Nonbinary Wiki

Module:Citation/CS1/Identifiers: Difference between revisions

Module Discussion

Module:Citation/CS1/Identifiers (edit)

Revision as of 11:41, 21 May 2021

37,103 bytes added ,  3 years ago
m
15 revisions imported from wikipedia:Module:Citation/CS1/Identifiers: see Topic:Vtixlm0q28eo6jtf
m>Trappist the monk
(Synch from sandbox;)
 
(29 intermediate revisions by 11 users not shown)
Line 1: Line 1:
local identifiers = {};


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


local is_set, in_array, set_error, select_one, add_maint_cat, substitute; -- functions in Module:Citation/CS1/Utilities
local has_accept_as_written, is_set, in_array, set_message, select_one, -- functions in Module:Citation/CS1/Utilities
substitute, make_wikilink;


local z; -- table of tables defined in Module:Citation/CS1/Utilities
local z; -- table of tables defined in Module:Citation/CS1/Utilities
Line 13: Line 11:




--[[--------------------------< E X T E R N A L _ L I N K _ I D >----------------------------------------------
--[[--------------------------< P A G E   S C O P E   V A R I A B L E S >--------------------------------------


Formats a wiki style external link
declare variables here that have page-wide scope that are not brought in from other modules; that are created here and used here


]]
]]


local function external_link_id(options)
local auto_link_urls = {}; -- holds identifier URLs for those identifiers that can auto-link |title=
local url_string = options.id;
local ext_link;
if options.encode == true or options.encode == nil then
url_string = mw.uri.encode( url_string );
end
ext_link = mw.ustring.format ('[%s%s%s %s]', options.prefix, url_string, options.suffix or "", mw.text.nowiki(options.id));
if options.free then
ext_link = substitute (cfg.presentation['free to read'], ext_link); -- add the free-to-read lock
end
return mw.ustring.format( '[[%s|%s]]%s%s', options.link, options.label, options.separator or "&nbsp;", ext_link);


-- return mw.ustring.format( '[[%s|%s]]%s[%s%s%s %s]',
-- options.link, options.label, options.separator or "&nbsp;",
-- options.prefix, url_string, options.suffix or "",
-- mw.text.nowiki(options.id)
-- );
end


--============================<< H E L P E R  F U N C T I O N S >>============================================


--[[--------------------------< I N T E R N A L _ L I N K _ I D >----------------------------------------------
--[[--------------------------< W I K I D A T A _ A R T I C L E _ N A M E _ G E T >----------------------------


Formats a wiki style internal link
as an aid to internationalizing identifier-label wikilinks, gets identifier article names from Wikidata.


]]
returns :<lang code>:<article title> when <q> has an <article title> for <lang code>; nil else


local function internal_link_id(options)
for identifiers that do not have q, returns nil
return mw.ustring.format( '[[%s|%s]]%s[[%s%s%s|%s]]',
options.link, options.label, options.separator or "&nbsp;",
options.prefix, options.id, options.suffix or "",
mw.text.nowiki(options.id)
);
end


for wikis that do not have mw.wikibase installed, returns nil


--[[--------------------------< IS _ V A L I D _ I S X N >-----------------------------------------------------
]]


ISBN-10 and ISSN validator code calculates checksum across all isbn/issn digits including the check digit.
local function wikidata_article_name_get (q)
ISBN-13 is checked in check_isbn().
if not is_set (q) or (q and not mw.wikibase) then -- when no q number or when a q number but mw.wikibase not installed on this wiki
return nil; -- abandon
end


If the number is valid the result will be 0. Before calling this function, issbn/issn must be checked for length
local wd_article;
and stripped of dashes, spaces and other non-isxn characters.
local this_wiki_code = cfg.this_wiki_code; -- Wikipedia subdomain; 'en' for en.wikipedia.org


]]
wd_article = mw.wikibase.getSitelink (q, this_wiki_code .. 'wiki'); -- fetch article title from WD; nil when no title available at this wiki


local function is_valid_isxn (isxn_str, len)
if wd_article then
local temp = 0;
wd_article = table.concat ({':', this_wiki_code, ':', wd_article}); -- interwiki-style link without brackets if taken from WD; leading colon required
isxn_str = { isxn_str:byte(1, len) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39, 'X' → 0x58
len = len+1; -- adjust to be a loop counter
for i, v in ipairs( isxn_str ) do -- loop through all of the bytes and calculate the checksum
if v == string.byte( "X" ) then -- if checkdigit is X (compares the byte value of 'X' which is 0x58)
temp = temp + 10*( len - i ); -- it represents 10 decimal
else
temp = temp + tonumber( string.char(v) )*(len-i);
end
end
end
return temp % 11 == 0; -- returns true if calculation result is zero
 
return wd_article; -- article title from WD; nil else
end
end




--[[--------------------------< IS _ V A L I D _ I S X N  _ 1 3 >----------------------------------------------
--[[--------------------------< L I N K _ L A B E L _ M A K E >------------------------------------------------


ISBN-13 and ISMN validator code calculates checksum across all 13 isbn/ismn digits including the check digit.
common function to create identifier link label from handler table or from Wikidata
If the number is valid, the result will be 0. Before calling this function, isbn-13/ismn must be checked for length
and stripped of dashes, spaces and other non-isxn-13 characters.


returns the first available of
1. redirect from local wiki's handler table (if enabled)
2. Wikidata (if there is a Wikidata entry for this identifier in the local wiki's language)
3. label specified in the local wiki's handler table
]]
]]


local function is_valid_isxn_13 (isxn_str)
local function link_label_make (handler)
local temp=0;
local wd_article;
isxn_str = { isxn_str:byte(1, 13) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39
if not (cfg.use_identifier_redirects and is_set (handler.redirect)) then -- redirect has priority so if enabled and available don't fetch from Wikidata because expensive
for i, v in ipairs( isxn_str ) do
wd_article = wikidata_article_name_get (handler.q); -- if Wikidata has an article title for this wiki, get it;
temp = temp + (3 - 2*(i % 2)) * tonumber( string.char(v) ); -- multiply odd index digits by 1, even index digits by 3 and sum; includes check digit
end
end
return temp % 10 == 0; -- sum modulo 10 is zero when isbn-13/ismn is correct
return (cfg.use_identifier_redirects and is_set (handler.redirect) and handler.redirect) or wd_article or handler.link;
end
end




