Module:Citation/CS1/Date validation/sandbox: Difference between revisions

Revision as of 10:58, 1 April 2021

Latest revision as of 16:19, 25 May 2025

(66 intermediate revisions by 9 users not shown)

Line 1:

--[[

History of changes since last sync: 20202024-1008-1017

2021-02-13: i18n support for year/date mismatch; see Help_talk:Citation_Style_1#Year%2Fdate_mismatch_error_config_to_be_moved_to_config_module

2025-05-25: make |date=Jan 0500, |date=0999–1000, |date=090–100 error, accept |date=Jan 500.

2025-05-25: make |date=2000-10-01|year=2010 error

]]

Line 50 ⟶ 54:

if good1 and good2 then -- lang.formatDate() returns a timestamp in the local script which which tonumber() may not understand

access_ts = tonumber (access_ts) or lang_object:parseFormattedNumber (access_ts); -- convert to numbers for the comparison;

tomorrow_ts = tonumber (tomorrow_ts) or lang_object:parseFormattedNumber (tomorrow_ts);

else

Line 61 ⟶ 65:

return false; -- accessdate out of range

end

end

--[[--------------------------< I S _ V A L I D _ E M B A R G O _ D A T E >------------------------------------

returns true and date value if that value has proper dmy, mdy, ymd format.

returns false and 9999 (embargoed forever) when date value is not proper format; assumes that when |pmc-embargo-date= is

set, the editor intended to embargo a PMC but |pmc-embargo-date= does not hold a single date.

]]

local function is_valid_embargo_date (v)

if v:match ('^%d%d%d%d%-%d%d%-%d%d$') or -- ymd

v:match ('^%d%d?%s+%a+%s+%d%d%d%d$') or -- dmy

v:match ('^%a+%s+%d%d?%s*,%s*%d%d%d%d$') then -- mdy

return true, v;

end

return false, '9999'; -- if here not good date so return false and set embargo date to long time in future

end

Line 91 ⟶ 76:

local function get_month_number (month)

return cfg.date_names['local'].long[month] or cfg.date_names['local'].short[month] or -- look for local names first

cfg.date_names['en'].long[month] or cfg.date_names['en'].short[month] or -- failing that, look for English names

0; -- not a recognized month name

end

Line 201 ⟶ 186:

Function gets current year from the server and compares it to year from a citation parameter. Years more than one

year in the future are not acceptable.

Special case for |pmc-embargo-date=: years more than two years in the future are not acceptable

]]

local function is_valid_year (year, param)

if not is_set (year_limit) then

year_limit = tonumber(os.date("%Y"))+1; -- global variable so we only have to fetch it once

end

year = tonumber (year) or lang_object:parseFormattedNumber (year); -- convert to numbersnumber for the comparison;

if year and (100 > year) then -- years less than 100 not supported

return false;

end

if 'pmc-embargo-date' == param then -- special case for |pmc-embargo-date=

return year and (year <= tonumber(os.date("%Y"))+2) or false; -- years more than two years in the future are not accepted

end

return year and (year <= year_limit) or false;

end

Line 225 ⟶ 219:

]]

local function is_valid_date (year, month, day, param)

local days_in_month = {31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31};

local month_length;

if not is_valid_year (year, param) then -- no farther into the future than next year except |pmc-embargo-date= no more than two years in the future

return false;

end

month = tonumber (month); -- required for YYYY-MM-DD dates

if (2 == month) then -- if February

month_length = 28; -- then 28 days unless

if 1582 > tonumber(year) then -- Julian calendar

if 0 == (year%4) then -- is a leap year?

month_length = 29; -- if leap year then 29 days in February

end

else -- Gregorian calendar

if (0 == (year%4) and (0 ~= (year%100) or 0 == (year%400))) then -- is a leap year?

month_length = 29; -- if leap year then 29 days in February

end

Line 259 ⟶ 253:

Months in a range are expected to have the same style: Jan–Mar or October–December but not February–Mar or Jul–August.

This function looks in cfg.date_names{} to see if both month names are listed in the long subtable or both are

There is a special test for May because it can be either short or long form.

listed in the short subtable. When both have the same style (both are listed in the same table), returns true; false else

Returns true when style for both months is the same

]]

local function is_valid_month_range_style (month1, month2)

if (cfg.date_names.en.long[month1] and cfg.date_names.en.long[month2]) or -- are both English names listed in the long subtable?

local len1 = month1:len();

(cfg.date_names.en.short[month1] and cfg.date_names.en.short[month2]) or -- are both English names listed in the short subtable?

local len2 = month2:len();

(cfg.date_names['local'].long[month1] and cfg.date_names['local'].long[month2]) or -- are both local names listed in the long subtable?

if len1 == len2 then

(cfg.date_names['local'].short[month1] and cfg.date_names['local'].short[month2]) then -- are both local names listed in the short subtable?

return true; -- both months are short form so return true

return true;

elseif 'May' == month1 or 'May'== month2 then -- ToDo: I18N

return true; -- both months are long form so return true

elseif 3 == len1 or 3 == len2 then

return false; -- months are mixed form so return false

else

return true; -- both months are long form so return true

end

return false; -- names are mixed

end

Line 306 ⟶ 295:

range_end_number = get_month_number (range_end); -- get end month number

if range_start_number < range_end_number and -- range_start is a month; does range_start precede range_end?

is_valid_month_range_style (range_start, range_end) then -- do months have the same style?

return true; -- proper order and same style

end

return false; -- range_start month number is greater than or equal to range end number; or range end isn't a month

Line 346 ⟶ 335:

local date; -- one date or first date in a range

local date2 = ''; -- end of range date

input.year = tonumber (input.year) or lang_object:parseFormattedNumber (input.year); -- language-aware tonumber()

