Difference between revisions of "Module:Unichar"

From blackwiki
Jump to navigation Jump to search
blackwiki>Erutuon
m (shorter?)
blackwiki>Erutuon
(some visible format characters)
Line 26: Line 26:
 
local U = mw.ustring.char
 
local U = mw.ustring.char
 
local function show(codepoint)
 
local function show(codepoint)
if Unicode_data.is_printable(codepoint) then
+
-- 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)
 
local printed_codepoint = U(codepoint)
 
if mw.ustring.toNFC(printed_codepoint) ~= printed_codepoint then
 
if mw.ustring.toNFC(printed_codepoint) ~= printed_codepoint then

Revision as of 04:12, 8 July 2018

Tests

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



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

function p.unichar(frame)
	local args = frame.args[1] and frame.args or frame:getParent().args
	
	local codepoint = args[1]
	codepoint = codepoint and tonumber(codepoint, 16)
		or errorf("code point in [[hexadecimal]] expected, got %s",
			type(codepoint) ~= "string" and type(codepoint)
				or ("%q"):format(codepoint))
	
	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