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

m
No edit summary
 
(16 intermediate revisions by 6 users not shown)
Line 1:
--[[--------------------------< F O R W A R D D E C L A R A T I O N S >--------------------------------------
]]
 
local add_prop_cat, is_set, in_array, set_message, substitute, wrap_style; -- imported functions from selected Module:Citation/CS1/Utilities
local p = {}
local cfg; -- table of tables imported from selected Module:Citation/CS1/Configuration
 
 
--[[--------------------------< F OI RL WE A- RS DC O P E D E C L A R A T I O N S >--------------------------------------
 
File-scope variables are declared here
 
]]
 
local lang_object = mw.getContentLanguage(); -- used by is_valid_accessdate(), is_valid_year(), date_name_xlate(); TODO: move to ~/Configuration?
local is_set, in_array; -- imported functions from selected Module:Citation/CS1/Utilities
local cfgyear_limit; -- table of tables imported fromused slectedby Module:Citation/CS1/Configurationis_valid_year()
 
 
--[=[-------------------------< I S _ V A L I D _ A C C E S S D A T E >----------------------------------------
Line 15 ⟶ 22:
 
Wikipedia start date is 2001-01-15T00:00:00 UTC which is 979516800 seconds after 1970-01-01T00:00:00 UTC (the start of Unix time)
accessdate is the date provided in |accessdateaccess-date= at time 00:00:00 UTC
today is the current date at time 00:00:00 UTC plus 48 hours
if today is 2015-01-01T00:00:00 then
Line 22 ⟶ 29:
 
This function does not work if it is fed month names for languages other than English. Wikimedia #time: parser
apparently doesn't understand non-EngishEnglish date month names. This function will always return false when the date
contains a non-English month name because good1 is false after the call to lang.formatDate(). To get around that
call this function with YYYY-MM-DD format dates.
Line 29 ⟶ 36:
 
local function is_valid_accessdate (accessdate)
local lang = mw.getContentLanguage();
local good1, good2;
local access_ts, tomorrow_ts; -- to hold unixUnix time stamps representing the dates
 
good1, access_ts = pcall( lang(lang_object.formatDate, langlang_object, 'U', accessdate ); -- convert accessdate value to unixUnix timesatmptimestamp
good2, tomorrow_ts = pcall( lang(lang_object.formatDate, langlang_object, 'U', 'today + 2 days' ); -- today midnight + 2 days is one second more than all day tomorrow
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 langlang_object:parseFormattedNumber (access_ts); -- convert to numbers for the comparison;
tomorrow_ts = tonumber (tomorrow_ts) or langlang_object:parseFormattedNumber (tomorrow_ts);
else
return false; -- one or both failed to convert to unixUnix time stamp
end
 
Line 53 ⟶ 59:
--[[--------------------------< G E T _ M O N T H _ N U M B E R >----------------------------------------------
 
returns a number according to the month in a date: 1 for January, etc. Capitalization and spelling must be correct. If not a valid month, returns 0
If not a valid month, returns 0
 
]]
 
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
 
 
--[[--------------------------< IG SE T _ VS E A LS IO DN _ EN U M B AE R G O _ D A T E >--------------------------------------------
 
returns a number according to the sequence of seasons in a year: 21 for Spring, etc. Capitalization and spelling
returns true and date value if that value has proper dmy, mdy, ymd format.
must be correct. If not a valid season, returns 0.
21-24 = Spring, Summer, Autumn, Winter, independent of “Hemisphere”
 
returns false and 9999 (embargoed forever)0 when date value<param> is not proper format; assumes that when |embargodate= is
 
set, the editor intended to embargo a pmc but |embargo= does not hold a single date.
Season numbering is defined by Extended Date/Time Format (EDTF) specification (https://www.loc.gov/standards/datetime/)
which became part of ISO 8601 in 2019. See '§Sub-year groupings'. The standard defines various divisions using
numbers 21-41. cs1|2 only supports generic seasons. EDTF does support the distinction between north and south
hemisphere seasons but cs1|2 has no way to make that distinction.
 
These additional divisions not currently supported:
25-28 = Spring - Northern Hemisphere, Summer- Northern Hemisphere, Autumn - Northern Hemisphere, Winter - Northern Hemisphere
29-32 = Spring – Southern Hemisphere, Summer– Southern Hemisphere, Autumn – Southern Hemisphere, Winter - Southern Hemisphere
33-36 = Quarter 1, Quarter 2, Quarter 3, Quarter 4 (3 months each)
37-39 = Quadrimester 1, Quadrimester 2, Quadrimester 3 (4 months each)
40-41 = Semestral 1, Semestral-2 (6 months each)
 
]]
 
local function is_valid_embargo_dateget_season_number (vseason, param)
if 'date' ~= param then
if v:match ('^%d%d%d%d%-%d%d%-%d%d$') or -- ymd
return 0; -- season dates only supported by |date=
v:match ('^%d%d?%s+%a+%s+%d%d%d%d$') or -- dmy
v:match ('^%a+%s+%d%d?%s*,%s*%d%d%d%d$') then -- mdy
return true, v;
end
return cfg.date_names['local'].season[season] or -- look for local names first
return false, '9999'; -- if here not good date so return false and set embargo date to long time in future
cfg.date_names['en'].season[season] or -- failing that, look for English names
0; -- not a recognized season name
end
 
 
--[[--------------------------< G E T _ SQ EU A SR OT NE R _ N U M B E R >--------------------------------------------
 
returns a number according to the sequence of seasonsquarters in a year: 133 for Winterfirst quarter, etc. Capitalization and spelling must be correct. If not a valid season, returns 0
must be correct. If not a valid quarter, returns 0.
33-36 = Quarter 1, Quarter 2, Quarter 3, Quarter 4 (3 months each)
 
returns 0 when <param> is not |date=
Uses ISO DIS 8601 2016 part 2 §4.7 Divisions of a year for hemishpere-independent seasons:
 
21-24 = Spring, Summer, Autumn, Winter, independent of “Hemisphere”
Quarter numbering is defined by Extended Date/Time Format (EDTF) specification (https://www.loc.gov/standards/datetime/)
which became part of ISO 8601 in 2019. See '§Sub-year groupings'. The standard defines various divisions using
numbers 21-41. cs1|2 only supports generic seasons and quarters.
 
