ISOdate

Documentation for this module may be created at Module:ISOdate/doc
--[[  
  __  __           _       _        ___ ____   ___      _       _       
 |  \/  | ___   __| |_   _| | ___ _|_ _/ ___| / _ \  __| | __ _| |_ ___ 
 | |\/| |/ _ \ / _` | | | | |/ _ (_)| |\___ \| | | |/ _` |/ _` | __/ _ \
 | |  | | (_) | (_| | |_| | |  __/_ | | ___) | |_| | (_| | (_| | ||  __/
 |_|  |_|\___/ \__,_|\__,_|_|\___(_)___|____/ \___/ \__,_|\__,_|\__\___|
     
This module is intended for processing of date strings.

Please do not modify this code without applying the changes first at Module:ISOdate/sandbox and testing 
at Module:ISOdate/sandbox/testcases and Module talk:ISOdate/sandbox/testcases.

Authors and maintainers:
* User:Parent5446 - original version of the function mimicking template:ISOdate
* User:Jarekt - original version of the functions mimicking template:Date and template:ISOyear
]]

-- =======================================
-- === Dependencies ======================
-- =======================================
local Date = require('Module:DateI18n')._Date

-- =======================================
-- === Local Functions ===================
-- =======================================

local function parse_ISOdate(datestr)
-- Core function of this module, which splits "datestr" contining date in ISO format into Year, month, day, ... components
-- Output is "datevec" array with numbers representing date components. We also return "tail" storing text following the date

	-- pattern: regexp - regular expresion to test; dlen - number of date elements; tail = which element is a "tail" if any
	-- regexp hints:
	--  1) Strings starting with "^" and ending with "$" indicate whole string match
	--  2) optional tail part copied as-is and following the main parsed part of the date have to be separated from the date by a whitespace, so "(\s.+)?"
	local patterns = {
		-- strings starting with YYYY-MM-DD HH:MM:SS. Year 4 digits (if we know seconds than it was within the last 100 years), the rest 1-2
		-- date and time can be separated by space or "T" and there could be a "Z" on the end indicating "Zulu" time zone
		{dlen=6, tail=7, regexp="^+?(%d%d%d%d)-(%d%d?)-(%d%d?)[ T](%d%d?):(%d%d?):(%d%d?)Z?(%s.*)"}, 
		{dlen=6, tail=0, regexp="^+?(%d%d%d%d)-(%d%d?)-(%d%d?)[ T](%d%d?):(%d%d?):(%d%d?)Z?$"}, 
		-- strings starting with YYYY-MM-DD HH:MM. Year 4 digits, the rest 1-2
		-- (if one knows hour and minute than it was probably after a year 1000)
		{dlen=5, tail=6, regexp="^+?(%d%d%d%d)-(%d%d?)-(%d%d?)[ T](%d%d?):(%d%d?)(%s.+)"},
		{dlen=5, tail=0, regexp="^+?(%d%d%d%d)-(%d%d?)-(%d%d?)[ T](%d%d?):(%d%d?)$"},
		-- strings starting with YYYY-MM-DD. Year 1-4 digits, the rest 1-2
		{dlen=3, tail=4, regexp="^+?(%d%d?%d?%d?)-(%d%d?)-(%d%d?)(%s.+)"},
		{dlen=3, tail=0, regexp="^+?(%d%d?%d?%d?)-(%d%d?)-(%d%d?)$"},
		-- strings starting with YYYY-MM. Year 3-4 digits, month 2 digits
		-- (want to avoit converting to dates strings like 10-5 = 5
		{dlen=2, tail=3, regexp="^+?(%d%d%d%d?)-(%d%d)(%s.+)"}, 
		-- if whole string is in YYYY-MM form: If Year 1-4 digits, month 1-2 digits
		{dlen=2, tail=0, regexp="^+?(%d%d?%d?%d?)-(%d%d?)$"}, 
		-- string starts with a number -> it has to be 3 or 4 digit long to be a year
		{dlen=1, tail=2, regexp="^+?(%d%d%d%d?)(%s.+)"},	
		 -- if whole string is a number (1-4 digit long) than it will be interpreted as a year
		{dlen=1, tail=0, regexp="^+?(%d%d?%d?%d?)$"},
	}
	
