Módulo:Lang

De Bestiario del Hypogripho

Progreso[editar]

Este módulo está en proceso de ser traducido al español, está siendo importado desde en:Módulo:Lang. Es bastante complejo y tiene muchas dependencias que también tendrán que importarse.
A continuación se muestra una lista de tareas a realizar antes de que el módulo se pueda utilizar:

  • Importar y traducir dependencias. ✓ Hecho
  • Traducir el script principal (este). ✓ Hecho
  • Documentar el módulo (importante). ✓ Hecho
  • Traducir los datos: ✓ Hecho
    • Idiomas (códigos en uso) ✓ Hecho
    • Idiomas (códigos en desuso) ✓ Hecho
    • Regiones ✓ Hecho
    • Sistemas de escritura ✓ Hecho
    • Variantes ✓ Hecho
  • Crear categorías de seguimiento ✓ Hecho
  • Importar todos los tests ✓ Hecho
  • Limpiar y publicar los scripts para generar las tablas de datos ✓ Hecho

Tras consultar a la UNE en Febrero/Marzo se publicará un listado oficial de los nombres en español, pero hasta ese momento hay que hacer apaños con lo que hay y completar el resto en inglés. Para generar los listados traducidos se están utilizando varias fuentes:

  • Wikcionario en español
  • Wikipedia en español (nombres de artículo)
  • Etiquetas en español de Wikidata

Documentación[editar]

Este módulo pretende generar el marcado html correcto para textos que no estén en español y se usen en la wiki. Tiene el propósito secundario de facilitar el correcto renderizado de textos en otros idiomas. El módulo agrupa el procesado de las plantillas {{Lang}}, {{lang-xx}} y {{Transl}} en un único módulo que utilice un conjunto de datos consolidado. También comprueba los posibles errores para asegurarse de que el html se renderiza correctamente para los navegadores y lectores de pantalla.

AVISO: Si vas a hacer cambios en este módulo, ponlos antes en la zona de pruebas y ejecuta los tests para comprobar que todo va bien.

Conjunto de datos[editar]

El conjunto de datos especificado en Módulo:Lang/data y sus módulos de datos incluidos:

Módulos de datos suplementarios:

Actualización[editar]

Para actualizar los datos se deben seguir los siguientes pasos:

Para generar los archivos mencionados se puede utilizar el código que se encuentra en este repositorio junto con las instrucciones.

Casos de prueba[editar]

Categorías de seguimiento[editar]


--[=[

Implementación en Lua de las plantillas {{lang}}, {{lang-xx}}, y {{transl}} reemplazando varias plantillas auxiliares. 

]=]

require('Módulo:No globals');

local getArgs = require ('Módulo:Arguments').getArgs;
local unicode = require ("Módulo:Unicode data");								-- for is_latin() and is_rtl()
local yesno = require ('Módulo:Yesno');

local lang_data =  mw.loadData ('Módulo:Lang/data');							-- language name override and transliteration tool-tip tables
local lang_name_table = lang_data.lang_name_table;								-- language codes, names, regions, scripts, suppressed scripts
local lang_table = lang_data.lang_name_table.lang;
local lang_dep_table = lang_data.lang_name_table.lang_dep;
local lang_es_table = lang_data.lang_name_table.lang_es;
local lang_es_dep_table = lang_data.lang_name_table.lang_es_dep;
local script_table = lang_data.lang_name_table.script;
local region_table = lang_data.lang_name_table.region;
local variant_table = lang_data.lang_name_table.variant;
local suppressed_table = lang_data.lang_name_table.suppressed;
local override_table = lang_data.override;

local synonym_table = mw.loadData ('Módulo:Lang/ISO 639 synonyms');				-- ISO 639-2/639-2T code translation to 639-1 code

local namespace = mw.title.getCurrentTitle().namespace;							-- used for categorization

local this_wiki_lang = mw.language.getContentLanguage().code;					-- get this wiki's language

local initial_style_state;														-- set by lang_xx_normal() and lang_xx_italic()

local maint_cats = {};															-- maintenance categories go here
local maint_msgs = {};															-- and their messages go here




--[[--------------------------< P A G E _ E X I S T S >--------------------------------------------------------

Returns true if the page exists; false otherwise. The page exists if it's contents can be transcluded.

]]

local function page_exists( name )
	local page = mw.title.makeTitle(0, name);
	return page.exists;
	--local transcluder = require("Módulo:Transcluder");
	--return pcall(transcluder.get, name);
end


--[[--------------------------< I S _ S E T >------------------------------------------------------------------

Returns true if argument is set; false otherwise. Argument is 'set' when it exists (not nil) or when it is not an empty string.

]]

local function is_set( var )
	return not (var == nil or var == '');
end


--[[--------------------------< I N V E R T  _ I T A L I C S >-------------------------------------------------

This function attempts to invert the italic markup a args.text by adding/removing leading/trailing italic markup
in args.text.  Like |italic=unset, |italic=invert disables automatic italic markup.  Individual leading/trailing
apostrophes are converted to their html numeric entity equivalent so that the new italic markup doesn't become
bold markup inadvertently.

Leading and trailing wiki markup is extracted from args.text into separate table elements.  Addition, removal,
replacement of wiki markup is handled by a string.gsub() replacement table operating only on these separate elements.
In the string.gsub() matching pattern, '.*' matches empty string as well as the three expected wiki markup patterns.

This function expects that markup in args.text is complete and correct; if it is not, oddness may result.

]]

local function invert_italics (source)
	local invert_pattern_table = {												-- leading/trailing markup add/remove/replace patterns
		[""]="\'\'",															-- empty string becomes italic markup
		["\'\'"]="",															-- italic markup becomes empty string
		["\'\'\'"]="\'\'\'\'\'",												-- bold becomes bold italic
		["\'\'\'\'\'"]="\'\'\'",												-- bold italic become bold
		};
	local seg = {};

	source = source:gsub ("%f[\']\'%f[^\']", '&#39;');							-- protect single quote marks from being interpreted as bold markup

	seg[1] = source:match ('^(\'\'+%f[^\']).+') or '';							-- get leading markup, if any; ignore single quote
	seg[3] = source:match ('.+(%f[\']\'\'+)$') or '';							-- get trailing markup, if any; ignore single quote

	if '' ~= seg[1] and '' ~= seg[3] then										-- extract the 'text' 
		seg[2] = source:match ('^\'\'+%f[^\'](.+)%f[\']\'\'+$')					-- from between leading and trailing markup
	elseif '' ~= seg[1] then
		seg[2] = source:match ('^\'\'+%f[^\'](.+)')								-- following leading markup
	elseif '' ~= seg[3] then
		seg[2] = source:match ('(.+)%f[\']\'\'+$')								-- preceding trailing markup
	else
		seg[2] = source															-- when there is no markup
	end

	seg[1] = invert_pattern_table[seg[1]] or seg[1];							-- replace leading markup according to pattern table
	seg[3] = invert_pattern_table[seg[3]] or seg[3];							-- replace leading markup according to pattern table

	return table.concat (seg);													-- put it all back together and done
end


--[[--------------------------< V A L I D A T E _ I T A L I C >------------------------------------------------

validates |italic= or |italics= assigned values.

When |italic= is set and has an acceptible assigned value, return the matching css font-style property value or,
for the special case 'default', return nil.

When |italic= is not set, or has an unacceptible assigned value, return nil and a nil error message.

When both |italic= and |italics= are set, returns nil and a 'conflicting' error message.

The return value nil causes the calling lang, lang_xx, or transl function to set args.italic according to the template's
defined default ('inherit' for {{lang}}, 'inherit' or 'italic' for {{lang-xx}} depending on
the individual template's requirements, 'italic' for {{transl}}) or to the value appropriate to |script=, if set ({{lang}}
and {{lang-xx}} only).

Accepted values and the values that this function returns are:
	nil		-	when |italic= absent or not set; returns nil
	default	-	for completeness, should rarely if ever be used; returns nil
	yes		-	force args.text to be rendered in italic font; returns 'italic'
	no		-	force args.text to be rendered in normal font; returns 'normal'
	unset	-	disables font control so that font-style applied to text is dictated by markup inside or outside the template; returns 'inherit'
	invert	-	disables font control so that font-style applied to text is dictated by markup outside or inverted inside the template; returns 'invert'

]]

local function validate_italic (args)
	local properties = {['yes'] = 'italic', ['no'] = 'normal', ['unset'] = 'inherit', ['invert'] = 'invert', ['default'] = nil};
	
	local count = 0
	for _, arg in pairs {'italic', 'italics', 'i'} do
		if args[arg] then
			count = count + 1
		end
	end

	if count > 1 then															-- return nil and an error message if more than one is set 
		return nil, 'solo se puede especificar uno entre |italic=, |italics=, o |i=';
	end
	
	return properties[args.italic or args.italics or args.i], nil;				-- return an appropriate value and a nil error message
end


--[=[--------------------------< V A L I D A T E _ C A T _ A R G S >----------------------------------------------------------

Default behavior of the {{lang}} and {{lang-xx}} templates is to add categorization when the templates are used in mainspace.
This default functionality may be suppressed by setting |nocat=yes or |cat=no.  This function selects one of these two parameters
to control categorization.

