Module:Lang/data/iana languages/make

require('strict');

--[=[< G E T _ V A R I A N T _ P A R T S >-

We get a record that looks more-or-less like this: %%\n Type: variant\n Subtag: bohoric\n Description: Slovene in Bohorič alphabet\n Added: 2012-06-27\n Prefix: sl\n

Each line is terminated with a \n character.

Type, for this function can only be 'variant'

Subtag is the code of Type

Prefix is a language code to which this variant applies; one language code per Prefix line. There can be more than one prefix line.

Description associates Subtag with a proper name or names; one name per Description line. There can be more than one Description line and Description lines can wrap to the next line. When they do, the first two characters of the continuation line are spaces.

Comments: lines can also be continued so once in a Comments line (which is otherwise ignored) all further continuations in the record are also ignored. This is a crude mechanism to prevent comment continuations from being concatenated onto the end of descriptions and relies on Description line occuring in the record before the Comments line.

Records with private use subtags are ignored.

]=]

local function get_variant_parts (record) local code; local descriptions = {}; local prefixes = {}; local in_comments = false;

if string.find (record, 'Deprecated', 1, true) or string.find (record, 'Preferred-Value', 1, true) or string.find (record, 'Private use', 1, true) then return 'skip'; end

for line in string.gmatch (record, '([^\n]+)\n') do						-- get a \n terminated line of text (without the \n) local label = string.match(line, "(.-):") if not label and string.find (line, '^ .+') and not in_comments then		-- if a continuation line but not a comments continuation descriptions[#descriptions] = string.gsub (descriptions[#descriptions], '\"$', '');		-- remove trailing quote mark from previous description			descriptions[#descriptions] = descriptions[#descriptions] .. ' ' .. string.match (line, '^ (.+)') .. '\"';	-- extract and save the continuation with new quote mark elseif label == 'Subtag' then						-- if this line is the subtag line code = string.match (line, 'Subtag: (%w+)');				-- extract and save to subtag's code elseif label == 'Description' then					-- if this line is a description line local desc = string.match (line, 'Description: (.+)');			-- extract the description desc = string.gsub (desc, '"', '\\"');							-- in case description contains quote marks (see 1959acad) table.insert (descriptions, '\"' .. desc .. '\"');					-- save the description wrapped in quote marks elseif label == 'Prefix' then						-- if this line is a prefix line table.insert (prefixes, '\"' .. string.match (line, 'Prefix: (.+)'):lower .. '\"');	-- extract and save the prefix wrapped in quote marks elseif label == 'Comments' then						-- if this line is a comments line in_comments = true; end end return code, table.concat (prefixes, ', '), table.concat (descriptions, ', '); end

--[=[< G E T _ L A N G _ S C R I P T _ R E G I O N _ P A R T S >---

We get a record that looks more-or-less like this: %%\n Type: language\n Subtag: aa\n Description: Afar\n Added: 2005-10-16\n

Each line is terminated with a \n character.

Type, for our purposes can be 'language', 'script', or 'region'

Subtag is the code of Type

Description associates Subtag with a proper name or names; one name per Description line. There can be more than one Description line and Description lines can wrap to the next line. When they do, the first two characters of the continuation line are spaces.

Comments: lines can also be continued so once in a Comments line (which is otherwise ignored) all further continuations in the record are also ignored. This is a crude mechanism to prevent comment continuations from being concatenated onto the end of descriptions and relies on Description line occuring in the record before the Comments line.

Records with private use subtags are ignored.

]=]

local function get_lang_script_region_parts (record) local code; local suppress;																-- Suppress script for this code if specified local deprecated;															-- boolean; true when subtag is deprecated local descriptions = {}; local in_comments = false;

if record:find ('Private use') then return 'skip'; end

for line in record:gmatch ('([^\n]+)\n') do									-- get a \n terminated line of text (without the \n) local label = line:match ('(.-):'); if 'Subtag' == label then												-- if this line is the subtag line code = line:match ('Subtag: (%w+)');								-- extract and save to subtag's code elseif 'Description' == label then										-- if this line is a description line table.insert (descriptions, '\"' .. line:match ('Description: (.+)') .. '\"');	-- extract and save the name wrapped in quote marks elseif 'Deprecated' == label then deprecated = true;													-- subtag is deprecated; set our flag elseif 'Suppress-Script' == label then suppress = line:match ('Suppress%-Script: (%S+)'); elseif 'Comments' == label then											-- if this line is a comments line in_comments = true; elseif line:find ('^ .+') and not in_comments then						-- if a continuation line but not a commnets continuation descriptions[#descriptions] = descriptions[#descriptions]:gsub ('\"$', '');		-- remove trailing quote mark from previous description			descriptions[#descriptions] = descriptions[#descriptions] .. ' ' .. line:match ('^ (.+)') .. '\"';	-- extract and save the continuation with new quote mark end end return code, table.concat (descriptions, ', '), suppress, deprecated; end

--[=[< I A N A _ E X T R A C T >---

read a local copy of the IANA language-subtag-registry file and from it build tables to replace the tables in: Module:Lang/data/iana languages Module:Lang/data/iana regions Module:Lang/data/iana scripts Module:Lang/data/iana supressed cripts Module:Lang/data/iana variants

current language-subtag-registry file can be found at: http://www.iana.org/assignments/language-subtag-registry archive.org has copies of previous versions see: https://web.archive.org/web/*/http://www.iana.org/assignments/language-subtag-registry