-- start temporary Julian / Gregorian calendar uncertainty detection

input.year2 = tonumber (input.year2) or lang_object:parseFormattedNumber (input.year2); -- COinS dates are pseudo-ISO 8601 so convert to Arabic numerals

local year = tonumber(input.year); -- this temporary code to determine the extent of sources dated to the Julian/Gregorian

local month = tonumber(input.month); -- interstice 1 October 1582 – 1 January 1926

if ((1582 == input.year) and (10 > tonumber(input.month))) or (1582 > input.year) then -- if a Julian calendar date

local day = tonumber (input.day);

tCOinS_date.rftdate = tostring (input.year); -- &rft.date gets year only

if (0 ~= day) and -- day must have a value for this to be a whole date

return; -- done

(((1582 == year) and (10 <= month) and (12 >= month)) or -- any whole 1582 date from 1 October to 31 December or

((1926 == year) and (1 == month) and (1 == input.day)) or -- 1 January 1926 or

((1582 < year) and (1925 >= year))) then -- any date 1 January 1583 – 31 December 1925

tCOinS_date.inter_cal_cat = true; -- set category flag true

end

-- here for all forms of Gregorian dates

-- end temporary Julian / Gregorian calendar uncertainty detection

if 20 < tonumber (input.month) then -- if season, quarter, or proper-name date

if 1582 date >= tonumber(input.year) or 20 < tonumber(input.month) then; -- Julian calendar or season so &rft.date gets year only

date = input.year;

if 0 ~= input.year2 and input.year ~= input.year2 then -- if a range, only the second year portion when not the same as range start year

date = string.format ('%.4d/%.4d', tonumber(input.year), tonumber(input.year2)) -- assemble the date range

end

if 20 < tonumber(input.month) then -- if season or proper-name date

local season = {[24] = 'winter', [21] = 'spring', [22] = 'summer', [23] = 'fall', [33] = '1', [34] = '2', [35] = '3', [36] = '4', [98] = 'Easter', [99] = 'Christmas'}; -- seasons lowercase, no autumn; proper-names use title case

if 0 == input.month2 then -- single season, quarter, or proper-name date

if 40 < tonumber(input.month) then

tCOinS_date.rftchron = season[input.month]; -- proper-name datesdate; used in journal metadata only

elseif 30 < tonumber(input.month) then

tCOinS_date.rftquarter = season[input.month]; -- quartersquarter date; used in journal metadata only

else

tCOinS_date.rftssn = season[input.month]; -- seasonsseason date; used in journal metadata only

end

else -- season rangeranges withare alumped secondinto season&rft.chron; &rft.ssn and &rft.quarter are left specifiedblank

if input.year ~= input.year2 then -- season year – season year range or season year–year

if 0 ~= input.month2 then

tCOinS_date.rftssn = season[input.month]; -- start of range season; keep this?

tCOinS_date.rftchron = string.format ('%s %s – %s %s', season[input.month], input.year, season[input.month2], input.year2); -- used in journal metadata only

if 0~= input.month2 then

tCOinS_date.rftchron = string.format ('%s %s – %s %s', season[input.month], input.year, season[input.month2], input.year2);

end

else -- season–season year range

tCOinS_date.rftssn = season[input.month]; -- start of range season; keep this?

tCOinS_date.rftchron = season[input.month] .. '–' .. season[input.month2]; -- season–season year range

end

else -- season–season year range

tCOinS_date.rftchron = season[input.month] .. '–' .. season[input.month2]; -- season–season year range; used in journal metadata only

end

end

tCOinS_date.rftdate = date;

tCOinS_date.rftdate = tostring (date);

return; -- done

end

-- here for gregorian calendar dates

if 0 ~= input.day then

date = string.format ('%s-%.2d-%.2d', input.year, tonumber(input.month), tonumber(input.day)); -- whole date

Line 412 ⟶ 395:

--[[--------------------------< P A T T E R N S _ T >--------------------------------------------------------------

this is the list of patterns for date formats that this module recognizes. Approximately the first half of these

Line 424 ⟶ 407:

]]

local patternspatterns_t = {

-- year-initial numerical year-month-day

['ymd'] = {'^(%d[1-9]%d%d%d)%-(%d%d)%-(%d%d)$', 'y', 'm', 'd'},

-- month-initial: month day, year

['Mdy'] = {'^(%D-) +([1-9]%d?), +((%d[1-9]%d%d%d?)%a?)$', 'm', 'd', 'a', 'y'},

-- month-initial day range: month day–day, year; days are separated by endash

['Md-dy'] = {'^(%D-) +([1-9]%d?)[%-–]([1-9]%d?), +(([1-9]%d%d%d%d?)%a?)$', 'm', 'd', 'd2', 'a', 'y'},

-- day-initial: day month year

['dMy'] = {'^([1-9]%d?) +(%D-) +((%d[1-9]%d%d%d?)%a?)$', 'd', 'm', 'a', 'y'},

-- year-initial: year month day; day: 1 or 2 two digits, leading zero allowed; not supported at en.wiki

-- ['yMd'] = {'^((%d[1-9]%d%d%d?)%a?) +(%D-) +(%d%d?)$', 'a', 'y', 'm', 'd'},

-- day-range-initial: day–day month year; days are separated by endash

['d-dMy'] = {'^([1-9]%d?)[%-–]([1-9]%d?) +(%D-) +(([1-9]%d%d%d%d?)%a?)$', 'd', 'd2', 'm', 'a', 'y'},

-- day initial month-day-range: day month - day month year; uses spaced endash

['dM-dMy'] = {'^([1-9]%d?) +(%D-) +[%-–] +([1-9]%d?) +(%D-) +(([1-9]%d%d%d%d?)%a?)$', 'd', 'm', 'd2', 'm2', 'a', 'y'},