These additional divisions not currently supported:
25-28 = Spring - Northern Hemisphere, Summer- Northern Hemisphere, Autumn - Northern Hemisphere, Winter - Northern Hemisphere
29-32 = Spring – Southern Hemisphere, Summer– Southern Hemisphere, Autumn – Southern Hemisphere, Winter - Southern Hemisphere
33-36 = Quarter 1, Quarter 2, Quarter 3, Quarter 4 (3 months each)
37-39 = Quadrimester 1, Quadrimester 2, Quadrimester 3 (4 months each)
40-41 = Semestral 1, Semestral-2 (6 months each)
 
 
]]
 
local function get_season_numberget_quarter_number (seasonquarter, param)
if 'date' ~= param then
return cfg.date_names['local'].season[season] or -- look for local names first
return 0; -- quarter dates only supported by |date=
cfg.date_names['en'].season[season] or -- failing that, look for English names
end
0; -- not a recognized season name
quarter = mw.ustring.gsub (quarter, ' +', ' '); -- special case replace multiple space chars with a single space char
return cfg.date_names['local'].quarter[quarter] or -- look for local names first
cfg.date_names['en'].quarter[quarter] or -- failing that, look for English names
0; -- not a recognized quarter name
end
 
--[[--------------------------< I S _ P R O P E R _ N A M E >--------------------------------------------------
 
--[[--------------------------< G E T _ P R O P E R _ N A M E _ N U M B E R >----------------------------------
returns a non-zero number if date contains a recognized proper name. Capitalization and spelling must be correct.
 
returns a non-zero number if date contains a recognized proper-name. Capitalization and spelling must be correct.
 
returns 0 when <param> is not |date=
 
]]
 
local function is_proper_nameget_proper_name_number (name, param)
if 'date' ~= param then
return 0; -- proper-name dates only supported by |date=
end
return cfg.date_names['local'].named[name] or -- look for local names dates first
cfg.date_names['en'].named[name] or -- failing that, look for English names
Line 118 ⟶ 149:
end
 
--[[--------------------------< I S _ V A L I D _ M O N T H _ O R _ S E A S O N >------------------------------
 
--[[--------------------------< G E T _ E L E M E N T _ N U M B E R <------------------------------------------
--returns true if month or season is valid (properly spelled, capitalized, abbreviated)
 
returns true if month or season or quarter or proper name is valid (properly spelled, capitalized, abbreviated)
 
]]
 
local function is_valid_month_or_seasonget_element_number (month_seasonelement, param)
local num;
if 0 == get_month_number (month_season) then -- if month text isn't one of the twelve months, might be a season
if 0 == get_season_number (month_season) then -- not a month, is it a season?
local funcs = {get_month_number, get_season_number, get_quarter_number, get_proper_name_number}; -- list of functions to execute in order
return false; -- return false not a month or one of the five seasons
for _, func in ipairs (funcs) do -- spin through the function list
num = func (element, param); -- call the function and get the returned number
if 0 ~= num then -- non-zero when valid month season quarter
return num; -- return that number
end
end
return truenil; -- not valid
end
 
 
--[[--------------------------< I S _ V A L I D _ Y E A R >----------------------------------------------------
 
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.
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 year_limit;
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
 
return tonumber(year) <= year_limit; -- false if year is in the future more than one year
year = tonumber (year) or lang_object:parseFormattedNumber (year); -- convert to number for the comparison;
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
 
 
--[[--------------------------< I S _ V A L I D _ D A T E >----------------------------------------------------
 
Returns true if day is less than or equal to the number of days in month and year is no farther into the future
than next year; else returns false.
Line 156 ⟶ 205:
]]
 
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
end
else
month_length = days_in_month[month];
end
 
Line 185 ⟶ 234:
return true;
end
 
 
--[[--------------------------< I S _ V A L I D _ M O N T H _ R A N G E _ S T Y L E >--------------------------
 
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
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 216 ⟶ 261:
Month pairs are expected to be left to right, earliest to latest in time.
 
All season ranges are accepted as valid because there are publishers out there who have published a Summer–Spring YYYY issue, sohence ...treat as ok
 
]]
 
local function is_valid_month_season_range(range_start, range_end, param)
local range_start_number = get_month_number (range_start);
local range_end_number;
 
if 0 == range_start_number then -- is this a month range?
range_start_number = get_season_number (range_start, param); -- not a month; is it a season? get start season number
range_end_number = get_season_number (range_end, param); -- get end season number
 
if (0 ~= range_start_number) and (0 ~= range_end_number) and (range_start_number ~= range_end_number) then
return true; -- any season pairing is accepted except when both are the same
end
return false; -- range_start and/or range_end is not a season
Line 235 ⟶ 280:
-- here when range_start is a month
range_end_number = get_month_number (range_end); -- get end month number
if range_start_number < range_end_number thenand -- range_start is a month; does range_start precede range_end?
if is_valid_month_range_style (range_start, range_end) then -- do months have the same style?
return true; -- proper order and same style
end
end
return false; -- range_start month number is greater than or equal to range end number; or range end isn't a month
Line 267 ⟶ 311:
the output table receives:
rftdate: an IS8601ISO 8601 formatted date
rftchron: a free-form version of the date, usually without year which is in rftdate (season ranges and propernameproper-name dates)
rftssn: one of four season keywords: winter, spring, summer, fall (lowercase)
rftquarter: one of four values: 1, 2, 3, 4
 
]]
Line 281 ⟶ 326:
local day = tonumber (input.day);
if (0 ~= day) and -- day must have a value for this to be a whole date
(((1582 == year) and (10 <= month) and (12 >= month)) or -- any whole 1582 date from 1 octoberOctober 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
-- end temporary Julian / GergorianGregorian calendar uncertainty detection
if 1582 > tonumber(input.year) or 20 < tonumber(input.month) then -- Julian calendar or season so &rft.date gets year only
Line 293 ⟶ 338:
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 propernameproper-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 date
if 3040 < tonumber(input.month) then
tCOinS_date.rftchron = season[input.month]; -- proper -name dates
elseif 30 < tonumber(input.month) then
tCOinS_date.rftquarter = season[input.month]; -- quarters
else
tCOinS_date.rftssn = season[input.month]; -- seasons
Line 337 ⟶ 384:
tCOinS_date.rftdate = date .. date2; -- date2 has the '/' separator
return;
end
 
 
--[[--------------------------< P A T T E R N S >--------------------------------------------------------------
 
this is the list of patterns for date formats that this module recognizes. Approximately the first half of these
patterns represent formats that might be reformatted into another format. Those that might be reformatted have
'indicator' letters that identify the content of the matching capture: 'd' (day), 'm' (month), 'a' (anchor year),
'y' (year); second day, month, year have a '2' suffix.
 