Because having two parameters with 'opposite' names and 'opposite' values is confusing, this function accepts only affirmative
values for |nocat= and only negative values for |cat=; in both cases the 'other' sense (and non-sense) is not accepted and the
parameter is treated as if it were not set in the template.

Sets args.nocat to true if categorization is to be turned off; to nil if the default behavior should apply.

Accepted values for |nocat= are the text strings:
	'yes', 'y', 'true', 't', on, '1'		-- [[Module:Yesno]] returns logical true for all of these; false or nil else
for |cat=
	'no', 'n', 'false', 'f', 'off', '0'		-- [[Module:Yesno]] returns logical false for all of these; true or nil else

]=]

local function validate_cat_args (args)
	if not (args.nocat or args.cat) then										-- both are nil, so categorize
		return;
	end
	
	if false == yesno (args.cat) or true == yesno (args.nocat) then
		args.nocat = true;														-- set to true when args.nocat is affirmative; nil else (as if the parameter were not set in the template)
	else																		-- args.nocat is the parameter actually used.
		args.nocat = nil;
	end
end


--[[--------------------------< I N _ A R R A Y >--------------------------------------------------------------

Whether needle is in haystack

]]

local function in_array ( needle, haystack )
	if needle == nil then
		return false;
	end
	for n,v in ipairs( haystack ) do
		if v == needle then
			return n;
		end
	end
	return false;
end


--[[--------------------------< F O R M A T _ I E T F _ T A G >------------------------------------------------

prettify ietf tags to use recommended subtag formats:
	code: lower case
	script: sentence case
	region: upper case
	variant: lower case
	private: lower case prefixed with -x-
	
]]

local function format_ietf_tag (code, extlang, script, region, variant, extension, private)
	local out = {};
	
	table.insert (out, code:lower());
	
	if is_set (extlang) then
		table.insert (out, extlang:lower());
	end
	
	if is_set (script) then
		script = script:lower():gsub ('^%a', string.upper);
		table.insert (out, script);
	end

	if is_set (region) then
		table.insert (out, region:upper());
	end
	
	if is_set (variant) then
		table.insert (out, variant:lower());
	end
	
	if is_set (extension) then
		table.insert (out, 'u-' .. extension:lower());
	end

	if is_set (private) then
		table.insert (out, 'x-' .. private:lower());
	end
	
	return table.concat (out, '-');
end


--[[--------------------------< G E T _ I E T F _ P A R T S >--------------------------------------------------

extracts and returns IETF language tag parts:
	primary language subtag (required) - 2 or 3 character IANA language code
	script subtag - four character IANA script code
	region subtag - two-letter or three digit IANA region code
	variant subtag - four digit or 5-8 alnum variant code; only one variant subtag supported
	private subtag - x- followed by 1-8 alnum private code; only supported with the primary language tag

in any one of these forms
	lang					lang-variant
	lang-script				lang-script-variant
	lang-region				lang-region-variant
	lang-script-region		lang-script-region-variant
	lang-x-private	
	
each of lang, script, region, variant, and private, when used, must be valid

Languages with both two- and three-character code synonyms are promoted to the two-character synonym because
the IANA registry file omits the synonymous three-character code; we cannot depend on browsers understanding
the synonymous three-character codes in the lang= attribute.

For {{lang-xx}} templates, the parameters |script=, |region=, and |variant= are supported (not supported in {{lang}}
because those parameters are superfluous to the IETF subtags in |code=)

returns six  values; all lower case.  Valid parts are returned as themselves; omitted parts are returned as empty strings, invalid
parts are returned as nil; the sixth returned item is an error message (if an error detected) or nil.

see http://www.rfc-editor.org/rfc/bcp/bcp47.txt section 2.1

]]

local function get_ietf_parts (source, args_script, args_region, args_variant)
	local code, script, region, variant, private, extlang, extension;								-- ietf tag parts

	if not is_set (source) then
		return nil, nil, nil, nil, nil, nil, nil, 'etiqueta de idioma faltante';
	end
	
	--extract private
	local privPattern = "^(.*)%-x%-(%w%w[%w%-]?[%w%-]?[%w%-]?[%w%-]?[%w%-]?[%w%-]?)$";
	local s2 = source;
	e1 = nil;
	local e1, e2 = s2:match(privPattern);
	if(is_set(e1)) then
		source = e1;
	end
	private = e2;
	
	--extract extension
	local extPattern = "^(.*)%-u%-(%w%w[%w%-]?[%w%-]?[%w%-]?[%w%-]?[%w%-]?[%w%-]?)$";
	s2 = source;
	local e1, e2, e3;
	e1, e2, e3 = s2:match(extPattern);
	if(is_set(e1)) then
		source = e1;
	end
	extension = e2;
	
	
	local pattern = {															-- table of tables holding acceptibe ietf tag patterns and short names of the ietf part captured by the pattern
		
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a%a%a)%-(%a%a)%-(%d%d%d%d)$', 's', 'r', 'v'}, 				-- 1 -  ll-ee-Ssss-RR-variant (where variant is 4 digits)
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a%a%a)%-(%d%d%d)%-(%d%d%d%d)$', 's', 'r', 'v'},				-- 2 -  ll-ee-Ssss-DDD-variant (where region is 3 digits; variant is 4 digits)
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a%a%a)%-(%a%a)%-(%w%w%w%w%w%w?%w?%w?)$', 's', 'r', 'v'},		-- 3 -  ll-ee-Ssss-RR-variant (where variant is 5-8 alnum characters)
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a%a%a)%-(%d%d%d)%-(%w%w%w%w%w%w?%w?%w?)$', 's', 'r', 'v'},	-- 4 -  ll-ee-Ssss-DDD-variant (where region is 3 digits; variant is 5-8 alnum characters)
		
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a%a%a)%-(%d%d%d%d)$', 's', 'v'},						-- 5 -  ll-ee-Ssss-variant (where variant is 4 digits)
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a%a%a)%-(%w%w%w%w%w%w?%w?%w?)$', 's', 'v'},			-- 6 -  ll-ee-Ssss-variant (where variant is 5-8 alnum characters)
		
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a)%-(%d%d%d%d)$', 'r', 'v'},							-- 7 -  ll-ee-RR-variant (where variant is 4 digits)
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%d%d%d)%-(%d%d%d%d)$', 'r', 'v'},						-- 8 -  ll-ee-DDD-variant (where region is 3 digits; variant is 4 digits)
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a)%-(%w%w%w%w%w%w?%w?%w?)$', 'r', 'v'},				-- 9 -  ll-ee-RR-variant (where variant is 5-8 alnum characters)
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%d%d%d)%-(%w%w%w%w%w%w?%w?%w?)$', 'r', 'v'},				-- 10 - ll-ee-DDD-variant (where region is 3 digits; variant is 5-8 alnum characters)
		
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%d%d%d%d)$', 'v'},										-- 11 - ll-ee-variant (where variant is 4 digits)
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%w%w%w%w%w%w?%w?%w?)$', 'v'},							-- 12 - ll-ee-variant (where variant is 5-8 alnum characters)
		
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a%a%a)%-(%a%a)$', 's', 'r'},							-- 13 - ll-ee-Ssss-RR
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a%a%a)%-(%d%d%d)$', 's', 'r'},						-- 14 - ll-ee-Ssss-DDD (region is 3 digits)
		
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a%a%a)$', 's'},										-- 15 - ll-ee-Ssss
		
		{'^(%a%a%a?%-?%a?%a?%a?)$'},														-- 18 - ll
		
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%a%a)$', 'r'},											-- 16 - ll-ee-RR
		{'^(%a%a%a?%-?%a?%a?%a?)%-(%d%d%d)$', 'r'},											-- 17 - ll-ee-DDD (region is 3 digits)
		
		
		}

	local t = {};																-- table of captures; serves as a translator between captured ietf tag parts and named variables

	for i, v in ipairs (pattern) do												-- spin through the pattern table looking for a match
		local c1, c2, c3, c4;													-- captures in the 'pattern' from the pattern table go here
	
		c1, c2, c3, c4 = source:match (pattern[i][1]);							-- one or more captures set if source matches pattern[i])
			if c1 then															-- c1 always set on match
				code, extlang = c1:match("^(%a%a%a?)%-?(%a?%a?%a?)$");				-- first capture is always code + extlang
				t = {
					[pattern[i][2] or 'x'] = c2,								-- fill the table of captures with the rest of the captures
					[pattern[i][3] or 'x'] = c3,								-- take index names from pattern table and assign sequential captures
					[pattern[i][4] or 'x'] = c4,								-- index name may be nil in pattern[i] table so "or 'x'" spoofs a name for this index in this table
				};
				script = t.s or '';												-- translate table contents to named variables;
				region = t.r or '';												-- absent table entries are nil so set named ietf parts to empty string for concatenation
				variant= t.v or '';
				break;															-- and done
			end
	end