-- month initial month-day-range: month day – month day, year; uses spaced endash

['Md-Mdy'] = {'^(%D-) +([1-9]%d?) +[%-–] +(%D-) +([1-9]%d?), +(([1-9]%d%d%d%d?)%a?)$','m', 'd', 'm2', 'd2', 'a', 'y'},

-- day initial month-day-year-range: day month year - day month year; uses spaced endash

['dMy-dMy'] = {'^([1-9]%d?) +(%D-) +([1-9]%d%d%d%d?) +[%-–] +([1-9]%d?) +(%D-) +(([1-9]%d%d%d%d?)%a?)$', 'd', 'm', 'y', 'd2', 'm2', 'a', 'y2'},

-- month initial month-day-year-range: month day, year – month day, year; uses spaced endash

['Mdy-Mdy'] = {'^(%D-) +([1-9]%d?), +([1-9]%d%d%d%d?) +[%-–] +(%D-) +([1-9]%d?), +(([1-9]%d%d%d%d?)%a?)$', 'm', 'd', 'y', 'm2', 'd2', 'a', 'y2'},

-- these date formats cannot be converted, per se, but month name can be rendered short or long

-- month/season year - month/season year; separated by spaced endash

['My-My'] = {'^(%D-) +([1-9]%d%d%d%d?) +[%-–] +(%D-) +(([1-9]%d%d%d%d?)%a?)$', 'm', 'y', 'm2', 'a', 'y2'},

-- month/season range year; months separated by endash

['M-My'] = {'^(%D-)[%-–](%D-) +(([1-9]%d%d%d%d?)%a?)$', 'm', 'm2', 'a', 'y'},

-- month/season year or proper-name year; quarter year when First Quarter YYYY etc.

['My'] = {'^([^%d–]-) +(([1-9]%d%d%d%d?)%a?)$', 'm', 'a', 'y'}, -- this way because endash is a member of %D; %D- will match January–March 2019 when it shouldn't

-- these date formats cannot be converted

['Sy4-y2'] = {'^(%D-) +((%d[1-9]%d)%d%d)[%-–]((%d%d)%a?)$'}, -- special case Winter/Summer year-year (YYYY-YY); year separated with unspaced endash

['Sy-y'] = {'^(%D-) +([1-9]%d%d%d%d?)[%-–](([1-9]%d%d%d%d?)%a?)$'}, -- special case Winter/Summer year-year; year separated with unspaced endash

['y-y'] = {'^(%d[1-9]%d%d%d?)[%-–]((%d[1-9]%d%d%d?)%a?)$'}, -- year range: YYY-YYY or YYY-YYYY or YYYY–YYYY; separated by unspaced endash; 100-9999

['y4-y2'] = {'^((%d[1-9]%d)%d%d)[%-–]((%d%d)%a?)$'}, -- year range: YYYY–YY; separated by unspaced endash

['ymxy'] = {'^(([1-9]%d%d%d%d?)%-(%d%da?)%-XX$', 'y', 'm'}, -- edtf year-initial; numericalhere year-month-XXaccept either YYY or YYYY

['y'] = {'^((%d%d%d%d?)%a?)$'}, -- year; here accept either YYY or YYYY

}

--[[--------------------------< I S _ V A L I D _ E M B A R G O _ D A T E >------------------------------------

returns true and date value if that value has proper dmy, mdy, ymd format.

returns false and 9999 (embargoed forever) when date value is not proper format; assumes that when |pmc-embargo-date= is

set, the editor intended to embargo a PMC but |pmc-embargo-date= does not hold a single date.

]]

local function is_valid_embargo_date (v)

if v:match (patterns_t['ymd'][1]) or -- ymd

v:match (patterns_t['Mdy'][1]) or -- dmy

v:match (patterns_t['dMy'][1]) then -- mdy

return true, v;

end

return false, '9999'; -- if here not good date so return false and set embargo date to long time in future

end

Line 488 ⟶ 489:

local function check_date (date_string, param, tCOinS_date)

local year; -- assume that year2, months, and days are not used;

local year2 = 0; -- second year in a year range

local month = 0;

local month2 = 0; -- second month in a month range

local day = 0;

local day2 = 0; -- second day in a day range

Line 496 ⟶ 497:

local coins_date;

if date_string:match (patternspatterns_t['ymd'][1]) then -- year-initial numerical year month day format

year, month, day = date_string:match (patternspatterns_t['ymd'][1]);

if 12 < tonumber(month) or 1 > tonumber(month) or 1582 > tonumber(year) or 0 == tonumber(day) then return false; end -- month or day number not valid or not Gregorian calendar

anchor_year = year;

elseif date_string:mw.ustring.match(date_string, (patternspatterns_t['ymxMdy'][1]) then -- yearmonth-initial numerical year: month edtfday, formatyear

yearmonth, monthday, anchor_year, year = date_string:mw.ustring.match (patternsdate_string, patterns_t['ymxMdy'][1]);

if 12 < tonumber(month) or 1 > tonumber(month) or 1582 > tonumber(year) then return false; end -- month number not valid or not Gregorian calendar

anchor_year = year;

elseif mw.ustring.match(date_string, patterns['Mdy'][1]) then -- month-initial: month day, year

month, day, anchor_year, year = mw.ustring.match(date_string, patterns['Mdy'][1]);

month = get_month_number (month);

if 0 == month then return false; end -- return false if month text isn't one of the twelve months

elseif mw.ustring.match(date_string, patternspatterns_t['Md-dy'][1]) then -- month-initial day range: month day–day, year; days are separated by endash

month, day, day2, anchor_year, year = mw.ustring.match(date_string, patternspatterns_t['Md-dy'][1]);

if tonumber(day) >= tonumber(day2) then return false; end -- date range order is left to right: earlier to later; dates may not be the same;

month = get_month_number (month);