These patterns are used for both date validation and for reformatting. This table should not be moved to ~/Configuration
because changes to this table require changes to check_date() and to reformatter() and reformat_date()
 
]]
 
local patterns = {
-- year-initial numerical year-month-day
['ymd'] = {'^(%d%d%d%d)%-(%d%d)%-(%d%d)$', 'y', 'm', 'd'},
-- month-initial: month day, year
['Mdy'] = {'^(%D-) +([1-9]%d?), +((%d%d%d%d?)%a?)$', '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?), +((%d%d%d%d)%a?)$', 'm', 'd', 'd2', 'a', 'y'},
-- day-initial: day month year
['dMy'] = {'^([1-9]%d?) +(%D-) +((%d%d%d%d?)%a?)$', '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%d%d%d?)%a?) +(%D-) +(%d%d?)$', 'a', 'y', 'm', 'd'},
-- day-range-initial: day–day month year; days are separated by endash
['d-dMy'] = {'^([1-9]%d?)[%-–]([1-9]%d?) +(%D-) +((%d%d%d%d)%a?)$', '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-) +((%d%d%d%d)%a?)$', 'd', 'm', 'd2', 'm2', 'a', 'y'},
-- month initial month-day-range: month day – month day, year; uses spaced endash
['Md-Mdy'] = {'^(%D-) +([1-9]%d?) +[%-–] +(%D-) +([1-9]%d?), +((%d%d%d%d)%a?)$','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-) +(%d%d%d%d) +[%-–] +([1-9]%d?) +(%D-) +((%d%d%d%d)%a?)$', 'd', 'm', 'y', 'd2', 'm2', 'a', 'y2'},
-- month initial month-day-year-range: month day, year – month day, year; uses spaced endash
['Mdy-Mdy'] = {'^(%D-) +([1-9]%d?), +(%d%d%d%d) +[%-–] +(%D-) +([1-9]%d?), +((%d%d%d%d)%a?)$', '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-) +(%d%d%d%d) +[%-–] +(%D-) +((%d%d%d%d)%a?)$', 'm', 'y', 'm2', 'a', 'y2'},
-- month/season range year; months separated by endash
['M-My'] = {'^(%D-)[%-–](%D-) +((%d%d%d%d)%a?)$', 'm', 'm2', 'a', 'y'},
-- month/season year or proper-name year; quarter year when First Quarter YYYY etc.
['My'] = {'^([^%d–]-) +((%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%d)%d%d)[%-–]((%d%d)%a?)$'}, -- special case Winter/Summer year-year (YYYY-YY); year separated with unspaced endash
['Sy-y'] = {'^(%D-) +(%d%d%d%d)[%-–]((%d%d%d%d)%a?)$'}, -- special case Winter/Summer year-year; year separated with unspaced endash
['y-y'] = {'^(%d%d%d%d?)[%-–]((%d%d%d%d?)%a?)$'}, -- year range: YYY-YYY or YYY-YYYY or YYYY–YYYY; separated by unspaced endash; 100-9999
['y4-y2'] = {'^((%d%d)%d%d)[%-–]((%d%d)%a?)$'}, -- year range: YYYY–YY; separated by unspaced endash
['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['ymd'][1]) or -- ymd
v:match (patterns['Mdy'][1]) or -- dmy
v:match (patterns['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 348 ⟶ 465:
 
If the date fails the format tests, this function returns false and does not return values for anchor_year and
COinS_date. When this happens, the date parameter is (DEBUG: not?) used in the COinS metadata and the CITEREF identifier gets
its year from the year parameter if present otherwise CITEREF does not get a date value.
 
Inputs:
date_string - date string from date-holding parameters (date, year, accessdatepublication-date, access-date, pmc-embargo-date, archivedatearchive-date, etc.lay-date)
 
Returns:
Line 362 ⟶ 479:
]]
 
local function check_date (date_string, tCOinS_dateparam, test_accessdatetCOinS_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
local anchor_year;
local coins_date;
 
if date_string:match ("^%d%d%d%d%-%d%d%-%d%d$"patterns['ymd'][1]) then -- year-initial numerical year month day format
year, month, day =string. date_string:match(date_string, "(%d%d%d%d)%-(%d%d)%-(%d%d)"patterns['ymd'][1]);
if 12 < tonumber(month) or 1 > tonumber(month) or 1582 > tonumber(year) or 0 == tonumber(day) then return false; end -- month or day number not valid or not Gregorian calendar
anchor_year = year;
elseif mw.ustring.match(date_string, "^%D- +[1-9]%d?, +[1-9]%d%d%d%a?$") then -- month-initial: month day, year
month, elseif day, anchor_year, year=mw.ustring.match(date_string, "(%D-patterns['Mdy'][1]) +(%d%d?)then -- month-initial: month day,%s*((%d%d%d%d?)%a?)"); 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, "^%D- +patterns[1'Md-9]%d?[%-–dy'][1-9]%d?, +[1-9]%d%d%d%a?$") 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, "(%D-) +(%d%d?)patterns[%'Md-dy'][1](%d%d?), +((%d%d%d%d)%a?)");
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);
if 0 == month then return false; end -- return false if month text isn't one of the twelve months
month2=month; -- for metadata
year2 = year;
 
elseif mw.ustring.match(date_string, "^patterns[1-9'dMy']%d? +%D- +[1-9]%d%d%d%a?$") then -- day-initial: day month year
day, month, anchor_year, year = mw.ustring.match(date_string, "(%d%d*)%s*(%D-) +((%d%d%d%d?)%a?)"patterns['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, "^[1-9]%d?[%-–][1-9]%d? +%D- +[1-9]%d%d%d%a?$") then -- day-range-initial: day–day month year; days are separated by endash
day, elseif day2, month, anchor_year, year=mw.ustring.match(date_string, "(%d%d?)patterns[%-–'yMd'][1](%d%d?) +(%Dthen --) +((%d%d%d%d)%a?)")year-initial: year month day; day: 1 or 2 two digits, leading zero allowed
anchor_year, year, month, day = mw.ustring.match(date_string, patterns['yMd'][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);
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 ]]
month2=month; -- for metadata
year2=year;
 
