Module:Citation/CS1

local cs1 ={};

--[[--------------------------< F O R W A R D D E C L A R A T I O N S >-------------------------------------- ]]

local dates, year_date_check, reformat_dates, date_hyphen_to_dash -- functions in Module:Citation/CS1/Date_validation

local is_set, in_array, substitute, error_comment, set_error, select_one, -- functions in Module:Citation/CS1/Utilities add_maint_cat, wrap_style, safe_for_italics, remove_wiki_link;

local z ={}; -- tables in Module:Citation/CS1/Utilities

local extract_ids, extract_id_access_levels, build_id_list, is_embargoed; -- functions in Module:Citation/CS1/Identifiers

local make_coins_title, get_coins_pages, COinS; -- functions in Module:Citation/CS1/COinS

local cfg = {}; -- table of configuration tables that are defined in Module:Citation/CS1/Configuration local whitelist = {}; -- table of tables listing valid template parameter names; defined in Module:Citation/CS1/Whitelist

--[[--------------------------< P A G E S C O P E V A R I A B L E S >--------------------------------------

delare variables here that have page-wide scope that are not brought in from other modules; thatare created here and used here

]]

local added_deprecated_cat; -- boolean flag so that the category is added only once local added_prop_cats = {}; -- list of property categories that have been added to z.properties_cats local added_vanc_errs; -- boolean flag so we only emit one Vancouver error / category

local Frame; -- holds the module's frame table

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

Locates and returns the first set value in a table of values where the order established in the table, left-to-right (or top-to-bottom), is the order in which the values are evaluated. Returns nil if none are set.

This version replaces the original 'for _, val in pairs do' and a similar version that used ipairs. With the pairs version the order of evaluation could not be guaranteed. With the ipairs version, a nil value would terminate the for-loop before it reached the actual end of the list.

]]

local function first_set (list, count) local i = 1; while i <= count do -- loop through all items in list if is_set( list[i] ) then return list[i]; -- return the first set list member end i = i + 1; -- point to next end end

--[[--------------------------< A D D _ P R O P _ C A T >--------------------------------------------------------

Adds a category to z.properties_cats using names from the configuration file with additional text if any.

added_prop_cats is a table declared in page scope variables above

]]

local function add_prop_cat (key, arguments) if not added_prop_cats [key] then added_prop_cats [key] = true; -- note that we've added this category table.insert( z.properties_cats, substitute (cfg.prop_cats [key], arguments)); -- make name then add to table end end

--[[--------------------------< A D D _ V A N C _ E R R O R >----------------------------------------------------

Adds a single Vancouver system error message to the template's output regardless of how many error actually exist. To prevent duplication, added_vanc_errs is nil until an error message is emitted.

added_vanc_errs is a boolean declared in page scope variables above

]]

local function add_vanc_error (source) if not added_vanc_errs then added_vanc_errs = true; -- note that we've added this category table.insert( z.message_tail, { set_error( 'vancouver', {source}, true ) } ); end end

--[[--------------------------< I S _ S C H E M E >------------------------------------------------------------

does this thing that purports to be a uri scheme seem to be a valid scheme? The scheme is checked to see if it is in agreement with http://tools.ietf.org/html/std66#section-3.1 which says: Scheme names consist of a sequence of characters beginning with a

  letter and followed by any combination of letters, digits, plus
  ("+"), period ("."), or hyphen ("-").

returns true if it does, else false

]]

local function is_scheme (scheme) return scheme and scheme:match ('^%a[%a%d%+%.%-]*:'); -- true if scheme is set and matches the pattern end

--[=[-------------------------< I S _ D O M A I N _ N A M E >--------------------------------------------------

Does this thing that purports to be a domain name seem to be a valid domain name?

Syntax defined here: http://tools.ietf.org/html/rfc1034#section-3.5 BNF defined here: https://tools.ietf.org/html/rfc4234 Single character names are generally reserved; see https://tools.ietf.org/html/draft-ietf-dnsind-iana-dns-01#page-15; see also Single-letter second-level domain list of tlds: https://www.iana.org/domains/root/db

rfc952 (modified by rfc 1123) requires the first and last character of a hostname to be a letter or a digit. Between the first and last characters the name may use letters, digits, and the hyphen.

Also allowed are IPv4 addresses. IPv6 not supported

domain is expected to be stripped of any path so that the last character in the last character of the tld. tld is two or more alpha characters. Any preceding '//' (from splitting a url with a scheme) will be stripped here. Perhaps not necessary but retained incase it is necessary for IPv4 dot decimal.

There are several tests: the first character of the whole domain name including subdomains must be a letter or a digit internationalized domain name (ascii characters with .xn-- ASCII Compatible Encoding (ACE) prefix xn-- in the tld) see https://tools.ietf.org/html/rfc3490 single-letter/digit second-level domains in the .org TLD q, x, and z SL domains in the .com TLD i and q SL domains in the .net TLD single-letter SL domains in the ccTLDs (where the ccTLD is two letters) two-character SL domains in gTLDs (where the gTLD is two or more letters) three-plus-character SL domains in gTLDs (where the gTLD is two or more letters) IPv4 dot-decimal address format; TLD not allowed

returns true if domain appears to be a proper name and tld or IPv4 address, else false

]=]

local function is_domain_name (domain) if not domain then return false; -- if not set, abandon end

domain = domain:gsub ('^//', ); -- strip '//' from domain name if present; done here so we only have to do it once

if not domain:match ('^[%a%d]') then -- first character must be letter or digit return false; end -- Do most common case first if domain:match ('%f[%a%d][%a%d][%a%d%-]+[%a%d]%.%a%a+$') then -- three or more character hostname.hostname or hostname.tld return true; elseif domain:match ('%f[%a%d][%a%d][%a%d%-]+[%a%d]%.xn%-%-[%a%d]+$') then -- internationalized domain name with ACE prefix return true; elseif domain:match ('%f[%a%d][%a%d]%.org$') then -- one character .org hostname return true; elseif domain:match ('%f[%a][qxz]%.com$') then -- assigned one character .com hostname (x.com times out 2015-12-10) return true; elseif domain:match ('%f[%a][iq]%.net$') then -- assigned one character .net hostname (q.net registered but not active 2015-12-10) return true; elseif domain:match ('%f[%a%d][%a%d]%.%a%a$') then -- one character hostname and cctld (2 chars) return true; elseif domain:match ('%f[%a%d][%a%d][%a%d]%.%a%a+$') then -- two character hostname and tld return true; elseif domain:match ('^%d%d?%d?%.%d%d?%d?%.%d%d?%d?%.%d%d?%d?') then -- IPv4 address return true; else return false; end end

--[[--------------------------< I S _ U R L >------------------------------------------------------------------

returns true if the scheme and domain parts of a url appear to be a valid url; else false.

This function is the last step in the validation process. This function is separate because there are cases that are not covered by split_url(), for example is_parameter_ext_wikilink() which is looking for bracketted external wikilinks.

]]

local function is_url (scheme, domain) if is_set (scheme) then -- if scheme is set check it and domain return is_scheme (scheme) and is_domain_name (domain); else return is_domain_name (domain); -- scheme not set when url is protocol relative end end

--[[--------------------------< S P L I T _ U R L >------------------------------------------------------------

Split a url into a scheme, authority indicator, and domain.

First remove Fully Qualified Domain Name terminator (a dot following tld) (if any) and any path(/), query(?) or fragment(#).

If protocol relative url, return nil scheme and domain else return nil for both scheme and domain.

When not protocol relative, get scheme, authority indicator, and domain. If there is an authority indicator (one or more '/' characters immediately following the scheme's colon), make sure that there are only 2.

Strip off any port and path;

]]

local function split_url (url_str) local scheme, authority, domain;

url_str = url_str:gsub ('([%a%d])%.?[/%?#].*$', '%1'); -- strip FQDN terminator and path(/), query(?), fragment (#) (the capture prevents false replacement of '//')

if url_str:match ('^//%S*') then -- if there is what appears to be a protocol relative url domain = url_str:match ('^//(%S*)') elseif url_str:match ('%S-:/*%S+') then -- if there is what appears to be a scheme, optional authority indicator, and domain name scheme, authority, domain = url_str:match ('(%S-:)(/*)(%S+)'); -- extract the scheme, authority indicator, and domain portions authority = authority:gsub ('//', , 1); -- replace place 1 pair of '/' with nothing; if is_set(authority) then -- if anything left (1 or 3+ '/' where authority should be) then return scheme; -- return scheme only making domain nil which will cause an error message end domain = domain:gsub ('(%a):%d+', '%1'); -- strip port number if present end

return scheme, domain; end

--[[--------------------------< L I N K _ P A R A M _ O K >---------------------------------------------------

checks the content of |title-link=, |series-link=, |author-link= etc for properly formatted content: no wikilinks, no urls

Link parameters are to hold the title of a wikipedia article so none of the WP:TITLESPECIALCHARACTERS are allowed: # < > [ ] | { } _ except the underscore which is used as a space in wiki urls and # which is used for section links

returns false when the value contains any of these characters.

When there are no illegal characters, this function returns TRUE if value DOES NOT appear to be a valid url (the |<param>-link= parameter is ok); else false when value appears to be a valid url (the |<param>-link= parameter is NOT ok).

]]

local function link_param_ok (value) local scheme, domain; if value:find ('[<>%[%]|{}]') then -- if any prohibited characters return false; end

scheme, domain = split_url (value); -- get scheme or nil and domain or nil from url; return not is_url (scheme, domain); -- return true if value DOES NOT appear to be a valid url end

--[[--------------------------< L I N K _ T I T L E _ O K >---------------------------------------------------

Use link_param_ok() to validate |<param>-link= value and its matching |<title>= value.

|<title>= may be wikilinked but not when |<param>-link= has a value. This function emits an error message when that condition exists

]]

local function link_title_ok (link, lorig, title, torig) local orig;

if is_set (link) then -- don't bother if <param>-link doesn't have a value if not link_param_ok (link) then -- check |<param>-link= markup orig = lorig; -- identify the failing link parameter elseif title:find ('%[%[') then -- check |title= for wikilink markup orig = torig; -- identify the failing |title= parameter end end

if is_set (orig) then table.insert( z.message_tail, { set_error( 'bad_paramlink', orig)}); -- url or wikilink in |title= with |title-link=; end end

--[[--------------------------< C H E C K _ U R L >------------------------------------------------------------

Determines whether a URL string appears to be valid.

First we test for space characters. If any are found, return false. Then split the url into scheme and domain portions, or for protocol relative (//example.com) urls, just the domain. Use is_url() to validate the two portions of the url. If both are valid, or for protocol relative if domain is valid, return true, else false.

Because it is different from a standard url, and because this module used external_link() to make external links that work for standard and news: links, we validate newsgroup names here. The specification for a newsgroup name is at https://tools.ietf.org/html/rfc5536#section-3.1.4

]]

local function check_url( url_str ) if nil == url_str:match ("^%S+$") then -- if there are any spaces in |url=value it can't be a proper url return false; end local scheme, domain;

scheme, domain = split_url (url_str); -- get scheme or nil and domain or nil from url;

if 'news:' == scheme then -- special case for newsgroups return domain:match('^[%a%d%+%-_]+%.[%a%d%+%-_%.]*[%a%d%+%-_]$'); end

return is_url (scheme, domain); -- return true if value appears to be a valid url end

--[=[-------------------------< I S _ P A R A M E T E R _ E X T _ W I K I L I N K >----------------------------

Return true if a parameter value has a string that begins and ends with square brackets [ and ] and the first non-space characters following the opening bracket appear to be a url. The test will also find external wikilinks that use protocol relative urls. Also finds bare urls.

The frontier pattern prevents a match on interwiki links which are similar to scheme:path urls. The tests that find bracketed urls are required because the parameters that call this test (currently |title=, |chapter=, |work=, and |publisher=) may have wikilinks and there are articles or redirects like '//Hus' so, while uncommon, |title=[[1]] is possible as might be en://Hus.

]=]

local function is_parameter_ext_wikilink (value) local scheme, domain;

if value:match ('%f[%[]%[%a%S*:%S+.*%]') then -- if ext wikilink with scheme and domain: [xxxx://yyyyy.zzz] scheme, domain = split_url (value:match ('%f[%[]%[(%a%S*:%S+).*%]')); elseif value:match ('%f[%[]%[2]') then -- if protocol relative ext wikilink: [3] scheme, domain = split_url (value:match ('%f[%[]%[(//%S+).*%]')); elseif value:match ('%a%S*:%S+') then -- if bare url with scheme; may have leading or trailing plain text scheme, domain = split_url (value:match ('(%a%S*:%S+)')); elseif value:match ('//%S+') then -- if protocol relative bare url: //yyyyy.zzz; may have leading or trailing plain text scheme, domain = split_url (value:match ('(//%S+)')); -- what is left should be the domain else return false; -- didn't find anything that is obviously a url end

return is_url (scheme, domain); -- return true if value appears to be a valid url end

--[[-------------------------< C H E C K _ F O R _ U R L >-----------------------------------------------------

loop through a list of parameters and their values. Look at the value and if it has an external link, emit an error message.

]]

local function check_for_url (parameter_list) local error_message = ; for k, v in pairs (parameter_list) do -- for each parameter in the list if is_parameter_ext_wikilink (v) then -- look at the value; if there is a url add an error message 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 .. "|" .. k .. "="; -- add the failed parameter end end if is_set (error_message) then -- done looping, if there is an error message, display it table.insert( z.message_tail, { set_error( 'param_has_ext_link', {error_message}, true ) } ); end end

--[[--------------------------< S A F E _ F O R _ U R L >------------------------------------------------------

Escape sequences for content that will be used for URL descriptions

]]

local function safe_for_url( str ) if str:match( "%[%[.-%]%]" ) ~= nil then table.insert( z.message_tail, { set_error( 'wikilink_in_url', {}, true ) } ); end

return str:gsub( '[%[%]\n]', { ['['] = '[', [']'] = ']', ['\n'] = ' ' } ); end

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

Format an external link with error checking

]]

local function external_link( URL, label, source, access) local error_str = ""; local domain; local path; local base_url;

if not is_set( label ) then label = URL; if is_set( source ) then error_str = set_error( 'bare_url_missing_title', { wrap_style ('parameter', source) }, false, " " ); else error( cfg.messages["bare_url_no_origin"] ); end end if not check_url( URL ) then error_str = set_error( 'bad_url', {wrap_style ('parameter', source)}, false, " " ) .. error_str; end

domain, path = URL:match ('^([/%.%-%+:%a%d]+)([/%?#].*)$'); -- split the url into scheme plus domain and path if path then -- if there is a path portion path = path:gsub ('[%[%]]', {['[']='%5b',[']']='%5d'}); -- replace '[' and ']' with their percent encoded values URL=domain..path; -- and reassemble end

if is_set (access) then -- access level (subscription, registration, limited) local label_head = ; local label_tail; local markup = ; -- can't start a span inside italic markup and end it outside the italic markup

label = safe_for_url (label); -- replace square brackets and newlines (is this necessary? already done above?) if label:match ("(.*)%s+(.+)(?)$") then -- for italicized titles (cite book, etc) label_head, label_tail, markup = label:match ("(.*)%s+(.+)(?)$"); -- split the label at the right-most space; separate the markup elseif label:match ("(.*)%s+(.+)$") then -- for upright titles (journal, news, magazine, etc) label_head, label_tail = label:match ("(.*)%s+(.+)$"); -- split the label at the right-most space; no markup elseif label:match ("(.+)(?)$") then -- single word label with markup label_tail, markup = label:match ("(.+)(?)$"); -- save label text as label tail; separate the markup else label_tail = label; end

base_url = table.concat ( { '[', -- opening css URL, -- the url ' ', -- the required space label_head, -- all but the last word of the label ' ', -- nowrap css for the last word and the signal icon label_tail, -- last (or only) word of the label inside the span '', -- signal spacing css cfg.presentation[access], -- the appropriate icon '', -- close signal spacing and nowrap spans markup, -- insert italic markup if any ']' -- close the plain links span }); else base_url = table.concat({ "[", URL, " ", safe_for_url( label ), "]" }); -- no signal markup end

return table.concat({ base_url, error_str }); end

--[[--------------------------< D E P R E C A T E D _ P A R A M E T E R >--------------------------------------

Categorize and emit an error message when the citation contains one or more deprecated parameters. The function includes the offending parameter name to the error message. Only one error message is emitted regardless of the number of deprecated parameters in the citation.

added_deprecated_cat is a boolean declared in page scope variables above

]]

local function deprecated_parameter(name) if not added_deprecated_cat then added_deprecated_cat = true; -- note that we've added this category table.insert( z.message_tail, { set_error( 'deprecated_params', {name}, true ) } ); -- add error message end end

--[[--------------------------< K E R N _ Q U O T E S >--------------------------------------------------------

Apply kerning to open the space between the quote mark provided by the Module and a leading or trailing quote mark contained in a |title= or |chapter= parameter's value. This function will positive kern either single or double quotes: "'Unkerned title with leading and trailing single quote marks'" " 'Kerned title with leading and trailing single quote marks' " (in real life the kerning isn't as wide as this example) Double single quotes (italic or bold wikimarkup) are not kerned.

Replaces unicode quotemarks with typewriter quote marks regardless of the need for kerning.

Call this function for chapter titles, for website titles, etc; not for book titles.