Line 519 ⟶ 515:

year2 = year;

elseif mw.ustring.match(date_string, patternspatterns_t['dMy'][1]) then -- day-initial: day month year

day, month, anchor_year, year = mw.ustring.match(date_string, patternspatterns_t['dMy'][1]);

month = get_month_number (month);

if 0 == month then return false; end -- return false if month text isn't one of the twelve months

--[[ NOT supported at en.wiki

elseif mw.ustring.match(date_string, patternspatterns_t['yMd'][1]) then -- year-initial: year month day; day: 1 or 2 two digits, leading zero allowed

anchor_year, year, month, day = mw.ustring.match(date_string, patternspatterns_t['yMd'][1]);

month = get_month_number (month);

if 0 == month then return false; end -- return false if month text isn't one of the twelve months

-- end NOT supported at en.wiki ]]

elseif mw.ustring.match(date_string, patternspatterns_t['d-dMy'][1]) then -- day-range-initial: day–day month year; days are separated by endash

day, day2, month, anchor_year, year = mw.ustring.match(date_string, patternspatterns_t['d-dMy'][1]);

if tonumber(day) >= tonumber(day2) then return false; end -- date range order is left to right: earlier to later; dates may not be the same;

month = get_month_number (month);

Line 539 ⟶ 535:

year2 = year;

elseif mw.ustring.match(date_string, patternspatterns_t['dM-dMy'][1]) then -- day initial month-day-range: day month - day month year; uses spaced endash

day, month, day2, month2, anchor_year, year = mw.ustring.match(date_string, patternspatterns_t['dM-dMy'][1]);

if (not is_valid_month_season_range(month, month2)) or not is_valid_year(year) then return false; end -- date range order is left to right: earlier to later;

month = get_month_number (month); -- for metadata

Line 546 ⟶ 542:

year2 = year;

elseif mw.ustring.match(date_string, patternspatterns_t['Md-Mdy'][1]) then -- month initial month-day-range: month day – month day, year; uses spaced endash

month, day, month2, day2, anchor_year, year = mw.ustring.match(date_string, patternspatterns_t['Md-Mdy'][1]);

if (not is_valid_month_season_range(month, month2, param)) or not is_valid_year(year) then return false; end

month = get_month_number (month); -- for metadata

Line 553 ⟶ 549:

year2 = year;

elseif mw.ustring.match(date_string, patternspatterns_t['dMy-dMy'][1]) then -- day initial month-day-year-range: day month year - day month year; uses spaced endash

day, month, year, day2, month2, anchor_year, year2 = mw.ustring.match(date_string, patternspatterns_t['dMy-dMy'][1]);

if tonumber(year2) <= tonumber(year) then return false; end -- must be sequential years, left to right, earlier to later

if not is_valid_year(year2) or not is_valid_month_range_style(month, month2) then return false; end -- year2 no more than one year in the future; months same style

Line 561 ⟶ 557:

if 0 == month or 0 == month2 then return false; end -- both must be valid

elseif mw.ustring.match(date_string, patternspatterns_t['Mdy-Mdy'][1]) then -- month initial month-day-year-range: month day, year – month day, year; uses spaced endash

month, day, year, month2, day2, anchor_year, year2 = mw.ustring.match(date_string, patternspatterns_t['Mdy-Mdy'][1]);

if tonumber(year2) <= tonumber(year) then return false; end -- must be sequential years, left to right, earlier to later

if not is_valid_year(year2) or not is_valid_month_range_style(month, month2) then return false; end -- year2 no more than one year in the future; months same style

Line 569 ⟶ 565:

if 0 == month or 0 == month2 then return false; end -- both must be valid

elseif mw.ustring.match(date_string, patternspatterns_t['Sy4-y2'][1]) then -- special case Winter/Summer year-year (YYYY-YY); year separated with unspaced endash

local century;

month, year, century, anchor_year, year2 = mw.ustring.match(date_string, patternspatterns_t['Sy4-y2'][1]);

if 'Winter' ~= month and 'Summer' ~= month then return false end; -- 'month' can only be Winter or Summer

anchor_year = year .. '–' .. anchor_year; -- assemble anchor_year from both years

year2 = century..year2; -- add the century to year2 for comparisons

if 1 ~= tonumber(year2) - tonumber(year) then return false; end -- must be sequential years, left to right, earlier to later

Line 579 ⟶ 575:

month = get_season_number(month, param);

elseif mw.ustring.match(date_string, patternspatterns_t['Sy-y'][1]) then -- special case Winter/Summer year-year; year separated with unspaced endash

month, year, anchor_year, year2 = mw.ustring.match(date_string, patternspatterns_t['Sy-y'][1]);

if 'Winter' ~= month and 'Summer' ~= get_season_number (month, then return false endparam); -- '<month'> can only be Winterwinter or Summersummer; also for metadata

if (month ~= cfg.date_names['en'].season['Winter']) and (month ~= cfg.date_names['en'].season['Summer']) then

anchor_year = year .. '–' .. anchor_year; -- assemble anchor_year from both years

return false; -- not Summer or Winter; abandon

end

anchor_year = year .. '–' .. anchor_year; -- assemble anchor_year from both years

if 1 ~= tonumber(year2) - tonumber(year) then return false; end -- must be sequential years, left to right, earlier to later

if not is_valid_year(year2) then return false; end -- no year farther in the future than next year

month = get_season_number (month, param); -- for metadata

elseif mw.ustring.match(date_string, patternspatterns_t['My-My'][1]) then -- month/season year - month/season year; separated by spaced endash

month, year, month2, anchor_year, year2 = mw.ustring.match(date_string, patternspatterns_t['My-My'][1]);

anchor_year = year .. '–' .. anchor_year; -- assemble anchor_year from both years

if tonumber(year) >= tonumber(year2) then return false; end -- left to right, earlier to later, not the same

if not is_valid_year(year2) then return false; end -- no year farther in the future than next year