	-- create datevec based on which variables are provided
	local datevec = {}
	local tail    = ''
	for i, pat in ipairs( patterns ) do
		local vec = {datestr:match( pat.regexp )}
		if vec and vec[1]~=nil then
			for j=1, pat.dlen do
				datevec[j] = vec[j]
			end
			if pat.tail>0 and vec[pat.tail]~=nil then
				tail = mw.ustring.gsub(' ' .. vec[pat.tail], ' +', ' ')
			end
			break
		end
	end
	if not datevec[1] or datevec[1]=='' then
		-- quickly return if datestr does not look like date (it could be a template)
		return nil, nil
	end
	return datevec, tail
end

-- ==================================================
-- === External functions ===========================
-- ==================================================
local p = {}

-- ===========================================================================
-- === Version of the function to be called from other LUA codes
-- ===========================================================================

function p._ISOyear( datestr )
    
	-- if empty string then return it
	datestr = mw.text.trim(datestr or '' )
	if datestr == '' then
		return ''
	end
    
	-- if number then return it
	if tonumber( datestr ) then
		return mw.ustring.format( '%04i', datestr )
	end
    
	-- otherwise use regular expression match
	datestr = mw.ustring.match( datestr, '^+?(-?%d%d?%d?%d?)-' )
	if datestr and tonumber( datestr ) then
		return mw.ustring.format( '%04i', datestr )
	else
		return ''
	end
end

function p._ISOdate(datestr, lang, case, class, trim_year)
	datestr = mw.text.trim(datestr or '' )
	local datevec, tail = parse_ISOdate(datestr)
	if not datevec then	
		return datestr, false -- quickly return if datestr does not look like date (it could be a template)
	end

	-- call p._Date function to format date string
	local datestr2 = Date(datevec, lang, case, class, trim_year)
	if datestr2~='' then
		return mw.text.trim( datestr2 .. tail), true
	else -- in case of errors return the original string
		return datestr, false
	end	
end

-- ===========================================================================
-- === Versions of the function to be called from template namespace
-- ===========================================================================
--[[
ISOdate
 
This function is the core part of the ISOdate template. 
 
Usage:
{{#invoke:ISOdate|ISOdate|target_string|lang=}}
 
Parameters:
     1: The date string 
  lang: The language to display it in
  form: Language format (genitive, etc.) for some languages
 class: CSS class for the <time> node

 Error Handling:
   If the string does not look like it contain the proper ISO date than the function will return the original string.
   
   That is the preferred treatment for the template:Information (and similar templates) which calling it.
]]
function p.ISOdate(frame)
	local args = {}
	for name, value in pairs( frame.args ) do 
		if value ~= '' then -- nuke empty strings
			if type(name)=='string' then 
				name = string.gsub( string.lower(name), ' ', '_')
			end
			args[name] = value
		end
	end


	local datestr, succeded = p._ISOdate(
		args[1] or '',
		args.lang,                  -- language
		args.case,                  -- allows to specify grammatical case for the month for languages that use them
		args.class or 'dtstart',    -- allows to set the html class of the time node where the date is included. 
		args.trim_year or '100-999' -- by default pad one and 2 digit years to be 4 digit long, while keeping 3 digit years as is	
	)
	return datestr
end

--[[
ISOyear
 
This function returns year part of date string.
 
Usage:
{{#invoke:ISOdate|ISOyear|target_string}}
 
Parameters
    1: The date string 
 
Error Handling:
   If the string does not look like it contain the year than the function will not return anything.
   That is the preferred treatment for the template:Creator which is the main (only?) template calling it.
]]
function p.ISOyear( frame )
	 return p._ISOyear( frame.args[1] )
end

return p