--[[--------------------------< C H E C K _ I S B N >------------------------------------------------------------
--[[--------------------------< E X T E R N A L _ L I N K _ I D >----------------------------------------------


Determines whether an ISBN string is valid
Formats a wiki-style external link


]]
]]


local function check_isbn( isbn_str )
local function external_link_id (options)
if nil ~= isbn_str:match("[^%s-0-9X]") then
local url_string = options.id;
return false, 'invalid character'; -- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
local ext_link;
local this_wiki_code = cfg.this_wiki_code; -- Wikipedia subdomain; 'en' for en.wikipedia.org
local wd_article; -- article title from Wikidata
if options.encode == true or options.encode == nil then
url_string = mw.uri.encode (url_string, 'PATH');
end
end
isbn_str = isbn_str:gsub( "-", "" ):gsub( " ", "" ); -- remove hyphens and spaces
 
local len = isbn_str:len();
if options.auto_link and is_set (options.access) then
auto_link_urls[options.auto_link] = table.concat ({options.prefix, url_string, options.suffix});
if len ~= 10 and len ~= 13 then
return false, 'length'; -- fail if incorrect length
end
end


if len == 10 then
ext_link = mw.ustring.format ('[%s%s%s %s]', options.prefix, url_string, options.suffix or "", mw.text.nowiki (options.id));
if isbn_str:match( "^%d*X?$" ) == nil then -- fail if isbn_str has 'X' anywhere but last position
if is_set (options.access) then
return false, 'invalid form';
ext_link = substitute (cfg.presentation['ext-link-access-signal'], {cfg.presentation[options.access].class, cfg.presentation[options.access].title, ext_link}); -- add the free-to-read / paywall lock
end
return is_valid_isxn(isbn_str, 10), 'checksum';
else
if isbn_str:match( "^%d+$" ) == nil then
return false, 'invalid character'; -- fail if isbn13 is not all digits
end
if isbn_str:match( "^97[89]%d*$" ) == nil then
return false, 'invalid prefix'; -- fail when isbn13 does not begin with 978 or 979
end
return is_valid_isxn_13 (isbn_str), 'checksum';
end
end
return table.concat ({
make_wikilink (link_label_make (options), options.label), -- redirect, Wikidata link, or locally specified link (in that order)
options.separator or '&nbsp;',
ext_link
});
end
end




--[[--------------------------< I S M N >----------------------------------------------------------------------
--[[--------------------------< I N T E R N A L _ L I N K _ I D >----------------------------------------------
 
Formats a wiki-style internal link


Determines whether an ISMN string is valid. Similar to isbn-13, ismn is 13 digits begining 979-0-... and uses the
TODO: Does not currently need to support options.access, options.encode, auto-linking and COinS (as in external_link_id),
same check digit calculations.  See http://www.ismn-international.org/download/Web_ISMN_Users_Manual_2008-6.pdf
but may be needed in the future for :m:Interwiki_map custom-prefixes like :arxiv:, :bibcode:, :DOI:, :hdl:, :ISSN:,
section 2, pages 9–12.
:JSTOR:, :Openlibrary:, :PMID:, :RFC:.


]]
]]


local function ismn (id)
local function internal_link_id (options)
local handler = cfg.id_handlers['ISMN'];
local id = mw.ustring.gsub (options.id, '%d', cfg.date_names.local_digits); -- translate 'local' digits to Western 0-9
local text;
 
local valid_ismn = true;
return table.concat (
{
make_wikilink (link_label_make (options), options.label), -- wiki-link the identifier label
options.separator or '&nbsp;', -- add the separator
make_wikilink (
table.concat (
{
options.prefix,
id, -- translated to Western digits
options.suffix or ''
}),
substitute (cfg.presentation['bdi'], {'', mw.text.nowiki (options.id)}) -- bdi tags to prevent Latin script identifiers from being reversed at RTL language wikis
); -- nowiki because MediaWiki still has magic links for ISBN and the like; TODO: is it really required?
});
end
 


id=id:gsub( "[%s-–]", "" ); -- strip spaces, hyphens, and endashes from the ismn
--[[--------------------------< I S _ E M B A R G O E D >------------------------------------------------------


if 13 ~= id:len() or id:match( "^9790%d*$" ) == nil then -- ismn must be 13 digits and begin 9790
Determines if a PMC identifier's online version is embargoed. Compares the date in |pmc-embargo-date= against
valid_ismn = false;
today's date.  If embargo date is in the future, returns the content of |pmc-embargo-date=; otherwise, returns
else
an empty string because the embargo has expired or because |pmc-embargo-date= was not set in this cite.
valid_ismn=is_valid_isxn_13 (id); -- validate ismn
end


-- text = internal_link_id({link = handler.link, label = handler.label, -- use this (or external version) when there is some place to link to
]]
-- prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
text="[[" .. handler.link .. "|" .. handler.label .. "]]" .. handler.separator .. id; -- because no place to link to yet


if false == valid_ismn then
local function is_embargoed (embargo)
text = text .. ' ' .. set_error( 'bad_ismn' ) -- add an error message if the ismn is invalid
if is_set (embargo) then
end
local lang = mw.getContentLanguage();
local good1, embargo_date, todays_date;
good1, embargo_date = pcall (lang.formatDate, lang, 'U', embargo);
todays_date = lang:formatDate ('U');
return text;
if good1 then -- if embargo date is a good date
if tonumber (embargo_date) >= tonumber (todays_date) then -- is embargo date is in the future?
return embargo; -- still embargoed
else
set_message ('maint_pmc_embargo'); -- embargo has expired; add main cat
return ''; -- unset because embargo has expired
end
end
end
return ''; -- |pmc-embargo-date= not set return empty string
end
end




--[[--------------------------< I S S N >----------------------------------------------------------------------
--[=[-------------------------< I S _ V A L I D _ B I O R X I V _ D A T E >------------------------------------
 
returns true if:
2019-12-11T00:00Z <= biorxiv_date < today + 2 days
The dated form of biorxiv identifier has a start date of 2019-12-11.  The Unix timestamp for that date is {{#time:U|2019-12-11}} = 1576022400
 
biorxiv_date is the date provided in those |biorxiv= parameter values that are dated at time 00:00:00 UTC
today is the current date at time 00:00:00 UTC plus 48 hours
if today is 2015-01-01T00:00:00 then
adding 24 hours gives 2015-01-02T00:00:00 – one second more than today
adding 24 hours gives 2015-01-03T00:00:00 – one second more than tomorrow