]=]

local function iana_extract (frame) local page = mw.title.getCurrentTitle;									-- get a page object for this page local content = page:getContent;											-- get unparsed content local lang_table = {};														-- languages go here local lang_dep_table = {};													-- deprecated languages go here local script_table = {};													-- scripts go here local region_table = {};													-- regions go here local variant_table = {};													-- variants go here local suppress_table = {};													-- here we collect suppressed scripts and associated language codes local iso_639_1_table = {};													-- ISO 639-1 languages; not used by Module:Lang but included here to ensure Module:Lang/data/ISO_639-1 gets updated local file_date;															-- first line

local code; local descriptions; local prefixes;																-- used for language variants only local suppress;																-- a code's suppress script local deprecated;															-- boolean: true when subtag is deprecated

file_date = content:match ('(File%-Date: %d%d%d%d%-%d%d%-%d%d)');			-- get the file date line from this version of the source file

for record in string.gmatch (content, '%%%%([^%%]+)') do					-- get a %% delimited 'record' from the file; leave off the delimiters local record_type = string.match(record, 'Type: (%w+)') if record_type == 'language' then										-- if a language record code, descriptions, suppress, deprecated = get_lang_script_region_parts (record);	-- get the code, description(s), suppress script, and deprecated flag if code and ('skip' ~= code) then if deprecated then table.insert (lang_dep_table, "[\"" .. code .. "\"] = {" .. descriptions .. "}");		-- make table entries else table.insert (lang_table, "[\"" .. code .. "\"] = {" .. descriptions .. "}");			-- make table entries if 2 == code:len then table.insert (iso_639_1_table, "[\"" .. code .. "\"] = {" .. descriptions .. "}");	-- make table entries end end elseif not code then table.insert (lang_table, "[\"error\"] = {" .. record .. "}");	-- code should never be nil, but inserting an error entry in the final output can be helpful end -- here we collect suppress stript tags and their associated language codes; -- prettigying the data in this table must wait until all language codes have been read if suppress then													-- if this code has a suppressed script local suppressed_code = table.concat ({'\"', code, '\"'});		-- wrap the code in quotes if suppress_table[suppress] then								-- if there is an entry for this script table.insert (suppress_table[suppress], suppressed_code);	-- insert the new code else suppress_table[suppress] = {};								-- add new script and empty table table.insert (suppress_table[suppress], suppressed_code);	-- insert the new code end end

elseif record_type == 'script' then					-- if a script record code, descriptions = get_lang_script_region_parts (record);			-- get the code and description(s) if code and ('skip' ~= code) then table.insert (script_table, "[\"" .. code .. "\"] = {" .. descriptions .. "}");	-- make table entries elseif not code then table.insert (script_table, "[\"error\"] = {" .. record .. "}");	-- code should never be nil, but ... end

elseif record_type == 'region' then					-- if a region record code, descriptions = get_lang_script_region_parts (record);			-- get the code and description(s) if code and ('skip' ~= code) then table.insert (region_table, "[\"" .. code .. "\"] = {" .. descriptions .. "}");	-- make table entries elseif not code then table.insert (region_table, "[\"error\"] = {" .. record .. "}");	-- code should never be nil, but ... end

elseif record_type == 'variant' then					-- if a variant record code, prefixes, descriptions = get_variant_parts (record);			-- get the code, prefix(es), and description(s)

if code and ('skip' ~= code) then table.insert (variant_table,					table.concat ({ "[\"",						code,						"\"] = { &#9;&#9;[\"descriptions\"] = {", descriptions, "}, &#9;&#9;[\"prefixes\"] = {", prefixes, "}, &#9;&#9;}"						})					);			elseif not code then table.insert (variant_table, "[\"error\"] = {" .. record .. "}");	-- code should never be nil, but ... end end end -- now prettify the supressed script table local pretty_suppressed = {}; for script, code_tbl in pairs (suppress_table) do		local LIMIT = 11;														-- max number of subtags on a line before a line break local fragment_tbl = {};												-- groups of LIMIT number of subtags collected here for i=1, #code_tbl, LIMIT do			local stop = ((i+LIMIT-1) > #code_tbl) and #code_tbl or i+LIMIT-1;	-- calculate a table.concat stop position table.insert (fragment_tbl, table.concat (code_tbl, ', ', i, stop));	-- get the fragment and save it		end table.insert (pretty_suppressed,										-- and make all pretty			table.concat ({'[\"', script, '\"] = {', table.concat (fragment_tbl, ',\n\t\t\t\t'), '}'})			); end table.sort (pretty_suppressed);

-- make final output pretty return ' --< I A N A  L A N G U A G E S >-- --' ..			file_date .. " local active = { &#9;" .. table.concat (lang_table, ', &#9;') .. " &#9;}

" .. 			"local deprecated = { &#9;" .. table.concat (lang_dep_table, ', &#9;') .. " &#9;}

" .. 			"return { &#9;active = active, &#9;deprecated = deprecated, &#9;}

" ..			'--< I A N A  S C R I P T S >-- --' ..			file_date .. " return { &#9;" .. table.concat (script_table, ', &#9;') .. " &#9;}

" ..			'--< I A N A  R E G I O N S >-- --' ..			file_date .. " return { &#9;" .. table.concat (region_table, ', &#9;') .. " &#9;}

" ..			'--< I A N A  V A R I A N T S > --' ..			file_date .. " return { &#9;" .. table.concat (variant_table, ', &#9;') .. " &#9;}

" ..			'--< I A N A  S U P P R E S S E D   S C R I P T S > --' ..			file_date .. " return { &#9;" .. table.concat (pretty_suppressed, ', &#9;') .. " &#9;}

" ..			'--< I S O  6 3 9 - 1 > --' ..			file_date .. " return { &#9;" .. table.concat (iso_639_1_table, ', &#9;') .. " &#9;}

" .. " "; end

----< E X P O R T E D  F U N C T I O N >

return { iana_extract = iana_extract, }