	--Fix extlang vs region confusion
	if is_set(extlang) and is_set(extlang:find("[A-Z]")) and not is_set(region) then
		region = extlang;
		extlang = nil;
	end

	if not code then
		return nil, nil, nil, nil, nil, nil, nil, table.concat ({'etiqueta de idioma no reconocida: ', source});		-- don't know what we got but it is malformed
	end

	code = code:lower();														-- ensure that we use and return lower case version of this
	
	if not (override_table[code] or lang_table[code] or lang_es_table[code] or synonym_table[code] or lang_dep_table[code] or lang_es_dep_table[code]) then
		return nil, nil, nil, nil, nil, nil, nil, table.concat ({'código de idioma no reconocido: ', code});		-- invalid language code, don't know about the others (don't care?)
	end
	
	if synonym_table[code] then													-- if 639-2/639-2T code has a 639-1 synonym
		table.insert (maint_cats, table.concat ({'Lang y lang-xx con códigos promovidos a ISO 639-1|', code}));
		table.insert (maint_msgs, table.concat ({'código: ', code, ' promovido al código: ', synonym_table[code]}));
		code = synonym_table[code];												-- use the synonym
	end

	if is_set (script) then
		if is_set (args_script) then
			return code, nil, nil, nil, nil, nil, nil, 'etiqueta script redundante';			-- both code with script and |script= not allowed
		end
	else
		script = args_script or '';												-- use args.script if provided
	end 

	if is_set (script) then
		script = script:lower();												-- ensure that we use and return lower case version of this
		if not script_table[script] then
			return code, nil, nil, nil, nil, nil, nil, table.concat ({'sistema de escritura no reconocido: ', script, ' para el código: ', code});	-- language code ok, invalid script, don't know about the others (don't care?)
		end
	end
	if suppressed_table[script] then											-- ensure that code-script does not use a suppressed script
		if in_array (code, suppressed_table[script]) then
			return code, nil, nil, nil, nil, nil, nil, table.concat ({'sistema de escritura: ', script, ' suprimido para el código: ', code});	-- language code ok, script is suppressed for this code
		end
	end

	if is_set (region) then
		if is_set (args_region) then
			return code, nil, nil, nil, nil, nil, nil, 'etiqueta de región redundante';			-- both code with region and |region= not allowed
		end
	else
		region = args_region or '';												-- use args.region if provided
	end 

	if is_set (region) then
		region = region:lower();												-- ensure that we use and return lower case version of this
		if not region_table[region] then
			return code, script, nil, nil, nil, nil, nil, table.concat ({'región no reconocida: ', region, ' para el código: ', code});
		end
	end
	
	if is_set (variant) then
		if is_set (args_variant) then
			return code, nil, nil, nil, nil, nil, nil, 'etiqueta de variante redundante';			-- both code with variant and |variant= not allowed
		end
	else
		variant = args_variant or '';											-- use args.variant if provided
	end 

	if is_set (variant) then
		variant = variant:lower();												-- ensure that we use and return lower case version of this
		if not variant_table[variant] then										-- make sure variant is valid
			return code, extlang, script, region, nil, nil, nil, table.concat ({'variante no reconocida: ', variant});
		end																		-- does this duplicate/replace tests in lang() and lang_xx()?
		if is_set (script) then													-- if script set it must be part of the 'prefix'
			if not in_array (table.concat ({code, '-', script}), variant_table[variant]['prefixes']) then
				return code, extlang, script, region, nil, nil, nil, table.concat ({'variante no reconocida: ', variant, ' para la pareja código-escritura: ', code, '-', script});
			end
		elseif is_set (region) then												-- if region set, there are some prefixes that require lang code and region (en-CA-newfound)
			if not in_array (code, variant_table[variant]['prefixes']) then		-- first see if lang code is all that's required (en-oxendict though en-GB-oxendict is preferred)
				if not in_array (table.concat ({code, '-', region}), variant_table[variant]['prefixes']) then		-- now try for lang code and region (en-CA-newfound)
					return code, extlang, script, region, nil, nil, nil, table.concat ({'variante no reconocida: ', variant, ' para la pareja código-región: ', code, '-', region});
				end
			end
		else
			if not in_array (code, variant_table[variant]['prefixes']) then
				return code, extlang, script, region, nil, nil, nil, table.concat ({'variante no reconocida: ', variant, ' para el código: ', code});
			end				
		end
	end
	
	if is_set (private) then
		private = private:lower();												-- ensure that we use and return lower case version of this
		cd = code .. (is_set (extlang) and '-'..extlang:lower() or '') .. (is_set (script) and '-'..script:lower() or '') .. (is_set (region) and '-'..region:lower() or '') .. (is_set (variant) and '-'..variant:lower() or '') .. (is_set (extension) and '-u-'..extension:lower() or '') .. '-x-' .. private
		if not override_table[cd] then																								-- make sure private tag is valid; note that index 
			return code, extlang, script, region, nil, nil, nil, table.concat ({'etiqueta privada no reconocida: ', private});
		end
	end

	if is_set (extlang) or is_set (extension) then
		cd = code .. (is_set (extlang) and '-'..extlang:lower() or '') .. (is_set (script) and '-'..script:lower() or '') .. (is_set (region) and '-'..region:lower() or '') .. (is_set (variant) and '-'..variant:lower() or '') .. (is_set (extension) and '-u-'..extension:lower() or '') .. (is_set (private) and '-x-'.. private:lower() or '')
		
		if not override_table[cd] then		-- make sure extension tag is valid; note that index 
			return code, extlang, script, region, nil, extension, private, table.concat ({'etiqueta compleja no reconocida: ', cd});
		end
	end
	
	return code, extlang, script, region, variant, extension, private, nil;							-- return the good bits; make sure that msg is nil
end


--[[--------------------------< M A K E _ E R R O R _ M S G >--------------------------------------------------

assembles an error message from template name, message text, help link, and error category.

]]

local function make_error_msg (msg, args, template)
	local out = {};
	local category;
	
	if 'Transl' == template then
		category = 'Transl';
	else
		category = 'Lang y lang-xx'
	end
	
	table.insert (out, table.concat ({'[', args.text or 'indefinido', '] '}));	-- for error messages output args.text if available
	table.insert (out, table.concat ({'<span style=\"font-size: 100%; font-style: normal;\" class=\"error\">Error: {{', template, '}}: '}));
	table.insert (out, msg);
	table.insert (out, table.concat ({' ([[:Categoría:Errores de ', category, '|ayuda]])'}));
	table.insert (out, '</span>');
	
	if (0 == namespace or 10 == namespace) and not args.nocat then						-- categorize in article space (and template space to take care of broken usages)
		table.insert (out, table.concat ({'[[Categoría:Errores de ', category, ']]'}));
	end

	return table.concat (out);
end
	

--[=[-------------------------< M A K E _ W I K I L I N K >----------------------------------------------------

Makes a wikilink; when both link and display text is provided, returns a wikilink in the form [[L|D]]; if only
link is provided, returns a wikilink in the form [[L]]; if neither are provided or link is omitted, returns an
empty string.

]=]

local function make_wikilink (link, display)
	if is_set (link) then
		if is_set (display) then
			return table.concat ({'[[', link, '|', display, ']]'});
		else
			return table.concat ({'[[', link, ']]'});
		end
	else
		return '';
	end
end


--[[--------------------------< D I V _ M A R K U P _ A D D >--------------------------------------------------

adds <i> and </i> tags to list-item text or to implied <p>..</p> text.  mixed not supported

]]

local function div_markup_add (text, style)
local implied_p = {};

	if text:find ('^\n[%*:;#]') then											-- look for list markup; list markup must begin at start of text
		if 'italic' == style then
			return mw.ustring.gsub (text, '(\n[%*:;#]+)([^\n]+)', '%1<i>%2</i>');	-- insert italic markup at each list item
		else
			return text;
		end
	end

	if text:find ('\n+') then													-- look for any number of \n characters in text
		text = text:gsub ('([^\n])\n([^\n])', '%1 %2');							-- replace single newline characters with a space character which mimics mediawiki

		if 'italic' == style then
			text = text:gsub('[^\n]+', '<p><i>%1</i></p>');						-- insert p and italic markup tags at each impled p (two or more consecutive '\n\n' sequences)
		else
			text = text:gsub ('[^\n]+', '<p>%1</p>');							-- insert p markup at each impled p
			text = text:gsub ('\n', '');										-- strip newline characters
		end
	end

	return text;
end


--[[--------------------------< M A K E _ T E X T _ H T M L >--------------------------------------------------

Add the html markup to text according to the type of content that it is: <span> or <i> tags for inline content or
<div> tags for block content