Validate and format an issn.  This code fixes the case where an editor has included an ISSN in the citation but
This function does not work if it is fed month names for languages other than English.  Wikimedia #time: parser
has separated the two groups of four digits with a space.  When that condition occurred, the resulting link looked
apparently doesn't understand non-English date month names. This function will always return false when the date
like this:
contains a non-English month name because good1 is false after the call to lang.formatDate().  To get around that
call this function with YYYY-MM-DD format dates.


|issn=0819 4327 gives: [http://www.worldcat.org/issn/0819 4327 0819 4327] -- can't have spaces in an external link
]=]
This code now prevents that by inserting a hyphen at the issn midpoint.  It also validates the issn for length
and makes sure that the checkdigit agrees with the calculated value.  Incorrect length (8 digits), characters
other than 0-9 and X, or checkdigit / calculated value mismatch will all cause a check issn error message.  The
issn is always displayed with a hyphen, even if the issn was given as a single group of 8 digits.


]]
local function is_valid_biorxiv_date (biorxiv_date)
local good1, good2;
local biorxiv_ts, tomorrow_ts; -- to hold Unix timestamps representing the dates
local lang_object = mw.getContentLanguage();


local function issn(id, e)
good1, biorxiv_ts = pcall (lang_object.formatDate, lang_object, 'U', biorxiv_date); -- convert biorxiv_date value to Unix timestamp
local issn_copy = id; -- save a copy of unadulterated issn; use this version for display if issn does not validate
good2, tomorrow_ts = pcall (lang_object.formatDate, lang_object, 'U', 'today + 2 days' ); -- today midnight + 2 days is one second more than all day tomorrow
local handler;
local text;
local valid_issn = true;
if e then
if good1 and good2 then -- lang.formatDate() returns a timestamp in the local script which tonumber() may not understand
handler = cfg.id_handlers['EISSN'];
biorxiv_ts = tonumber (biorxiv_ts) or lang_object:parseFormattedNumber (biorxiv_ts); -- convert to numbers for the comparison;
tomorrow_ts = tonumber (tomorrow_ts) or lang_object:parseFormattedNumber (tomorrow_ts);
else
else
handler = cfg.id_handlers['ISSN'];
return false; -- one or both failed to convert to Unix timestamp
end
end


id=id:gsub( "[%s-–]", "" ); -- strip spaces, hyphens, and endashes from the issn
return ((1576022400 <= biorxiv_ts) and (biorxiv_ts < tomorrow_ts)) -- 2012-12-11T00:00Z <= biorxiv_date < tomorrow's date
end
 
 
--[[--------------------------< IS _ V A L I D _ I S X N >-----------------------------------------------------


if 8 ~= id:len() or nil == id:match( "^%d*X?$" ) then -- validate the issn: 8 digits long, containing only 0-9 or X in the last position
ISBN-10 and ISSN validator code calculates checksum across all ISBN/ISSN digits including the check digit.
valid_issn=false; -- wrong length or improper character
ISBN-13 is checked in isbn().
else
 
valid_issn=is_valid_isxn(id, 8); -- validate issn
If the number is valid the result will be 0. Before calling this function, ISBN/ISSN must be checked for length
and stripped of dashes, spaces and other non-ISxN characters.
 
]]
 
local function is_valid_isxn (isxn_str, len)
local temp = 0;
isxn_str = { isxn_str:byte(1, len) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39, 'X' → 0x58
len = len + 1; -- adjust to be a loop counter
for i, v in ipairs (isxn_str) do -- loop through all of the bytes and calculate the checksum
if v == string.byte ("X" ) then -- if checkdigit is X (compares the byte value of 'X' which is 0x58)
temp = temp + 10 * (len - i); -- it represents 10 decimal
else
temp = temp + tonumber (string.char (v) )*(len-i);
end
end
end
return temp % 11 == 0; -- returns true if calculation result is zero
end
--[[--------------------------< IS _ V A L I D _ I S X N _ 1 3 >-----------------------------------------------
ISBN-13 and ISMN validator code calculates checksum across all 13 ISBN/ISMN digits including the check digit.
If the number is valid, the result will be 0. Before calling this function, ISBN-13/ISMN must be checked for length
and stripped of dashes, spaces and other non-ISxN-13 characters.


if true == valid_issn then
]]
id = string.sub( id, 1, 4 ) .. "-" .. string.sub( id, 5 ); -- if valid, display correctly formatted version
 
else
local function is_valid_isxn_13 (isxn_str)
id = issn_copy; -- if not valid, use the show the invalid issn with error message
local temp=0;
isxn_str = { isxn_str:byte(1, 13) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39
for i, v in ipairs (isxn_str) do
temp = temp + (3 - 2*(i % 2)) * tonumber (string.char (v) ); -- multiply odd index digits by 1, even index digits by 3 and sum; includes check digit
end
end
return temp % 10 == 0; -- sum modulo 10 is zero when ISBN-13/ISMN is correct
text = external_link_id({link = handler.link, label = handler.label,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
if false == valid_issn then
text = text .. ' ' .. set_error( 'bad_issn', e and 'e' or '' ) -- add an error message if the issn is invalid
end
return text
end
end




--[[--------------------------< A M A Z O N >------------------------------------------------------------------
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------


Formats a link to Amazon. Do simple error checking: asin must be mix of 10 numeric or uppercase alpha
LCCN normalization (http://www.loc.gov/marc/lccn-namespace.html#normalization)
characters. If a mix, first character must be uppercase alpha; if all numeric, asins must be 10-digit
1. Remove all blanks.
isbn. If 10-digit isbn, add a maintenance category so a bot or awb script can replace |asin= with |isbn=.
2. If there is a forward slash (/) in the string, remove it, and remove all characters to the right of the forward slash.
Error message if not 10 characters, if not isbn10, if mixed and first character is a digit.
3. If there is a hyphen in the string:
a. Remove it.
b. Inspect the substring following (to the right of) the (removed) hyphen. Then (and assuming that steps 1 and 2 have been carried out):
1. All these characters should be digits, and there should be six or less. (not done in this function)
2. If the length of the substring is less than 6, left-fill the substring with zeroes until the length is six.