]]

local function kern_quotes (str) local cap=; local cap2=; -- TODO: move this elswhere so that all title-holding elements get these quote marks replaced? -- str= mw.ustring.gsub (str, '[“”]', '\"'); -- replace “” (U+201C & U+201D) with " (typewriter double quote mark) -- str= mw.ustring.gsub (str, '[‘’]', '\); -- replace ‘’ (U+2018 & U+2019) with ' (typewriter single quote mark)

cap, cap2 = str:match ("^([\"\'])([^\'].+)"); -- match leading double or single quote but not double single quotes if is_set (cap) then str = substitute (cfg.presentation['kern-left'], {cap, cap2}); end

cap, cap2 = str:match ("^(.+[^\'])([\"\'])$") if is_set (cap) then str = substitute (cfg.presentation['kern-right'], {cap, cap2}); end return str; end

--[[--------------------------< F O R M A T _ S C R I P T _ V A L U E >----------------------------------------

|script-title= holds title parameters that are not written in Latin based scripts: Chinese, Japanese, Arabic, Hebrew, etc. These scripts should not be italicized and may be written right-to-left. The value supplied by |script-title= is concatenated onto Title after Title has been wrapped in italic markup.

Regardless of language, all values provided by |script-title= are wrapped in ... tags to isolate rtl languages from the English left to right.

|script-title= provides a unique feature. The value in |script-title= may be prefixed with a two-character ISO639-1 language code and a colon: |script-title=ja:*** *** (where * represents a Japanese character) Spaces between the two-character code and the colon and the colon and the first script character are allowed: |script-title=ja : *** *** |script-title=ja: *** *** |script-title=ja :*** *** Spaces preceding the prefix are allowed: |script-title = ja:*** ***

The prefix is checked for validity. If it is a valid ISO639-1 language code, the lang attribute (lang="ja") is added to the tag so that browsers can know the language the tag contains. This may help the browser render the script more correctly. If the prefix is invalid, the lang attribute is not added. At this time there is no error message for this condition.

Supports |script-title= and |script-chapter=

TODO: error messages when prefix is invalid ISO639-1 code; when script_value has prefix but no script; ]]

local function format_script_value (script_value) local lang=; -- initialize to empty string local name; if script_value:match('^%l%l%s*:') then -- if first 3 non-space characters are script language prefix lang = script_value:match('^(%l%l)%s*:%s*%S.*'); -- get the language prefix or nil if there is no script if not is_set (lang) then return ; -- script_value was just the prefix so return empty string end -- if we get this far we have prefix and script name = mw.language.fetchLanguageName( lang, "en" ); -- get language name so that we can use it to categorize if is_set (name) then -- is prefix a proper ISO 639-1 language code? script_value = script_value:gsub ('^%l%l%s*:%s*', ); -- strip prefix from script -- is prefix one of these language codes? if in_array (lang, cfg.script_lang_codes) then add_prop_cat ('script_with_name', {name, lang}) else add_prop_cat ('script') end lang = ' lang="' .. lang .. '" '; -- convert prefix into a lang attribute else lang = ; -- invalid so set lang to empty string end end script_value = substitute (cfg.presentation['bdi'], {lang, script_value}); -- isolate in case script is rtl

return script_value; end

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

Initially for |title= and |script-title=, this function concatenates those two parameter values after the script value has been wrapped in tags. ]]

local function script_concatenate (title, script) if is_set (script) then script = format_script_value (script); -- tags, lang atribute, categorization, etc; returns empty string on error if is_set (script) then title = title .. ' ' .. script; -- concatenate title and script title end end return title; end

--[[--------------------------< W R A P _ M S G >--------------------------------------------------------------

Applies additional message text to various parameter values. Supplied string is wrapped using a message_list configuration taking one argument. Supports lower case text for templates. Additional text taken from citation_config.messages - the reason this function is similar to but separate from wrap_style().

]]

local function wrap_msg (key, str, lower) if not is_set( str ) then return ""; end if true == lower then local msg; msg = cfg.messages[key]:lower(); -- set the message to lower case before return substitute( msg, str ); -- including template text else return substitute( cfg.messages[key], str ); end end

--[[--------------------------< F O R M A T _ C H A P T E R _ T I T L E >--------------------------------------

Format the four chapter parameters: |script-chapter=, |chapter=, |trans-chapter=, and |chapter-url= into a single Chapter meta- parameter (chapter_url_source used for error messages).

]]

local function format_chapter_title (scriptchapter, chapter, transchapter, chapterurl, chapter_url_source, no_quotes) local chapter_error = ;

if not is_set (chapter) then chapter = ; -- to be safe for concatenation else if false == no_quotes then chapter = kern_quotes (chapter); -- if necessary, separate chapter title's leading and trailing quote marks from Module provided quote marks chapter = wrap_style ('quoted-title', chapter); end end

chapter = script_concatenate (chapter, scriptchapter) -- tags, lang atribute, categorization, etc; must be done after title is wrapped

if is_set (transchapter) then transchapter = wrap_style ('trans-quoted-title', transchapter); if is_set (chapter) then chapter = chapter .. ' ' .. transchapter; else -- here when transchapter without chapter or script-chapter chapter = transchapter; -- chapter_error = ' ' .. set_error ('trans_missing_title', {'chapter'}); end end

if is_set (chapterurl) then chapter = external_link (chapterurl, chapter, chapter_url_source, nil); -- adds bare_url_missing_title error if appropriate end

return chapter .. chapter_error; end

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

This function searches a parameter's value for nonprintable or invisible characters. The search stops at the first match.

This function will detect the visible replacement character when it is part of the wikisource.

Detects but ignores nowiki and math stripmarkers. Also detects other named stripmarkers (gallery, math, pre, ref) and identifies them with a slightly different error message. See also coins_cleanup().

Detects but ignores the character pattern that results from the transclusion of Template:' templates.

Output of this function is an error message that identifies the character or the Unicode group, or the stripmarker that was detected along with its position (or, for multi-byte characters, the position of its first byte) in the parameter value.

]]

local function has_invisible_chars (param, v) local position = ; -- position of invisible char or starting position of stripmarker local dummy; -- end of matching string; not used but required to hold end position when a capture is returned local capture; -- used by stripmarker detection to hold name of the stripmarker local i=1; local stripmarker, apostrophe;

capture = string.match (v, '[%w%p ]*'); -- Test for values that are simple ASCII text and bypass other tests if true if capture == v then -- if same there are no unicode characters return; end

while cfg.invisible_chars[i] do local char=cfg.invisible_chars[i][1] -- the character or group name local pattern=cfg.invisible_chars[i][2] -- the pattern used to find it position, dummy, capture = mw.ustring.find (v, pattern) -- see if the parameter value contains characters that match the pattern

if position then if 'nowiki' == capture or 'math' == capture then -- nowiki, math stripmarker (not an error condition) stripmarker = true; -- set a flag elseif true == stripmarker and 'delete' == char then -- because stripmakers begin and end with the delete char, assume that we've found one end of a stripmarker position = nil; -- unset else local err_msg; if capture then err_msg = capture .. ' ' .. char; else err_msg = char .. ' ' .. 'character'; end

table.insert( z.message_tail, { set_error( 'invisible_char', {err_msg, wrap_style ('parameter', param), position}, true ) } ); -- add error message return; -- and done with this parameter end end i=i+1; -- bump our index end end

--[[--------------------------< A R G U M E N T _ W R A P P E R >----------------------------------------------

Argument wrapper. This function provides support for argument mapping defined in the configuration file so that multiple names can be transparently aliased to single internal variable.

]]

local function argument_wrapper( args ) local origin = {};

return setmetatable({ ORIGIN = function( self, k ) local dummy = self[k]; --force the variable to be loaded. return origin[k]; end }, { __index = function ( tbl, k ) if origin[k] ~= nil then return nil; end

local args, list, v = args, cfg.aliases[k];

if type( list ) == 'table' then v, origin[k] = select_one( args, list, 'redundant_parameters' ); if origin[k] == nil then origin[k] = ; -- Empty string, not nil end elseif list ~= nil then v, origin[k] = args[list], list; else -- maybe let through instead of raising an error? -- v, origin[k] = args[k], k; error( cfg.messages['unknown_argument_map'] ); end

-- Empty strings, not nil; if v == nil then v = cfg.defaults[k] or ; origin[k] = ; end

tbl = rawset( tbl, k, v ); return v; end, }); end

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

Looks for a parameter's name in one of several whitelists.

Parameters in the whitelist can have three values: true - active, supported parameters false - deprecated, supported parameters nil - unsupported parameters

]]

--local function validate( name ) local function validate( name, cite_class ) local name = tostring( name ); local state;

if in_array (cite_class, {'arxiv', 'biorxiv', 'citeseerx'}) then -- limited parameter sets allowed for these templates state = whitelist.limited_basic_arguments[ name ]; if true == state then return true; end -- valid actively supported parameter if false == state then deprecated_parameter (name); -- parameter is deprecated but still supported return true; end

if 'arxiv' == cite_class then -- basic parameters unique to these templates state = whitelist.arxiv_basic_arguments[name]; end if 'biorxiv' == cite_class then state = whitelist.biorxiv_basic_arguments[name]; end if 'citeseerx' == cite_class then state = whitelist.citeseerx_basic_arguments[name]; end

if true == state then return true; end -- valid actively supported parameter if false == state then deprecated_parameter (name); -- parameter is deprecated but still supported return true; end -- limited enumerated parameters list name = name:gsub( "%d+", "#" ); -- replace digit(s) with # (last25 becomes last#) state = whitelist.limited_numbered_arguments[ name ]; if true == state then return true; end -- valid actively supported parameter if false == state then deprecated_parameter (name); -- parameter is deprecated but still supported return true; end

return false; -- not supported because not found or name is set to nil end -- end limited parameter-set templates

state = whitelist.basic_arguments[ name ]; -- all other templates; all normal parameters allowed

if true == state then return true; end -- valid actively supported parameter if false == state then deprecated_parameter (name); -- parameter is deprecated but still supported return true; end -- all enumerated parameters allowed name = name:gsub( "%d+", "#" ); -- replace digit(s) with # (last25 becomes last# state = whitelist.numbered_arguments[ name ];

if true == state then return true; end -- valid actively supported parameter if false == state then deprecated_parameter (name); -- parameter is deprecated but still supported return true; end

return false; -- not supported because not found or name is set to nil end

--[[--------------------------< N O W R A P _ D A T E >--------------------------------------------------------

When date is YYYY-MM-DD format wrap in nowrap span: YYYY-MM-DD. When date is DD MMMM YYYY or is MMMM DD, YYYY then wrap in nowrap span: DD MMMM YYYY or MMMM DD, YYYY

DOES NOT yet support MMMM YYYY or any of the date ranges.

]]

local function nowrap_date (date) local cap=; local cap2=;

if date:match("^%d%d%d%d%-%d%d%-%d%d$") then date = substitute (cfg.presentation['nowrap1'], date);

elseif date:match("^%a+%s*%d%d?,%s+%d%d%d%d$") or date:match ("^%d%d?%s*%a+%s+%d%d%d%d$") then cap, cap2 = string.match (date, "^(.*)%s+(%d%d%d%d)$"); date = substitute (cfg.presentation['nowrap2'], {cap, cap2}); end

return date; end

--[[--------------------------< S E T _ T I T L E T Y P E >----------------------------------------------------

This function sets default title types (equivalent to the citation including |type=<default value>) for those templates that have defaults. Also handles the special case where it is desirable to omit the title type from the rendered citation (|type=none).

]]

local function set_titletype (cite_class, title_type) if is_set(title_type) then if "none" == title_type then title_type = ""; -- if |type=none then type parameter not displayed end return title_type; -- if |type= has been set to any other value use that value end

return cfg.title_types [cite_class] or ; -- set template's default title type; else empty string for concatenation end

--[[--------------------------< H Y P H E N _ T O _ D A S H >--------------------------------------------------

Converts a hyphen to a dash

]]

local function hyphen_to_dash( str ) if not is_set(str) or str:match( "[%[%]{}<>]" ) ~= nil then return str; end return str:gsub( '-', '–' ); end

--[[--------------------------< S A F E _ J O I N >------------------------------------------------------------

Joins a sequence of strings together while checking for duplicate separation characters.

]]

local function safe_join( tbl, duplicate_char ) --[[ Note: we use string functions here, rather than ustring functions.

This has considerably faster performance and should work correctly as long as the duplicate_char is strict ASCII. The strings in tbl may be ASCII or UTF8. ]]

local str = ; -- the output string local comp = ; -- what does 'comp' mean? local end_chr = ; local trim; for _, value in ipairs( tbl ) do if value == nil then value = ; end

if str == then -- if output string is empty str = value; -- assign value to it (first time through the loop) elseif value ~= then if value:sub(1,1) == '<' then -- Special case of values enclosed in spans and other markup. comp = value:gsub( "%b<>", "" ); -- remove html markup (string -> string) else comp = value; end -- typically duplicate_char is sepc if comp:sub(1,1) == duplicate_char then -- is first charactier same as duplicate_char? why test first character? -- Because individual string segments often (always?) begin with terminal punct for th -- preceding segment: 'First element' .. 'sepc next element' .. etc? trim = false; end_chr = str:sub(-1,-1); -- get the last character of the output string -- str = str .. "<HERE(enchr=" .. end_chr.. ")" -- debug stuff? if end_chr == duplicate_char then -- if same as separator str = str:sub(1,-2); -- remove it elseif end_chr == "'" then -- if it might be wikimarkup if str:sub(-3,-1) == duplicate_char .. "" then -- if last three chars of str are sepc str = str:sub(1, -4) .. ""; -- remove them and add back elseif str:sub(-5,-1) == duplicate_char .. "]]" then -- if last five chars of str are sepc]] trim = true; -- why? why do this and next differently from previous? elseif str:sub(-4,-1) == duplicate_char .. "]" then -- if last four chars of str are sepc] trim = true; -- same question end elseif end_chr == "]" then -- if it might be wikimarkup if str:sub(-3,-1) == duplicate_char .. "]]" then -- if last three chars of str are sepc]] wikilink trim = true; elseif str:sub(-2,-1) == duplicate_char .. "]" then -- if last two chars of str are sepc] external link trim = true; elseif str:sub(-4,-1) == duplicate_char .. "]" then -- normal case when |url=something & |title=Title. trim = true; end elseif end_chr == " " then -- if last char of output string is a space if str:sub(-2,-1) == duplicate_char .. " " then -- if last two chars of str are <sepc><space> str = str:sub(1,-3); -- remove them both end end

if trim then if value ~= comp then -- value does not equal comp when value contains html markup local dup2 = duplicate_char; if dup2:match( "%A" ) then dup2 = "%" .. dup2; end -- if duplicate_char not a letter then escape it

value = value:gsub( "(%b<>)" .. dup2, "%1", 1 ) -- remove duplicate_char if it follows html markup else value = value:sub( 2, -1 ); -- remove duplicate_char when it is first character end end end str = str .. value; --add it to the output string end end return str; end

--[[--------------------------< I S _ S U F F I X >------------------------------------------------------------

returns true is suffix is properly formed Jr, Sr, or ordinal in the range 2–9. Puncutation not allowed.

]]

local function is_suffix (suffix) if in_array (suffix, {'Jr', 'Sr', '2nd', '3rd'}) or suffix:match ('^%dth$') then return true; end return false; end

--[[--------------------------< I S _ G O O D _ V A N C _ N A M E >--------------------------------------------

For Vancouver Style, author/editor names are supposed to be rendered in Latin (read ASCII) characters. When a name uses characters that contain diacritical marks, those characters are to converted to the corresponding Latin character. When a name is written using a non-Latin alphabet or logogram, that name is to be transliterated into Latin characters. These things are not currently possible in this module so are left to the editor to do.

This test allows |first= and |last= names to contain any of the letters defined in the four Unicode Latin character sets C0 Controls and Basic Latin 0041–005A, 0061–007A C1 Controls and Latin-1 Supplement 00C0–00D6, 00D8–00F6, 00F8–00FF Latin Extended-A 0100–017F Latin Extended-B 0180–01BF, 01C4–024F

|lastn= also allowed to contain hyphens, spaces, and apostrophes. (http://www.ncbi.nlm.nih.gov/books/NBK7271/box/A35029/) |firstn= also allowed to contain hyphens, spaces, apostrophes, and periods

This original test: if nil == mw.ustring.find (last, "^[A-Za-zÀ-ÖØ-öø-ƿDŽ-ɏ%-%s%']*$") or nil == mw.ustring.find (first, "^[A-Za-zÀ-ÖØ-öø-ƿDŽ-ɏ%-%s%'%.]+[2-6%a]*$") then was written ouside of the code editor and pasted here because the code editor gets confused between character insertion point and cursor position. The test has been rewritten to use decimal character escape sequence for the individual bytes of the unicode characters so that it is not necessary to use an external editor to maintain this code.

\195\128-\195\150 – À-Ö (U+00C0–U+00D6 – C0 controls) \195\152-\195\182 – Ø-ö (U+00D8-U+00F6 – C0 controls) \195\184-\198\191 – ø-ƿ (U+00F8-U+01BF – C0 controls, Latin extended A & B) \199\132-\201\143 – DŽ-ɏ (U+01C4-U+024F – Latin extended B)

]]

local function is_good_vanc_name (last, first) local first, suffix = first:match ('(.-),?%s*([%dJS][%drndth]+)%.?$') or first; -- if first has something that looks like a generational suffix, get it

if is_set (suffix) then if not is_suffix (suffix) then add_vanc_error ('suffix'); return false; -- not a name with an appropriate suffix end end if nil == mw.ustring.find (last, "^[A-Za-z\195\128-\195\150\195\152-\195\182\195\184-\198\191\199\132-\201\143%-%s%']*$") or nil == mw.ustring.find (first, "^[A-Za-z\195\128-\195\150\195\152-\195\182\195\184-\198\191\199\132-\201\143%-%s%'%.]*$") then add_vanc_error ('non-Latin character'); return false; -- not a string of latin characters; Vancouver requires Romanization end; return true; end

--[[--------------------------< R E D U C E _ T O _ I N I T I A L S >------------------------------------------

Attempts to convert names to initials in support of |name-list-format=vanc.

Names in |firstn= may be separated by spaces or hyphens, or for initials, a period. See http://www.ncbi.nlm.nih.gov/books/NBK7271/box/A35062/.