Line 602 ⟶ 600:

end

elseif mw.ustring.match(date_string, patternspatterns_t['M-My'][1]) then -- month/season range year; months separated by endash

month, month2, anchor_year, year = mw.ustring.match(date_string, patternspatterns_t['M-My'][1]);

if (not is_valid_month_season_range(month, month2, param)) or (not is_valid_year(year)) then return false; end

if 0 ~= get_month_number(month) then -- determined to be a valid range so just check this one to know if month or season

Line 615 ⟶ 613:

year2 = year;

elseif mw.ustring.match(date_string, patternspatterns_t['My'][1]) then -- month/season/quarter/proper-name year

month, anchor_year, year = mw.ustring.match(date_string, patternspatterns_t['My'][1]);

if not is_valid_year(year) then return false; end

month = get_element_number(month, param); -- get month season quarter proper-name number or nil

if not month then return false; end -- not valid whatever it is

elseif mw.ustring.match(date_string, patternspatterns_t['y-y'][1]) then -- Year range: YYY-YYY or YYY-YYYY or YYYY–YYYY; separated by unspaced endash; 100-9999

year, anchor_year, year2 = mw.ustring.match(date_string, patternspatterns_t['y-y'][1]);

anchor_year = year .. '–' .. anchor_year; -- assemble anchor year from both years

if tonumber(year) >= tonumber(year2) then return false; end -- left to right, earlier to later, not the same

if not is_valid_year(year2) then return false; end -- no year farther in the future than next year

elseif mw.ustring.match(date_string, patternspatterns_t['y4-y2'][1]) then -- Year range: YYYY–YY; separated by unspaced endash

local century;

year, century, anchor_year, year2 = mw.ustring.match(date_string, patternspatterns_t['y4-y2'][1]);

anchor_year = year .. '–' .. anchor_year; -- assemble anchor year from both years

if in_array (param, {'date', 'publication-date', 'year'}) then

add_prop_cat ('year_range_abbreviated');

end

if 13 > tonumber(year2) then return false; end -- don't allow 2003-05 which might be May 2003

year2 = century .. year2; -- add the century to year2 for comparisons

if tonumber(year) >= tonumber(year2) then return false; end -- left to right, earlier to later, not the same

if not is_valid_year(year2) then return false; end -- no year farther in the future than next year

if in_array (param, {'date', 'publication-date', 'year'}) then -- here when 'valid' abbreviated year range; if one of these parameters

elseif mw.ustring.match(date_string, patterns['y'][1]) then -- year; here accept either YYY or YYYY

add_prop_cat ('year-range-abbreviated'); -- add properties cat

anchor_year, year = mw.ustring.match(date_string, patterns['y'][1]);

end

elseif mw.ustring.match(date_string, patterns_t['y'][1]) then -- year; here accept either YYY or YYYY

anchor_year, year = mw.ustring.match(date_string, patterns_t['y'][1]);

if false == is_valid_year(year) then

return false;

Line 651 ⟶ 649:

end

if param ~= 'access-date' == param then -- testCITEREF accessdatedisambiguation hereonly becauseallowed wein have|date=; numerical|year= & |publication-date= partspromote to date

if anchor_year:match ('%l$') then

return false;

end

end

if 'access-date' == param then -- test access-date here because we have numerical date parts

if 0 ~= year and 0 ~= month and 0 ~= day and -- all parts of a single date required

0 == year2 and 0 == month2 and 0 == day2 then -- none of these; accessdateaccess-date must not be a range

if not is_valid_accessdate(year .. '-' .. month .. '-' .. day) then

return false; -- return false when accessdateaccess-date out of bounds

end

else

return false; -- return false when accessdateaccess-date is a range of two dates

end

end

if 'archive-date' == param then -- test archive-date here because we have numerical date parts

if not (0 ~= year and 0 ~= month and 0 ~= day and -- all parts of a single date required

0 == year2 and 0 == month2 and 0 == day2) then -- none of these; archive-date must not be a range

return false; -- return false when archive-date is a range of two dates

end

end

Line 664 ⟶ 675:

local result=true; -- check whole dates for validity; assume true because not all dates will go through this test

if 0 ~= year and 0 ~= month and 0 ~= day and 0 == year2 and 0 == month2 and 0 == day2 then -- YMD (simple whole date)

result = is_valid_date (year, month, day, param); -- <param> for |pmc-embargo-date=

elseif 0 ~= year and 0 ~= month and 0 ~= day and 0 == year2 and 0 == month2 and 0 ~= day2 then -- YMD-d (day range)

result = is_valid_date (year, month, day);

result = result and is_valid_date (year, month, day2);

elseif 0 ~= year and 0 ~= month and 0 ~= day and 0 == year2 and 0 ~= month2 and 0 ~= day2 then -- YMD-md (day month range)

result = is_valid_date (year, month, day);

result = result and is_valid_date (year, month2, day2);

elseif 0 ~= year and 0 ~= month and 0 ~= day and 0 ~= year2 and 0 ~= month2 and 0 ~= day2 then -- YMD-ymd (day month year range)

Line 729 ⟶ 740:

good_date, anchor_year, COinS_date = true, v.val:match("((%d+)%a?)");

end

elseif 'pmc-embargo-date' == k then -- if the parameter is |pmc-embargo-date=

good_date = check_date (v.val, k); -- go test the date

if true == good_date then -- if the date is a valid date

good_date, embargo_date = is_valid_embargo_date (v.val); -- is |pmc-embargo-date= date a single dmy, mdy, or ymd formatted date? yes: returns embargo date; no: returns 9999

end

else -- any other date-holding parameter

Line 743 ⟶ 754:

end

return anchor_year, embargo_date; -- and done

end

--[[--------------------------< Y E A R _ C H E C K >----------------------------------------------------------

Temporary function to test |year= for acceptable values:

YYY, YYYY, year-only ranges, their circa forms, with or without CITEREF disambiguators.