Returns a normalized LCCN for lccn() to validate.  There is no error checking (step 3.b.1) performed in this function.
]]
]]


local function amazon(id, domain)
local function normalize_lccn (lccn)
local err_cat = ""
lccn = lccn:gsub ("%s", ""); -- 1. strip whitespace


if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
if nil ~= string.find (lccn, '/') then
err_cat = ' ' .. set_error ('bad_asin'); -- asin is not a mix of 10 uppercase alpha and numeric characters
lccn = lccn:match ("(.-)/"); -- 2. remove forward slash and all character to the right of it
else
end
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
 
if check_isbn( id ) then -- see if asin value is isbn10
local prefix
add_maint_cat ('ASIN');
local suffix
elseif not is_set (err_cat) then
prefix, suffix = lccn:match ("(.+)%-(.+)"); -- 3.a remove hyphen by splitting the string into prefix and suffix
err_cat =  ' ' .. set_error ('bad_asin'); -- asin is not isbn10
 
end
if nil ~= suffix then -- if there was a hyphen
elseif not id:match("^%u[%d%u]+$") then
suffix = string.rep("0", 6-string.len (suffix)) .. suffix; -- 3.b.2 left fill the suffix with 0s if suffix length less than 6
err_cat = ' ' .. set_error ('bad_asin'); -- asin doesn't begin with uppercase alpha
lccn = prefix..suffix; -- reassemble the LCCN
end
end
end
if not is_set(domain) then
domain = "com";
return lccn;
elseif in_array (domain, {'jp', 'uk'}) then -- Japan, United Kingdom
domain = "co." .. domain;
elseif in_array (domain, {'au', 'br', 'mx'}) then -- Australia, Brazil, Mexico
domain = "com." .. domain;
end
end
local handler = cfg.id_handlers['ASIN'];
return external_link_id({link=handler.link,
label=handler.label, prefix=handler.prefix .. domain .. "/dp/",
id=id, encode=handler.encode, separator = handler.separator}) .. err_cat;
end


--============================<< I D E N T I F I E R  F U N C T I O N S >>====================================


--[[--------------------------< A R X I V >--------------------------------------------------------------------
--[[--------------------------< A R X I V >--------------------------------------------------------------------
Line 265: Line 289:


format and error check arXiv identifier.  There are three valid forms of the identifier:
format and error check arXiv identifier.  There are three valid forms of the identifier:
the first form, valid only between date codes 9108 and 0703 is:
the first form, valid only between date codes 9107 and 0703, is:
arXiv:<archive>.<class>/<date code><number><version>
arXiv:<archive>.<class>/<date code><number><version>
where:
where:
<archive> is a string of alpha characters - may be hyphenated; no other punctuation
<archive> is a string of alpha characters - may be hyphenated; no other punctuation
<class> is a string of alpha characters - may be hyphenated; no other punctuation
<class> is a string of alpha characters - may be hyphenated; no other punctuation; not the same as |class= parameter which is not supported in this form
<date code> is four digits in the form YYMM where YY is the last two digits of the four-digit year and MM is the month number January = 01
<date code> is four digits in the form YYMM where YY is the last two digits of the four-digit year and MM is the month number January = 01
first digit of YY for this form can only 9 and 0
first digit of YY for this form can only 9 and 0
Line 289: Line 313:
]]
]]


local function arxiv (id, class)
local function arxiv (options)
local handler = cfg.id_handlers['ARXIV'];
local id = options.id;
local class = options.Class; -- TODO: lowercase?
local handler = options.handler;
local year, month, version;
local year, month, version;
local err_cat = '';
local err_cat = false; -- assume no error message
local text;
local text; -- output text
if id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%d$") or id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%dv%d+$") then -- test for the 9108-0703 format w/ & w/o version
if id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%d$") or id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%dv%d+$") then -- test for the 9107-0703 format with or without version
year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
year = tonumber(year);
year = tonumber (year);
month = tonumber(month);
month = tonumber (month);
if ((not (90 < year or 8 > year)) or (1 > month or 12 < month)) or -- if invalid year or invalid month
if ((not (90 < year or 8 > year)) or (1 > month or 12 < month)) or -- if invalid year or invalid month
((91 == year and 7 > month) or (7 == year and 3 < month)) then -- if years ok, are starting and ending months ok?
((91 == year and 7 > month) or (7 == year and 3 < month)) then -- if years ok, are starting and ending months ok?
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- set error message
err_cat = true; -- flag for error message
end
end
elseif id:match("^%d%d[01]%d%.%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%dv%d+$") then -- test for the 0704-1412 w/ & w/o version
 
elseif id:match("^%d%d[01]%d%.%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%dv%d+$") then -- test for the 0704-1412 with or without version
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
year = tonumber(year);
year = tonumber (year);
month = tonumber(month);
month = tonumber (month);
if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or -- is year invalid or is month invalid? (doesn't test for future years)
if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or -- is year invalid or is month invalid? (doesn't test for future years)
((7 == year) and (4 > month)) then --or -- when year is 07, is month invalid (before April)?
((7 == year) and (4 > month)) then -- when year is 07, is month invalid (before April)?
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- set error message
err_cat = true; -- flag for error message
end
end
elseif id:match("^%d%d[01]%d%.%d%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%d%dv%d+$") then -- test for the 1501- format w/ & w/o version
 
elseif id:match("^%d%d[01]%d%.%d%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%d%dv%d+$") then -- test for the 1501- format with or without version
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d%d[v%d]*$");
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d%d[v%d]*$");
year = tonumber(year);
year = tonumber (year);
month = tonumber(month);
month = tonumber (month);
if ((15 > year) or (1 > month or 12 < month)) then -- is year invalid or is month invalid? (doesn't test for future years)
if ((15 > year) or (1 > month or 12 < month)) then -- is year invalid or is month invalid? (doesn't test for future years)
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- set error message
err_cat = true; -- flag for error message
end
end
else
else
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- arXiv id doesn't match any format
err_cat = true; -- not a recognized format; flag for error message
end
end


text = external_link_id({link = handler.link, label = handler.label,
if err_cat then
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, free=handler.free}) .. err_cat;
options.coins_list_t['ARXIV'] = nil; -- when error, unset so not included in COinS
end
err_cat = err_cat and table.concat ({' ', set_message ('err_bad_arxiv')}) or ''; -- set error message if flag is true
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;


if is_set (class) then
if is_set (class) then
class = ' [[' .. '//arxiv.org/archive/' .. class .. ' ' .. class .. ']]'; -- external link within square brackets, not wikilink
if id:match ('^%d+') then
text = table.concat ({text, ' [[//arxiv.org/archive/', class, ' ', class, ']]'}); -- external link within square brackets, not wikilink
else
text = table.concat ({text, ' ', set_message ('err_class_ignored')});
end
end
 
return text;
end
 
 
--[[--------------------------< B I B C O D E >--------------------------------------------------------------------
 
Validates (sort of) and formats a bibcode ID.
 