Vancouver style requires family rank designations (Jr, II, III, etc) to be rendered as Jr, 2nd, 3rd, etc. See http://www.ncbi.nlm.nih.gov/books/NBK7271/box/A35085/. This code only accepts and understands generational suffix in the Vancouver format because Roman numerals look like, and can be mistaken for, initials.

This function uses ustring functions because firstname initials may be any of the unicode Latin characters accepted by is_good_vanc_name ().

]]

local function reduce_to_initials(first) local name, suffix = mw.ustring.match(first, "^(%u+) ([%dJS][%drndth]+)$");

if not name then -- if not initials and a suffix name = mw.ustring.match(first, "^(%u+)$"); -- is it just intials? end

if name then -- if first is initials with or without suffix if 3 > mw.ustring.len (name) then -- if one or two initials if suffix then -- if there is a suffix if is_suffix (suffix) then -- is it legitimate? return first; -- one or two initials and a valid suffix so nothing to do else add_vanc_error ('suffix'); -- one or two initials with invalid suffix so error message return first; -- and return first unmolested end else return first; -- one or two initials without suffix; nothing to do end end end -- if here then name has 3 or more uppercase letters so treat them as a word

local initials, names = {}, {}; -- tables to hold name parts and initials local i = 1; -- counter for number of initials

names = mw.text.split (first, '[%s,]+'); -- split into a table of names and possible suffix

while names[i] do -- loop through the table if 1 < i and names[i]:match ('[%dJS][%drndth]+%.?$') then -- if not the first name, and looks like a suffix (may have trailing dot) names[i] = names[i]:gsub ('%.', ); -- remove terminal dot if present if is_suffix (names[i]) then -- if a legitimate suffix table.insert (initials, ' ' .. names[i]); -- add a separator space, insert at end of initials table break; -- and done because suffix must fall at the end of a name end -- no error message if not a suffix; possibly because of Romanization end if 3 > i then table.insert (initials, mw.ustring.sub(names[i],1,1)); -- insert the intial at end of initials table end i = i+1; -- bump the counter end

return table.concat(initials) -- Vancouver format does not include spaces. end

--[[--------------------------< L I S T _ P E O P L E >-------------------------------------------------------

Formats a list of people (e.g. authors / editors)

]]

local function list_people(control, people, etal) local sep; local namesep; local format = control.format local maximum = control.maximum local lastauthoramp = control.lastauthoramp; local text = {}

if 'vanc' == format then -- Vancouver-like author/editor name styling? sep = ','; -- name-list separator between authors is a comma namesep = ' '; -- last/first separator is a space elseif 'mla' == control.mode then sep = ','; -- name-list separator between authors is a comma namesep = ', ' -- last/first separator is <comma><space> else sep = ';' -- name-list separator between authors is a semicolon namesep = ', ' -- last/first separator is <comma><space> end

if sep:sub(-1,-1) ~= " " then sep = sep .. " " end if is_set (maximum) and maximum < 1 then return "", 0; end -- returned 0 is for EditorCount; not used for authors

for i,person in ipairs(people) do if is_set(person.last) then local mask = person.mask local one local sep_one = sep; if is_set (maximum) and i > maximum then etal = true; break; elseif (mask ~= nil) then local n = tonumber(mask) if (n ~= nil) then one = string.rep("—",n) else one = mask; sep_one = " "; end else one = person.last local first = person.first if is_set(first) then if 'mla' == control.mode then if i == 1 then -- for mla one = one .. namesep .. first; -- first name last, first else -- all other names one = first .. ' ' .. one; -- first last end else if ( "vanc" == format ) then -- if vancouver format one = one:gsub ('%.', ); -- remove periods from surnames (http://www.ncbi.nlm.nih.gov/books/NBK7271/box/A35029/) if not person.corporate and is_good_vanc_name (one, first) then -- and name is all Latin characters; corporate authors not tested first = reduce_to_initials(first) -- attempt to convert first name(s) to initials end end one = one .. namesep .. first; end end if is_set(person.link) and person.link ~= control.page_name then one = "" .. one .. "" -- link author/editor if this page is not the author's/editor's page end end table.insert( text, one ) table.insert( text, sep_one ) end end

local count = #text / 2; -- (number of names + number of separators) divided by 2 if count > 0 then if count > 1 and is_set(lastauthoramp) and not etal then if 'mla' == control.mode then text[#text-2] = ", and "; -- replace last separator with ', and ' text else text[#text-2] = " & "; -- replace last separator with ampersand text end end text[#text] = nil; -- erase the last separator end

local result = table.concat(text) -- construct list if etal and is_set (result) then -- etal may be set by |display-authors=etal but we might not have a last-first list result = result .. sep .. ' ' .. cfg.messages['et al']; -- we've go a last-first list and etal so add et al. end

return result, count end

--[[--------------------------< A N C H O R _ I D >------------------------------------------------------------

Generates a CITEREF anchor ID if we have at least one name or a date. Otherwise returns an empty string.

namelist is one of the contributor-, author-, or editor-name lists chosen in that order. year is Year or anchor_year.

]] local function anchor_id (namelist, year) local names={}; -- a table for the one to four names and year for i,v in ipairs (namelist) do -- loop through the list and take up to the first four last names names[i] = v.last if i == 4 then break end -- if four then done end table.insert (names, year); -- add the year at the end local id = table.concat(names); -- concatenate names and year for CITEREF id if is_set (id) then -- if concatenation is not an empty string return "CITEREF" .. id; -- add the CITEREF portion else return ; -- return an empty string; no reason to include CITEREF id in this citation end end

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

Evaluates the content of author and editor name parameters for variations on the theme of et al. If found, the et al. is removed, a flag is set to true and the function returns the modified name and the flag.

This function never sets the flag to false but returns it's previous state because it may have been set by previous passes through this function or by the parameters |display-authors=etal or |display-editors=etal

]]

local function name_has_etal (name, etal, nocat)

if is_set (name) then -- name can be nil in which case just return local etal_pattern = "[;,]? *[\"']*%f[%a][Ee][Tt] *[Aa][Ll][%.\"']*$" -- variations on the 'et al' theme local others_pattern = "[;,]? *%f[%a]and [Oo]thers"; -- and alternate to et al.

if name:match (etal_pattern) then -- variants on et al. name = name:gsub (etal_pattern, ); -- if found, remove etal = true; -- set flag (may have been set previously here or by |display-authors=etal) if not nocat then -- no categorization for |vauthors= add_maint_cat ('etal'); -- and add a category if not already added end elseif name:match (others_pattern) then -- if not 'et al.', then 'and others'? name = name:gsub (others_pattern, ); -- if found, remove etal = true; -- set flag (may have been set previously here or by |display-authors=etal) if not nocat then -- no categorization for |vauthors= add_maint_cat ('etal'); -- and add a category if not already added end end end return name, etal; -- end

--[[--------------------------< N A M E _ H A S _ E D _ M A R K U P >------------------------------------------

Evaluates the content of author and editor parameters for extranious editor annotations: ed, ed., eds, (Ed.), etc. These annotation do not belong in author parameters and are redundant in editor parameters. If found, the function adds the editor markup maintenance category.

]]

local function name_has_ed_markup (name, list_name) local _, pattern; local patterns = { -- these patterns match annotations at end of name '%f[%(%[][%(%[]%s*[Ee][Dd][Ss]?%.?%s*[%)%]]?$', -- (ed) or (eds): leading '(', case insensitive 'ed', optional 's', '.' and/or ')' '[,%.%s]%f[e]eds?%.?$', -- ed or eds: without '('or ')'; case sensitive (ED could be initials Ed could be name) '%f[%(%[][%(%[]%s*[Ee][Dd][Ii][Tt][Oo][Rr][Ss]?%.?%s*[%)%]]?$', -- (editor) or (editors): leading '(', case insensitive, optional '.' and/or ')' '[,%.%s]%f[Ee][Ee][Dd][Ii][Tt][Oo][Rr][Ss]?%.?$', -- editor or editors: without '('or ')'; case insensitive

-- these patterns match annotations at beginning of name '^eds?[%.,;]', -- ed. or eds.: lower case only, optional 's', requires '.' '^[%(%[]%s*[Ee][Dd][Ss]?%.?%s*[%)%]]', -- (ed) or (eds): also sqare brackets, case insensitive, optional 's', '.' '^[%(%[]?%s*[Ee][Dd][Ii][Tt][Oo][Rr][Ss]?%A', -- (editor or (editors: also sq brackets, case insensitive, optional brackets, 's' '^[%(%[]?%s*[Ee][Dd][Ii][Tt][Ee][Dd]%A', -- (edited: also sq brackets, case insensitive, optional brackets

}

if is_set (name) then for _, pattern in ipairs (patterns) do -- spin through patterns table and if name:match (pattern) then add_maint_cat ('extra_text_names', cfg.special_case_translation [list_name]); -- add a maint cat for this template break; end end end return name; -- and done end

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

Evaluates the content of author and editor (surnames only) parameters for multiple names. Multiple names are indicated if there is more than one comma and or semicolon. If found, the function adds the multiple name (author or editor) maintenance category.

]]

local function name_has_mult_names (name, list_name) local count, _; if is_set (name) then _, count = name:gsub ('[;,]', ); -- count the number of separator-like characters

if 1 < count then -- param could be |author= or |editor= so one separator character is acceptable add_maint_cat ('mult_names', cfg.special_case_translation [list_name]); -- more than one separator indicates multiple names so add a maint cat for this template end end return name; -- and done end

--[[--------------------------< N A M E _ C H E C K S >-------------------------------------------------------- This function calls various name checking functions used to validate the content of the various name-holding parameters.

]]

local function name_checks (last, first, list_name) if is_set (last) then if last:match ('^%(%(.*%)%)$') then -- if wrapped in doubled parentheses, accept as written last = last:match ('^%(%((.*)%)%)$'); -- strip parens else last = name_has_mult_names (last, list_name); -- check for multiple names in the parameter (last only) last = name_has_ed_markup (last, list_name); -- check for extraneous 'editor' annotation end end if is_set (first) then if first:match ('^%(%(.*%)%)$') then -- if wrapped in doubled parentheses, accept as written first = first:match ('^%(%((.*)%)%)$'); -- strip parens else first = name_has_ed_markup (first, list_name); -- check for extraneous 'editor' annotation end end return last, first; -- done end

--[[--------------------------< E X T R A C T _ N A M E S >---------------------------------------------------- Gets name list from the input arguments

Searches through args in sequential order to find |lastn= and |firstn= parameters (or their aliases), and their matching link and mask parameters. Stops searching when both |lastn= and |firstn= are not found in args after two sequential attempts: found |last1=, |last2=, and |last3= but doesn't find |last4= and |last5= then the search is done.

This function emits an error message when there is a |firstn= without a matching |lastn=. When there are 'holes' in the list of last names, |last1= and |last3= are present but |last2= is missing, an error message is emitted. |lastn= is not required to have a matching |firstn=.

When an author or editor parameter contains some form of 'et al.', the 'et al.' is stripped from the parameter and a flag (etal) returned that will cause list_people() to add the static 'et al.' text from Module:Citation/CS1/Configuration. This keeps 'et al.' out of the template's metadata. When this occurs, the page is added to a maintenance category.

]]

local function extract_names(args, list_name) local names = {}; -- table of names local last; -- individual name components local first; local link; local mask; local i = 1; -- loop counter/indexer local n = 1; -- output table indexer local count = 0; -- used to count the number of times we haven't found a |last= (or alias for authors, |editor-last or alias for editors) local etal=false; -- return value set to true when we find some form of et al. in an author parameter

local err_msg_list_name = list_name:match ("(%w+)List") .. 's list'; -- modify AuthorList or EditorList for use in error messages if necessary while true do last = select_one( args, cfg.aliases[list_name .. '-Last'], 'redundant_parameters', i ); -- search through args for name components beginning at 1 first = select_one( args, cfg.aliases[list_name .. '-First'], 'redundant_parameters', i ); link = select_one( args, cfg.aliases[list_name .. '-Link'], 'redundant_parameters', i ); mask = select_one( args, cfg.aliases[list_name .. '-Mask'], 'redundant_parameters', i );

last, etal = name_has_etal (last, etal, false); -- find and remove variations on et al. first, etal = name_has_etal (first, etal, false); -- find and remove variations on et al. last, first= name_checks (last, first, list_name); -- multiple names, extraneous annotation, etc checks

if first and not last then -- if there is a firstn without a matching lastn table.insert( z.message_tail, { set_error( 'first_missing_last', {err_msg_list_name, i}, true ) } ); -- add this error message elseif not first and not last then -- if both firstn and lastn aren't found, are we done? count = count + 1; -- number of times we haven't found last and first if 2 <= count then -- two missing names and we give up break; -- normal exit or there is a two-name hole in the list; can't tell which end else -- we have last with or without a first link_title_ok (link, list_name:match ("(%w+)List"):lower() .. '-link' .. i, last, list_name:match ("(%w+)List"):lower() .. '-last' .. i); -- check for improper wikimarkup

names[n] = {last = last, first = first, link = link, mask = mask, corporate=false}; -- add this name to our names list (corporate for |vauthors= only) n = n + 1; -- point to next location in the names table if 1 == count then -- if the previous name was missing table.insert( z.message_tail, { set_error( 'missing_name', {err_msg_list_name, i-1}, true ) } ); -- add this error message end count = 0; -- reset the counter, we're looking for two consecutive missing names end i = i + 1; -- point to next args location end

return names, etal; -- all done, return our list of names end

--[[--------------------------< G E T _ I S O 6 3 9 _ C O D E >------------------------------------------------

Validates language names provided in |language= parameter if not an ISO639-1 or 639-2 code.

Returns the language name and associated two- or three-character code. Because case of the source may be incorrect or different from the case that WikiMedia uses, the name comparisons are done in lower case and when a match is found, the Wikimedia version (assumed to be correct) is returned along with the code. When there is no match, we return the original language name string.

mw.language.fetchLanguageNames(<local wiki language>, 'all') return a list of languages that in some cases may include extensions. For example, code 'cbk-zam' and its associated name 'Chavacano de Zamboanga' (MediaWiki does not support code 'cbk' or name 'Chavacano'.

Names but that are included in the list will be found if that name is provided in the |language= parameter. For example, if |language=Chavacano de Zamboanga, that name will be found with the associated code 'cbk-zam'. When names are found and the associated code is not two or three characters, this function returns only the Wikimedia language name.

Adapted from code taken from Module:Check ISO 639-1.

]]

local function get_iso639_code (lang, this_wiki_code) if 'bangla' == lang:lower() then -- special case related to Wikimedia remap of code 'bn' at mw:Extension:CLDR return 'Bengali', 'bn'; -- make sure rendered version is properly capitalized end

local languages = mw.language.fetchLanguageNames(this_wiki_code, 'all') -- get a list of language names known to Wikimedia -- ('all' is required for North Ndebele, South Ndebele, and Ojibwa) local langlc = mw.ustring.lower(lang); -- lower case version for comparisons

for code, name in pairs(languages) do -- scan the list to see if we can find our language if langlc == mw.ustring.lower(name) then if 2 ~= code:len() and 3 ~= code:len() then -- two- or three-character codes only; extensions not supported return name; -- so return the name but not the code end return name, code; -- found it, return name to ensure proper capitalization and the the code end end return lang; -- not valid language; return language in original case and nil for the code end

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

Gets language name from a provided two- or three-character ISO 639 code. If a code is recognized by MediaWiki, use the returned name; if not, then use the value that was provided with the language parameter.

When |language= contains a recognized language (either code or name), the page is assigned to the category for that code: Category:Norwegian-language sources (no). For valid three-character code languages, the page is assigned to the single category for '639-2' codes: Category:CS1 ISO 639-2 language sources.

Languages that are the same as the local wiki are not categorized. MediaWiki does not recognize three-character equivalents of two-character codes: code 'ar' is recognized bit code 'ara' is not.

This function supports multiple languages in the form |language=nb, French, th where the language names or codes are separated from each other by commas.

]]

local function language_parameter (lang) local code; -- the two- or three-character language code local name; -- the language name local language_list = {}; -- table of language names to be rendered local names_table = {}; -- table made from the value assigned to |language=

local this_wiki = mw.getContentLanguage(); -- get a language object for this wiki local this_wiki_code = this_wiki:getCode() -- get this wiki's language code local this_wiki_name = mw.language.fetchLanguageName(this_wiki_code, this_wiki_code); -- get this wiki's language name

names_table = mw.text.split (lang, '%s*,%s*'); -- names should be a comma separated list

for _, lang in ipairs (names_table) do -- reuse lang

if lang:match ('^%a%a%-') then -- strip ietf language tags from code; TODO: is there a need to support 3-char with tag? lang = lang:match ('(%a%a)%-') -- keep only 639-1 code portion to lang; TODO: do something with 3166 alpha 2 country code? end if 2 == lang:len() or 3 == lang:len() then -- if two-or three-character code name = mw.language.fetchLanguageName( lang:lower(), this_wiki_code); -- get language name if |language= is a proper code end

if is_set (name) then -- if |language= specified a valid code code = lang:lower(); -- save it else name, code = get_iso639_code (lang, this_wiki_code); -- attempt to get code from name (assign name here so that we are sure of proper capitalization) end

if is_set (code) then -- only 2- or 3-character codes if 'bn' == code then name = 'Bengali' end; -- override wikimedia when code is 'bn' if this_wiki_code ~= code then -- when the language is not the same as this wiki's language if 2 == code:len() then -- and is a two-character code add_prop_cat ('foreign_lang_source', {name, code}) -- categorize it else -- or is a recognized language (but has a three-character code) add_prop_cat ('foreign_lang_source_2', {code}) -- categorize it differently TODO: support mutliple three-character code categories per cs1|2 template end end else add_maint_cat ('unknown_lang'); -- add maint category if not already added end

table.insert (language_list, name); name = ; -- so we can reuse it end

code = #language_list -- reuse code as number of languages in the list if 2 >= code then name = table.concat (language_list, ' and ') -- insert '<space>and<space>' between two language names elseif 2 < code then language_list[code] = 'and ' .. language_list[code]; -- prepend last name with 'and<space>' name = table.concat (language_list, ', ') -- and concatenate with '<comma><space>' separators end if this_wiki_name == name then return ; -- if one language and that language is this wiki's return an empty string (no annotation) end return (" " .. wrap_msg ('language', name)); -- otherwise wrap with '(in ...)' --[[ TODO: should only return blank or name rather than full list so we can clean up the bunched parenthetical elements Language, Type, Format ]] end

--[[--------------------------< S E T _ C S 1 _ S T Y L E >----------------------------------------------------

Set style settings for CS1 citation templates. Returns separator and postscript settings

]]

local function set_cs1_style (ps) if not is_set (ps) then -- unless explicitely set to something ps = '.'; -- terminate the rendered citation with a period end return '.', ps; -- separator is a full stop end

--[[--------------------------< S E T _ C S 2 _ S T Y L E >----------------------------------------------------

Set style settings for CS2 citation templates. Returns separator, postscript, ref settings

]]

local function set_cs2_style (ps, ref) if not is_set (ps) then -- if |postscript= has not been set, set cs2 default ps = ; -- make sure it isn't nil end if not is_set (ref) then -- if |ref= is not set ref = "harv"; -- set default |ref=harv end return ',', ps, ref; -- separator is a comma end

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

When |mode= is not set or when its value is invalid, use config.CitationClass and parameter values to establish rendered style.

