_codepoint(args, arg)
	local val = args[arg]
	local is_negative = false
	
	if type(val) ~= "string" then
		errorf("code point in [[hexadecimal]] expected, got %s", type(val))
	elseif val:find('^%s*%-') then
		-- Negative number strings yield a bizarre value:
		-- tonumber("-1", 16) -> 1.844674407371e+19.
		-- Strip initial minus.
		val = val:match("%-(.+)")
		is_negative = true
	end
	
	local codepoint = tonumber(val, 16)
		or errorf("code point in [[hexadecimal]] expected, got %q", val)
	
	if is_negative then
		codepoint = -codepoint
	end
	
	if not (0 <= codepoint and codepoint <= 0x10FFFF) then
		errorf("code point %d out of range", codepoint)
	end
	
	return codepoint
end

function p.unichar(frame)
	local args = frame.args[1] and frame.args or frame:getParent().args
	local codepoint = get_codepoint(args, 1)
	
	local name_or_label = Unicode_data.lookup_name(codepoint)
	local is_label = name_or_label:sub(1, 1) == "<"
	
	return ("%s %s %s"):format(
		style(u_plus(codepoint), "monospace"),
		show(codepoint),
		is_label and name_or_label or style(name_or_label, "smallcaps"))
end

return p