Format for bibcodes is specified here: http://adsabs.harvard.edu/abs_doc/help_pages/data.html#bibcodes
 
But, this: 2015arXiv151206696F is apparently valid so apparently, the only things that really matter are length, 19 characters
and first four digits must be a year.  This function makes these tests:
length must be 19 characters
characters in position
1–4 must be digits and must represent a year in the range of 1000 – next year
5 must be a letter
6–8 must be letter, digit, ampersand, or dot (ampersand cannot directly precede a dot; &. )
9–18 must be letter, digit, or dot
19 must be a letter or dot
 
]]
 
local function bibcode (options)
local id = options.id;
local access = options.access;
local handler = options.handler;
local err_type;
local year;
 
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode,
access = access});
if 19 ~= id:len() then
err_type = cfg.err_msg_supl.length;
else
else
class = ''; -- empty string for concatenation
year = id:match ("^(%d%d%d%d)[%a][%w&%.][%w&%.][%w&%.][%w.]+[%a%.]$");
if not year then -- if nil then no pattern match
err_type = cfg.err_msg_supl.value; -- so value error
else
local next_year = tonumber (os.date ('%Y')) + 1; -- get the current year as a number and add one for next year
year = tonumber (year); -- convert year portion of bibcode to a number
if (1000 > year) or (year > next_year) then
err_type = cfg.err_msg_supl.year; -- year out of bounds
end
if id:find('&%.') then
err_type = cfg.err_msg_supl.journal; -- journal abbreviation must not have '&.' (if it does it's missing a letter)
end
end
end
end
 
return text .. class;
if is_set (err_type) then -- if there was an error detected
text = text .. ' ' .. set_message ('err_bad_bibcode', {err_type});
options.coins_list_t['BIBCODE'] = nil; -- when error, unset so not included in COinS
 
end
return text;
end
end




--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------
--[[--------------------------< B I O R X I V >-----------------------------------------------------------------
 
Format bioRxiv ID and do simple error checking.  Before 2019-12-11, biorXiv IDs were 10.1101/ followed by exactly
6 digits.  After 2019-12-11, biorXiv IDs retained the six-digit identifier but prefixed that with a yyyy.mm.dd.
date and suffixed with an optional version identifier.


lccn normalization (http://www.loc.gov/marc/lccn-namespace.html#normalization)
The bioRxiv ID is the string of characters:
1. Remove all blanks.
https://doi.org/10.1101/078733 -> 10.1101/078733
2. If there is a forward slash (/) in the string, remove it, and remove all characters to the right of the forward slash.
or a date followed by a six-digit number followed by an optional version indicator 'v' and one or more digits:
3. If there is a hyphen in the string:
https://www.biorxiv.org/content/10.1101/2019.12.11.123456v2 -> 10.1101/2019.12.11.123456v2
a. Remove it.
b. Inspect the substring following (to the right of) the (removed) hyphen. Then (and assuming that steps 1 and 2 have been carried out):
see https://www.biorxiv.org/about-biorxiv
1. All these characters should be digits, and there should be six or less. (not done in this function)
2. If the length of the substring is less than 6, left-fill the substring with zeroes until the length is six.


Returns a normalized lccn for lccn() to validate.  There is no error checking (step 3.b.1) performed in this function.
]]
]]


local function normalize_lccn (lccn)
local function biorxiv (options)
lccn = lccn:gsub ("%s", ""); -- 1. strip whitespace
local id = options.id;
local handler = options.handler;
local err_cat = true; -- flag; assume that there will be an error
local patterns = {
'^10.1101/%d%d%d%d%d%d$', -- simple 6-digit identifier (before 2019-12-11)
'^10.1101/(20[1-9]%d)%.([01]%d)%.([0-3]%d)%.%d%d%d%d%d%dv%d+$', -- y.m.d. date + 6-digit identifier + version (after 2019-12-11)
'^10.1101/(20[1-9]%d)%.([01]%d)%.([0-3]%d)%.%d%d%d%d%d%d$', -- y.m.d. date + 6-digit identifier (after 2019-12-11)
}
for _, pattern in ipairs (patterns) do -- spin through the patterns looking for a match
if id:match (pattern) then
local y, m, d = id:match (pattern); -- found a match, attempt to get year, month and date from the identifier
 
if m then -- m is nil when id is the six-digit form
if not is_valid_biorxiv_date (y .. '-' .. m .. '-' .. d) then -- validate the encoded date; TODO: don't ignore leap-year and actual month lengths ({{#time:}} is a poor date validator)
break; -- date fail; break out early so we don't unset the error message
end
end
err_cat = nil; -- we found a match so unset the error message
break; -- and done
end
end -- err_cat remains set here when no match


if nil ~= string.find (lccn,'/') then
if err_cat then
lccn = lccn:match ("(.-)/"); -- 2. remove forward slash and all character to the right of it
options.coins_list_t['BIORXIV'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator,
encode = handler.encode, access = handler.access}) .. (err_cat and (' ' .. set_message ('err_bad_biorxiv')) or '');
end


local prefix
--[[--------------------------< C I T E S E E R X >------------------------------------------------------------
local suffix
 
prefix, suffix = lccn:match ("(.+)%-(.+)"); -- 3.a remove hyphen by splitting the string into prefix and suffix
CiteSeerX use their own notion of "doi" (not to be confused with the identifiers resolved via doi.org).
 
The description of the structure of this identifier can be found at Help_talk:Citation_Style_1/Archive_26#CiteSeerX_id_structure
]]


if nil ~= suffix then -- if there was a hyphen
local function citeseerx (options)
suffix=string.rep("0", 6-string.len (suffix)) .. suffix; -- 3.b.2 left fill the suffix with 0s if suffix length less than 6
local id = options.id;
lccn=prefix..suffix; -- reassemble the lccn
local handler = options.handler;
end
local matched;
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode,
access = handler.access});
return lccn;
matched = id:match ("^10%.1%.1%.[1-9]%d?%d?%d?%.[1-9]%d?%d?%d?$");
if not matched then
text = text .. ' ' .. set_message ('err_bad_citeseerx' );
options.coins_list_t['CITESEERX'] = nil; -- when error, unset so not included in COinS
end
end
return text;
end
--[[--------------------------< D O I >------------------------------------------------------------------------


Formats a DOI and checks for DOI errors.