]]

local function get_settings_from_cite_class (ps, ref, cite_class) local sep; if (cite_class == "citation") then -- for citation templates (CS2) sep, ps, ref = set_cs2_style (ps, ref); else -- not a citation template so CS1 sep, ps = set_cs1_style (ps); end

return sep, ps, ref -- return them all end

--[[--------------------------< S E T _ S T Y L E >------------------------------------------------------------

Establish basic style settings to be used when rendering the citation. Uses |mode= if set and valid or uses config.CitationClass from the template's #invoke: to establish style.

]]

local function set_style (mode, ps, ref, cite_class) local sep; if 'cs2' == mode then -- if this template is to be rendered in CS2 (citation) style sep, ps, ref = set_cs2_style (ps, ref); elseif 'cs1' == mode then -- if this template is to be rendered in CS1 (cite xxx) style sep, ps = set_cs1_style (ps); elseif 'mla' == mode then -- if this template is to be rendered in mla style use cs1 for bot cs1 & cs2 templates sep, ps = set_cs1_style (ps); else -- anything but cs1 or cs2 sep, ps, ref = get_settings_from_cite_class (ps, ref, cite_class); -- get settings based on the template's CitationClass end if 'none' == ps:lower() then -- if assigned value is 'none' then ps = ; -- set to empty string end

return sep, ps, ref end

--[=[-------------------------< I S _ P D F >------------------------------------------------------------------

Determines if a url has the file extension that is one of the pdf file extensions used by MediaWiki:Common.css when applying the pdf icon to external links.

returns true if file extension is one of the recognized extension, else false

]=]

local function is_pdf (url) return url:match ('%.pdf[%?#]?') or url:match ('%.PDF[%?#]?'); end

--[[--------------------------< S T Y L E _ F O R M A T >------------------------------------------------------

Applies css style to |format=, |chapter-format=, etc. Also emits an error message if the format parameter does not have a matching url parameter. If the format parameter is not set and the url contains a file extension that is recognized as a pdf document by MediaWiki's commons.css, this code will set the format parameter to (PDF) with the appropriate styling.

]]

local function style_format (format, url, fmt_param, url_param) if is_set (format) then format = wrap_style ('format', format); -- add leading space, parenthases, resize if not is_set (url) then format = format .. set_error( 'format_missing_url', {fmt_param, url_param} ); -- add an error message end elseif is_pdf (url) then -- format is not set so if url is a pdf file then format = wrap_style ('format', 'PDF'); -- set format to pdf else format = ; -- empty string for concatenation end return format; end

--[[--------------------------< G E T _ D I S P L A Y _ A U T H O R S _ E D I T O R S >------------------------

Returns a number that defines the number of names displayed for author and editor name lists and a boolean flag to indicate when et al. should be appended to the name list.

When the value assigned to |display-xxxxors= is a number greater than or equal to zero, return the number and the previous state of the 'etal' flag (false by default but may have been set to true if the name list contains some variant of the text 'et al.').

When the value assigned to |display-xxxxors= is the keyword 'etal', return a number that is one greater than the number of authors in the list and set the 'etal' flag true. This will cause the list_people() to display all of the names in the name list followed by 'et al.'

In all other cases, returns nil and the previous state of the 'etal' flag.

inputs: max: A['DisplayAuthors'] or A['DisplayEditors']; a number or some flavor of etal count: #a or #e list_name: 'authors' or 'editors' etal: author_etal or editor_etal

]]

local function get_display_authors_editors (max, count, list_name, etal) if is_set (max) then if 'etal' == max:lower():gsub("[ '%.]", ) then -- the :gsub() portion makes 'etal' from a variety of 'et al.' spellings and stylings max = count + 1; -- number of authors + 1 so display all author name plus et al. etal = true; -- overrides value set by extract_names() elseif max:match ('^%d+$') then -- if is a string of numbers max = tonumber (max); -- make it a number if max >= count then -- if |display-xxxxors= value greater than or equal to number of authors/editors add_maint_cat ('disp_auth_ed', cfg.special_case_translation [list_name]); end else -- not a valid keyword or number table.insert( z.message_tail, { set_error( 'invalid_param_val', {'display-' .. list_name, max}, true ) } ); -- add error message max = nil; -- unset; as if |display-xxxxors= had not been set end end

return max, etal; end

--[[--------------------------< E X T R A _ T E X T _ I N _ P A G E _ C H E C K >------------------------------

Adds page to Category:CS1 maint: extra text if |page= or |pages= has what appears to be some form of p. or pp. abbreviation in the first characters of the parameter content.

check Page and Pages for extraneous p, p., pp, and pp. at start of parameter value: good pattern: '^P[^%.P%l]' matches when |page(s)= begins PX or P# but not Px where x and X are letters and # is a dgiit bad pattern: '^[Pp][Pp]' matches matches when |page(s)= begins pp or pP or Pp or PP

]]

local function extra_text_in_page_check (page) -- local good_pattern = '^P[^%.P%l]'; local good_pattern = '^P[^%.Pp]'; -- ok to begin with uppercase P: P7 (pg 7 of section P) but not p123 (page 123) TODO: add Gg for PG or Pg? -- local bad_pattern = '^[Pp][Pp]'; local bad_pattern = '^[Pp]?[Pp]%.?[ %d]';

if not page:match (good_pattern) and (page:match (bad_pattern) or page:match ('^[Pp]ages?')) then add_maint_cat ('extra_text'); end -- if Page:match ('^[Pp]?[Pp]%.?[ %d]') or Page:match ('^[Pp]ages?[ %d]') or -- Pages:match ('^[Pp]?[Pp]%.?[ %d]') or Pages:match ('^[Pp]ages?[ %d]') then -- add_maint_cat ('extra_text'); -- end end

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

split apart a |vautthors= or |veditors= parameter. This function allows for corporate names, wrapped in doubled parentheses to also have commas; in the old version of the code, the doubled parnetheses were included in the rendered citation and in the metadata.

|vauthors=Jones AB, White EB, ((Black, Brown, and Co.))

This code is experimental and may not be retained.

]] local function get_v_name_table (vparam, output_table) local name_table = mw.text.split(vparam, "%s*,%s*"); -- names are separated by commas

local i = 1;

while name_table[i] do if name_table[i]:match ('^%(%(.*[^%)][^%)]$') then -- first segment of corporate with one or more commas; this segment has the opening doubled parens local name = name_table[i]; i=i+1; -- bump indexer to next segment while name_table[i] do name = name .. ', ' .. name_table[i]; -- concatenate with previous segments if name_table[i]:match ('^.*%)%)$') then -- if this table member has the closing doubled parens break; -- and done reassembling so end i=i+1; -- bump indexer end table.insert (output_table, name); -- and add corporate name to the output table else table.insert (output_table, name_table[i]); -- add this name end i = i+1; end return output_table; end

--[[--------------------------< P A R S E _ V A U T H O R S _ V E D I T O R S >--------------------------------

This function extracts author / editor names from |vauthors= or |veditors= and finds matching |xxxxor-maskn= and |xxxxor-linkn= in args. It then returns a table of assembled names just as extract_names() does.

Author / editor names in |vauthors= or |veditors= must be in Vancouver system style. Corporate or institutional names may sometimes be required and because such names will often fail the is_good_vanc_name() and other format compliance tests, are wrapped in doubled paranethese ((corporate name)) to suppress the format tests.

Supports generational suffixes Jr, 2nd, 3rd, 4th–6th.

This function sets the vancouver error when a reqired comma is missing and when there is a space between an author's initials.

]]

local function parse_vauthors_veditors (args, vparam, list_name) local names = {}; -- table of names assembled from |vauthors=, |author-maskn=, |author-linkn= local v_name_table = {}; local etal = false; -- return value set to true when we find some form of et al. vauthors parameter local last, first, link, mask, suffix; local corporate = false;

vparam, etal = name_has_etal (vparam, etal, true); -- find and remove variations on et al. do not categorize (do it here because et al. might have a period) if vparam:find ('%[%[') or vparam:find ('%]%]') then -- no wikilinking vauthors names add_vanc_error ('wikilink'); end v_name_table = get_v_name_table (vparam, v_name_table); -- names are separated by commas

for i, v_name in ipairs(v_name_table) do if v_name:match ('^%(%(.+%)%)$') then -- corporate authors are wrapped in doubled parentheses to supress vanc formatting and error detection first = ; -- set to empty string for concatenation and because it may have been set for previous author/editor last = v_name:match ('^%(%((.+)%)%)$') -- remove doubled parntheses corporate = true; -- flag used in list_people() elseif string.find(v_name, "%s") then if v_name:find('[;%.]') then -- look for commonly occurring punctuation characters; add_vanc_error ('punctuation'); end local lastfirstTable = {} lastfirstTable = mw.text.split(v_name, "%s") first = table.remove(lastfirstTable); -- removes and returns value of last element in table which should be author intials if is_suffix (first) then -- if a valid suffix suffix = first -- save it as a suffix and first = table.remove(lastfirstTable); -- get what should be the initials from the table end -- no suffix error message here because letter combination may be result of Romanization; check for digits? last = table.concat(lastfirstTable, " ") -- returns a string that is the concatenation of all other names that are not initials if mw.ustring.match (last, '%a+%s+%u+%s+%a+') then add_vanc_error ('missing comma'); -- matches last II last; the case when a comma is missing end if mw.ustring.match (v_name, ' %u %u$') then -- this test is in the wrong place TODO: move or replace with a more appropriate test add_vanc_error ('name'); -- matches a space between two intiials end else first = ; -- set to empty string for concatenation and because it may have been set for previous author/editor last = v_name; -- last name or single corporate name? Doesn't support multiword corporate names? do we need this? end

if is_set (first) then if not mw.ustring.match (first, "^%u?%u$") then -- first shall contain one or two upper-case letters, nothing else add_vanc_error ('initials'); -- too many initials; mixed case initials (which may be ok Romanization); hyphenated initials end is_good_vanc_name (last, first); -- check first and last before restoring the suffix which may have a non-Latin digit if is_set (suffix) then first = first .. ' ' .. suffix; -- if there was a suffix concatenate with the initials suffix = ; -- unset so we don't add this suffix to all subsequent names end else if not corporate then is_good_vanc_name (last, ); end end -- this from extract_names () link = select_one( args, cfg.aliases[list_name .. '-Link'], 'redundant_parameters', i ); mask = select_one( args, cfg.aliases[list_name .. '-Mask'], 'redundant_parameters', i ); names[i] = {last = last, first = first, link = link, mask = mask, corporate=corporate}; -- add this assembled name to our names list end return names, etal; -- all done, return our list of names end

--[[--------------------------< S E L E C T _ A U T H O R _ E D I T O R _ S O U R C E >------------------------

Select one of |authors=, |authorn= / |lastn / firstn=, or |vauthors= as the source of the author name list or select one of |editors=, |editorn= / editor-lastn= / |editor-firstn= or |veditors= as the source of the editor name list.

Only one of these appropriate three will be used. The hierarchy is: |authorn= (and aliases) highest and |authors= lowest and similarly, |editorn= (and aliases) highest and |editors= lowest

When looking for |authorn= / |editorn= parameters, test |xxxxor1= and |xxxxor2= (and all of their aliases); stops after the second test which mimicks the test used in extract_names() when looking for a hole in the author name list. There may be a better way to do this, I just haven't discovered what that way is.

Emits an error message when more than one xxxxor name source is provided.

In this function, vxxxxors = vauthors or veditors; xxxxors = authors or editors as appropriate.

]]

local function select_author_editor_source (vxxxxors, xxxxors, args, list_name) local lastfirst = false; if select_one( args, cfg.aliases[list_name .. '-Last'], 'none', 1 ) or -- do this twice incase we have a |first1= without a |last1=; this ... select_one( args, cfg.aliases[list_name .. '-First'], 'none', 1 ) or -- ... also catches the case where |first= is used with |vauthors= select_one( args, cfg.aliases[list_name .. '-Last'], 'none', 2 ) or select_one( args, cfg.aliases[list_name .. '-First'], 'none', 2 ) then lastfirst=true; end

if (is_set (vxxxxors) and true == lastfirst) or -- these are the three error conditions (is_set (vxxxxors) and is_set (xxxxors)) or (true == lastfirst and is_set (xxxxors)) then local err_name; if 'AuthorList' == list_name then -- figure out which name should be used in error message err_name = 'author'; else err_name = 'editor'; end table.insert( z.message_tail, { set_error( 'redundant_parameters', {err_name .. '-name-list parameters'}, true ) } ); -- add error message end

if true == lastfirst then return 1 end; -- return a number indicating which author name source to use if is_set (vxxxxors) then return 2 end; if is_set (xxxxors) then return 3 end; return 1; -- no authors so return 1; this allows missing author name test to run in case there is a first without last end

--[[--------------------------< I S _ V A L I D _ P A R A M E T E R _ V A L U E >------------------------------

This function is used to validate a parameter's assigned value for those parameters that have only a limited number of allowable values (yes, y, true, no, etc). When the parameter value has not been assigned a value (missing or empty in the source template) the function refurns true. If the parameter value is one of the list of allowed values returns true; else, emits an error message and returns false.

]]

local function is_valid_parameter_value (value, name, possible, cite_class) -- begin hack to limit |mode=mla to a specific set of templates if ('mode' == name) and ('mla' == value) and not in_array (cite_class, {'book', 'journal', 'news'}) then table.insert( z.message_tail, { set_error( 'invalid_param_val', {name, value}, true ) } ); -- not an allowed value so add error message return false end -- end hack

if not is_set (value) then return true; -- an empty parameter is ok elseif in_array(value:lower(), possible) then return true; else table.insert( z.message_tail, { set_error( 'invalid_param_val', {name, value}, true ) } ); -- not an allowed value so add error message return false end end

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

This function terminates a name list (author, contributor, editor) with a separator character (sepc) and a space when the last character is not a sepc character or when the last three characters are not sepc followed by two closing square brackets (close of a wikilink). When either of these is true, the name_list is terminated with a single space character.

]]

local function terminate_name_list (name_list, sepc) if (string.sub (name_list,-3,-1) == sepc .. '. ') then -- if already properly terminated return name_list; -- just return the name list elseif (string.sub (name_list,-1,-1) == sepc) or (string.sub (name_list,-3,-1) == sepc .. ']]') then -- if last name in list ends with sepc char return name_list .. " "; -- don't add another else return name_list .. sepc .. ' '; -- otherwise terninate the name list end end

--[[-------------------------< F O R M A T _ V O L U M E _ I S S U E >----------------------------------------

returns the concatenation of the formatted volume and issue parameters as a single string; or formatted volume or formatted issue, or an empty string if neither are set.

]]

local function format_volume_issue (volume, issue, cite_class, origin, sepc, lower, mode) if not is_set (volume) and not is_set (issue) then return ; end

if ('mla' == mode) and ('journal' == cite_class) then -- same as cs1 for magazines lower = true; -- mla 8th edition; force these to lower case if is_set (volume) and is_set (issue) then return wrap_msg ('vol-no', {sepc, volume, issue}, lower); elseif is_set (volume) then return wrap_msg ('vol', {sepc, volume}, lower); else return ; end end

if 'magazine' == cite_class or (in_array (cite_class, {'citation', 'map'}) and 'magazine' == origin) then if is_set (volume) and is_set (issue) then return wrap_msg ('vol-no', {sepc, volume, issue}, lower); elseif is_set (volume) then return wrap_msg ('vol', {sepc, volume}, lower); else return wrap_msg ('issue', {sepc, issue}, lower); end end

local vol = ;

if is_set (volume) then if (4 < mw.ustring.len(volume)) then vol = substitute (cfg.messages['j-vol'], {sepc, volume}); else vol = substitute (cfg.presentation['vol-bold'], {sepc, hyphen_to_dash(volume)}); end end if is_set (issue) then return vol .. substitute (cfg.messages['j-issue'], issue); end return vol; end

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

not currently used

normalizes a comma, ampersand, and/or space separated list to be '<value>, <value>, ..., <value>' returns list unchanged if there are no commas else strips whitespace and then reformats the list

]] --[[ local function normalize_page_list (list) if not list:find ('[,& ]') then return list end -- if list is not delimited with commas, ampersands, or spaces; done

list = mw.text.split (list, '[,&%s]+'); -- make a table of values list = table.concat (list, ', '); -- and now make a normalized list return list; end ]]

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

adds static text to one of |page(s)= or |sheet(s)= values and returns it with all of the others set to empty strings. The return order is: page, pages, sheet, sheets

Singular has priority over plural when both are provided.