When |year= holds some form of date that is not one of these year-only dates, emit a maintenance message.

This function necessary because many non-cs1|2 templates have a |year= parameter so cirrus searches are more-or-

less useless

]]

local function year_check (year)

year = year:gsub ('c%. *', ''); -- remove circa annotation (if present) before testing <year>

for _, index in ipairs ({'y-y', 'y4-y2', 'y'}) do -- spin through these indexes into patterns_t

if mw.ustring.match (year, patterns_t[index][1]) then

return; -- if a match then |year= holds a valid 'year'

end

end

set_message ('maint_year'); -- if here, |year= value is not an accepted value; add a maint cat

end

Line 753 ⟶ 789:

2 - year value matches the year value in date when date is in the form YYYY-MM-DD and year is disambiguated (|year=YYYYx)

the numernicnumeric value in <result> determines the 'output' if any from this function:

0 – adds error message to error_list sequence table

1 – adds maint cat

Line 768 ⟶ 804:

year = year_string:match ('(%d%d%d%d?)');

if date_string:match ('%d%d%d%d%-%d%d%-%d%d') and year_string:match ('%d%d%d%d%a') then --special case where both date and year are required YYYY-MM-DD and YYYYx

date1 = date_string:match ('(%d%d%d%d)');

year = year_string:match ('(%d%d%d%d)');

if year ~= date1 then

result = 0; -- years don't match

elseif year_string:match ('%d%d%d%d%a') then

else

result = 2; -- years match; but because disambiguated, don't add to maint cat

end

Line 816 ⟶ 852:

format string used by string.format()

identifier letters ('d', 'm', 'y', 'd2', 'm2', 'y2') that serve as indexes into a table t{} that holds captures

from mw.ustring.match() for the various date parts specified by patternspatterns_t[pattern_idx][1]

Items in patternspatterns_t{} have the general form:

['ymd'] = {'^(%d%d%d%d)%-(%d%d)%-(%d%d)$', 'y', 'm', 'd'}, where:

['ymd'] is pattern_idx

patternspatterns_t['ymd'][1] is the match pattern with captures for mw.ustring.match()

patternspatterns_t['ymd'][2] is an indicator letter identifying the content of the first capture

patternspatterns_t['ymd'][3] ... the second capture etc.

when a pattern matches a date, the captures are loaded into table t{} in capture order using the idemtifier

Line 832 ⟶ 868:

format_param set to the desired format. This function loads table t{} as described and then calls string.format()

with the format string specified by re_format[pattern_idx][format_param][1] using values taken from t{} according

to the capture identifier letters specified by patternspatterns_t[pattern_idx][format_param][n] where n is 2..

]]

Line 840 ⟶ 876:

['mdy'] = {'%s %s, %s', 'm', 'd', 'y'}, -- |df=mdy

['dmy'] = {'%s %s %s', 'd', 'm', 'y'}, -- |df=dmy

-- ['yMd'] = {'%s %s %s', 'y', 'm', 'd'}, -- |df=yMd; not supported at en.wiki

},

['Mdy'] = { -- date format is Mdy; reformat to:

Line 846 ⟶ 882:

['dmy'] = {'%s %s %s', 'd', 'm', 'y'}, -- |df=dmy

['ymd'] = {'%s-%s-%s', 'y', 'm', 'd'}, -- |df=ymd

-- ['yMd'] = {'%s %s %s', 'y', 'm', 'd'}, -- |df=yMd; not supported at en.wiki

},

['dMy'] = { -- date format is dMy; reformat to:

Line 852 ⟶ 888:

['mdy'] = {'%s %s, %s', 'm', 'd', 'y'}, -- |df=mdy

['ymd'] = {'%s-%s-%s', 'y', 'm', 'd'}, -- |df=ymd

-- ['yMd'] = {'%s %s %s', 'y', 'm', 'd'}, -- |df=yMd; not supported at en.wiki

},

['Md-dy'] = { -- date format is Md-dy; reformat to:

Line 887 ⟶ 923:

['any'] = {'%s %s', 'm', 'y'}, -- dmy/mdy agnostic

},

-- ['yMd'] = { -- not supported at en.wiki

-- ['mdy'] = {'%s %s, %s', 'm', 'd', 'y'}, -- |df=mdy

-- ['dmy'] = {'%s %s %s', 'd', 'm', 'y'}, -- |df=dmy

-- ['ymd'] = {'%s-%s-%s', 'y', 'm', 'd'}, -- |df=ymd

-- },

}

Line 897 ⟶ 933:

local function reformatter (date, pattern_idx, format_param, mon_len)

if not in_array (pattern_idx, {'ymd', 'Mdy', 'Md-dy', 'dMy', 'yMd', 'd-dMy', 'dM-dMy', 'Md-Mdy', 'dMy-dMy', 'Mdy-Mdy', 'My-My', 'M-My', 'My'}) then

return; -- not in this set of date format patternspatterns_t then not a reformattable date

end

Line 908 ⟶ 944:

end

-- yMd is not supported at en.wiki; when yMd is supported at your wiki, uncomment the next line