--[[--------------------------< L C C N >----------------------------------------------------------------------
DOI names contain two parts: prefix and suffix separated by a forward slash.
Prefix: directory indicator '10.' followed by a registrant code
Suffix: character string of any length chosen by the registrant


Format LCCN link and do simple error checking.  LCCN is a character string 8-12 characters long. The length of
This function checks a DOI name for: prefix/suffix.  If the DOI name contains spaces or endashes, or, if it ends
the LCCN dictates the character type of the first 1-3 characters; the rightmost eight are always digits.
with a period or a comma, this function will emit a bad_doi error message.
http://info-uri.info/registry/OAIHandler?verb=GetRecord&metadataPrefix=reg&identifier=info:lccn/


length = 8 then all digits
DOI names are case-insensitive and can incorporate any printable Unicode characters so the test for spaces, endash,
length = 9 then lccn[1] is lower case alpha
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
length = 10 then lccn[1] and lccn[2] are both lower case alpha or both digits
if ever used in DOI names.
length = 11 then lccn[1] is lower case alpha, lccn[2] and lccn[3] are both lower case alpha or both digits
length = 12 then lccn[1] and lccn[2] are both lower case alpha


]]
]]


local function lccn(lccn)
local function doi (options)
local handler = cfg.id_handlers['LCCN'];
local id = options.id;
local err_cat = ''; -- presume that LCCN is valid
local inactive = options.DoiBroken
local id = lccn; -- local copy of the lccn
local access = options.access;
local ignore_invalid = options.accept;
local handler = options.handler;
local err_cat;


id = normalize_lccn (id); -- get canonical form (no whitespace, hyphens, forward slashes)
local text;
local len = id:len(); -- get the length of the lccn
if is_set (inactive) then
local inactive_year = inactive:match("%d%d%d%d") or ''; -- try to get the year portion from the inactive date
local inactive_month, good;


if 8 == len then
if is_set (inactive_year) then
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits)
if 4 < inactive:len() then -- inactive date has more than just a year (could be anything)
err_cat = ' ' .. set_error( 'bad_lccn' ); -- set an error message
local lang_obj = mw.getContentLanguage(); -- get a language object for this wiki
good, inactive_month = pcall (lang_obj.formatDate, lang_obj, 'F', inactive); -- try to get the month name from the inactive date
if not good then
inactive_month = nil; -- something went wrong so make sure this is unset
end
end
else
inactive_year = nil; -- |doi-broken-date= has something but it isn't a date
end
end
elseif 9 == len then -- LCCN should be adddddddd
if nil == id:match("%l%d%d%d%d%d%d%d%d") then -- does it match our pattern?
if is_set (inactive_year) and is_set (inactive_month) then
err_cat = ' ' .. set_error( 'bad_lccn' ); -- set an error message
set_message ('maint_doi_inactive_dated', {inactive_year, inactive_month, ' '});
elseif is_set (inactive_year) then
set_message ('maint_doi_inactive_dated', {inactive_year, '', ''});
else
set_message ('maint_doi_inactive');
end
end
elseif 10 == len then -- LCCN should be aadddddddd or dddddddddd
inactive = " (" .. cfg.messages['inactive'] .. ' ' .. inactive .. ')';
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits) ...
end
if nil == id:match("^%l%l%d%d%d%d%d%d%d%d") then -- ... see if it matches our pattern
 
err_cat = ' ' .. set_error( 'bad_lccn' ); -- no match, set an error message
local registrant = id:match ('^10%.([^/]+)/[^%s–]-[^%.,]$'); -- registrant set when DOI has the proper basic form
local registrant_err_patterns = { -- these patterns are for code ranges that are not supported
'^[^1-3]%d%d%d%d%.%d%d*$', -- 5 digits with subcode (0xxxx, 40000+); accepts: 10000–39999
'^[^1-5]%d%d%d%d$', -- 5 digits without subcode (0xxxx, 60000+); accepts: 10000–59999
'^[^1-9]%d%d%d%.%d%d*$', -- 4 digits with subcode (0xxx); accepts: 1000–9999
'^[^1-9]%d%d%d$', -- 4 digits without subcode (0xxx); accepts: 1000–9999
'^%d%d%d%d%d%d+', -- 6 or more digits
'^%d%d?%d?$', -- less than 4 digits without subcode (with subcode is legitimate)
'^5555$', -- test registrant will never resolve
'[^%d%.]', -- any character that isn't a digit or a dot
}
 
if not ignore_invalid then
if registrant then -- when DOI has proper form
for i, pattern in ipairs (registrant_err_patterns) do -- spin through error patterns
if registrant:match (pattern) then -- to validate registrant codes
err_cat = ' ' .. set_message ('err_bad_doi'); -- when found, mark this DOI as bad
break; -- and done
end
end
end
end
else
elseif 11 == len then -- LCCN should be aaadddddddd or adddddddddd
err_cat = ' ' .. set_message ('err_bad_doi'); -- invalid directory or malformed
if not (id:match("^%l%l%l%d%d%d%d%d%d%d%d") or id:match("^%l%d%d%d%d%d%d%d%d%d%d")) then -- see if it matches one of our patterns
err_cat = ' ' .. set_error( 'bad_lccn' ); -- no match, set an error message
end
elseif 12 == len then -- LCCN should be aadddddddddd
if not id:match("^%l%l%d%d%d%d%d%d%d%d%d%d") then -- see if it matches our pattern
err_cat = ' ' .. set_error( 'bad_lccn' ); -- no match, set an error message
end
end
else
else
err_cat = ' ' .. set_error( 'bad_lccn' ); -- wrong length, set an error message
set_message ('maint_doi_ignore');
end
end


if not is_set (err_cat) and nil ~= lccn:find ('%s') then
if err_cat then
err_cat = ' ' .. set_error( 'bad_lccn' ); -- lccn contains a space, set an error message
options.coins_list_t['DOI'] = nil; -- when error, unset so not included in COinS
end
end
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access,
auto_link = not (err_cat or is_set (inactive) or ignore_invalid) and 'doi' or nil -- do not auto-link when |doi-broken-date= has a value or when there is a DOI error or (to play it safe, after all, auto-linking is not essential) when invalid DOIs are ignored
}) .. (inactive or '');


return external_link_id({link = handler.link, label = handler.label,
return text .. (err_cat and err_cat or ''); -- parentheses required
prefix=handler.prefix,id=lccn,separator=handler.separator, encode=handler.encode}) .. err_cat;
end
end




--[[--------------------------< P M I D >----------------------------------------------------------------------
--[[--------------------------< H D L >------------------------------------------------------------------------
 
Formats an HDL with minor error checking.
 