]]

local function format_pages_sheets (page, pages, sheet, sheets, cite_class, origin, sepc, nopp, lower, mode) if 'map' == cite_class then -- only cite map supports sheet(s) as in-source locators if is_set (sheet) then if 'journal' == origin then return , , wrap_msg ('j-sheet', sheet, lower), ; else return , , wrap_msg ('sheet', {sepc, sheet}, lower), ; end elseif is_set (sheets) then if 'journal' == origin then return , , , wrap_msg ('j-sheets', sheets, lower); else return , , , wrap_msg ('sheets', {sepc, sheets}, lower); end end end

local is_journal = 'journal' == cite_class or (in_array (cite_class, {'citation', 'map'}) and 'journal' == origin);

if is_journal and 'mla' == mode then is_journal = false; -- mla always uses p & pp end

if is_set (page) then if is_journal then return substitute (cfg.messages['j-page(s)'], page), , , ; elseif not nopp then return substitute (cfg.messages['p-prefix'], {sepc, page}), , , ; else return substitute (cfg.messages['nopp'], {sepc, page}), , , ; end elseif is_set(pages) then if is_journal then return substitute (cfg.messages['j-page(s)'], pages), , , ; elseif tonumber(pages) ~= nil and not nopp then -- if pages is only digits, assume a single page number return , substitute (cfg.messages['p-prefix'], {sepc, pages}), , ; elseif not nopp then return , substitute (cfg.messages['pp-prefix'], {sepc, pages}), , ; else return , substitute (cfg.messages['nopp'], {sepc, pages}), , ; end end

return , , , ; -- return empty strings end

--[=[-------------------------< A R C H I V E _ U R L _ C H E C K >--------------------------------------------

Check archive.org urls to make sure they at least look like they are pointing at valid archives and not to the save snapshot url or to calendar pages. When the archive url is 'https://web.archive.org/save/' (or http://...) archive.org saves a snapshot of the target page in the url. That is something that Wikipedia should not allow unwitting readers to do.

When the archive.org url does not have a complete timestamp, archive.org chooses a snapshot according to its own algorithm or provides a calendar 'search' result. WP:ELNO discourages links to search results.

This function looks at the value assigned to |archive-url= and returns empty strings for |archive-url= and |archive-date= and an error message when: |archive-url= holds an archive.org save command url |archive-url= is an archive.org url that does not have a complete timestamp (YYYYMMDDhhmmss 14 digits) in the correct place otherwise returns |archive-url= and |archive-date=

There are two mostly compatible archive.org urls: //web.archive.org/<timestamp>... -- the old form //web.archive.org/web/<timestamp>... -- the new form

The old form does not support or map to the new form when it contains a display flag. There are four identified flags ('id_', 'js_', 'cs_', 'im_') but since archive.org ignores others following the same form (two letters and an underscore) we don't check for these specific flags but we do check the form.

This function supports a preview mode. When the article is rendered in preview mode, this funct may return a modified archive url: for save command errors, return undated wildcard (/*/) for timestamp errors when the timestamp has a wildcard, return the url unmodified for timestamp errors when the timestamp does not have a wildcard, return with timestamp limited to six digits plus wildcard (/yyyymm*/)

]=]

local function archive_url_check (url, date) local err_msg = ; -- start with the error message empty local path, timestamp, flag; -- portions of the archive.or url

if (not url:match('//web%.archive%.org/')) and (not url:match('//liveweb%.archive%.org/')) then -- also deprecated liveweb Wayback machine url return url, date; -- not an archive.org archive, return ArchiveURL and ArchiveDate end

if url:match('//web%.archive%.org/save/') then -- if a save command url, we don't want to allow saving of the target page err_msg = 'save command'; url = url:gsub ('(//web%.archive%.org)/save/', '%1/*/', 1); -- for preview mode: modify ArchiveURL elseif url:match('//liveweb%.archive%.org/') then err_msg = 'liveweb'; else path, timestamp, flag = url:match('//web%.archive%.org/([^%d]*)(%d+)([^/]*)/'); -- split out some of the url parts for evaluation

if not is_set(timestamp) or 14 ~= timestamp:len() then -- path and flag optional, must have 14-digit timestamp here err_msg = 'timestamp'; if '*' ~= flag then url=url:gsub ('(//web%.archive%.org/[^%d]*%d?%d?%d?%d?%d?%d?)[^/]*', '%1*', 1) -- for preview, modify ts to be yearmo* max (0-6 digits plus splat) end elseif is_set(path) and 'web/' ~= path then -- older archive urls do not have the extra 'web/' path element err_msg = 'path'; elseif is_set (flag) and not is_set (path) then -- flag not allowed with the old form url (without the 'web/' path element) err_msg = 'flag'; elseif is_set (flag) and not flag:match ('%a%a_') then -- flag if present must be two alpha characters and underscore (requires 'web/' path element) err_msg = 'flag'; else return url, date; -- return archiveURL and ArchiveDate end end -- if here, something not right so table.insert( z.message_tail, { set_error( 'archive_url', {err_msg}, true ) } ); -- add error message and if is_set (Frame:preprocess('393')) then return , ; -- return empty strings for archiveURL and ArchiveDate else return url, date; -- preview mode so return archiveURL and ArchiveDate end end

--[[--------------------------< M I S S I N G _ P I P E _ C H E C K >------------------------------------------

Look at the contents of a parameter. If the content has a string of characters and digits followed by an equal sign, compare the alphanumeric string to the list of cs1|2 parameters. If found, then the string is possibly a parameter that is missing its pipe: Template:Cite ...

cs1|2 shares some parameter names with xml/html atributes: class=, title=, etc. To prevent false positives xml/html tags are removed before the search.

If a missing pipe is detected, this function adds the missing pipe maintenance category.

]]

local function missing_pipe_check (value) local capture; value = value:gsub ('%b<>', ); -- remove xml/html tags because attributes: class=, title=, etc

capture = value:match ('%s+(%a[%a%d]+)%s*=') or value:match ('^(%a[%a%d]+)%s*='); -- find and categorize parameters with possible missing pipes if capture and validate (capture) then -- if the capture is a valid parameter name add_maint_cat ('missing_pipe'); end end

--[[--------------------------< C I T A T I O N 0 >------------------------------------------------------------

This is the main function doing the majority of the citation formatting.

]]

local function citation0( config, args) --[[ Load Input Parameters The argument_wrapper facilitates the mapping of multiple aliases to single internal variable. ]] local A = argument_wrapper( args ); local i

-- Pick out the relevant fields from the arguments. Different citation templates -- define different field names for the same underlying things.

-- set default parameter values defined by |mode= parameter. If |mode= is empty or omitted, use CitationClass to set these values local Mode = A['Mode']; if not is_valid_parameter_value (Mode, 'mode', cfg.keywords['mode'], config.CitationClass) then Mode = ; end

local author_etal; local a = {}; -- authors list from |lastn= / |firstn= pairs or |vauthors= local Authors; local NameListFormat = A['NameListFormat']; local Collaboration = A['Collaboration'];

do -- to limit scope of selected local selected = select_author_editor_source (A['Vauthors'], A['Authors'], args, 'AuthorList'); if 1 == selected then a, author_etal = extract_names (args, 'AuthorList'); -- fetch author list from |authorn= / |lastn= / |firstn=, |author-linkn=, and |author-maskn= elseif 2 == selected then NameListFormat = 'vanc'; -- override whatever |name-list-format= might be a, author_etal = parse_vauthors_veditors (args, args.vauthors, 'AuthorList'); -- fetch author list from |vauthors=, |author-linkn=, and |author-maskn= elseif 3 == selected then Authors = A['Authors']; -- use content of |authors= if 'authors' == A:ORIGIN('Authors') then -- but add a maint cat if the parameter is |authors= add_maint_cat ('authors'); -- because use of this parameter is discouraged; what to do about the aliases is a TODO: end end if is_set (Collaboration) then author_etal = true; -- so that |display-authors=etal not required end end

local Others = A['Others'];

local editor_etal; local e = {}; -- editors list from |editor-lastn= / |editor-firstn= pairs or |veditors= local Editors;

do -- to limit scope of selected local selected = select_author_editor_source (A['Veditors'], A['Editors'], args, 'EditorList'); if 1 == selected then e, editor_etal = extract_names (args, 'EditorList'); -- fetch editor list from |editorn= / |editor-lastn= / |editor-firstn=, |editor-linkn=, and |editor-maskn= elseif 2 == selected then NameListFormat = 'vanc'; -- override whatever |name-list-format= might be e, editor_etal = parse_vauthors_veditors (args, args.veditors, 'EditorList'); -- fetch editor list from |veditors=, |editor-linkn=, and |editor-maskn= elseif 3 == selected then Editors = A['Editors']; -- use content of |editors= add_maint_cat ('editors'); -- but add a maint cat because use of this parameter is discouraged end end

local t = {}; -- translators list from |translator-lastn= / translator-firstn= pairs local Translators; -- assembled translators name list t = extract_names (args, 'TranslatorList'); -- fetch translator list from |translatorn= / |translator-lastn=, -firstn=, -linkn=, -maskn=

local interviewers_list = {}; local Interviewers = A['Interviewers'] if is_set (Interviewers) then -- add a maint cat if the |interviewers= is used add_maint_cat ('interviewers'); -- because use of this parameter is discouraged else interviewers_list = extract_names (args, 'InterviewerList'); -- else, process preferred interviewers parameters end

local c = {}; -- contributors list from |contributor-lastn= / contributor-firstn= pairs local Contributors; -- assembled contributors name list local Contribution = A['Contribution']; if in_array(config.CitationClass, {"book","citation"}) and not is_set(A['Periodical']) then -- |contributor= and |contribution= only supported in book cites c = extract_names (args, 'ContributorList'); -- fetch contributor list from |contributorn= / |contributor-lastn=, -firstn=, -linkn=, -maskn=

if 0 < #c then if not is_set (Contribution) then -- |contributor= requires |contribution= table.insert( z.message_tail, { set_error( 'contributor_missing_required_param', 'contribution')}); -- add missing contribution error message c = {}; -- blank the contributors' table; it is used as a flag later end if 0 == #a then -- |contributor= requires |author= table.insert( z.message_tail, { set_error( 'contributor_missing_required_param', 'author')}); -- add missing author error message c = {}; -- blank the contributors' table; it is used as a flag later end end else -- if not a book cite if select_one (args, cfg.aliases['ContributorList-Last'], 'redundant_parameters', 1 ) then -- are there contributor name list parameters? table.insert( z.message_tail, { set_error( 'contributor_ignored')}); -- add contributor ignored error message end Contribution = nil; -- unset end

if not is_valid_parameter_value (NameListFormat, 'name-list-format', cfg.keywords['name-list-format']) then -- only accepted value for this parameter is 'vanc' NameListFormat = ; -- anything else, set to empty string end

local Year = A['Year']; local PublicationDate = A['PublicationDate']; local OrigYear = A['OrigYear']; local Date = A['Date']; local LayDate = A['LayDate']; ------------------------------------------------- Get title data local Title = A['Title']; local ScriptTitle = A['ScriptTitle']; local BookTitle = A['BookTitle']; local Conference = A['Conference']; local TransTitle = A['TransTitle']; local TitleNote = A['TitleNote']; local TitleLink = A['TitleLink']; link_title_ok (TitleLink, A:ORIGIN ('TitleLink'), Title, 'title'); -- check for wikimarkup in |title-link= or wikimarkup in |title= when |title-link= is set

local Chapter = A['Chapter']; local ScriptChapter = A['ScriptChapter']; local ChapterLink -- = A['ChapterLink']; -- deprecated as a parameter but still used internally by cite episode local TransChapter = A['TransChapter']; local TitleType = A['TitleType']; local Degree = A['Degree']; local Docket = A['Docket']; local ArchiveFormat = A['ArchiveFormat'];

local ArchiveDate; local ArchiveURL;

ArchiveURL, ArchiveDate = archive_url_check (A['ArchiveURL'], A['ArchiveDate'])

local DeadURL = A['DeadURL'] if not is_valid_parameter_value (DeadURL, 'dead-url', cfg.keywords ['deadurl']) then -- set in config.defaults to 'yes' DeadURL = ; -- anything else, set to empty string end

local URL = A['URL'] local URLorigin = A:ORIGIN('URL'); -- get name of parameter that holds URL local ChapterURL = A['ChapterURL']; local ChapterURLorigin = A:ORIGIN('ChapterURL'); -- get name of parameter that holds ChapterURL local ConferenceFormat = A['ConferenceFormat']; local ConferenceURL = A['ConferenceURL']; local ConferenceURLorigin = A:ORIGIN('ConferenceURL'); -- get name of parameter that holds ConferenceURL local Periodical = A['Periodical']; local Periodical_origin = A:ORIGIN('Periodical'); -- get the name of the periodical parameter

local Series = A['Series'];

local Volume; local Issue; local Page; local Pages; local At; -- previously conference books did not support volume -- if in_array (config.CitationClass, cfg.templates_using_volume) and not ('conference' == config.CitationClass and not is_set (Periodical)) then if in_array (config.CitationClass, cfg.templates_using_volume) then Volume = A['Volume']; end -- conference & map books do not support issue if in_array (config.CitationClass, cfg.templates_using_issue) and not (in_array (config.CitationClass, {'conference', 'map'}) and not is_set (Periodical))then Issue = A['Issue']; end local Position = ; if not in_array (config.CitationClass, cfg.templates_not_using_page) then Page = A['Page']; Pages = hyphen_to_dash( A['Pages'] ); At = A['At']; end

local Edition = A['Edition']; local PublicationPlace = A['PublicationPlace'] local Place = A['Place'];

local PublisherName = A['PublisherName']; local RegistrationRequired = A['RegistrationRequired']; if not is_valid_parameter_value (RegistrationRequired, 'registration', cfg.keywords ['yes_true_y']) then RegistrationRequired=nil; end local SubscriptionRequired = A['SubscriptionRequired']; if not is_valid_parameter_value (SubscriptionRequired, 'subscription', cfg.keywords ['yes_true_y']) then SubscriptionRequired=nil; end local UrlAccess = A['UrlAccess']; if not is_valid_parameter_value (UrlAccess, 'url-access', cfg.keywords ['url-access']) then UrlAccess = nil; end if not is_set(URL) and is_set(UrlAccess) then UrlAccess = nil; table.insert( z.message_tail, { set_error( 'param_access_requires_param', {'url'}, true ) } ); end

if is_set (UrlAccess) and is_set (SubscriptionRequired) then -- while not aliases, these are much the same so if both are set table.insert( z.message_tail, { set_error( 'redundant_parameters', {wrap_style ('parameter', 'url-access') .. ' and ' .. wrap_style ('parameter', 'subscription')}, true ) } ); -- add error message SubscriptionRequired = nil; -- unset; prefer |access= over |subscription= end if is_set (UrlAccess) and is_set (RegistrationRequired) then -- these are not the same but contradictory so if both are set table.insert( z.message_tail, { set_error( 'redundant_parameters', {wrap_style ('parameter', 'url-access') .. ' and ' .. wrap_style ('parameter', 'registration')}, true ) } ); -- add error message RegistrationRequired = nil; -- unset; prefer |access= over |registration= end

local Via = A['Via']; local AccessDate = A['AccessDate']; local Agency = A['Agency'];

local Language = A['Language']; local Format = A['Format']; local ChapterFormat = A['ChapterFormat']; local DoiBroken = A['DoiBroken']; local ID = A['ID']; local ASINTLD = A['ASINTLD']; local IgnoreISBN = A['IgnoreISBN']; if not is_valid_parameter_value (IgnoreISBN, 'ignore-isbn-error', cfg.keywords ['yes_true_y']) then IgnoreISBN = nil; -- anything else, set to empty string end local Embargo = A['Embargo']; local Class = A['Class']; -- arxiv class identifier

local ID_list = extract_ids( args ); local ID_access_levels = extract_id_access_levels( args, ID_list );

local Quote = A['Quote'];

local LayFormat = A['LayFormat']; local LayURL = A['LayURL']; local LaySource = A['LaySource']; local Transcript = A['Transcript']; local TranscriptFormat = A['TranscriptFormat']; local TranscriptURL = A['TranscriptURL'] local TranscriptURLorigin = A:ORIGIN('TranscriptURL'); -- get name of parameter that holds TranscriptURL

local LastAuthorAmp = A['LastAuthorAmp']; if not is_valid_parameter_value (LastAuthorAmp, 'last-author-amp', cfg.keywords ['yes_true_y']) then LastAuthorAmp = nil; -- set to empty string end if 'mla' == Mode then LastAuthorAmp = 'yes'; -- replaces last author/editor separator with ' and ' text end local no_tracking_cats = A['NoTracking']; if not is_valid_parameter_value (no_tracking_cats, 'no-tracking', cfg.keywords ['yes_true_y']) then no_tracking_cats = nil; -- set to empty string end

--local variables that are not cs1 parameters local use_lowercase; -- controls capitalization of certain static text local this_page = mw.title.getCurrentTitle(); -- also used for COinS and for language local anchor_year; -- used in the CITEREF identifier local COinS_date = {}; -- holds date info extracted from |date= for the COinS metadata by Module:Date verification

local DF = A['DF']; -- date format set in cs1|2 template if not is_valid_parameter_value (DF, 'df', cfg.keywords['date-format']) then -- validate reformatting keyword DF = ; -- not valid, set to empty string end

local sepc; -- separator between citation elements for CS1 a period, for CS2, a comma local PostScript; local Ref; sepc, PostScript, Ref = set_style (Mode:lower(), A['PostScript'], A['Ref'], config.CitationClass); use_lowercase = ( sepc == ',' ); -- used to control capitalization for certain static text