elseif mw.ustring.match(date_string, "^patterns[1-9]%'d? +%D- +[%-–dMy'] +[1-9]%d? +%D- +[1-9]%d%d%d%a?$") then -- day initial month-dayrange-rangeinitial: dayday–day month -year; daydays month year;are usesseparated spacedby endash
day, month, day2, month2month, anchor_year, year = mw.ustring.match(date_string, "(%patterns['d%d?) +(%D-) +dMy'][%-–1] +(%d%d?) +(%D-) +((%d%d%d%d)%a?)");
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);
if 0 == month then return false; end -- return false if month text isn't one of the twelve months
month2 = month; -- for metadata
year2 = year;
 
elseif mw.ustring.match(date_string, patterns['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, patterns['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
month2 = get_month_number (month2);
year2 = year;
 
elseif mw.ustring.match(date_string, "^%D- +patterns[1'Md-9Mdy']%d? +[%-–] +%D- +[1-9]%d?, +[1-9]%d%d%d%a?$") 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, "(%D-) +(%d%d?) +patterns[%'Md-Mdy'][1] +(%D-) +(%d%d?), +((%d%d%d%d)%a?)");
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
month2 = get_month_number (month2);
year2 = year;
 
elseif mw.ustring.match(date_string, "^patterns[1'dMy-9dMy']%d? +%D- +[1-9]%d%d%d +[%-–] +[1-9]%d? +%D- +[1-9]%d%d%d%a?$") 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, "(%d%d?) +(%D-) +(%d%d%d%d) +patterns[%'dMy-dMy'][1] +(%d%d?) +(%D-) +((%d%d%d%d)%a?)");
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
month = get_month_number (month); -- for metadata
month2 = get_month_number (month2);
if 0 == month or 0 == month2 then return false; end -- both must be valid
 
elseif mw.ustring.match(date_string, "^%D- +patterns[1'Mdy-9Mdy']%d?, +[1-9]%d%d%d +[%-–] +%D- +[1-9]%d?, +[1-9]%d%d%d%a?$") 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, "(%D-) +(%d%d?), +(%d%d%d%d) +patterns[%'Mdy-Mdy'][1] +(%D-) +(%d%d?), +((%d%d%d%d)%a?)");
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
month = get_month_number (month); -- for metadata
month2 = get_month_number (month2);
if 0 == month or 0 == month2 then return false; end -- both must be valid
 
elseif mw.ustring.match(date_string, "^%D- +patterns[1'Sy4-9y2']%d%d%d[%-–1]%d%d%a?$") 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, "(%D-) +((%d%d)%d%d)patterns[%'Sy4-y2'][1]((%d%d)%a?)");
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
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);
 
elseif mw.ustring.match(date_string, "^%D- +patterns[1'Sy-9]%d%d%d[%-–y'][1-9]%d%d%d%a?$") then -- special case Winter/Summer year-year; year separated with unspaced endash
month, year, anchor_year, year2 = mw.ustring.match(date_string, "(%D-) +(%d%d%d%d)patterns[%'Sy-y'][1]((%d%d%d%d)%a?)");
ifmonth '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
if 1 ~= tonumber(year2) - tonumber(year) then return false; end -- must be sequential years, left to right, earlier to later
end
if not is_valid_year(year2) then return false; end -- no year farther in the future than next year
anchor_year = year .. '–' .. anchor_year; -- assemble anchor_year from both years
month = get_season_number (month); -- for metadata
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
 
elseif mw.ustring.match(date_string, "^%D- +patterns[1'My-9My']%d%d%d +[%-–] +%D- +[1-9]%d%d%d%a?$") then -- month/season year - month/season year; separated by spaced endash
month, year, month2, anchor_year, year2 = mw.ustring.match(date_string, "(%D-) +(%d%d%d%d) +patterns[%'My-My'][1] +(%D-) +((%d%d%d%d)%a?)");
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
if 0 ~= get_month_number(month) and 0 ~= get_month_number(month2) and is_valid_month_range_style(month, month2) then -- both must be month year, same month style
month = get_month_number(month);
month2 = get_month_number(month2);
elseif 0 ~= get_season_number(month, param) and 0 ~= get_season_number(month2, param) then -- both must be or season year, not mixed
month = get_season_number(month, param);
month2 = get_season_number(month2, param);
else
return false;
end
 
elseif mw.ustring.match(date_string, "^%D-patterns[%'M-My']%D- +[1-9]%d%d%d%a?$") then -- month/season range year; months separated by endash
month, month2, anchor_year, year = mw.ustring.match(date_string, "(%D-)patterns[%'M-My'][1](%D-)%s*((%d%d%d%d)%a?)");
if (not is_valid_month_season_range(month, month2, param)) or (not is_valid_year(year)) then return false; end
if 0 ~= get_month_number(month) then -- determined to be a valid range so just check this one to know if month or season
month = get_month_number(month);
month2 = get_month_number(month2);
if 0 == month or 0 == month2 then return false; end
else
month = get_season_number(month, param);
month2 = get_season_number(month2, param);
end
year2 = year;
elseif mw.ustring.match(date_string, "^%D- +%d%d%d%d%a?$"patterns['My'][1]) then -- month/season year or /quarter/proper-name year
month, anchor_year, year = mw.ustring.match(date_string, "(%D-)%s*((%d%d%d%d)%a?)"patterns['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 is_valid_month_or_season (month) and 0 == is_proper_name (month) then return false; end
if 0not ~= get_month_number(month) then return false; end -- determined to be anot valid range so just check this one to know if monthwhatever orit seasonis
month = get_month_number(month);
elseif 0 ~= get_season_number(month) then
month = get_season_number(month);
else
month = is_proper_name (month); -- must be proper name; not supported in COinS
end
 