]]

local function make_text_html (code, text, tag, rtl, style, size, language, subtags)
	local out = {};

	if text:match ('^%*') then
		table.insert (out, '&#42;');											-- move proto language text prefix outside of italic markup if any; use numeric entity because plain splat confuses MediaWiki
		text = text:gsub ('^%*', '');											-- remove the splat from the text
	end

	if 'span' == tag then														-- default html tag for inline content
		if 'italic' == style then												-- but if italic
			tag = 'i';															-- change to <i> tags
		end
	else																		-- must be div so go
		text = div_markup_add (text, style);									-- handle implied <p>, implied <p> with <i>, and list markup (*;:#) with <i>
	end
	
	html = mw.html.create( tag );
	
	code = code:gsub ('%-x%-.*', '');											-- strip private use subtag from code tag because meaningless outside of wikipedia
	html:attr( 'lang', code);

	if rtl or unicode.is_rtl(text) then
		html:attr( 'dir', 'rtl');									        	-- add direction attribute for right to left languages
	end

	if 'normal' == style then													-- when |italic=no
		html:css( 'font-style', 'normal');              						-- override external markup, if any
	end

	if is_set (size) then														-- when |size=<something>
		html:css( 'font-size', size);
	end

	if is_set (language) then
		local txt = ""
		if language:find ('Lenguas') then
			txt = txt .. 'Texto colectivo en ';						        	-- for collective languages
		else
			txt = txt .. 'Texto en idioma ';						        	-- for individual languages
		end
		txt = txt .. language;
		html:attr('title', txt);                                				-- close the style attribute and close opening html tag
	end
	html:wikitext(text);		    											-- insert the text
	table.insert (out, tostring( html ));

	if rtl then																	-- legacy; shouldn't be necessary because all of the rtl text is wrapped inside an html tag with dir="rtl" attribute 
		table.insert (out, '&lrm;');											-- make sure the browser knows that we're at the end of the rtl
	end
	return table.concat (out);													-- put it all together and done
end


--[=[-------------------------< M A K E _ C A T E G O R Y >----------------------------------------------------

Para idiomas, <lengua>, devuelve:
	[[Categoría:Artículos con texto en <lengua>]]

Para español:
	[[Categoría:Artículos con texto citado explícitamente en español]]
	
Para lenguas colectivas listasdas en el ISO 639-2 (y para 639-1 bh):
	[[Categoría:Artículos con texto en <lengua colectiva>]]
	
En el caso de que no exista la categoría ideal:
    [[Categoría:Artículos con texto en otros idiomas]]

]=]

local function make_category (code, language_name, nocat, name_get)
	local cat = {};
	local retval;

	if ((0 ~= namespace) or nocat) and not name_get then						-- only categorize in article space
		return '';																-- return empty string for concatenation
	end

	if language_name:find ('Lenguas') then
		if(page_exists('Categoría:Artículos con texto en ' .. language_name)) then
			return table.concat ({'[[Categoría:Artículos con texto en ', language_name, ']]'});
		end
		return "[[Categoría:Artículos con texto en otros idiomas]]";
	end
		
	table.insert (cat, '[[Categoría:Artículos con ');

	if 'es' == code then
		table.insert (cat, 'texto citado explícitamente en ' .. language_name .. ']]');				-- falls back to English if regional name not available
	else
		table.insert (cat, 'texto en ' .. language_name .. ']]');
	end
	
	local final = table.concat (cat);
	
	if(page_exists(final:gsub ('[%[%]]', ''))) then
		return final;	
	end

	return "[[Categoría:Artículos con texto en otros idiomas]]";
	
end


--[[--------------------------< M A K E _ T R A N S L I T >----------------------------------------------------

return translit <i lang=xx-Latn>...</i> where xx is the language code; else return empty string

The value |script= is not used in {{transl}} for this purpose; instead it uses |code.  Because language scripts
are listed in the {{transl}} switches they are included in the data tables.  The script parameter is introduced
at {{Language with name and transliteration}}.  If |script= is set, this function uses it in preference to code.

To avoid confusion, in this module and the templates that use it, the transliteration script parameter is renamed
to be |translit-script= (in this function, tscript)

This function is used by both lang_xx() and transl()
	lang_xx() always provides code, language_name, and translit; may provide tscript; never provides style
	transl() always provides language_name, translit, and one of code or tscript, never both; always provides style

For {{transl}}, style only applies when a language code is provided
]]

local function make_translit (code, language_name, translit, std, tscript, style)
	local title;
	local msg;
	local tout = {};
	local title_table = lang_data.translit_title_table;							-- table of transliteration standards and the language codes and scripts that apply to those standards
	
	if is_set (code) then														-- when a language code is provided (always with {{lang-xx}} templates, not always with {{transl}})
		if not style then                                                       -- nil for the default italic style
			tout = mw.html.create('i');                                         -- so use <i> tag
		else
			tout = mw.html.create('span');
			tout:css('font-style', style);										-- non-standard style, construct a span tag for it
		end
		tout:attr('lang', code .. "-Latn");										-- transliterations are always Latin script
	else
		tout = mw.html.create('span');											-- when no language code: no lang= attribute, not italic ({{transl}} only)
	end
	
	std = std and std:lower();													-- lower case for table indexing
	
	local texto = {}
	
	if not is_set (std) and not is_set (tscript) then							-- when neither standard nor script specified
		table.insert (texto, 'Romanización de');									-- write a generic tool tip
		if not language_name:find ('Lenguas') then						  	    -- collective language names (plural 'languages' is part of the name)
			table.insert (texto, 'l idioma ')									-- skip this text (individual and macro languages only)
		else
			table.insert (texto, ' las ')
		end	
		table.insert (texto, language_name);							            -- finish the tool tip; use romanization when neither script nor standard supplied
	elseif is_set (std) and is_set (tscript) then								-- when both are specified
		if title_table[std] then												-- and if standard is legitimate
			if title_table[std][tscript] and script_table[tscript] then									-- and if script for that standard is legitimate
				table.insert (texto, table.concat ({'Transliteración según ', title_table[std][tscript:lower()], ' (Sistema de escritura ', script_table[tscript][1], ')'}));	-- add the appropriate text to the tool tip
			elseif title_table[std][tscript] then
				msg = table.concat ({'translit-script no válido: ', tscript, ' para el código: ', code});
				table.insert (texto, table.concat ({'Transliteración según ', title_table[std][tscript:lower()]}));
			else
				table.insert (texto, title_table[std]['default']);				-- use the default if script not in std table; TODO: maint cat? error message because script not found for this standard?
			end
		else
			return nil, table.concat ({'translit-std no válido: ', std or '[faltante]'});															-- invalid standard, setup for error message
		end

	elseif is_set (std) then													-- translit-script not set, use language code
		if not title_table[std] then 
			return nil, table.concat ({'translit-std no válido: ', std or '[faltante]'});
		end								-- invalid standard, setup for error message
		
		if title_table[std][code] then											-- if language code is in the table (transl may not provide a language code)
--			table.insert (texto, table.concat ({title_table[std][code:lower()], ' (', lang_table[code][1], ' language) transliteration'}));	-- add the appropriate text to the tool tip
			table.insert (texto, table.concat ({'Transliteración según ', title_table[std][code:lower()], ' (Idioma ', language_name, ')'}));	-- add the appropriate text to the tool tip
		else																	-- code doesn't match
			table.insert (texto, title_table[std]['default']);					-- so use the standard's default
		end
	else																		-- here if translit-script set but translit-std not set
		if title_table['no_std'][tscript] then
			table.insert (texto, title_table['no_std'][tscript]);				-- use translit-script if set
		elseif title_table['no_std'][code] then
			table.insert (texto, title_table['no_std'][code]);					-- use language code
		else
			if is_set (tscript) then
				table.insert (texto, table.concat ({'Transliteración del sistema de escritura del ', language_name}));	-- write a script tool tip
			elseif is_set (code) then
				if not language_name:find ('Lenguas') then					-- collective language names (plural 'languages' is part of the name)
					table.insert (texto, '-language')							-- skip this text (individual and macro languages only)
				end
				table.insert (texto, ' transliteration');						-- finish the tool tip
			else
				table.insert (texto, ' transliteration');						-- generic tool tip (can we ever get here?)
			end
		end
	end
	tout:attr('title', table.concat (texto));
	
	tout:wikitext(translit);
	
	return tostring(tout), msg;
end


--[[--------------------------< V A L I D A T E _ T E X T >----------------------------------------------------

This function checks the content of args.text and returns empty string if nothing is amiss else it returns an
error message.  The tests are for empty or missing text and for improper or disallowed use of apostrophe markup.

Italic rendering is controlled by the |italic= template parameter so italic markup should never appear in args.text
either as ''itself'' or as '''''bold italic''''' unless |italic=unset or |italic=invert.