--check this page to see if it is in one of the namespaces that cs1 is not supposed to add to the error categories if not is_set (no_tracking_cats) then -- ignore if we are already not going to categorize this page if in_array (this_page.nsText, cfg.uncategorized_namespaces) then no_tracking_cats = "true"; -- set no_tracking_cats end for _,v in ipairs (cfg.uncategorized_subpages) do -- cycle through page name patterns if this_page.text:match (v) then -- test page name against each pattern no_tracking_cats = "true"; -- set no_tracking_cats break; -- bail out if one is found end end end

-- check for extra |page=, |pages= or |at= parameters. (also sheet and sheets while we're at it) select_one( args, {'page', 'p', 'pp', 'pages', 'at', 'sheet', 'sheets'}, 'redundant_parameters' ); -- this is a dummy call simply to get the error message and category

local NoPP = A['NoPP'] if is_set (NoPP) and is_valid_parameter_value (NoPP, 'nopp', cfg.keywords ['yes_true_y']) then NoPP = true; else NoPP = nil; -- unset, used as a flag later end

if is_set(Page) then if is_set(Pages) or is_set(At) then Pages = ; -- unset the others At = ; end extra_text_in_page_check (Page); -- add this page to maint cat if |page= value begins with what looks like p. or pp. elseif is_set(Pages) then if is_set(At) then At = ; -- unset end extra_text_in_page_check (Pages); -- add this page to maint cat if |pages= value begins with what looks like p. or pp. end

-- both |publication-place= and |place= (|location=) allowed if different if not is_set(PublicationPlace) and is_set(Place) then PublicationPlace = Place; -- promote |place= (|location=) to |publication-place end

if PublicationPlace == Place then Place = ; end -- don't need both if they are the same

--[[ Parameter remapping for cite encyclopedia: When the citation has these parameters: |encyclopedia and |title then map |title to |article and |encyclopedia to |title |encyclopedia and |article then map |encyclopedia to |title |encyclopedia then map |encyclopedia to |title

|trans_title maps to |trans_chapter when |title is re-mapped |url maps to |chapterurl when |title is remapped

All other combinations of |encyclopedia, |title, and |article are not modified

]]

local Encyclopedia = A['Encyclopedia'];

if ( config.CitationClass == "encyclopaedia" ) or ( config.CitationClass == "citation" and is_set (Encyclopedia)) then -- test code for citation if is_set(Periodical) then -- Periodical is set when |encyclopedia is set if is_set(Title) or is_set (ScriptTitle) then if not is_set(Chapter) then Chapter = Title; -- |encyclopedia and |title are set so map |title to |article and |encyclopedia to |title ScriptChapter = ScriptTitle; TransChapter = TransTitle; ChapterURL = URL; if not is_set (ChapterURL) and is_set (TitleLink) then Chapter= '' .. Chapter .. ''; end Title = Periodical; ChapterFormat = Format; Periodical = ; -- redundant so unset TransTitle = ; URL = ; Format = ; TitleLink = ; ScriptTitle = ; end else -- |title not set Title = Periodical; -- |encyclopedia set and |article set or not set so map |encyclopedia to |title Periodical = ; -- redundant so unset end end end

-- Special case for cite techreport. if (config.CitationClass == "techreport") then -- special case for cite techreport if is_set(A['Number']) then -- cite techreport uses 'number', which other citations alias to 'issue' if not is_set(ID) then -- can we use ID for the "number"? ID = A['Number']; -- yes, use it else -- ID has a value so emit error message table.insert( z.message_tail, { set_error('redundant_parameters', {wrap_style ('parameter', 'id') .. ' and ' .. wrap_style ('parameter', 'number')}, true )}); end end end

-- special case for cite mailing list if (config.CitationClass == "mailinglist") then Periodical = A ['MailingList']; elseif 'mailinglist' == A:ORIGIN('Periodical') then Periodical = ; -- unset because mailing list is only used for cite mailing list end

-- Account for the oddity that is , before generation of COinS data. if 'conference' == config.CitationClass then if is_set(BookTitle) then Chapter = Title; -- ChapterLink = TitleLink; -- |chapterlink= is deprecated ChapterURL = URL; ChapterURLorigin = URLorigin; URLorigin = ; ChapterFormat = Format; TransChapter = TransTitle; Title = BookTitle; Format = ; -- TitleLink = ; TransTitle = ; URL = ; end elseif 'speech' ~= config.CitationClass then Conference = ; -- not cite conference or cite speech so make sure this is empty string end

-- cite map oddities local Cartography = ""; local Scale = ""; local Sheet = A['Sheet'] or ; local Sheets = A['Sheets'] or ; if config.CitationClass == "map" then Chapter = A['Map']; ChapterURL = A['MapURL']; TransChapter = A['TransMap']; ChapterURLorigin = A:ORIGIN('MapURL'); ChapterFormat = A['MapFormat'];

Cartography = A['Cartography']; if is_set( Cartography ) then Cartography = sepc .. " " .. wrap_msg ('cartography', Cartography, use_lowercase); end Scale = A['Scale']; if is_set( Scale ) then Scale = sepc .. " " .. Scale; end end

-- Account for the oddities that are and Template:Cite serial, before generation of COinS data. if 'episode' == config.CitationClass or 'serial' == config.CitationClass then local AirDate = A['AirDate']; local SeriesLink = A['SeriesLink'];

link_title_ok (SeriesLink, A:ORIGIN ('SeriesLink'), Series, 'series'); -- check for wikimarkup in |series-link= or wikimarkup in |series= when |series-link= is set

local Network = A['Network']; local Station = A['Station']; local s, n = {}, {}; -- do common parameters first if is_set(Network) then table.insert(n, Network); end if is_set(Station) then table.insert(n, Station); end ID = table.concat(n, sepc .. ' ');

if not is_set (Date) and is_set (AirDate) then -- promote airdate to date Date = AirDate; end

if 'episode' == config.CitationClass then -- handle the oddities that are strictly local Season = A['Season']; local SeriesNumber = A['SeriesNumber'];

if is_set (Season) and is_set (SeriesNumber) then -- these are mutually exclusive so if both are set table.insert( z.message_tail, { set_error( 'redundant_parameters', {wrap_style ('parameter', 'season') .. ' and ' .. wrap_style ('parameter', 'seriesno')}, true ) } ); -- add error message SeriesNumber = ; -- unset; prefer |season= over |seriesno= end -- assemble a table of parts concatenated later into Series if is_set(Season) then table.insert(s, wrap_msg ('season', Season, use_lowercase)); end if is_set(SeriesNumber) then table.insert(s, wrap_msg ('series', SeriesNumber, use_lowercase)); end if is_set(Issue) then table.insert(s, wrap_msg ('episode', Issue, use_lowercase)); end Issue = ; -- unset because this is not a unique parameter

Chapter = Title; -- promote title parameters to chapter ScriptChapter = ScriptTitle; ChapterLink = TitleLink; -- alias episodelink TransChapter = TransTitle; ChapterURL = URL; ChapterURLorigin = A:ORIGIN('URL');

Title = Series; -- promote series to title TitleLink = SeriesLink; Series = table.concat(s, sepc .. ' '); -- this is concatenation of season, seriesno, episode number

if is_set (ChapterLink) and not is_set (ChapterURL) then -- link but not URL Chapter = '' .. Chapter .. ''; -- ok to wikilink elseif is_set (ChapterLink) and is_set (ChapterURL) then -- if both are set, URL links episode; Series = '' .. Series .. ''; -- series links with ChapterLink (episodelink -> TitleLink -> ChapterLink) ugly end URL = ; -- unset TransTitle = ; ScriptTitle = ;

else -- now oddities that are cite serial Issue = ; -- unset because this parameter no longer supported by the citation/core version of cite serial Chapter = A['Episode']; -- TODO: make |episode= available to cite episode someday? if is_set (Series) and is_set (SeriesLink) then Series = '' .. Series .. ''; end Series = wrap_style ('italic-title', Series); -- series is italicized end end -- end of stuff

-- Account for the oddities that are Template:Cite arxiv, Template:Cite biorxiv, Template:Cite citeseerx, before generation of COinS data. do if in_array (config.CitationClass, {'arxiv', 'biorxiv', 'citeseerx'}) then if not is_set (ID_list[config.CitationClass:upper()]) then -- |arxiv= or |eprint= required for cite arxiv; |biorxiv= & |citeseerx= required for their templates table.insert( z.message_tail, { set_error( config.CitationClass .. '_missing', {}, true ) } ); -- add error message end

if 'arxiv' == config.CitationClass then Periodical = 'arXiv'; -- set to arXiv for COinS; after that, must be set to empty string end if 'biorxiv' == config.CitationClass then Periodical = 'bioRxiv'; -- set to bioRxiv for COinS; after that, must be set to empty string end if 'citeseerx' == config.CitationClass then Periodical = 'CiteSeerX'; -- set to CiteSeerX for COinS; after that, must be set to empty string end end end

-- handle type parameter for those CS1 citations that have default values if in_array(config.CitationClass, {"AV-media-notes", "interview", "mailinglist", "map", "podcast", "pressrelease", "report", "techreport", "thesis"}) then TitleType = set_titletype (config.CitationClass, TitleType); if is_set(Degree) and "Thesis" == TitleType then -- special case for cite thesis TitleType = Degree .. ' ' .. cfg.title_types ['thesis']:lower(); end end

if is_set(TitleType) then -- if type parameter is specified TitleType = substitute( cfg.messages['type'], TitleType); -- display it in parentheses -- TODO: Hack on TitleType to fix bunched parentheses problem end

-- legacy: promote PublicationDate to Date if neither Date nor Year are set. if not is_set (Date) then Date = Year; -- promote Year to Date Year = nil; -- make nil so Year as empty string isn't used for CITEREF if not is_set (Date) and is_set(PublicationDate) then -- use PublicationDate when |date= and |year= are not set Date = PublicationDate; -- promote PublicationDate to Date PublicationDate = ; -- unset, no longer needed end end

if PublicationDate == Date then PublicationDate = ; end -- if PublicationDate is same as Date, don't display in rendered citation

--[[ Go test all of the date-holding parameters for valid MOS:DATE format and make sure that dates are real dates. This must be done before we do COinS because here is where we get the date used in the metadata.

Date validation supporting code is in Module:Citation/CS1/Date_validation ]] do -- create defined block to contain local variables error_message, date_parameters_list, mismatch local error_message = ; -- AirDate has been promoted to Date so not necessary to check it local date_parameters_list = {['access-date']=AccessDate, ['archive-date']=ArchiveDate, ['date']=Date, ['doi-broken-date']=DoiBroken, ['embargo']=Embargo, ['lay-date']=LayDate, ['publication-date']=PublicationDate, ['year']=Year};

anchor_year, Embargo, error_message = dates(date_parameters_list, COinS_date);

if is_set (Year) and is_set (Date) then -- both |date= and |year= not normally needed; local mismatch = year_date_check (Year, Date) if 0 == mismatch then -- |year= does not match a year-value in |date= if is_set (error_message) then -- if there is already an error message error_message = error_message .. ', '; -- tack on this additional message end error_message = error_message .. '|year= / |date= mismatch'; elseif 1 == mismatch then -- |year= matches year-value in |date= add_maint_cat ('date_year'); end end

if not is_set(error_message) then -- error free dates only local modified = false; -- flag if is_set (DF) then -- if we need to reformat dates modified = reformat_dates (date_parameters_list, DF, false); -- reformat to DF format, use long month names if appropriate end

if true == date_hyphen_to_dash (date_parameters_list) then -- convert hyphens to dashes where appropriate modified = true; add_maint_cat ('date_format'); -- hyphens were converted so add maint category end

if modified then -- if the date_parameters_list values were modified AccessDate = date_parameters_list['access-date']; -- overwrite date holding parameters with modified values ArchiveDate = date_parameters_list['archive-date']; Date = date_parameters_list['date']; DoiBroken = date_parameters_list['doi-broken-date']; LayDate = date_parameters_list['lay-date']; PublicationDate = date_parameters_list['publication-date']; end else table.insert( z.message_tail, { set_error( 'bad_date', {error_message}, true ) } ); -- add this error message end end -- end of do

-- Account for the oddity that is with |pmc= set and |url= not set. Do this after date check but before COInS. -- Here we unset Embargo if PMC not embargoed (|embargo= not set in the citation) or if the embargo time has expired. Otherwise, holds embargo date Embargo = is_embargoed (Embargo); --

if config.CitationClass == "journal" and not is_set(URL) and is_set(ID_list['PMC']) then if not is_set (Embargo) then -- if not embargoed or embargo has expired URL=cfg.id_handlers['PMC'].prefix .. ID_list['PMC']; -- set url to be the same as the PMC external link if not embargoed URLorigin = cfg.id_handlers['PMC'].parameters[1]; -- set URLorigin to parameter name for use in error message if citation is missing a |title= if is_set(AccessDate) then -- access date requires |url=; pmc created url is not |url= table.insert( z.message_tail, { set_error( 'accessdate_missing_url', {}, true ) } ); AccessDate = ; -- unset end

end end

-- At this point fields may be nil if they weren't specified in the template use. We can use that fact. -- Test if citation has no title if not is_set(Title) and not is_set(TransTitle) and not is_set(ScriptTitle) then if 'episode' == config.CitationClass then -- special case for cite episode; TODO: is there a better way to do this? table.insert( z.message_tail, { set_error( 'citation_missing_title', {'series'}, true ) } ); else table.insert( z.message_tail, { set_error( 'citation_missing_title', {'title'}, true ) } ); end end

if 'none' == Title and in_array (config.CitationClass, {'journal', 'citation'}) and is_set (Periodical) and 'journal' == A:ORIGIN('Periodical') then -- special case for journal cites Title = ; -- set title to empty string add_maint_cat ('untitled'); end

check_for_url ({ -- add error message when any of these parameters contains a URL ['title']=Title, [A:ORIGIN('Chapter')]=Chapter, [A:ORIGIN('Periodical')]=Periodical, [A:ORIGIN('PublisherName')] = PublisherName });

-- COinS metadata (see <http://ocoins.info/>) for automated parsing of citation information. -- handle the oddity that is cite encyclopedia and . Here we presume that -- when Periodical, Title, and Chapter are all set, then Periodical is the book (encyclopedia) title, Title -- is the article title, and Chapter is a section within the article. So, we remap

local coins_chapter = Chapter; -- default assuming that remapping not required local coins_title = Title; -- et tu if 'encyclopaedia' == config.CitationClass or ('citation' == config.CitationClass and is_set (Encyclopedia)) then if is_set (Chapter) and is_set (Title) and is_set (Periodical) then -- if all are used then coins_chapter = Title; -- remap coins_title = Periodical; end end local coins_author = a; -- default for coins rft.au if 0 < #c then -- but if contributor list coins_author = c; -- use that instead end

-- this is the function call to COinS() local OCinSoutput = COinS({ ['Periodical'] = Periodical, ['Encyclopedia'] = Encyclopedia, ['Chapter'] = make_coins_title (coins_chapter, ScriptChapter), -- Chapter and ScriptChapter stripped of bold / italic wikimarkup ['Degree'] = Degree; -- cite thesis only ['Title'] = make_coins_title (coins_title, ScriptTitle), -- Title and ScriptTitle stripped of bold / italic wikimarkup ['PublicationPlace'] = PublicationPlace, ['Date'] = COinS_date.rftdate, -- COinS_date has correctly formatted date if Date is valid; ['Season'] = COinS_date.rftssn, ['Chron'] = COinS_date.rftchron or (not COinS_date.rftdate and Date) or , -- chron but if not set and invalid date format use Date; keep this last bit? ['Series'] = Series, ['Volume'] = Volume, ['Issue'] = Issue, ['Pages'] = get_coins_pages (first_set ({Sheet, Sheets, Page, Pages, At}, 5)), -- pages stripped of external links ['Edition'] = Edition, ['PublisherName'] = PublisherName, ['URL'] = first_set ({ChapterURL, URL}, 2), ['Authors'] = coins_author, ['ID_list'] = ID_list, ['RawPage'] = this_page.prefixedText, }, config.CitationClass);

-- Account for the oddities that are Template:Cite arxiv, AFTER generation of COinS data. -- if 'arxiv' == config.CitationClass then -- we have set rft.jtitle in COinS to arXiv, now unset so it isn't displayed if in_array (config.CitationClass, {'arxiv', 'biorxiv', 'citeseerx'}) then -- we have set rft.jtitle in COinS to arXiv, bioRxiv, or CiteSeerX now unset so it isn't displayed Periodical = ; -- periodical not allowed in these templates; if article has been published, use cite journal end

-- special case for cite newsgroup. Do this after COinS because we are modifying Publishername to include some static text if 'newsgroup' == config.CitationClass then if is_set (PublisherName) then PublisherName = substitute (cfg.messages['newsgroup'], external_link( 'news:' .. PublisherName, PublisherName, A:ORIGIN('PublisherName'), nil )); end end