elseif mw.ustring.match(date_string, "^patterns[1'y-9]%d%d%d?[%-–y'][1-9]%d%d%d?%a?$") 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, "(%d%d%d%d?)patterns[%'y-y'][1]((%d%d%d%d?)%a?)");
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, "^patterns[1'y4-9y2']%d%d%d[%-–1]%d%d%a?$") then -- Year range: YYYY–YY; separated by unspaced endash
local century;
year, century, anchor_year, year2 = mw.ustring.match(date_string, "((%d%d)%d%d)patterns[%'y4-y2'][1]((%d%d)%a?)");
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
 
elseif date_string:mw.ustring.match("^date_string, patterns['y'][1-9]%d%d%d?%a?$") then -- year; here accept either YYY or YYYY
anchor_year, year =date_string: mw.ustring.match("((%d%d%d%d?)%a?)"date_string, patterns['y'][1]);
if false == is_valid_year(year) then
return false;
Line 512 ⟶ 641:
end
 
if test_accessdateparam ~= 'date' 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 accessdate 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; accessdate must not be a range
if not is_valid_accessdate (year .. '-' .. month .. '-' .. day) then
return false; -- return false when accessdate out of bounds
end
Line 525 ⟶ 660:
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)
result = is_valid_date(year, month, day);
result = result and is_valid_date(year2, month2, day2);
end
Line 543 ⟶ 678:
 
if nil ~= tCOinS_date then -- this table only passed into this function when testing |date= parameter values
make_COinS_date ({year = year, month = month, day = day, year2 = year2, month2 = month2, day2 = day2}, tCOinS_date); -- make an ISO 8601 date string for COinS
end
Line 557 ⟶ 692:
 
Unlike most error messages created in this module, only one error message is created by this function. Because all of the date holding parameters are processed serially,
aparameters singlewith errorerrors messageare isadded createdto the <error_list> sequence table as the dates are tested.
 
]]
 
local function dates(date_parameters_list, tCOinS_date, error_list)
local anchor_year; -- will return as nil if the date being tested is not |date=
local COinS_date; -- will return as nil if the date being tested is not |date=
local embargo_date; -- if embargo date is a good dmy, mdy, ymd date then holds original value else reset to 9999
local error_message = "";
local good_date = false;
 
for k, v in pairs(date_parameters_list) do -- for each date-holding parameter in the list
if is_set(v.val) then -- if the parameter has a value
v.val = mw.ustring.gsub(v.val, '%d', cfg.date_names.local_digits); -- translate 'local' digits to Western 0-9
if v:match("^c%. [1-9]%d%d%d?%a?$") then -- special case for c. year or with or without CITEREF disambiguator - only |date= and |year=
if v.val:match("^c%. [1-9]%d%d%d?%a?$") then -- special case for c. year or with or without CITEREF disambiguator - only |date= and |year=
v = mw.ustring.gsub (v, '%d', cfg.date_names.local_digits);
local year = v.val:match("c%. ([1-9]%d%d%d?)%a?"); -- get the year portion so it can be tested
if 'date' == k then
anchor_year, COinS_date = v.val:match("((c%. [1-9]%d%d%d?)%a?)"); -- anchor year and COinS_date only from |date= parameter
good_date = is_valid_year(year);
elseif 'year' == k then
good_date = is_valid_year(year);
end
elseif 'date' == k then -- if the parameter is |date=
if v.val:match("^n%.d%.%a?$") then -- ToDo: I18N -- if |date=n.d. with or without a CITEREF disambiguator
good_date, anchor_year, COinS_date = true, v.val:match("((n%.d%.)%a?)"); -- ToDo: I18N -- "n.d."; no error when date parameter is set to no date
elseif v.val:match("^nd%a?$") then -- ToDo: I18N -- if |date=nd with or without a CITEREF disambiguator
good_date, anchor_year, COinS_date = true, v.val:match("((nd)%a?)"); -- ToDo: I18N -- "nd"; no error when date parameter is set to no date
else
good_date, anchor_year, COinS_date = check_date (v.val, k, tCOinS_date); -- go test the date
end
elseif 'year' == k then -- if the parameter is |year= it should hold only a year value
if v.val:match("^[1-9]%d%d%d?%a?$") then -- if |year = 3 or 4 digits only with or without a CITEREF disambiguator
good_date, anchor_year, COinS_date = true, v.val:match("((%d+)%a?)");
end
elseif 'accesspmc-embargo-date' == k then -- if the parameter is |pmc-embargo-date=
good_date = check_date (v.val, nil, truek); -- go test the date; nil is a placeholder; true is the test_accessdate flag
elseif 'embargo'==k then -- if the parameter is |embargo=
good_date = check_date (v); -- 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
good_date = check_date (v.val, k); -- go test the date
end
if false == good_date then -- assemble one error message so we don't add the tracking category multiple times
table.insert (error_list, wrap_style ('parameter', v.name)); -- make parameter name suitable for error message list
if is_set(error_message) then -- once we've added the first portion of the error message ...
error_message=error_message .. ", "; -- ... add a comma space separator
end
error_message=error_message .. "&#124;" .. k .. "="; -- add the failed parameter
end
end
end
return anchor_year, embargo_date, error_message; -- and done
end
 
Line 615 ⟶ 744:
--[[--------------------------< Y E A R _ D A T E _ C H E C K >------------------------------------------------
 
Compare the value provided in |year= with the year value(s) provided in |date=. This function returnssets a local numeric value:
0 - year value does not match the year value in date
1 - (default) year value matches the year value in date or one of the year values when date contains two years
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 numeric value in <result> determines the 'output' if any from this function:
0 – adds error message to error_list sequence table
1 – adds maint cat
2 – does nothing
 
]]
 
local function year_date_check (year_string, year_origin, date_string, date_origin, error_list)
local year;
local date1;
local date2;
local result = 1; -- result of the test; assume that the test passes
 
year = year_string:match ('(%d%d%d%d?)');
 
Line 645 ⟶ 779:
end
 
elseif mw.ustring.match(date_string, "%d%d%d%d[%-–]%d%d") then -- YYYY-YY date ranges
local century;
date1, century, date2 = mw.ustring.match(date_string, "((%d%d)%d%d)[%-–]+(%d%d)");
Line 658 ⟶ 792:
result = 0;
end
else -- should never get here; this function called only when no other date errors
else
result = 0; -- no recognizable year in date
end
 
return result;
if 0 == result then -- year / date mismatch
table.insert (error_list, substitute (cfg.messages['mismatch'], {year_origin, date_origin})); -- add error message to error_list sequence table
elseif 1 == result then -- redundant year / date
set_message ('maint_date_year'); -- add a maint cat
end
end
 
 
--[[--------------------------< R E F O R M A T T A B L E SR >--------------------------------------------------------
 
reformat 'date' into new format specified by format_param if pattern_idx (the current format of 'date') can be
These table are used exclusively for reformatting dates
reformatted. Does the grunt work for reformat_dates().
 