HDL names contain two parts: prefix and suffix separated by a forward slash.
Prefix: character string using any character in the UCS-2 character set except '/'
Suffix: character string of any length using any character in the UCS-2 character set chosen by the registrant
 
This function checks a HDL name for: prefix/suffix.  If the HDL name contains spaces, endashes, or, if it ends
with a period or a comma, this function will emit a bad_hdl error message.
 
HDL names are case-insensitive and can incorporate any printable Unicode characters so the test for endashes and
terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in HDLs.


Format PMID and do simple error checking.  PMIDs are sequential numbers beginning at 1 and counting up.  This
Query string parameters are named here: http://www.handle.net/proxy_servlet.html.  query strings are not displayed
code checks the PMID to see that it contains only digits and is less than test_limit; the value in local variable
but since '?' is an allowed character in an HDL, '?' followed by one of the query parameters is the only way we
test_limit will need to be updated periodically as more PMIDs are issued.
have to detect the query string so that it isn't URL-encoded with the rest of the identifier.


]]
]]


local function pmid(id)
local function hdl (options)
local test_limit = 30000000; -- update this value as PMIDs approach
local id = options.id;
local handler = cfg.id_handlers['PMID'];
local access = options.access;
local err_cat = ''; -- presume that PMID is valid
local handler = options.handler;
local query_params = { -- list of known query parameters from http://www.handle.net/proxy_servlet.html
'noredirect',
'ignore_aliases',
'auth',
'cert',
'index',
'type',
'urlappend',
'locatt',
'action',
}
if id:match("[^%d]") then -- if PMID has anything but digits
local hdl, suffix, param = id:match ('(.-)(%?(%a+).+)$'); -- look for query string
err_cat = ' ' .. set_error( 'bad_pmid' ); -- set an error message
local found;
else -- PMID is only digits
 
local id_num = tonumber(id); -- convert id to a number for range testing
if hdl then -- when there are query strings, this is the handle identifier portion
if 1 > id_num or test_limit < id_num then -- if PMID is outside test limit boundaries
for _, q in ipairs (query_params) do -- spin through the list of query parameters
err_cat = ' ' .. set_error( 'bad_pmid' ); -- set an error message
if param:match ('^' .. q) then -- if the query string begins with one of the parameters
found = true; -- announce a find
break; -- and stop looking
end
end
end
end
end
 
return external_link_id({link = handler.link, label = handler.label,
if found then
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode}) .. err_cat;
id = hdl; -- found so replace id with the handle portion; this will be URL-encoded, suffix will not
else
suffix = ''; -- make sure suffix is empty string for concatenation else
end
 
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, suffix = suffix, separator = handler.separator, encode = handler.encode, access = access})
 
if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then -- HDL must contain a forward slash, must not contain spaces, endashes, and must not end with period or comma
text = text .. ' ' .. set_message ('err_bad_hdl' );
options.coins_list_t['HDL'] = nil; -- when error, unset so not included in COinS
end
return text;
end
end




--[[--------------------------< I S _ E M B A R G O E D >------------------------------------------------------
--[[--------------------------< I S B N >----------------------------------------------------------------------


Determines if a PMC identifier's online version is embargoed. Compares the date in |embargo= against today's date.  If embargo date is
Determines whether an ISBN string is valid
in the future, returns the content of |embargo=; otherwise, returns and empty string because the embargo has expired or because
|embargo= was not set in this cite.


]]
]]


local function is_embargoed (embargo)
local function isbn (options)
if is_set (embargo) then
local isbn_str = options.id;
local lang = mw.getContentLanguage();
local ignore_invalid = options.accept;
local good1, embargo_date, good2, todays_date;
local handler = options.handler;
good1, embargo_date = pcall( lang.formatDate, lang, 'U', embargo );
 
good2, todays_date = pcall( lang.formatDate, lang, 'U' );
local function return_result (check, err_type) -- local function to handle the various returns
local ISBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
if good1 and good2 then -- if embargo date and today's date are good dates
prefix = handler.prefix, id = isbn_str, separator = handler.separator});
if tonumber( embargo_date ) >= tonumber( todays_date ) then -- is embargo date is in the future?
if ignore_invalid then -- if ignoring ISBN errors
return embargo; -- still embargoed
set_message ('maint_isbn_ignore'); -- add a maint category even when there is no error
else
else -- here when not ignoring
add_maint_cat ('embargo')
if not check then -- and there is an error
return ''; -- unset because embargo has expired
options.coins_list_t['ISBN'] = nil; -- when error, unset so not included in COinS
return ISBN .. set_message ('err_bad_isbn', {err_type}, false, ' '); -- display an error message
end
end
end
end
return ISBN;
end
if nil ~= isbn_str:match ('[^%s-0-9X]') then
return return_result (false, cfg.err_msg_supl.char); -- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
end
local id = isbn_str:gsub ('[%s-]', ''); -- remove hyphens and whitespace
local len = id:len();
if len ~= 10 and len ~= 13 then
return return_result (false, cfg.err_msg_supl.length); -- fail if incorrect length
end
if len == 10 then
if id:match ('^%d*X?$') == nil then -- fail if isbn_str has 'X' anywhere but last position
return return_result (false, cfg.err_msg_supl.form);
end
if not is_valid_isxn (id, 10) then -- test isbn-10 for numerical validity
return return_result (false, cfg.err_msg_supl.check); -- fail if isbn-10 is not numerically valid
end
if id:find ('^63[01]') then -- 630xxxxxxx and 631xxxxxxx are (apparently) not valid isbn group ids but are used by amazon as numeric identifiers (asin)
return return_result (false, cfg.err_msg_supl.group); -- fail if isbn-10 begins with 630/1
end
return return_result (true, cfg.err_msg_supl.check); -- pass if isbn-10 is numerically valid
else
if id:match ('^%d+$') == nil then
return return_result (false, cfg.err_msg_supl.char); -- fail if ISBN-13 is not all digits
end
if id:match ('^97[89]%d*$') == nil then
return return_result (false, cfg.err_msg_supl.prefix); -- fail when ISBN-13 does not begin with 978 or 979
end
if id:match ('^9790') then
return return_result (false, cfg.err_msg_supl.group); -- group identifier '0' is reserved to ISMN
end
return return_result (is_valid_isxn_13 (id), cfg.err_msg_supl.check);
end
end
return ''; -- |embargo= not set return empty string
end
end




--[[--------------------------< P M C >------------------------------------------------------------------------
--[[--------------------------< A S I N >----------------------------------------------------------------------
 
Format a PMC, do simple error checking, and check for embargoed articles.