-- Now perform various field substitutions. -- We also add leading spaces and surrounding markup and punctuation to the -- various parts of the citation, but only when they are non-nil. local EditorCount; -- used only for choosing {ed.) or (eds.) annotation at end of editor name-list do local last_first_list; local control = { format = NameListFormat, -- empty string or 'vanc' maximum = nil, -- as if display-authors or display-editors not set lastauthoramp = LastAuthorAmp, page_name = this_page.text, -- get current page name so that we don't wikilink to it via editorlinkn mode = Mode };

do -- do editor name list first because the now unsupported coauthors used to modify control table control.maximum , editor_etal = get_display_authors_editors (A['DisplayEditors'], #e, 'editors', editor_etal); last_first_list, EditorCount = list_people(control, e, editor_etal);

if is_set (Editors) then if editor_etal then Editors = Editors .. ' ' .. cfg.messages['et al']; -- add et al. to editors parameter beause |display-editors=etal EditorCount = 2; -- with et al., |editors= is multiple names; spoof to display (eds.) annotation else EditorCount = 2; -- we don't know but assume |editors= is multiple names; spoof to display (eds.) annotation end else Editors = last_first_list; -- either an author name list or an empty string end

if 1 == EditorCount and (true == editor_etal or 1 < #e) then -- only one editor displayed but includes etal then EditorCount = 2; -- spoof to display (eds.) annotation end end do -- now do interviewers control.maximum = #interviewers_list; -- number of interviewerss Interviewers = list_people(control, interviewers_list, false); -- et al not currently supported end do -- now do translators control.maximum = #t; -- number of translators Translators = list_people(control, t, false); -- et al not currently supported end do -- now do contributors control.maximum = #c; -- number of contributors Contributors = list_people(control, c, false); -- et al not currently supported end do -- now do authors control.maximum , author_etal = get_display_authors_editors (A['DisplayAuthors'], #a, 'authors', author_etal);

last_first_list = list_people(control, a, author_etal);

if is_set (Authors) then Authors, author_etal = name_has_etal (Authors, author_etal, false); -- find and remove variations on et al. if author_etal then Authors = Authors .. ' ' .. cfg.messages['et al']; -- add et al. to authors parameter end else Authors = last_first_list; -- either an author name list or an empty string end end -- end of do

if is_set (Authors) and is_set (Collaboration) then Authors = Authors .. ' (' .. Collaboration .. ')'; -- add collaboration after et al. end

end

-- apply |[xx-]format= styling; at the end, these parameters hold correctly styled format annotation, -- an error message if the associated url is not set, or an empty string for concatenation ArchiveFormat = style_format (ArchiveFormat, ArchiveURL, 'archive-format', 'archive-url'); ConferenceFormat = style_format (ConferenceFormat, ConferenceURL, 'conference-format', 'conference-url'); Format = style_format (Format, URL, 'format', 'url'); LayFormat = style_format (LayFormat, LayURL, 'lay-format', 'lay-url'); TranscriptFormat = style_format (TranscriptFormat, TranscriptURL, 'transcript-format', 'transcripturl');

-- special case for chapter format so no error message or cat when chapter not supported if not (in_array(config.CitationClass, {'web', 'news', 'journal', 'magazine', 'pressrelease', 'podcast', 'newsgroup', 'arxiv', 'biorxiv', 'citeseerx'}) or ('citation' == config.CitationClass and is_set (Periodical) and not is_set (Encyclopedia))) then ChapterFormat = style_format (ChapterFormat, ChapterURL, 'chapter-format', 'chapter-url'); end

if not is_set(URL) then if in_array(config.CitationClass, {"web","podcast", "mailinglist"}) then -- |url= required for cite web, cite podcast, and cite mailinglist table.insert( z.message_tail, { set_error( 'cite_web_url', {}, true ) } ); end

-- do we have |accessdate= without either |url= or |chapter-url=? if is_set(AccessDate) and not is_set(ChapterURL)then -- ChapterURL may be set when URL is not set; table.insert( z.message_tail, { set_error( 'accessdate_missing_url', {}, true ) } ); AccessDate = ; end end

local OriginalURL, OriginalURLorigin, OriginalFormat, OriginalAccess; DeadURL = DeadURL:lower(); -- used later when assembling archived text if is_set( ArchiveURL ) then if is_set (ChapterURL) then -- URL not set so if chapter-url is set apply archive url to it OriginalURL = ChapterURL; -- save copy of source chapter's url for archive text OriginalURLorigin = ChapterURLorigin; -- name of chapter-url parameter for error messages OriginalFormat = ChapterFormat; -- and original |format= if 'no' ~= DeadURL then ChapterURL = ArchiveURL -- swap-in the archive's url ChapterURLorigin = A:ORIGIN('ArchiveURL') -- name of archive-url parameter for error messages ChapterFormat = ArchiveFormat or ; -- swap in archive's format end elseif is_set (URL) then OriginalURL = URL; -- save copy of original source URL OriginalURLorigin = URLorigin; -- name of url parameter for error messages OriginalFormat = Format; -- and original |format= OriginalAccess = UrlAccess; if 'no' ~= DeadURL then -- if URL set then archive-url applies to it URL = ArchiveURL -- swap-in the archive's url URLorigin = A:ORIGIN('ArchiveURL') -- name of archive url parameter for error messages Format = ArchiveFormat or ; -- swap in archive's format UrlAccess = nil; -- restricted access levels do not make sense for archived urls end

		end

end

if in_array(config.CitationClass, {'web','news','journal', 'magazine', 'pressrelease', 'podcast', 'newsgroup', 'arxiv', 'biorxiv', 'citeseerx'}) or -- if any of the 'periodical' cites except encyclopedia ('citation' == config.CitationClass and is_set (Periodical) and not is_set (Encyclopedia)) then local chap_param; if is_set (Chapter) then -- get a parameter name from one of these chapter related meta-parameters chap_param = A:ORIGIN ('Chapter') elseif is_set (TransChapter) then chap_param = A:ORIGIN ('TransChapter') elseif is_set (ChapterURL) then chap_param = A:ORIGIN ('ChapterURL') elseif is_set (ScriptChapter) then chap_param = A:ORIGIN ('ScriptChapter') else is_set (ChapterFormat) chap_param = A:ORIGIN ('ChapterFormat') end

if is_set (chap_param) then -- if we found one table.insert( z.message_tail, { set_error( 'chapter_ignored', {chap_param}, true ) } ); -- add error message Chapter = ; -- and set them to empty string to be safe with concatenation TransChapter = ; ChapterURL = ; ScriptChapter = ; ChapterFormat = ; end else -- otherwise, format chapter / article title local no_quotes = false; -- default assume that we will be quoting the chapter parameter value if is_set (Contribution) and 0 < #c then -- if this is a contribution with contributor(s) if in_array (Contribution:lower(), cfg.keywords.contribution) then -- and a generic contribution title no_quotes = true; -- then render it unquoted end end

Chapter = format_chapter_title (ScriptChapter, Chapter, TransChapter, ChapterURL, ChapterURLorigin, no_quotes); -- Contribution is also in Chapter if is_set (Chapter) then Chapter = Chapter .. ChapterFormat ; if 'map' == config.CitationClass and is_set (TitleType) then Chapter = Chapter .. ' ' .. TitleType; -- map annotation here; not after title end Chapter = Chapter.. sepc .. ' '; elseif is_set (ChapterFormat) then -- |chapter= not set but |chapter-format= is so ... Chapter = ChapterFormat .. sepc .. ' '; -- ... ChapterFormat has error message, we want to see it end end

-- Format main title. if is_set(TitleLink) and is_set(Title) then Title = "" .. Title .. "" end

if in_array(config.CitationClass, {'web', 'news', 'journal', 'magazine', 'pressrelease', 'podcast', 'newsgroup', 'mailinglist', 'interview', 'arxiv', 'biorxiv', 'citeseerx'}) or ('citation' == config.CitationClass and is_set (Periodical) and not is_set (Encyclopedia)) or ('map' == config.CitationClass and is_set (Periodical)) then -- special case for cite map when the map is in a periodical treat as an article Title = kern_quotes (Title); -- if necessary, separate title's leading and trailing quote marks from Module provided quote marks Title = wrap_style ('quoted-title', Title); Title = script_concatenate (Title, ScriptTitle); -- tags, lang atribute, categorization, etc; must be done after title is wrapped TransTitle= wrap_style ('trans-quoted-title', TransTitle ); elseif 'report' == config.CitationClass then -- no styling for cite report Title = script_concatenate (Title, ScriptTitle); -- tags, lang atribute, categorization, etc; must be done after title is wrapped TransTitle= wrap_style ('trans-quoted-title', TransTitle ); -- for cite report, use this form for trans-title else Title = wrap_style ('italic-title', Title); Title = script_concatenate (Title, ScriptTitle); -- tags, lang atribute, categorization, etc; must be done after title is wrapped TransTitle = wrap_style ('trans-italic-title', TransTitle); end

local TransError = ""; if is_set(TransTitle) then if is_set(Title) then TransTitle = " " .. TransTitle; else TransError = " " .. set_error( 'trans_missing_title', {'title'} ); end end

if is_set(Title) then if not is_set(TitleLink) and is_set(URL) then

Title = external_link( URL, Title, URLorigin, UrlAccess ) .. TransTitle .. TransError .. Format; -- this experiment hidden 2016-04-10; see Help_talk:Citation_Style_1#Recycled_urls -- local temp_title = external_link( URL, Title, URLorigin ) .. TransError .. Format; -- do this so we get error message even if url is usurped no archive -- if in_array (DeadURL, {'unfit no archive', 'usurped no archive'}) then -- when url links to inappropriate location and there is no archive of original source available -- local err_msg -- if temp_title:match ('%[%S+%s+(.+)%](<.+)') then -- if there is an error message -- Title, err_msg = temp_title:match ('%[%S+%s+(.+)%](<.+)'); -- strip off external link; TODO: find a better to do this -- Title = Title .. (err_msg or ); -- end -- else -- Title = temp_title; -- end

URL = ; -- unset these because no longer needed Format = ""; else Title = Title .. TransTitle .. TransError; end end

if is_set(Place) then Place = " " .. wrap_msg ('written', Place, use_lowercase) .. sepc .. " "; end

if is_set (Conference) then if is_set (ConferenceURL) then Conference = external_link( ConferenceURL, Conference, ConferenceURLorigin, nil ); end Conference = sepc .. " " .. Conference .. ConferenceFormat; elseif is_set(ConferenceURL) then Conference = sepc .. " " .. external_link( ConferenceURL, nil, ConferenceURLorigin, nil ); end

if not is_set(Position) then local Minutes = A['Minutes']; local Time = A['Time'];

if is_set(Minutes) then if is_set (Time) then table.insert( z.message_tail, { set_error( 'redundant_parameters', {wrap_style ('parameter', 'minutes') .. ' and ' .. wrap_style ('parameter', 'time')}, true ) } ); end Position = " " .. Minutes .. " " .. cfg.messages['minutes']; else if is_set(Time) then local TimeCaption = A['TimeCaption'] if not is_set(TimeCaption) then TimeCaption = cfg.messages['event']; if sepc ~= '.' then TimeCaption = TimeCaption:lower(); end end Position = " " .. TimeCaption .. " " .. Time; end end else Position = " " .. Position; At = ; end

Page, Pages, Sheet, Sheets = format_pages_sheets (Page, Pages, Sheet, Sheets, config.CitationClass, Periodical_origin, sepc, NoPP, use_lowercase, Mode);

At = is_set(At) and (sepc .. " " .. At) or ""; Position = is_set(Position) and (sepc .. " " .. Position) or ""; if config.CitationClass == 'map' then local Section = A['Section']; local Sections = A['Sections']; local Inset = A['Inset'];

if is_set( Inset ) then Inset = sepc .. " " .. wrap_msg ('inset', Inset, use_lowercase); end

if is_set( Sections ) then Section = sepc .. " " .. wrap_msg ('sections', Sections, use_lowercase); elseif is_set( Section ) then Section = sepc .. " " .. wrap_msg ('section', Section, use_lowercase); end At = At .. Inset .. Section; end

if is_set (Language) then Language = language_parameter (Language); -- format, categories, name from ISO639-1, etc else Language=""; -- language not specified so make sure this is an empty string; --[[ TODO: need to extract the wrap_msg from language_parameter so that we can solve parentheses bunching problem with Format/Language/TitleType ]] end

Others = is_set(Others) and (sepc .. " " .. Others) or "";

if is_set (Translators) then if 'mla' == Mode then Others = sepc .. ' Trans. ' .. Translators .. Others; else Others = sepc .. ' ' .. wrap_msg ('translated', Translators, use_lowercase) .. Others; end end if is_set (Interviewers) then Others = sepc .. ' ' .. wrap_msg ('interview', Interviewers, use_lowercase) .. Others; end

TitleNote = is_set(TitleNote) and (sepc .. " " .. TitleNote) or ""; if is_set (Edition) then if Edition:match ('%f[%a][Ee]d%.?$') or Edition:match ('%f[%a][Ee]dition$') then add_maint_cat ('extra_text', 'edition'); end if 'mla' == Mode then Edition = '. ' .. Edition .. ' ed.'; else Edition = " " .. wrap_msg ('edition', Edition); end else Edition = ; end

Series = is_set(Series) and (sepc .. " " .. Series) or ""; if 'mla' == Mode then -- not in brackets for mla OrigYear = is_set(OrigYear) and (". " .. OrigYear) or ""; else OrigYear = is_set(OrigYear) and (" [" .. OrigYear .. "]") or ""; end Agency = is_set(Agency) and (sepc .. " " .. Agency) or "";

Volume = format_volume_issue (Volume, Issue, config.CitationClass, Periodical_origin, sepc, use_lowercase, Mode);

------------------------------------ totally unrelated data if is_set(Via) then Via = " " .. wrap_msg ('via', Via); end

--[[ Subscription implies paywall; Registration does not. If both are used in a citation, the subscription required link note is displayed. There are no error messages for this condition.

]] if is_set (SubscriptionRequired) then SubscriptionRequired = sepc .. " " .. cfg.messages['subscription']; -- subscription required message elseif is_set (RegistrationRequired) then SubscriptionRequired = sepc .. " " .. cfg.messages['registration']; -- registration required message else SubscriptionRequired = ; -- either or both might be set to something other than yes true y end

if is_set(AccessDate) then local retrv_text = " " .. cfg.messages['retrieved']

AccessDate = nowrap_date (AccessDate); -- wrap in nowrap span if date in appropriate format if 'mla' == Mode then -- retrieved text not used in mla AccessDate = ' ' .. AccessDate; else if (sepc ~= ".") then retrv_text = retrv_text:lower() end -- if mode is cs2, lower case AccessDate = substitute (retrv_text, AccessDate); -- add retrieved text end AccessDate = substitute (cfg.presentation['accessdate'], {sepc, AccessDate}); -- allow editors to hide accessdates end

if is_set(ID) then ID = sepc .." ".. ID; end

  	if "thesis" == config.CitationClass and is_set(Docket) then

ID = sepc .." Docket ".. Docket .. ID; end

  	if "report" == config.CitationClass and is_set(Docket) then					-- for cite report when |docket= is set

ID = sepc .. ' ' .. Docket; -- overwrite ID even if |id= is set end

ID_list = build_id_list( ID_list, {IdAccessLevels=ID_access_levels, DoiBroken = DoiBroken, ASINTLD = ASINTLD, IgnoreISBN = IgnoreISBN, Embargo=Embargo, Class = Class} );

if is_set(URL) then URL = " " .. external_link( URL, nil, URLorigin, UrlAccess ); end

if is_set(Quote) then if Quote:sub(1,1) == '"' and Quote:sub(-1,-1) == '"' then -- if first and last characters of quote are quote marks Quote = Quote:sub(2,-2); -- strip them off end Quote = sepc .." " .. wrap_style ('quoted-text', Quote ); -- wrap in ... tags PostScript = ""; -- cs1|2 does not supply terminal punctuation when |quote= is set end

local Archived if is_set(ArchiveURL) then if not is_set(ArchiveDate) then ArchiveDate = set_error('archive_missing_date'); end if "no" == DeadURL then local arch_text = cfg.messages['archived']; if sepc ~= "." then arch_text = arch_text:lower() end Archived = sepc .. " " .. substitute( cfg.messages['archived-not-dead'], { external_link( ArchiveURL, arch_text, A:ORIGIN('ArchiveURL'), nil ) .. ArchiveFormat, ArchiveDate } ); if not is_set(OriginalURL) then Archived = Archived .. " " .. set_error('archive_missing_url'); end elseif is_set(OriginalURL) then -- DeadURL is empty, 'yes', 'true', 'y', 'unfit', 'usurped' local arch_text = cfg.messages['archived-dead']; if sepc ~= "." then arch_text = arch_text:lower() end if in_array (DeadURL, {'unfit', 'usurped', 'bot: unknown'}) then Archived = sepc .. " " .. 'Archived from the original on ' .. ArchiveDate; -- format already styled if 'bot: unknown' == DeadURL then add_maint_cat ('bot:_unknown'); -- and add a category if not already added else add_maint_cat ('unfit'); -- and add a category if not already added end else -- DeadURL is empty, 'yes', 'true', or 'y' Archived = sepc .. " " .. substitute( arch_text, { external_link( OriginalURL, cfg.messages['original'], OriginalURLorigin, OriginalAccess ) .. OriginalFormat, ArchiveDate } ); -- format already styled end else local arch_text = cfg.messages['archived-missing']; if sepc ~= "." then arch_text = arch_text:lower() end Archived = sepc .. " " .. substitute( arch_text, { set_error('archive_missing_url'), ArchiveDate } ); end elseif is_set (ArchiveFormat) then Archived = ArchiveFormat; -- if set and ArchiveURL not set ArchiveFormat has error message else Archived = "" end

local Lay = ; if is_set(LayURL) then if is_set(LayDate) then LayDate = " (" .. LayDate .. ")" end if is_set(LaySource) then LaySource = " – " .. safe_for_italics(LaySource) .. ""; else LaySource = ""; end if sepc == '.' then Lay = sepc .. " " .. external_link( LayURL, cfg.messages['lay summary'], A:ORIGIN('LayURL'), nil ) .. LayFormat .. LaySource .. LayDate else Lay = sepc .. " " .. external_link( LayURL, cfg.messages['lay summary']:lower(), A:ORIGIN('LayURL'), nil ) .. LayFormat .. LaySource .. LayDate end elseif is_set (LayFormat) then -- Test if |lay-format= is given without giving a |lay-url= Lay = sepc .. LayFormat; -- if set and LayURL not set, then LayFormat has error message end

if is_set(Transcript) then if is_set(TranscriptURL) then Transcript = external_link( TranscriptURL, Transcript, TranscriptURLorigin, nil ); end Transcript = sepc .. ' ' .. Transcript .. TranscriptFormat; elseif is_set(TranscriptURL) then Transcript = external_link( TranscriptURL, nil, TranscriptURLorigin, nil ); end

local Publisher; if is_set(PublicationDate) then PublicationDate = wrap_msg ('published', PublicationDate); end if is_set(PublisherName) then if is_set(PublicationPlace) then Publisher = sepc .. " " .. PublicationPlace .. ": " .. PublisherName .. PublicationDate; else Publisher = sepc .. " " .. PublisherName .. PublicationDate; end elseif is_set(PublicationPlace) then Publisher= sepc .. " " .. PublicationPlace .. PublicationDate; else Publisher = PublicationDate; end