The table re_formats maps pattern_idx (current format) and format_param (desired format) to a table that holds:
]]
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 patterns[pattern_idx][1]
 
Items in patterns{} have the general form:
local source_patterns = { -- this table holds patterns that match allowed date formats used to extract date components
['dmyymd'] = {'^(%d%d?)%s+(d%a+d)%s+-(%d%d)%-(%d%d)$', 'y', 'm', 'd'}, where:
['ymd'] is pattern_idx
['mdy'] = '^(%a+)%s+(%d%d?),%s+(%d%d%d%d)$',
patterns['ymd'][1] is the match pattern with captures for mw.ustring.match()
['ymd'] = '^(%d%d%d%d)%-(%d%d)-(%d%d)$',
patterns['ymd'][2] is an indicator letter identifying the content of the first capture
}
patterns['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
local short_formats = { -- this table holds format strings used by os.date() for short month names
characters as indexes into t{} For the above, a ymd date is in t{} as:
['dmy'] = '%e %b %Y',
t.y = first capture (year), t.m = second capture (month), t.d = third capture (day)
['mdy'] = '%b %e, %Y',
['ymd'] = '%F',
}
 
To reformat, this function is called with the pattern_idx that matches the current format of the date and with
local long_formats = { -- this table holds format strings used by os.date() for long month names
format_param set to the desired format. This function loads table t{} as described and then calls string.format()
['dmy'] = '%e %B %Y',
with the format string specified by re_format[pattern_idx][format_param][1] using values taken from t{} according
['mdy'] = '%B %e, %Y',
to the capture identifier letters specified by patterns[pattern_idx][format_param][n] where n is 2..
['ymd'] = '%F',
}
 
 
--[[-------------------------< G E T _ D M Y _ D A T E _ P A R T S >------------------------------------------
 
extracts year, month and day from DMY formatted date, places them in the source_date table, and returns.
 
]]
 
local re_formats = {
local function get_dmy_date_parts (date, source_date)
['ymd'] = { -- date format is ymd; reformat to:
source_date.day, source_date.month, source_date.year = date:match (source_patterns['dmy']); -- get date components as strings
['mdy'] = {'%s %s, %s', 'm', 'd', 'y'}, -- |df=mdy
source_date.month = get_month_number (source_date.month); -- get month number
['dmy'] = {'%s %s %s', 'd', 'm', 'y'}, -- |df=dmy
end
-- ['yMd'] = {'%s %s %s', 'y', 'm', 'd'}, -- |df=yMd; not supported at en.wiki
},
['Mdy'] = { -- date format is Mdy; reformat to:
['mdy'] = {'%s %s, %s', 'm', 'd', 'y'}, -- for long/short reformatting
['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:
['dmy'] = {'%s %s %s', 'd', 'm', 'y'}, -- for long/short reformatting
['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:
['mdy'] = {'%s %s–%s, %s', 'm', 'd', 'd2', 'y'}, -- for long/short reformatting
['dmy'] = {'%s–%s %s %s', 'd', 'd2', 'm', 'y'}, -- |df=dmy -> d-dMy
},
['d-dMy'] = { -- date format is d-d>y; reformat to:
['dmy'] = {'%s–%s %s %s', 'd', 'd2', 'm', 'y'}, -- for long/short reformatting
['mdy'] = {'%s %s–%s, %s', 'm', 'd', 'd2', 'y'}, -- |df=mdy -> Md-dy
},
['dM-dMy'] = { -- date format is dM-dMy; reformat to:
['dmy'] = {'%s %s – %s %s %s', 'd', 'm', 'd2', 'm2', 'y'}, -- for long/short reformatting
['mdy'] = {'%s %s – %s %s, %s', 'm', 'd', 'm2', 'd2', 'y'}, -- |df=mdy -> Md-Mdy
},
['Md-Mdy'] = { -- date format is Md-Mdy; reformat to:
['mdy'] = {'%s %s – %s %s, %s', 'm', 'd', 'm2', 'd2', 'y'}, -- for long/short reformatting
['dmy'] = {'%s %s – %s %s %s', 'd', 'm', 'd2', 'm2', 'y'}, -- |df=dmy -> dM-dMy
},
['dMy-dMy'] = { -- date format is dMy-dMy; reformat to:
['dmy'] = {'%s %s %s – %s %s %s', 'd', 'm', 'y', 'd2', 'm2', 'y2'}, -- for long/short reformatting
['mdy'] = {'%s %s, %s – %s %s, %s', 'm', 'd', 'y', 'm2', 'd2', 'y2'}, -- |df=mdy -> Mdy-Mdy
},
['Mdy-Mdy'] = { -- date format is Mdy-Mdy; reformat to:
['mdy'] = {'%s %s, %s – %s %s, %s', 'm', 'd', 'y', 'm2', 'd2', 'y2'}, -- for long/short reformatting
['dmy'] = {'%s %s %s – %s %s %s', 'd', 'm', 'y', 'd2', 'm2', 'y2'}, -- |df=dmy -> dMy-dMy
},
['My-My'] = { -- these for long/short reformatting
['any'] = {'%s %s – %s %s', 'm', 'y', 'm2', 'y2'}, -- dmy/mdy agnostic
},
['M-My'] = { -- these for long/short reformatting
['any'] = {'%s–%s %s', 'm', 'm2', 'y'}, -- dmy/mdy agnostic
},
['My'] = { -- these for long/short reformatting
['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
-- },
}
 
 
local function reformatter (date, pattern_idx, format_param, mon_len)
--[[-------------------------< G E T _ M D Y _ D A T E _ P A R T S >------------------------------------------
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 patterns then not a reformattable date
end
if 'ymd' == format_param and in_array (pattern_idx, {'ymd', 'Md-dy', 'd-dMy', 'dM-dMy', 'Md-Mdy', 'dMy-dMy', 'Mdy-Mdy', 'My-My', 'M-My', 'My'}) then
return; -- ymd date ranges not supported at en.wiki; no point in reformatting ymd to ymd
end
 