-- if 'yMd' == format_param and in_array (pattern_idx, {'yMd', 'Md-dy', 'd-dMy', 'dM-dMy', 'Md-Mdy', 'dMy-dMy', 'Mdy-Mdy'}) then -- these formats not convertable; yMd not supported at en.wiki

-- if yMd is supported at your wiki, uncomment the next line

-- if 'yMd' == format_param andthen -- in_array (pattern_idx, {'yMd', 'Md-dy',not 'd-dMy',supported 'dM-dMy',at 'Md-Mdy',en.wiki; 'dMy-dMy',when 'Mdy-Mdy'})yMd then --is thesesupported formatsat notyour convertable;wiki, yMdremove notor supportedcomment-out atthis en.wikiline

-- if yMd is supported at your wiki, remove or comment-out the next line

if 'yMd' == format_param then -- yMd not supported at en.wiki

return; -- not a reformattable date

end

local c1, c2, c3, c4, c5, c6, c7; -- these hold the captures specified in patternspatterns_t[pattern_idx][1]

c1, c2, c3, c4, c5, c6, c7 = mw.ustring.match (date, patternspatterns_t[pattern_idx][1]); -- get the captures

local t = { -- table that holds k/v pairs of date parts from the captures and patternspatterns_t[pattern_idx][2..]

[patternspatterns_t[pattern_idx][2]] = c1; -- at minimum there is always one capture with a matching indicator letter

[patternspatterns_t[pattern_idx][3] or 'x'] = c2; -- patternspatterns_t can have a variable number of captures; each capture requires an indicator letter;

[patternspatterns_t[pattern_idx][4] or 'x'] = c3; -- where there is no capture, there is no indicator letter so n in patternspatterns_t[pattern_idx][n] will be nil;

[patternspatterns_t[pattern_idx][5] or 'x'] = c4; -- the 'x' here spoofs an indicator letter to prevent 'table index is nil' error

[patternspatterns_t[pattern_idx][6] or 'x'] = c5;

[patternspatterns_t[pattern_idx][7] or 'x'] = c6;

[patternspatterns_t[pattern_idx][8] or 'x'] = c7;

};

if t.a then -- if this date has an anchor year capture (all convertable date formats except ymd)

t.y =if t.a; y2 then -- use the anchorfor year capture when reassembling therange date formats

t.y2 = t.a; -- use the anchor year capture when reassembling the date

else -- here for single date formats (except ymd)

t.y = t.a; -- use the anchor year capture when reassembling the date

end

end

if tonumber(t.m) then -- if raw month is a number (converting from ymd)

if 's' == mon_len then -- if we are to use abbreviated month names

t.m = cfg.date_names['inv_local_sinv_local_short'][tonumber(t.m)]; -- convert it to a month name

else

t.m = cfg.date_names['inv_local_linv_local_long'][tonumber(t.m)]; -- convert it to a month name

end

t.d = t.d:gsub ('0(%d)', '%1'); -- strip leading '0' from day if present

elseif 'ymd' == format_param then -- when converting to ymd

t.y = t.y:gsub ('%a', ''); -- strip CITREF disambiguator if present; anchor year already known so process can proceed; TODO: maint message?

if 1582 > tonumber(t.y) then -- ymd format dates not allowed before 1582

if 1582 > tonumber (t.y) then -- ymd format dates not allowed before 1582

return;

end

Line 951 ⟶ 990:

t[mon] = get_month_number (t[mon]); -- get the month number for this month (is length agnostic)

if 0 == t[mon] then return; end -- seasons and named dates can't be converted

t[mon] = (('s' == mon_len) and cfg.date_names['inv_local_sinv_local_short'][t[mon]]) or cfg.date_names['inv_local_linv_local_long'][t[mon]]; -- fetch month name according to length

end

end

Line 1,021 ⟶ 1,060:

if is_set (param_val.val) then -- if the parameter has a value

if not (not all and in_array (param_name, {'access-date', 'archive-date'})) then -- skip access- or archive-date unless format is xxx-all; yeah, ugly; TODO: find a better way

for pattern_idx, pattern in pairs (patternspatterns_t) do

if mw.ustring.match (param_val.val, pattern[1]) then

if all and in_array (param_name, {'access-date', 'archive-date'}) then -- if this date is an access- or archive-date

Line 1,032 ⟶ 1,071:

date_parameters_list[param_name].val = new_date; -- update date in date list

result = true; -- and announce that changes have been made

break;

end

end -- if

Line 1,038 ⟶ 1,078:

end -- if

end -- for

return result; -- declare boolean result and done

end

Line 1,056 ⟶ 1,096:

for param_name, param_val in pairs(date_parameters_list) do -- for each date-holding parameter in the list

if is_set (param_val.val) and

not mw.ustring.match (param_val.val, '%d%d%d%d%-%d%d%-%d%d'patterns_t.ymd[1]) then -- for those that are not ymd dates (ustring because here digits may not be Western)

param_val.val, n = param_val.val:gsub ('%-', '–'); -- replace any hyphen with ndash

if 0 ~= n then

date_parameters_list[param_name].val = param_val.val; -- update the list

result = true;

end

end

end

return result; -- so we know if any hyphens were replaced

end

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

Loops through the list of date-holding parameters and converts any EDTF formatted dates to MOS compliant dates.

Only YYY-MM-XX supported at this time. Not called if the cs1|2 template has any date errors.

must be done before reformat_dates()

Modifies the date_parameters_list and returns true if transformation is performed, else returns false.