-- Several of the above rely upon detecting this as nil, so do it last. if is_set(Periodical) then if is_set(Title) or is_set(TitleNote) then Periodical = sepc .. " " .. wrap_style ('italic-title', Periodical) else Periodical = wrap_style ('italic-title', Periodical) end end

--[[ Handle the oddity that is cite speech. This code overrides whatever may be the value assigned to TitleNote (through |department=) and forces it to be " (Speech)" so that the annotation directly follows the |title= parameter value in the citation rather than the |event= parameter value (if provided). ]] if "speech" == config.CitationClass then -- cite speech only TitleNote = " (Speech)"; -- annotate the citation if is_set (Periodical) then -- if Periodical, perhaps because of an included |website= or |journal= parameter if is_set (Conference) then -- and if |event= is set Conference = Conference .. sepc .. " "; -- then add appropriate punctuation to the end of the Conference variable before rendering end end end

-- Piece all bits together at last. Here, all should be non-nil. -- We build things this way because it is more efficient in LUA -- not to keep reassigning to the same string variable over and over.

local tcommon; local tcommon2; -- used for book cite when |contributor= is set

if in_array(config.CitationClass, {"journal","citation"}) and is_set(Periodical) then if is_set(Others) then Others = Others .. sepc .. " " end if 'mla' == Mode then tcommon = safe_join( {Conference, Periodical, Format, TitleType, Series, Language, Edition, Publisher, Agency, Volume}, sepc ); else tcommon = safe_join( {Others, Title, TitleNote, Conference, Periodical, Format, TitleType, Series, Language, Edition, Publisher, Agency, Volume}, sepc ); end elseif in_array(config.CitationClass, {"book","citation"}) and not is_set(Periodical) then -- special cases for book cites if is_set (Contributors) then -- when we are citing foreword, preface, introduction, etc tcommon = safe_join( {Title, TitleNote}, sepc ); -- author and other stuff will come after this and before tcommon2 if 'mla' == Mode then tcommon2 = safe_join( {Conference, Periodical, Format, TitleType, Series, Language, Volume, Edition, Publisher, Agency}, sepc ); else tcommon2 = safe_join( {Conference, Periodical, Format, TitleType, Series, Language, Volume, Others, Edition, Publisher, Agency}, sepc ); end elseif 'mla' == Mode then tcommon = safe_join( {TitleNote, Conference, Periodical, Format, TitleType, Series, Language, Volume, Publisher, Agency}, sepc ); else tcommon = safe_join( {Title, TitleNote, Conference, Periodical, Format, TitleType, Series, Language, Volume, Others, Edition, Publisher, Agency}, sepc ); end

elseif 'map' == config.CitationClass then -- special cases for cite map if is_set (Chapter) then -- map in a book; TitleType is part of Chapter tcommon = safe_join( {Title, Format, Edition, Scale, Series, Language, Cartography, Others, Publisher, Volume}, sepc ); elseif is_set (Periodical) then -- map in a periodical tcommon = safe_join( {Title, TitleType, Format, Periodical, Scale, Series, Language, Cartography, Others, Publisher, Volume}, sepc ); else -- a sheet or stand-alone map tcommon = safe_join( {Title, TitleType, Format, Edition, Scale, Series, Language, Cartography, Others, Publisher}, sepc ); end

elseif 'episode' == config.CitationClass then -- special case for cite episode tcommon = safe_join( {Title, TitleNote, TitleType, Series, Transcript, Language, Edition, Publisher}, sepc );

elseif ('news' == config.CitationClass) and ('mla' == Mode) then -- special case for cite news in MLA mode tcommon = safe_join( {Periodical, Format, TitleType, Series, Language, Edition, Agency}, sepc );

elseif ('web' == config.CitationClass) and ('mla' == Mode) then -- special case for cite web in MLA mode tcommon = safe_join( {Periodical, Format, TitleType, Series, Language, Edition, Publisher, Agency}, sepc );

else -- all other CS1 templates tcommon = safe_join( {Title, TitleNote, Conference, Periodical, Format, TitleType, Series, Language, Volume, Others, Edition, Publisher, Agency}, sepc ); end

if #ID_list > 0 then ID_list = safe_join( { sepc .. " ", table.concat( ID_list, sepc .. " " ), ID }, sepc ); else ID_list = ID; end

local idcommon = safe_join( { ID_list, URL, Archived, AccessDate, Via, SubscriptionRequired, Lay, Quote }, sepc ); local text; local pgtext = Position .. Sheet .. Sheets .. Page .. Pages .. At;

if is_set(Date) then if ('mla' == Mode) then if in_array (config.CitationClass, {'book', 'news', 'web'}) then Date = ', ' .. Date; -- origyear follows title in mla elseif 'journal' == config.CitationClass then Date = ', (' .. Date .. ')'; end elseif is_set (Authors) or is_set (Editors) then -- date follows authors or editors when authors not set Date = " (" .. Date ..")" .. OrigYear .. sepc .. " "; -- in paranetheses else -- neither of authors and editors set if (string.sub(tcommon,-1,-1) == sepc) then -- if the last character of tcommon is sepc Date = " " .. Date .. OrigYear; -- Date does not begin with sepc else Date = sepc .. " " .. Date .. OrigYear; -- Date begins with sepc end end end if is_set(Authors) then if (not is_set (Date)) or ('mla' == Mode) then -- when date is set it's in parentheses; no Authors termination Authors = terminate_name_list (Authors, sepc); -- when no date, terminate with 0 or 1 sepc and a space end if is_set(Editors) then local in_text = " "; local post_text = ""; if is_set(Chapter) and 0 == #c and 'mla' ~= Mode then in_text = in_text .. cfg.messages['in'] .. " " if (sepc ~= '.') then in_text = in_text:lower() end -- lowercase for cs2 elseif is_set(Chapter) and 'mla' == Mode then if EditorCount <= 1 then in_text = '. Ed. '; else in_text = '. Eds. '; end else if EditorCount <= 1 then post_text = ", " .. cfg.messages['editor']; else post_text = ", " .. cfg.messages['editors']; end end Editors = terminate_name_list (in_text .. Editors .. post_text, sepc); -- terminate with 0 or 1 sepc and a space end if is_set (Contributors) then -- book cite and we're citing the intro, preface, etc local by_text = sepc .. ' ' .. cfg.messages['by'] .. ' '; if (sepc ~= '.') then by_text = by_text:lower() end -- lowercase for cs2 Authors = by_text .. Authors; -- author follows title so tweak it here if is_set (Editors) and is_set (Date) and ('mla' ~= Mode) then -- when Editors make sure that Authors gets terminated Authors = terminate_name_list (Authors, sepc); -- terminate with 0 or 1 sepc and a space end if (not is_set (Date)) or ('mla' == Mode) then -- when date is set it's in parentheses; no Contributors termination Contributors = terminate_name_list (Contributors, sepc); -- terminate with 0 or 1 sepc and a space end if 'mla' == Mode then text = safe_join( {Contributors, Chapter, tcommon, OrigYear, Authors, Place, Others, Editors, tcommon2, Date, pgtext, idcommon }, sepc ); else text = safe_join( {Contributors, Date, Chapter, tcommon, Authors, Place, Editors, tcommon2, pgtext, idcommon }, sepc ); end elseif 'mla' == Mode then tcommon = tcommon .. Date; -- hack to avoid duplicate separators text = safe_join( {Authors, Chapter, Title, OrigYear, Others, Editors, Edition, Place, tcommon, pgtext, idcommon }, sepc ); else text = safe_join( {Authors, Date, Chapter, Place, Editors, tcommon, pgtext, idcommon }, sepc ); end elseif is_set(Editors) then if is_set(Date) then if EditorCount <= 1 then Editors = Editors .. ", " .. cfg.messages['editor']; else Editors = Editors .. ", " .. cfg.messages['editors']; end else if EditorCount <= 1 then Editors = Editors .. " (" .. cfg.messages['editor'] .. ")" .. sepc .. " " else Editors = Editors .. " (" .. cfg.messages['editors'] .. ")" .. sepc .. " " end end if 'mla' == Mode then if in_array(config.CitationClass, {'journal', 'news', 'web'}) and is_set(Periodical) then text = safe_join( {Editors, Title, Place, tcommon, pgtext, Date, idcommon}, sepc ); else text = safe_join( {Editors, Chapter, Title, Place, tcommon, Date, pgtext, idcommon}, sepc ); end else text = safe_join( {Editors, Date, Chapter, Place, tcommon, pgtext, idcommon}, sepc ); end elseif 'mla' == Mode then if in_array(config.CitationClass, {'journal', 'news', 'web'}) and is_set(Periodical) then text = safe_join( {Title, Place, tcommon, pgtext, Date, idcommon}, sepc ); else text = safe_join( {Chapter, Title, Place, tcommon, Date, pgtext, idcommon}, sepc ); end else if in_array(config.CitationClass, {"journal","citation"}) and is_set(Periodical) then text = safe_join( {Chapter, Place, tcommon, pgtext, Date, idcommon}, sepc ); else text = safe_join( {Chapter, Place, tcommon, Date, pgtext, idcommon}, sepc ); end end

if is_set(PostScript) and PostScript ~= sepc then text = safe_join( {text, sepc}, sepc ); --Deals with italics, spaces, etc. text = text:sub(1,-sepc:len()-1); end

text = safe_join( {text, PostScript}, sepc );

-- Now enclose the whole thing in a element local options = {};

if is_set(config.CitationClass) and config.CitationClass ~= "citation" then options.class = config.CitationClass; options.class = "citation " .. config.CitationClass; -- class=citation required for blue highlight when used with |ref= else options.class = "citation"; end

if is_set(Ref) and Ref:lower() ~= "none" then -- set reference anchor if appropriate local id = Ref if ('harv' == Ref ) then local namelist = {}; -- holds selected contributor, author, editor name list local year = first_set ({Year, anchor_year}, 2); -- Year first for legacy citations and for YMD dates that require disambiguation

if #c > 0 then -- if there is a contributor list namelist = c; -- select it elseif #a > 0 then -- or an author list namelist = a; elseif #e > 0 then -- or an editor list namelist = e; end if #namelist > 0 then -- if there are names in namelist id = anchor_id (namelist, year); -- go make the CITEREF anchor else id = ; -- unset end end options.id = id; end

if string.len(text:gsub("<span[^>/]*>(.-)", "%1"):gsub("%b<>","")) <= 2 then -- remove tags and other html-like markup; then get length of what remains z.error_categories = {}; text = set_error('empty_citation'); z.message_tail = {}; end

if is_set(options.id) then -- here we wrap the rendered citation in ... tags text = substitute (cfg.presentation['cite-id'], {mw.uri.anchorEncode(options.id), mw.text.nowiki(options.class), text}); -- when |ref= is set else text = substitute (cfg.presentation['cite'], {mw.text.nowiki(options.class), text}); -- all other cases end

text = text .. substitute (cfg.presentation['ocins'], {OCinSoutput}); -- append metadata to the citation

if #z.message_tail ~= 0 then text = text .. " "; for i,v in ipairs( z.message_tail ) do if is_set(v[1]) then if i == #z.message_tail then text = text .. error_comment( v[1], v[2] ); else text = text .. error_comment( v[1] .. "; ", v[2] ); end end end end

if #z.maintenance_cats ~= 0 then text = text .. ''; for _, v in ipairs( z.maintenance_cats ) do -- append maintenance categories text = text .. v .. ' (link)'; end text = text .. ''; -- maintenance mesages (realy just the names of the categories for now) end

no_tracking_cats = no_tracking_cats:lower(); if in_array(no_tracking_cats, {"", "no", "false", "n"}) then for _, v in ipairs( z.error_categories ) do text = text .. ; end for _, v in ipairs( z.maintenance_cats ) do -- append maintenance categories text = text .. ; end for _, v in ipairs( z.properties_cats ) do -- append maintenance categories text = text .. ; end end

return text end

--[[--------------------------< C S 1 . C I T A T I O N >------------------------------------------------------

This is used by templates such as to create the actual citation text.

]]

function cs1.citation(frame) Frame = frame; -- save a copy incase we need to display an error message in preview mode local pframe = frame:getParent() local validation, utilities, identifiers, metadata;

if nil ~= string.find (frame:getTitle(), 'sandbox', 1, true) then -- did the use sandbox version? cfg = mw.loadData ('Module:Citation/CS1/Configuration/sandbox'); -- load sandbox versions of support modules whitelist = mw.loadData ('Module:Citation/CS1/Whitelist/sandbox'); utilities = require ('Module:Citation/CS1/Utilities/sandbox'); validation = require ('Module:Citation/CS1/Date_validation/sandbox'); identifiers = require ('Module:Citation/CS1/Identifiers/sandbox'); metadata = require ('Module:Citation/CS1/COinS/sandbox');

else -- otherwise cfg = mw.loadData ('Module:Citation/CS1/Configuration'); -- load live versions of support modules whitelist = mw.loadData ('Module:Citation/CS1/Whitelist'); utilities = require ('Module:Citation/CS1/Utilities'); validation = require ('Module:Citation/CS1/Date_validation'); identifiers = require ('Module:Citation/CS1/Identifiers'); metadata = require ('Module:Citation/CS1/COinS'); end

utilities.set_selected_modules (cfg); -- so that functions in Utilities can see the cfg tables identifiers.set_selected_modules (cfg, utilities); -- so that functions in Identifiers can see the selected cfg tables and selected Utilities module validation.set_selected_modules (utilities); -- so that functions in Date validataion can see the selected Utilities module metadata.set_selected_modules (cfg, utilities); -- so that functions in COinS can see the selected cfg tables and selected Utilities module

dates = validation.dates; -- imported functions from Module:Citation/CS1/Date validation year_date_check = validation.year_date_check; reformat_dates = validation.reformat_dates; date_hyphen_to_dash = validation.date_hyphen_to_dash;

is_set = utilities.is_set; -- imported functions from Module:Citation/CS1/Utilities in_array = utilities.in_array; substitute = utilities.substitute; error_comment = utilities.error_comment; set_error = utilities.set_error; select_one = utilities.select_one; add_maint_cat = utilities.add_maint_cat; wrap_style = utilities.wrap_style; safe_for_italics = utilities.safe_for_italics; remove_wiki_link = utilities.remove_wiki_link;

z = utilities.z; -- table of error and category tables in Module:Citation/CS1/Utilities

extract_ids = identifiers.extract_ids; -- imported functions from Module:Citation/CS1/Identifiers build_id_list = identifiers.build_id_list; is_embargoed = identifiers.is_embargoed; extract_id_access_levels = identifiers.extract_id_access_levels;

make_coins_title = metadata.make_coins_title; -- imported functions from Module:Citation/CS1/COinS get_coins_pages = metadata.get_coins_pages; COinS = metadata.COinS;

local args = {}; -- table where we store all of the template's arguments local suggestions = {}; -- table where we store suggestions if we need to loadData them local error_text, error_state;

local config = {}; -- table to store parameters from the module for k, v in pairs( frame.args ) do config[k] = v; -- args[k] = v; -- debug tool that allows us to render a citation from module end

local capture; -- the single supported capture when matching unknown parameters using patterns for k, v in pairs( pframe.args ) do if v ~= then if not validate( k, config.CitationClass ) then error_text = ""; if type( k ) ~= 'string' then -- Exclude empty numbered parameters if v:match("%S+") ~= nil then error_text, error_state = set_error( 'text_ignored', {v}, true ); end elseif validate( k:lower(), config.CitationClass ) then error_text, error_state = set_error( 'parameter_ignored_suggest', {k, k:lower()}, true ); else if nil == suggestions.suggestions then -- if this table is nil then we need to load it if nil ~= string.find (frame:getTitle(), 'sandbox', 1, true) then -- did the use sandbox version? suggestions = mw.loadData( 'Module:Citation/CS1/Suggestions/sandbox' ); -- use the sandbox version else suggestions = mw.loadData( 'Module:Citation/CS1/Suggestions' ); -- use the live version end end for pattern, param in pairs (suggestions.patterns) do -- loop through the patterns to see if we can suggest a proper parameter capture = k:match (pattern); -- the whole match if no caputre in pattern else the capture if a match if capture then -- if the pattern matches param = substitute( param, capture ); -- add the capture to the suggested parameter (typically the enumerator) error_text, error_state = set_error( 'parameter_ignored_suggest', {k, param}, true ); -- set the error message end end if not is_set (error_text) then -- couldn't match with a pattern, is there an expicit suggestion? if suggestions.suggestions[ k:lower() ] ~= nil then error_text, error_state = set_error( 'parameter_ignored_suggest', {k, suggestions.suggestions[ k:lower() ]}, true ); else error_text, error_state = set_error( 'parameter_ignored', {k}, true ); v = ; -- unset value assigned to unrecognized parameters (this for the limited parameter lists) end end end if error_text ~= then table.insert( z.message_tail, {error_text, error_state} ); end end missing_pipe_check (v); -- do we think that there is a parameter that is missing a pipe?

args[k] = v; elseif args[k] ~= nil or (k == 'postscript') then args[k] = v; end end

for k, v in pairs( args ) do if 'string' == type (k) then -- don't evaluate positional parameters has_invisible_chars (k, v); end end return citation0( config, args) end

return cs1;