]]

local function validate_text (template, args)
	if not is_set (args.text) then
		return make_error_msg ('no hay texto', args, template);
	end

	if args.text:find ("%f[\']\'\'\'\'%f[^\']") or args.text:find ("\'\'\'\'\'[\']+") then	-- because we're looking, look for 4 appostrophes or 6+ appostrophes
		return make_error_msg ('el texto tiene marcado incorrecto', args, template);
	end

	local style = args.italic;

	if ('unset' ~= style) and ('invert' ~=style) then
		if args.text:find ("%f[\']\'\'%f[^\']") or args.text:find ("%f[\']\'\'\'\'\'%f[^\']") then	-- italic but not bold, or bold italic
			return make_error_msg ('el texto tiene marcado de cursiva', args, template);
		end
	end
end


--[[--------------------------< C L E A N _ T E X T >----------------------------------------------------

Eliminar formato de cursivas

]]

local function clean_text (args)

	local style = args.italic;

	if ('unset' ~= style) and ('invert' ~=style) and args.text then
		if args.text:find ("%f[\']\'\'%f[^\']") or args.text:find ("%f[\']\'\'\'\'\'%f[^\']") then	-- italic but not bold, or bold italic
			return args.text:gsub("\'\'", "");
		end
	end
	return args.text;
end


--[[--------------------------< R E N D E R _ M A I N T >------------------------------------------------------

render mainenance messages and categories

]]

local function render_maint(nocat, args)
	local maint = {};
	
	local style = args.italic;
	if ('unset' ~= style) and ('invert' ~=style) then
		if args.text:find ("%f[\']\'\'%f[^\']") or args.text:find ("%f[\']\'\'\'\'\'%f[^\']") then	-- italic but not bold, or bold italic
			--table.insert (maint, table.concat ({'[[Categoría:Errores_de_Lang_y_lang-xx]]'}));
		end
	end
	
	
	if 0 < #maint_msgs then														-- when there are maintenance messages
		table.insert (maint, table.concat ({'<span class="lang-comment" style="font-style: normal; display: none; color: #33aa33; margin-left: 0.3em;">'}));	-- opening <span> tag
		for _, msg in ipairs (maint_msgs) do
			table.insert (maint, table.concat ({msg, ' '}));					-- add message strings
		end
		table.insert (maint, '</span>');										-- close the span
	end
	
	if (0 < #maint_cats) and (0 == namespace) and not nocat then				-- when there are maintenance categories; article namespace only
		for _, cat in ipairs (maint_cats) do
			table.insert (maint, table.concat ({'[[Categoría:', cat, ']]'}));	-- format and add the categories
		end
	end
	
	return table.concat (maint);
end


--[[--------------------------< P R O T O _ P R E F I X >------------------------------------------------------

for proto languages, text is prefixed with a splat.  We do that here as a flag for make_text_html() so that a splat
will be rendered outside of italic markup (if used).  If the first character in text here is already a splat, we
do nothing

proto_param is boolean or nil; true adds splat prefix regardless of language name; false removes and / or inhibits
regardless of language name; nil does nothing; presumes that the value in text is correct but removes extra splac

]]

local function proto_prefix (text, language_name, proto_param)
	if false == proto_param then												-- when forced by |proto=no
		return text:gsub ('^%**', '');											-- return text without splat prefix regardless of language name or existing splat prefix in text
	elseif (language_name:find ('^Proto%-') or (true == proto_param)) then		-- language is a proto or forced by |proto=yes
		return text:gsub ('^%**', '*');											-- prefix proto-language text with a splat; also removes duplicate prefixing splats
	end
	
	return text:gsub ('^%*+', '*');												-- return text unmolested except multiple splats reduced to one splat
end


--[[--------------------------< H A S _ P O E M _ T A G >------------------------------------------------------

looks for a poem strip marker in text; returns true when found; false else

auto-italic detection disabled when text has poem stripmarker because it is not possible for this code to know
the content that will replace the stripmarker.

]]

local function has_poem_tag (text)
	return text:find ('\127[^\127]*UNIQ%-%-poem%-[%a%d]+%-QINU[^\127]*\127') and true or false;
end


--[[--------------------------< H T M L _ T A G _ S E L E C T >------------------------------------------------

Inspects content of and selectively trims text.  Returns text and the name of an appropriate html tag for text.

If text contains:
	\n\n	text has implied <p>..</p> tags - trim leading and trailing whitespace and return 
If text begins with list markup:
	\n*		unordered
	\n;		definition
	\n:		definition
	\n#		ordered
trim all leading whitespace except  \n and trim all trailing whitespace

If text contains <poem>...</poem> stripmarker, return text unmodified and choose <div>..</div> tags because 
the stripmarker is replaced with text wrapped in <div>..</div> tags.

]]

local function html_tag_select (text)
	local tag;
	
	if has_poem_tag (text) then													-- contains poem stripmarker (we can't know the content of that)
		tag = 'div';															-- poem replacement is in div tags so lang must use div tags
	elseif mw.text.trim (text):find ('\n\n+') then								-- contains implied p tags
		text = mw.text.trim (text);												-- trim leading and trailing whitespace characters
		tag = 'div';															-- must be div because span may not contain p tags (added later by MediaWiki); poem replacement is in div tags
	elseif text:find ('\n[%*:;%#]') then										-- if text has list markup
		text = text:gsub ('^[\t\r\f ]*', ''):gsub ('%s*$', '');					-- trim all whitespace except leading newline character '\n'
		tag = 'div';															-- must be div because span may not contain ul, dd, dl, ol tags (added later by MediaWiki)
	else
		text = mw.text.trim (text);												-- plain text
		tag = 'span';															-- so span is fine
	end
	
	return text, tag;
end


--[[--------------------------< V A L I D A T E _ P R O T O >--------------------------------------------------

valida el valor asignado al parámetro |proto=; se permiten yes y no; yes retorna true, no retorna false,
una cadena vacía (o si el parámetro se ha omitido) retorna como nil; cualquier otro valor retorna como 
nil con un segundo valor de retorno como true que indica que algún otro valor se ha asignado a |proto=

]]

local function validate_proto (proto_param)
	if 'yes' == proto_param then
		return true;
	elseif 'no' == proto_param then
		return false;
	elseif is_set (proto_param) then
		return nil, true;														-- |proto= something other than 'yes' or 'no'
	else
		return nil;																-- missing or empty
	end
end


--[[--------------------------< L A N G U A G E _ N A M E _ G E T >--------------------------------------------

función común para retornar el nombre de idioma partiendo de los datos según una etiqueta IETF

retorna el nombre de idioma si se encuentra en las tablas; nil en caso contrario

]]

local function language_name_get (ietf, code, cat)
	ietf = ietf:lower();														-- ietf:lower() because format_ietf_tag() returns mixed case
	
	local name;																		-- remains nil if not found
	
	local code;
	local msg;
	local subtags = {};
	code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private, msg = get_ietf_parts (ietf, nil, nil, nil);
	
	if override_table[ietf] then												-- look for whole IETF tag in override table
		name = override_table[ietf][1];	
	else
		if override_table[code] then											-- not there so try basic language code
			name = override_table[code][1];
		elseif lang_es_table[code] then												-- shift to iana code/name table
			name = lang_es_table[code][1];	
		elseif lang_table[code] then												-- shift to iana code/name table
			name = lang_table[code][1];												-- table entries sometimes have multiple names, always take the first one
		elseif lang_es_dep_table[code] then
			name = lang_es_dep_table[code][1];											-- table entries sometimes have multiple names, always take the first one
		elseif lang_dep_table[code] then
			name = lang_dep_table[code][1];											-- table entries sometimes have multiple names, always take the first one
		end
		
		if is_set(subtags.region) then
			name = mw.ustring.lower(name) .. ' de ' .. region_table[subtags.region][1]:gsub("^%l", mw.ustring.upper);
			--name = name .. ' de ' .. region_table[subtags.region][1]:gsub("^%l", mw.ustring.upper);
		end
	end

	if lang_dep_table[code] and cat then										-- because deprecated code may have been overridden to en.wiki preferred name
		table.insert (maint_cats, table.concat ({'Lang y lang-xx con códigos ISO 639 obsoletos|', code}));
		table.insert (maint_msgs, table.concat ({'el código: ', code, ' está obsoleto'}));
	end

	return name;																-- name from data tables or nil
end


--[[--------------------------< A R T I C L E _ N A M E _ G E T >--------------------------------------------

función común para retornar el nombre del artículo partiendo de los datos según un código de idioma

retorna el título estándar del idioma si se encuentra o los datos de override

]]

local function article_name_get (ietf, code, language_name)
	ietf = ietf:lower();														-- ietf:lower() because format_ietf_tag() returns mixed case
	
	if not is_set(language_name) then
		return nil
	end
	
	local link;																	-- remains nil if not found
	local link_text;																	-- remains nil if not found
	
	local code;
	local msg;
	local subtags = {};
	code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private, msg = get_ietf_parts (ietf, nil, nil, nil);
	
	if lang_data.article_name[ietf] then												-- look for whole IETF tag in override table
		link = lang_data.article_name[ietf][1];
		link_text = language_name;
	elseif lang_data.article_name[code] then											-- not there so try basic language code
		link = lang_data.article_name[code][1];
		link_text = language_name;
	else
		if language_name:find ('lenguas') then
			link = language_name;
			link_text = language_name;
		elseif is_set(subtags.region) then
			link = language_name;
			link_text = language_name;
		else
			link = 'Idioma ' .. language_name;
			link_text = language_name;
		end
	end
	
	
	return link, link_text;																-- name from data tables or nil
end


--[[--------------------------< _ L A N G >--------------------------------------------------------------------

punto de entrada para {{lang}}

no debería existir ninguna razón para configurar parámetros en el {{#invoke:}} de {{lang}}
	<includeonly>{{#invoke:lang|lang}}</includeonly>

los parámetros se reciben del frame de la plantilla (el frame padre)