if in_array (pattern_idx, {'My', 'M-My', 'My-My'}) then -- these are not dmy/mdy so can't be 'reformatted' into either
extracts year, month and day from MDY formatted date, places them in the source_date table, and returns.
format_param = 'any'; -- so format-agnostic
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' == format_param then -- yMd not supported at en.wiki; when yMd is supported at your wiki, remove or comment-out this line
return; -- not a reformattable date
end
local c1, c2, c3, c4, c5, c6, c7; -- these hold the captures specified in patterns[pattern_idx][1]
c1, c2, c3, c4, c5, c6, c7 = mw.ustring.match (date, patterns[pattern_idx][1]); -- get the captures
 
local t = { -- table that holds k/v pairs of date parts from the captures and patterns[pattern_idx][2..]
local function get_mdy_date_parts (date, source_date)
[patterns[pattern_idx][2]] = c1; -- at minimum there is always one capture with a matching indicator letter
source_date.month, source_date.day, source_date.year = date:match (source_patterns['mdy']); -- get date components as strings
[patterns[pattern_idx][3] or 'x'] = c2; -- patterns can have a variable number of captures; each capture requires an indicator letter;
source_date.month = get_month_number (source_date.month); -- get month number
[patterns[pattern_idx][4] or 'x'] = c3; -- where there is no capture, there is no indicator letter so n in patterns[pattern_idx][n] will be nil;
end
[patterns[pattern_idx][5] or 'x'] = c4; -- the 'x' here spoofs an indicator letter to prevent 'table index is nil' error
[patterns[pattern_idx][6] or 'x'] = c5;
[patterns[pattern_idx][7] or 'x'] = c6;
[patterns[pattern_idx][8] or 'x'] = c7;
};
 
if t.a then -- if this date has an anchor year capture (all convertable date formats except ymd)
if t.y2 then -- for year range 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)
--[[-------------------------< G E T _ Y M D _ D A T E _ P A R T S >------------------------------------------
if 's' == mon_len then -- if we are to use abbreviated month names
t.m = cfg.date_names['inv_local_short'][tonumber(t.m)]; -- convert it to a month name
else
t.m = cfg.date_names['inv_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
return;
end
t.m = string.format ('%02d', get_month_number (t.m)); -- make sure that month and day are two digits
t.d = string.format ('%02d', t.d);
elseif mon_len then -- if mon_len is set to either 'short' or 'long'
for _, mon in ipairs ({'m', 'm2'}) do -- because there can be two month names, check both
if t[mon] then
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_short'][t[mon]]) or cfg.date_names['inv_local_long'][t[mon]]; -- fetch month name according to length
end
end
end
 
local new_date = string.format (re_formats[pattern_idx][format_param][1], -- format string
extracts year, month and day from YMD formatted date, places them in the source_date table, and returns.
t[re_formats[pattern_idx][format_param][2]], -- named captures from t{}
t[re_formats[pattern_idx][format_param][3]],
t[re_formats[pattern_idx][format_param][4]],
t[re_formats[pattern_idx][format_param][5]],
t[re_formats[pattern_idx][format_param][6]],
t[re_formats[pattern_idx][format_param][7]],
t[re_formats[pattern_idx][format_param][8]]
);
 
return new_date;
]]
 
local function get_ymd_date_parts (date, source_date)
source_date.year, source_date.month, source_date.day = date:match (source_patterns['ymd']); -- get date components as strings
end
 
Line 727 ⟶ 971:
--[[-------------------------< R E F O R M A T _ D A T E S >--------------------------------------------------
 
Reformats existing dates into the format specified by format and short.
 
format is one of several manual keywords: dmy, dmy-all, mdy, mdy-all, ymd, ymd-all. The -all version includes access- and
access- and archive-dates; otherwise these dates are not reformatted.
 
This function allows automatic date formatting. In ~/Configuration, the article source is searched for one of
Date ranges, season dates, proper name dates are not currently supported.
the {{use xxx dates}} templates. If found, xxx becomes the global date format as xxx-all. If |cs1-dates= in
{{use xxx dates}} has legitimate value then that value determines how cs1|2 dates will be rendered. Legitimate
values for |cs1-dates= are:
l - all dates are rendered with long month names
ls - publication dates use long month names; access-/archive-dates use abbreviated month names
ly - publication dates use long month names; access-/archive-dates rendered in ymd format
s - all dates are rendered with abbreviated (short) month names
sy - publication dates use abbreviated month names; access-/archive-dates rendered in ymd format
y - all dates are rendered in ymd format
 
the format argument for automatic date formatting will be the format specified by {{use xxx dates}} with the
For i18n: This code works only at en.wiki because os.date() doesn't support any languages other than English.
value supplied by |cs1-dates so one of: xxx-l, xxx-ls, xxx-ly, xxx-s, xxx-sy, xxx-y, or simply xxx (|cs1-dates=
mw.getContentLanguage():formatDate() will work at non-English wikis only when the date format is yyyy-mm-dd. This is
empty, omitted, or invalid) where xxx shall be either of dmy or mdy.
the same issue that plagues is_valid_accessdate()
 
dates are extracted from date_parameters_list, reformatted (if appropriate), and then written back into the
It is possible that a solution like that written for ht:Module:Citation/CS1/Date_validation date_name_xlate() could be applied to this problem
list in the new format. Dates in date_parameters_list are presumed here to be valid (no errors). This function
returns true when a date has been reformatted, false else. Actual reformatting is done by reformatter().
 