]]

local function edtf_transform (date_parameters_list)

local result = false;

local source_date = {};

for param_name, param_val in pairs(date_parameters_list) do -- for each date-holding parameter in the list

if is_set(param_val.val) and param_val.val:match (patterns.ymx[1]) then -- if parameter is set and is an EDTF dates

source_date.year, source_date.month = param_val.val:match (patterns.ymx[1]); -- get year and month number

source_date.day = 1; -- required by os.time()

date_parameters_list[param_name].val = mw.text.trim (os.date ('%B %Y', os.time (source_date)));

result = true;

end

end

return result; -- so we know if a transform was done

end

Line 1,097 ⟶ 1,110:

--[[-------------------------< D A T E _ N A M E _ X L A T E >------------------------------------------------

Attempts to translate English monthdate names to local-language monthdate names using names supplied by MediaWiki's

date parser function. This is simple name-for-name replacement and may not work for all languages.

Line 1,111 ⟶ 1,124:

local date;

local sources_t = {

{cfg.date_names.en.long, cfg.date_names.inv_local_long}, -- for translating long English month names to long local month names

{cfg.date_names.en.short, cfg.date_names.inv_local_short}, -- short month names

{cfg.date_names.en.quarter, cfg.date_names.inv_local_quarter}, -- quarter date names

{cfg.date_names.en.season, cfg.date_names.inv_local_season}, -- season date nam

{cfg.date_names.en.named, cfg.date_names.inv_local_named}, -- named dates

}

local function is_xlateable (month) -- local function to get local date name that replaces existing English-language date name

for _, date_names_t in ipairs (sources_t) do -- for each sequence table in date_names_t

if date_names_t[1][month] then -- if date name is English month (long or short), quarter, season or named and

if date_names_t[2][date_names_t[1][month]] then -- if there is a matching local date name

return date_names_t[2][date_names_t[1][month]]; -- return the local date name

end

end

end

end

for param_name, param_val in pairs(date_parameters_list) do -- for each date-holding parameter in the list

if is_set(param_val.val) then -- if the parameter has a value

date = param_val.val;

for month in mw.ustring.gmatch (date, '[%a ]+') do -- iterate through all datesdate names in the date (single date or date range)

month = mw.text.trim (month); -- this because quarterly dates contain whitespace

if cfg.date_names.en.long[month] then

modexlate = 'F'is_xlateable (month); -- Englishget nametranslate is long so<month>; usereturns longtranslation localor namenil

elseif cfg.date_names.en.short[month] then

if xlate then

mode = 'M'; -- English name is short so use short local name

else

mode = nil; -- not an English month name; could be local language month name or an English season name

end

if mode then -- might be a season

xlate = lang_object:formatDate(mode, '1' .. month); -- translate the month name to this local language

date = mw.ustring.gsub (date, month, xlate); -- replace the English with the translation

date_parameters_list[param_name].val = date; -- save the translated date

Line 1,158 ⟶ 1,183:

cfg = cfg_table_ptr; -- import tables from selected Module:Citation/CS1/Configuration

end

--[[--------------------------< A R C H I V E _ D A T E _ C H E C K >------------------------------------------

Compare value in |archive-date= with the timestamp in Wayback machine urls. Emits an error message with suggested

date from the |archive-url= timestamp in an appropriate format when the value in |archive-date= does not match

the timestamp.

this function never called when any date in a cs1|2 template has errors

error message suggests new |archive-date= value in an appropriate format specified by <df>. <df> is either

|df= or cfg.global_df in that order. If <df> is nil, suggested date has format from |archive-date=. There is

a caveat: when |df=dmy or |df=mdy, the reformatter leaves |access-date= and |archive-date= formats as they are.

The error message suggested date is passed to the formatter as YYYY-MM-DD so when |df=dmy or |df=mdy, the format

is not changed.

]]

local function archive_date_check (archive_date, archive_url_timestamp, df)

local archive_date_format = 'dmy-y'; -- holds the date format of date in |archive-date; default to ymd; 'dmy' used here to spoof reformat_dates()

for _, v_t in ipairs ({{'dMy', 'dmy-all'}, {'Mdy', 'mdy-all'}}) do -- is |archive-date= format dmy or mdy?

if archive_date:match (patterns_t[v_t[1]][1]) then -- does the pattern match?

archive_date_format = cfg.keywords_xlate[v_t[2]]; -- get appropriate |df= supported keyword from the i18n translator table

break;

end

end

local dates_t = {};

dates_t['archive-date'] = {val=archive_date, name=''}; -- setup to call reformat_dates(); never called when errors so <name> unset as not needed

reformat_dates (dates_t, 'dmy-y'); -- reformat |archive-date= to ymd; 'dmy' used here to spoof reformat_dates()

local archive_url_date = archive_url_timestamp:gsub ('(%d%d%d%d)(%d%d)(%d%d)%d*', '%1-%2-%3'); -- make ymd format date from timestamp

if dates_t['archive-date'].val == archive_url_date then -- are the two dates the same

return; -- yes, done

else

dates_t['archive-date'] = {val=archive_url_date, name=''}; -- setup to call reformat_dates() with the timestamp date

reformat_dates (dates_t, df or archive_date_format); -- reformat timestamp to format specified by <df> or format used in |archive-date=

archive_url_date = dates_t['archive-date'].val;

set_message ('err_archive_date_url_ts_mismatch', archive_url_date); -- emit an error message

end

end

Line 1,165 ⟶ 1,232:

return { -- return exported functions

archive_date_check = archive_date_check,

dates = dates,

year_date_check = year_date_check,

reformat_dates = reformat_dates,

date_hyphen_to_dash = date_hyphen_to_dash,

date_name_xlate = date_name_xlate,

dates = dates,

edtf_transform = edtf_transform,

reformat_dates = reformat_dates,

set_selected_modules = set_selected_modules

set_selected_modules = set_selected_modules,

year_check = year_check,

year_date_check = year_date_check,

}