]]

local function _lang (args)
	local out = {};
	local language_name;														-- used to make category names
	local category_name;														-- same as language_name except that it retains any parenthetical disambiguators (if any) from the data set
	local subtags = {};															-- IETF subtags script, region, variant, and private
	local code;																	-- the language code
	local msg;																	-- for error messages
	local tag = 'span';															-- initial value for make_text_html()
	local template = args.template or 'Lang';

	validate_cat_args (args);													-- determine if categorization should be suppressed

	if args[1] and args.code then
		return make_error_msg ('conflicto: {{{1}}} y |code=', args, template);
	else
		args.code = args[1] or args.code;										-- prefer args.code
	end

	if args[2] and args.text then
		return make_error_msg ('conflicto: {{{2}}} y |text=', args, template);
	else
		args.text = args[2] or args.text;										-- prefer args.text
	end
	
	args.text = clean_text(args);
	msg = validate_text (template, args);										-- ensure that |text= is set
	if is_set (msg) then														-- msg is an already-formatted error message
		return msg;
	end
	
	args.text, tag = html_tag_select (args.text);								-- inspects text; returns appropriate html tag with text trimmed accordingly

	args.rtl = args.rtl == 'yes';												-- convert to boolean: 'yes' -> true, other values -> false

	args.proto, msg = validate_proto (args.proto);								-- return boolean, or nil, or nil and error message flag
	if msg then
		return make_error_msg (table.concat ({'|proto= no válido: ', args.proto}), args, template);
	end

	code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private, msg = get_ietf_parts (args.code);	-- |script=, |region=, |variant= not supported because they should be part of args.code ({{{1}}} in {{lang}})

	if msg then
		return make_error_msg ( msg, args, template);
	end

	args.italic, msg = validate_italic (args);
	if msg then
		return make_error_msg (msg, args, template);
	end

	if nil == args.italic then													-- nil when |italic= absent or not set or |italic=default; args.italic controls
		if ('latn' == subtags.script) or								 		-- script is latn
			(this_wiki_lang ~= code and not is_set (subtags.script) and not has_poem_tag (args.text) and unicode.is_Latin (args.text)) then -- text not this wiki's language, no script specified and not in poem markup but is wholly latn script (auto-italics)
				args.italic = 'italic';											-- DEFAULT for {{lang}} templates is upright; but if latn script set for font-style:italic
		else
			args.italic = 'inherit';											-- italic not set; script not latn; inherit current style
		end
	end
	
	if is_set (subtags.script) then												-- if script set, override rtl setting
		if in_array (subtags.script, lang_data.rtl_scripts) then
			args.rtl = true;													-- script is an rtl script
		else
			args.rtl = false;													-- script is not an rtl script
		end
	end

	args.code = format_ietf_tag (code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private);	-- format to recommended subtag styles
	language_name = language_name_get (args.code, code, true);					-- get language name; try ietf tag first, then code w/o variant then code w/ variant

	if 'invert' == args.italic and 'span' == tag then							-- invert only supported for in-line content
		args.text = invert_italics (args.text)
	end

	args.text = proto_prefix (args.text, language_name, args.proto);			-- prefix proto-language text with a splat

	table.insert (out, make_text_html (args.code, args.text, tag, args.rtl, args.italic, args.size, language_name, subtags));
	table.insert (out, make_category (code, language_name, args.nocat));
	table.insert (out, render_maint(args.nocat, args));								-- maintenance messages and categories

	return table.concat (out);													-- put it all together and done
end


--[[--------------------------< L A N G >----------------------------------------------------------------------

punto de entrada para {{lang}}

no debería existir ninguna razón para configurar parámetrosen el {{#invoke:}} de {{lang}}
	<includeonly>{{#invoke:lang|lang}}</includeonly>

los parámetros se reciben del frame de la plantilla (el frame padre)

]]

local function lang (frame)
	local args = getArgs (frame, {												-- this code so that we can detect and handle wiki list markup in text
		valueFunc = function (key, value)
			if 2 == key or 'text' == key then									-- the 'text' parameter; do not trim wite space
				return value;													-- return untrimmed 'text'
			elseif value then													-- all other values: if the value is not nil
				value = mw.text.trim (value);									-- trim whitespace
				if '' ~= value then												-- empty string when value was only whitespace
					return value;
				end
			end
			return nil;															-- value was empty or contained only whitespace
		end																		-- end of valueFunc
		});

	return _lang (args);
end


--[[--------------------------< L A N G _ X X >----------------------------------------------------------------

For the {{lang-xx}} templates, the only parameter required to be set in the template is the language code.  All
other parameters can, usually should, be written in the template call.  For {{lang-xx}} templates for languages
that can have multiple writing systems, it may be appropriate to set |script= as well.

For each {{lang-xx}} template choose the appropriate entry-point function so that this function knows the default
styling that should be applied to text.

For normal, upright style:
	<includeonly>{{#invoke:lang|lang_xx_inherit|code=xx}}</includeonly>
For italic style:
	<includeonly>{{#invoke:lang|lang_xx_italic|code=xx}}</includeonly>

All other parameters should be received from the template's frame (parent frame)

Supported parameters are:
	|code = (required) the IANA language code
	|script = IANA script code; especially for use with languages that use multiple writing systems
	|region = IANA region code
	|variant = IANA variant code
	|text = (required) the displayed text in language specified by code
	|link = boolean false ('no') does not link code-spcified language name to associated language article
	|rtl = boolean true ('yes') identifies the language specified by code as a right-to-left language
	|nocat = boolean true ('yes') inhibits normal categorization; error categories are not affected
	|cat = boolian false ('no') opposite form of |nocat=
	|italic = boolean true ('yes') renders displayed text in italic font; boolean false ('no') renders displayed text in normal font; not set renders according to initial_style_state
	|lit = text that is a literal translation of text
	|label = 'none' to suppress all labeling (language name, 'translit.', 'lit.')
				any other text replaces language-name label - automatic wikilinking disabled
	
	for those {{lang-xx}} templates that support transliteration (those templates where |text= is not entirely latn script):
	|translit = text that is a transliteration of text
	|translit-std = the standard that applies to the transliteration
	|translit-script = ISO 15924 script name; falls back to code

For {{lang-xx}}, the positional parameters are:
	{{{1}}}	text
	{{{2}}}	transliterated text
	{{{3}}}	literal translation text
no other positional parameters are allowed

]]

local function _lang_xx (args)
	local out = {};
	local language_name;														-- used to make display text, article links
	local category_name;														-- same as language_name except that it retains any parenthetical disambiguators (if any) from the data set
	local subtags = {};															-- IETF subtags script, region, and variant
	local code;																	-- the language code

	local translit_script_name;													-- name associated with IANA (ISO 15924) script code
	local translit;
	local translit_title;
	local msg;																	-- for error messages
	local tag = 'span';															-- initial value for make_text_html()
	local template = args.template or 'Lang-xx';

	if args[1] and args.text then
		return make_error_msg ('conflicto: {{{1}}} y |text=', args, template);
	else
		args.text = args[1] or args.text;										-- prefer args.text
	end
	
	args.text = clean_text(args);
	msg = validate_text (template, args);										-- ensure that |text= is set, does not contain italic markup and is protected from improper bolding
	if is_set (msg) then
		return msg;
	end

	args.text, tag = html_tag_select (args.text);								-- inspects text; returns appropriate html tag with text trimmed accordingly

	if args[2] and args.translit then
		return make_error_msg ('conflicto: {{{2}}} y |translit=', args, template);
	else
		args.translit = args[2] or args.translit								-- prefer args.translit
	end
	
	if args[3] and (args.translation or args.lit) then
		return make_error_msg ('conflicto: {{{3}}} y |lit= o |translation=', args, template);
	elseif args.translation and args.lit then
		return make_error_msg ('conflicto: |lit= y |translation=', args, template);
	else
		args.translation = args[3] or args.translation or args.lit;				-- prefer args.translation
	end

	if args.links and args.link then
		return make_error_msg ('conflicto: |links= y |link=', args, template);
	else
		args.link = args.link or args.links;									-- prefer args.link
	end

	validate_cat_args (args);													-- determine if categorization should be suppressed

	args.rtl = args.rtl == 'yes';												-- convert to boolean: 'yes' -> true, other values -> false

	code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private, msg = get_ietf_parts (args.code, args.script, args.region, args.variant);	-- private omitted because private

	if msg then																	-- if an error detected then there is an error message
		return make_error_msg (msg, args, template);
	end
	
	args.italic, msg = validate_italic (args);
	if msg then
		return make_error_msg (msg, args, template);
	end

	if nil == args.italic then													-- args.italic controls
		if is_set (subtags.script) then
			if 'latn' == subtags.script then
				args.italic = 'italic';											-- |script=Latn; set for font-style:italic
			else
				args.italic = initial_style_state;								-- italic not set; script is not latn; set for font-style:<initial_style_state>
			end
		else
			args.italic = initial_style_state;									-- here when |italic= and |script= not set; set for font-style:<initial_style_state>
		end
	end
	
	if is_set (subtags.script) then												-- if script set override rtl setting
		if in_array (subtags.script, lang_data.rtl_scripts) then
			args.rtl = true;													-- script is an rtl script
		else
			args.rtl = false;													-- script is not an rtl script
		end
	end

	args.proto, msg = validate_proto (args.proto);								-- return boolean, or nil, or nil and error message flag
	if msg then
		return make_error_msg (table.concat ({'|proto= no válido: ', args.proto}), args, template);
	end