]]
 
local function reformat_dates (date_parameters_list, format, short)
local all = false; -- set to false to skip access- and archive-dates
local len_p = 'l'; -- default publication date length shall be long
local format_str;
local len_a = 'l'; -- default access-/archive-date length shall be long
local source_date = {};
local result = false;
local new_date;
if format:match('%a+%-all') then -- manual df keyword; auto df keyword when length not specified in {{use xxx dates}};
format = format:match('(%a+)%-all'); -- extract the format
all = true; -- setall todates trueare tolong format access-dates andbecause archive-datesthis keyword doesn't specify length
elseif format:match('%a+%-[lsy][sy]?') then -- auto df keywords; internal only
end
all = true; -- auto df applies to all dates; use length specified by capture len_p for all dates
format, len_p, len_a = format:match('(%a+)%-([lsy])([sy]?)'); -- extract the format and length keywords
for param_name, param_val in pairs (date_parameters_list) do -- for each date-holding parameter in the list
if is_set'y' (param_val)== len_p then -- ifbecause allowed by MOS:DATEUNIFY (sort of) range dates theand parameterMy hasdates anot valuereformatted
format = 'ymd'; -- override {{use xxx dates}}
if not all and in_array (param_name, {'access-date', 'archive-date'}) then -- if access- or archive-date and format not xxx-all
elseif (not is_set(len_a)) or (len_p == len_a) then -- no access-/archive-date length specified or same length as publication dates then
param_val = ''; -- set to empty string so we don't process this date
len_a = len_p; -- in case len_a not set
end
for source, pattern in pairs (source_patterns) do
if param_val:match (pattern) then
if 'ymd' == source then
get_ymd_date_parts (param_val, source_date); -- get the date parts into the source_date table
elseif 'dmy' == source then
get_dmy_date_parts (param_val, source_date); -- get the date parts into the source_date table
elseif 'mdy' == source then
get_mdy_date_parts (param_val, source_date); -- get the date parts into the source_date table
end
 
if 'ymd' == format and 1582 > tonumber(source_date.year) then -- ymd format dates not allowed before 1582
return false; -- abandon reformatting
end
if short then
format_str = short_formats[format];
else
format_str = long_formats[format];
end
-- convert date and save;
date_parameters_list[param_name] = mw.text.trim (os.date (format_str, os.time(source_date))); -- strip leading space when single digit day and %e is first format
end
end
end
end -- else only publication dates and they are long
end
 
return true; -- declare success and done
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
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 (patterns) 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
new_date = reformatter (param_val.val, pattern_idx, (('y' == len_a) and 'ymd') or format, len_a); -- choose ymd or dmy/mdy according to len_a setting
else -- all other dates
new_date = reformatter (param_val.val, pattern_idx, format, len_p);
end
if new_date then -- set when date was reformatted
date_parameters_list[param_name].val = new_date; -- update date in date list
result = true; -- and announce that changes have been made
end
end -- if
end -- for
end -- if
end -- if
end -- for
return result; -- declare boolean result and done
end
 
Line 799 ⟶ 1,054:
local n;
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
if not param_val:match ('%d%d%d%d%-%d%d%-%d%d') then -- for those that are not ymd dates
not mw.ustring.match (param_val.val, patterns.ymd[1]) then -- for those that are not ymd dates (ustring because here digits may not be Western)
param_val, n = param_val:gsub ('%-', '–'); -- replace any hyphen with ndash
param_val.val, n = param_val.val:gsub ('%-', '–'); -- replace any hyphen with ndash
if 0 ~= n then
if 0 ~= n then
date_parameters_list[param_name] = param_val; -- update the list
date_parameters_list[param_name].val = param_val.val; -- update the list
result = true;
result = true;
end
end
end
end
Line 813 ⟶ 1,069:
--[[-------------------------< 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.
 
if xlat_dig is true, this function will also translate Western (English) digits to the local language's digits.
This will also translate ymd dates.
 
]]
 
local function date_name_xlate (date_parameters_list, xlt_dig)
local xlate;
local mode; -- long or short month names
Line 824 ⟶ 1,083:
local date;
local sources_t = {
for param_name, param_val in pairs(date_parameters_list) do -- for each date-holding parameter in the list
{cfg.date_names.en.long, cfg.date_names.inv_local_long}, -- for translating long English month names to long local month names
if is_set(param_val) then -- if the parameter has a value
{cfg.date_names.en.short, cfg.date_names.inv_local_short}, -- short month names
date = param_val;
{cfg.date_names.en.quarter, cfg.date_names.inv_local_quarter}, -- quarter date names
for month in mw.ustring.gmatch (date, '%a+') do -- iterate through all dates in the date (single date or date range)
if {cfg.date_names.en.long[month]season, cfg.date_names.inv_local_season}, -- season date thennam
{cfg.date_names.en.named, cfg.date_names.inv_local_named}, -- named dates
mode = 'F'; -- English name is long so use long local name
}
elseif cfg.date_names.en.short[month] then
 
mode = 'M'; -- English name is short so use short local name
local function is_xlateable (month) -- local function to get local date name that replaces existing English-language date name
else
for _, date_names_t in ipairs (sources_t) do -- for each sequence table in date_names_t
mode = nil; -- not an English month name; could be local language month name or an English season name
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
if mode then -- might be a season
end
xlate = mw.getContentLanguage():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
for param_name, param_val in pairs(date_parameters_list) do -- for each date-holding parameter in the list
date_parameters_list[param_name] = date; -- save the translated date
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 date names in the date (single date or date range)
month = mw.text.trim (month); -- this because quarterly dates contain whitespace
xlate = is_xlateable (month); -- get translate <month>; returns translation or nil
if xlate then
date = mw.ustring.gsub (date, month, xlate); -- replace the English with the translation
date_parameters_list[param_name].val = date; -- save the translated date
modified = true;
end
end
 
if xlt_dig then -- shall we also translate digits?
date = date:gsub ('%d', cfg.date_names.xlate_digits); -- translate digits from Western to 'local digits'
date_parameters_list[param_name].val = date; -- save the translated date
modified = true;
end
end
end
return modified;
end
 
 
Line 858 ⟶ 1,134:
 
local function set_selected_modules (cfg_table_ptr, utilities_page_ptr)
is_setadd_prop_cat = utilities_page_ptr.is_setadd_prop_cat ; -- import functions from selected Module:Citation/CS1/Utilities module
is_set = utilities_page_ptr.is_set;
in_array = utilities_page_ptr.in_array; -- import functions from selected Module:Citation/CS1/Utilities module
in_array = utilities_page_ptr.in_array;
set_message = utilities_page_ptr.set_message;
substitute = utilities_page_ptr.substitute;
wrap_style = utilities_page_ptr.wrap_style;
 
cfg = cfg_table_ptr; -- import tables from selected Module:Citation/CS1/Configuration
end
 
 
--[[--------------------------< E X P O R T E D F U N C T I O N S >------------------------------------------
]]
 
return { -- return exported functions
verified
2,516

edits