	args.code = format_ietf_tag (code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private);	-- format to recommended subtag styles
	language_name = language_name_get (args.code, code, true);					-- get language name; try ietf tag first, then code w/o variant then code w/ variant

	category_name = language_name;												-- category names retain IANA parenthetical diambiguators (if any)
	language_name = language_name:gsub ('%s+%b()', '');							-- remove IANA parenthetical disambiguators or qualifiers from names that have them

	if args.label then
		if 'none' ~= args.label then
			table.insert (out, table.concat ({args.label, ': '}));				-- custom label
		end
	else
		table.insert (out, 'en ');
		if 'no' == args.link then
			table.insert (out, language_name);									-- language name without wikilink
		else
			local article_name;
			local link_text;
			article_name, link_text = article_name_get (args.code, code, language_name);
			if page_exists(article_name) then
				table.insert (out, make_wikilink (article_name, link_text));
			else
				table.insert (out, language_name);	
			end
		end
		table.insert (out, ', ');												-- separator
	end

	if 'invert' == args.italic then
		args.text = invert_italics (args.text)
	end
	
	args.text = proto_prefix (args.text, language_name, args.proto);			-- prefix proto-language text with a splat

	table.insert (out, make_text_html (args.code, args.text, tag, args.rtl, args.italic, args.size, ('none' == args.label) and language_name or nil, subtags))

	if is_set (args.translit) and not unicode.is_Latin (args.text) then			-- transliteration (not supported in {{lang}}); not supported when args.text is wholly latn text (this is an imperfect test)
		table.insert (out, ', ');												-- comma to separate text from translit
		if 'none' ~= args.label then
			table.insert (out, '<small>');
			if script_table[args['translit-script']] then						-- when |translit-script= is set, try to use the script's name
				translit_script_name = script_table[args['translit-script'][1]];
			else
				translit_script_name = mw.ustring.lower(language_name);							-- fall back on language name
			end
			translit_title = mw.title.makeTitle (0, table.concat ({'Romanización del ', translit_script_name}));		-- make a title object
			if translit_title.exists and ('no' ~= args.link) then
				table.insert (out, make_wikilink ('Romanización del ' .. translit_script_name or language_name, 'romanizado') .. ':');	-- make a wikilink if there is an article to link to
			else
				table.insert (out, 'romanizado:');								-- else plain text
			end
			table.insert (out, '&nbsp;</small>');								-- close the small tag
		end
		
		translit, msg = make_translit (args.code, language_name, args.translit, args['translit-std'], args['translit-script'])
		if msg then
			return make_error_msg (msg, args, template);
		end
		if is_set (translit) then
			table.insert (out, translit);
		end
	end
	
	if is_set (args.translation) then											-- translation (not supported in {{lang}})
		table.insert (out, ', ');
		if 'none' ~= args.label then
			test = mw.html.create('small')
			if 'no' == args.link then
				test
				 :tag('abbr')
				 :attr('title', 'traducción literal')
				 :wikitext('lit.'):done();
			else
				test:wikitext(make_wikilink ('Traducción metafrástica', 'lit.'));
			end
			test:wikitext('&nbsp;');
			table.insert (out, tostring(test));
		end
		table.insert (out, table.concat ({'&#39;', args.translation, '&#39;'}));	-- use html entities to avoid wiki markup confusion
	end
	
	table.insert (out, make_category (code, category_name, args.nocat));
	table.insert (out, render_maint(args.nocat, args));								-- maintenance messages and categories

	return table.concat (out);													-- put it all together and done
end


--[[--------------------------< L A N G _ X X _ A R G S _ G E T >----------------------------------------------

Función común para obtener la tabla de argumentos de todas las plantillas {{lang-??}} templates

retorna la tabla de argumentos

]]

local function lang_xx_args_get (frame)
	local args = getArgs(frame,
		{
		parentFirst= true,														-- parameters in the template override parameters set in the {{#invoke:}}
		valueFunc = function (key, value)
			if 1 == key then													-- the 'text' parameter; do not trim wite space
				return value;													-- return untrimmed 'text'
			elseif value then													-- all other values: if the value is not nil
				value = mw.text.trim (value);									-- trim whitespace
				if '' ~= value then												-- empty string when value was only whitespace
					return value;
				end
			end
			return nil;															-- value was empty or contained only whitespace
		end																		-- end of valueFunc
	});

	return args;
end


--[[--------------------------< L A N G _ X X _ I T A L I C >--------------------------------------------------

Punto de entrada para las plantillas {{lang-xx}} que llaman a lang_xx_italic(). Configura el estilo inicial como italic.

]]

local function lang_xx_italic (frame)
	local args = lang_xx_args_get (frame);
	
	initial_style_state = 'italic';
	return _lang_xx (args);
end


--[[--------------------------< _ L A N G _ X X _ I T A L I C >------------------------------------------------

Punto de entrada desde otro módulo. Configura el estilo inicial como italic.

]]

local function _lang_xx_italic (args)
	initial_style_state = 'italic';
	return _lang_xx (args);
end


--[[--------------------------< L A N G _ X X _ I N H E R I T >------------------------------------------------

Punto de entrada para las plantillas {{lang-xx}} que llaman a lang_xx_inherit(). Configura el estilo inicial como inherit.

]]

local function lang_xx_inherit (frame)
	local args = lang_xx_args_get (frame);

	initial_style_state = 'inherit';
	return _lang_xx (args);
end


--[[--------------------------< _ L A N G _ X X _ I N H E R I T >----------------------------------------------

Punto de entrada desde otro módulo. Configura el estilo inicial como inherit.

]]

local function _lang_xx_inherit (args)
	initial_style_state = 'inherit';
	return _lang_xx (args);
end


--[[--------------------------< _ I S _ I E T F _ T A G >------------------------------------------------------

Retorna true cuando existe un idioma asociado a la etiqueta IETF introducida; nil en caso contrario. La etiqueta IETF debe ser válida.

Todas las combinaciones de códigos soportadas por {{lang}} y las plantillas {{lang-xx}} son aceptadas por esta función.

Punto de entrada desde otro módulo

]]

local function _is_ietf_tag (tag)												-- entry point when this module is require()d into another
	local c, e, s, r, v, ext, p, err;													-- code, script, region, variant, private, error message
	c, e, s, r, v, ext, p, err = get_ietf_parts (tag);									-- disassemble tag into constituent part and validate

	return ((c and not err) and true) or nil;									-- return true when code portion has a value without error message; nil else
end


--[[--------------------------< I S _ I E T F _ T A G >--------------------------------------------------------

Punto de entrada desde un {{#invoke:}}

]]

local function is_ietf_tag (frame)
	return _is_ietf_tag (getArgs(frame)[1]);									-- args[1] is the ietf language tag to be tested; getArgs() so we also get parent frame
end


--[[--------------------------< _ N A M E _ F R O M _ T A G >--------------------------------------------------

Retorna el nombre del idioma asociado a una etiqueta de idioma IETF válida; en cualquier otro caso un mensaje de error.

Todas las combinaciones de códigos soportadas por {{lang}} y las plantillas {{lang-xx}} son soportadas por esta función.

Configura el parámetro |link= en el invoke para obtener la versión del nombre con wikilink.

Punto de entrada desde otro módulo

]]

local function _name_from_tag (args)
	local subtags = {};															-- IETF subtags script, region, variant, and private
	local raw_code = args[1];													-- save a copy of the input IETF subtag
	local link = 'yes' == args['link'];											-- make a boolean
	local label = args.label;
	local code;																	-- the language code
	local msg;																	-- gets an error message if IETF language tag is malformed or invalid
	local language_name = '';
	
	code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private, msg = get_ietf_parts (raw_code);
	if msg then
		local template = (args['template'] and table.concat ({'{{', args['template'], '}}: '})) or '';	-- make template name (if provided by the template)
		return table.concat ({'<span style=\"font-size: 100%; font-style: normal;\" class=\"error\">Error: ', template, msg, '</span>'});
	end

	raw_code = format_ietf_tag (code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private);	-- format to recommended subtag styles; private omitted because private
	language_name = language_name_get (raw_code, code);							-- get language name; try ietf tag first, then code w/o variant then code w/ variant

	language_name = language_name:gsub ('%s+%b()', '');							-- remove IANA parenthetical disambiguators or qualifiers from names that have them

	if link then																-- when |link=yes, wikilink the language name
		local article_name;
		local link_text;
		article_name, link_text = article_name_get (raw_code, code, language_name);
		language_name = make_wikilink (article_name, label or link_text);
	end
	
	return language_name;
end


--[[--------------------------< N A M E _ F R O M _ T A G >----------------------------------------------------

Punto de entrada desde un {{#invoke:}}

]]

local function name_from_tag (frame)											-- entry point from an {{#invoke:Lang|name_from_tag|<ietf tag>|link=<yes>|template=<template name>}}
	return _name_from_tag (getArgs(frame))										-- pass-on the args table, nothing else; getArgs() so we also get parent frame
end


--[[--------------------------< _ T A G _ F R O M _ N A M E >--------------------------------------------------

Retorna la etiqueta de idioma IETF asociada a un nombre concreto. El nombre debe estar bien escrito
de la misma forma que aparece en la tabla. Cuando un nombre está desambiguado se debe omitir la coletilla,
no están presentes en la tabla.

Punto de entrada desde otro módulo

]]

local function _tag_from_name (args)											-- entry point from another module
	local msg;

	if args[1] and '' ~= args[1] then
		local data = mw.loadData ('Módulo:Lang/tag from name');					-- get the reversed data tables TODO: change when going live
		local lang = mw.ustring.lower(args[1]);										-- allow any-case for the language name (speeling must till be correct)
		local tag = data.rev_override_table[lang] or data.rev_lang_es_table[lang] or data.rev_lang_table[lang] or data.rev_lang_es_dep_table[lang] or data.rev_lang_dep_table[lang];	-- get the code; look first in the override then in the standard

		if tag then
			return tag, true;													-- language name found so return tag and done; second return used by is_lang_name()
		else
			msg = 'idioma: ' .. args[1] .. ' no encontrado'						-- language name not found, error message
		end
	else
		msg = 'nombre de idioma faltante'										-- language name not provided, error message
	end

	local template = '';
	if args.template and '' ~= args.template then
		template = table.concat ({'{{', args['template'], '}}: '});				-- make template name (if provided by the template)
	end
	return table.concat ({'<span style=\"font-size: 100%; font-style: normal;\" class=\"error\">Error: ', template, msg, '</span>'});
end


--[[--------------------------< T A G _ F R O M _ N A M E >----------------------------------------------------

Punto de entrada desde un {{#invoke:}}

]]

local function tag_from_name (frame)											-- entry point from an {{#invoke:Lang|tag_from_name|<language name>|link=<yes>|template=<template name>}}
	local result, _ = _tag_from_name (getArgs(frame))							-- pass-on the args table, nothing else; getArgs() so we also get parent frame; supress second return used by is_lang_name()
	return result;		
end


--[[--------------------------< I S _ L A N G _ N A M E >------------------------------------------------------

Punto de entrada desde un {{#invoke:}}

]]

local function is_lang_name (frame)
	local _, result = _tag_from_name (getArgs(frame))							-- pass-on the args table, nothing else; getArgs() so we also get parent frame; supress second return used by tag_from_name()
	return result and true or nil;		
end


--[[--------------------------< _ T R A N S L >----------------------------------------------------------------

Punto de entrada desde otro módulo

]]

local function _transl (args)
	local title_table = lang_data.translit_title_table;							-- table of transliteration standards and the language codes and scripts that apply to those standards
	local language_name;														-- language name that matches language code; used for tool tip
	local translit;																-- translitterated text to display
	local script;																-- IANA script
	local msg;																	-- for when called functions return an error message

	if is_set (args[3]) then													-- [3] set when {{transl|code|standard|text}}
		args.text = args[3];													-- get the transliterated text
		args.translit_std = args[2] and args[2]:lower();						-- get the standard; lower case for table indexing

		if not title_table[args.translit_std] then
			return make_error_msg (table.concat ({'estándar de transliteración no reconocido: ', args.translit_std}), args, 'Transl');
		end
	else
		if is_set (args[2]) then												-- [2] set when {{transl|code|text}}
			args.text = args[2];												-- get the transliterated text
		else
			if args[1] and args[1]:match ('^%a%a%a?%a?$') then					-- args[2] missing; is args[1] a code or its it the transliterated text?
				return make_error_msg ('no hay texto', args, 'Transl');				-- args[1] is a code so we're missing text
			else
				args.text = args[1];											-- args[1] is not a code so we're missing that; assign args.text for error message
				return make_error_msg ('código de idioma / sistema de escritura faltante', args, 'Transl');
			end
		end
	end

	if is_set (args[1]) then													-- IANA language code used for html lang= attribute; or ISO 15924 script code
		if args[1]:match ('^%a%a%a?%a?$') then									-- args[1] has correct form?
			args.code = args[1]:lower();										-- use the language/script code; only (2, 3, or 4 alpha characters); lower case because table indexes are lower case
		else
			return make_error_msg (table.concat ({'código de idioma / sistema de escritura no reconocido: ', args[1]}), args, 'Transl');	-- invalid language / script code
		end
	else
		return make_error_msg ('código de idioma / sistema de escritura faltante', args, 'Transl');	-- missing language / script code so quit
	end

	args.italic, msg = validate_italic (args);
	if msg then
		return make_error_msg (msg, args, 'Transl');
	end
	
	if 'italic' == args.italic then												-- 'italic' when |italic=yes; because that is same as absent or not set and |italic=default
		args.italic = nil;														-- set to nil; 
	end

	if override_table[args.code] then											-- is code a language code defined in the override table?
		language_name = override_table[args.code][1];
	elseif lang_es_table[args.code] then											-- is code a language code defined in the standard language code tables?
		language_name = lang_es_table[args.code][1];
	elseif lang_table[args.code] then											-- is code a language code defined in the standard language code tables?
		language_name = lang_table[args.code][1];
	elseif lang_es_dep_table[args.code] then										-- is code a language code defined in the deprecated language code tables?
		language_name = lang_es_dep_table[args.code][1];
	elseif lang_dep_table[args.code] then										-- is code a language code defined in the deprecated language code tables?
		language_name = lang_dep_table[args.code][1];
	elseif script_table[args.code] then											-- if here, code is not a language code; is it a script code?
		language_name = script_table[args.code][1];
		script = args.code;														-- code was an ISO 15924 script so use that instead
		args.code = '';															-- unset because not a language code
	else
		return make_error_msg (table.concat ({'código de idioma / sistema de escritura no reconocido: ', args.code}), args, 'Transl');	-- invalid language / script code
	end
	
	local res;  																-- here only when all parameters passed to make_translit() are valid
	res, msg = make_translit (args.code, language_name, args.text, args.translit_std, script, args.italic);
	
	if msg then
		return make_error_msg (msg, args, 'Transl');
	end	
		
	return res;
end


--[[--------------------------< T R A N S L >------------------------------------------------------------------

Punto de entrada desde un {{#invoke:}}

]]

local function transl (frame)
	return _transl (getArgs(frame));
end


--[[--------------------------< C A T E G O R Y _ F R O M _ T A G >--------------------------------------------

Retorna el nombre de categoría asociado a una etiqueta de idioma IETF si es válida; error en caso contrario

Todas las combinaciones de códigos soportadas por {{lang}} y las plantillas {{lang-xx}} son soportadas por esta función.

Punto de entrada desde otro módulo

]]

local function _category_from_tag (args)
	local subtags = {};															-- IETF subtags script, region, variant, and private
	local raw_code = args[1];													-- save a copy of the input IETF subtag
	local link = 'yes' == args['link'];											-- make a boolean
	local label = args.label;
	local code;																	-- the language code
	local msg;																	-- gets an error message if IETF language tag is malformed or invalid
	local category_name = '';
	
	code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private, msg = get_ietf_parts (raw_code);
	if msg then
		local template = (args['template'] and table.concat ({'{{', args['template'], '}}: '})) or '';	-- make template name (if provided by the template)
		return table.concat ({'<span style=\"font-size: 100%; font-style: normal;\" class=\"error\">Error: ', template, msg, '</span>'});
	end

	raw_code = format_ietf_tag (code, subtags.extlang, subtags.script, subtags.region, subtags.variant, subtags.extension, subtags.private);	-- format to recommended subtag styles; private omitted because private
	category_name = language_name_get (raw_code, code);							-- get language name; try ietf tag first, then code w/o variant then code w/ variant
	category_name = make_category (code, category_name, nil, true):gsub ('[%[%]]', '');

	return category_name;
end


--[[--------------------------< C A T E G O R Y _ F R O M _ T A G >--------------------------------------------

Punto de entrada desde un {{#invoke:}}

]]

local function category_from_tag (frame)										-- entry point from an {{#invoke:Lang|category_from_tag|<ietf tag>|template=<template name>}}
	return _category_from_tag (getArgs (frame));								-- pass-on the args table, nothing else; getArgs() so we also get parent frame
end


--[[--------------------------< F U N C I O N E S   E X P O R T A D A S >------------------------------------------
]]

return {
	category_from_tag = category_from_tag,
	lang = lang,																-- entry point for {{lang}}
	lang_xx_inherit = lang_xx_inherit,											-- entry points for {{lang-??}}
	lang_xx_italic = lang_xx_italic,
	is_ietf_tag = is_ietf_tag,
	is_lang_name = is_lang_name,
	tag_from_name = tag_from_name,												-- returns ietf tag associated with language name
	name_from_tag = name_from_tag,												-- used for template documentation; possible use in ISO 639 name from code templates
	transl = transl,															-- entry point for {{transl}}
	
	_category_from_tag = _category_from_tag,									-- entry points when this module is require()d into other modules
	_lang = _lang,
	_lang_xx_inherit = _lang_xx_inherit,
	_lang_xx_italic = _lang_xx_italic,
	_is_ietf_tag = _is_ietf_tag,
	get_ietf_parts = get_ietf_parts,
	_tag_from_name = _tag_from_name, 
	_name_from_tag = _name_from_tag,
	_transl = _transl,
	};