The embargo parameter takes a date for a value. If the embargo date is in the future the PMC identifier will not
Formats a link to Amazon. Do simple error checking: ASIN must be mix of 10 numeric or uppercase alpha
be linked to the article.  If the embargo date is today or in the past, or if it is empty or omitted, then the
characters.  If a mix, first character must be uppercase alpha; if all numeric, ASINs must be 10-digit
PMC identifier is linked to the article through the link at cfg.id_handlers['PMC'].prefix.
ISBN. If 10-digit ISBN, add a maintenance category so a bot or AWB script can replace |asin= with |isbn=.
Error message if not 10 characters, if not ISBN-10, if mixed and first character is a digit.


PMC embargo date testing is done in function is_embargoed () which is called earlier because when the citation
|asin=630....... and |asin=631....... are (apparently) not a legitimate ISBN though it checksums as one; these
has |pmc=<value> but does not have a |url= then |title= is linked with the PMC link. Function is_embargoed ()
do not cause this function to emit the maint_asin message
returns the embargo date if the PMC article is still embargoed, otherwise it returns an empty string.


PMCs are sequential numbers beginning at 1 and counting up.  This code checks the PMC to see that it contains only digits and is less
This function is positioned here because it calls isbn()
than test_limit; the value in local variable test_limit will need to be updated periodically as more PMCs are issued.


]]
]]


local function pmc(id, embargo)
local function asin (options)
local test_limit = 5000000; -- update this value as PMCs approach
local id = options.id;
local handler = cfg.id_handlers['PMC'];
local domain = options.ASINTLD;
local err_cat =  ''; -- presume that PMC is valid
local text;
local err_cat = ''


if id:match("[^%d]") then -- if PMC has anything but digits
if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
err_cat = ' ' .. set_error( 'bad_pmc' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_asin'); -- ASIN is not a mix of 10 uppercase alpha and numeric characters
else -- PMC is only digits
else
local id_num = tonumber(id); -- convert id to a number for range testing
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
if 1 > id_num or test_limit < id_num then -- if PMC is outside test limit boundaries
if is_valid_isxn (id, 10) then -- see if ASIN value is or validates as ISBN-10
err_cat = ' ' .. set_error( 'bad_pmc' ); -- set an error message
if not id:find ('^63[01]') then -- 630xxxxxxx and 631xxxxxxx are (apparently) not a valid isbn prefixes but are used by amazon as a numeric identifier
err_cat = ' ' .. set_message ('err_bad_asin'); -- ASIN has ISBN-10 form but begins with something other than 630/1 so probably an isbn
end
elseif not is_set (err_cat) then
err_cat = ' ' .. set_message ('err_bad_asin'); -- ASIN is not ISBN-10
end
elseif not id:match("^%u[%d%u]+$") then
err_cat = ' ' .. set_message ('err_bad_asin'); -- asin doesn't begin with uppercase alpha
end
end
end
if (not is_set (domain)) or in_array (domain, {'us'}) then -- default: United States
domain = "com";
elseif in_array (domain, {'jp', 'uk'}) then -- Japan, United Kingdom
domain = "co." .. domain;
elseif in_array (domain, {'z.cn'}) then -- China
domain = "cn";
elseif in_array (domain, {'au', 'br', 'mx', 'sg', 'tr'}) then -- Australia, Brazil, Mexico, Singapore, Turkey
domain = "com." .. domain;
elseif not in_array (domain, {'ae', 'ca', 'cn', 'de', 'es', 'fr', 'in', 'it', 'nl', 'pl', 'sa', 'se', 'co.jp', 'co.uk', 'com', 'com.au', 'com.br', 'com.mx', 'com.sg', 'com.tr'}) then -- Arabic Emirates, Canada, China, Germany, Spain, France, Indonesia, Italy, Netherlands, Poland, Saudi Arabia, Sweden (as of 2021-03 Austria (.at), Liechtenstein (.li) and Switzerland (.ch) still redirect to the German site (.de) with special settings, so don't maintain local ASINs for them)
err_cat = ' ' .. set_message ('err_bad_asin_tld'); -- unsupported asin-tld value
end
local handler = options.handler;
if not is_set (err_cat) then
options.coins_list_t['ASIN'] = handler.prefix .. domain .. "/dp/" .. id; -- experiment for asin coins
else
options.coins_list_t['ASIN'] = nil; -- when error, unset so not included in COinS
end
end
if is_set (embargo) then -- is PMC is still embargoed?
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
text="[[" .. handler.link .. "|" .. handler.label .. "]]:" .. handler.separator .. id .. err_cat; -- still embargoed so no external link
prefix = handler.prefix .. domain .. "/dp/",
id = id, encode = handler.encode, separator = handler.separator}) .. err_cat;
end
 
 
--[[--------------------------< I S M N >----------------------------------------------------------------------
 
Determines whether an ISMN string is valid. Similar to ISBN-13, ISMN is 13 digits beginning 979-0-... and uses the
same check digit calculations. See http://www.ismn-international.org/download/Web_ISMN_Users_Manual_2008-6.pdf
section 2, pages 9–12.
 
ismn value not made part of COinS metadata because we don't have a url or isn't a COinS-defined identifier (rft.xxx)
or an identifier registered at info-uri.info (info:)
 
]]
 
local function ismn (options)
local id = options.id;
local handler = options.handler;
local text;
local valid_ismn = true;
local id_copy;
 
id_copy = id; -- save a copy because this testing is destructive
id = id:gsub ('[%s-]', ''); -- remove hyphens and white space
 
if 13 ~= id:len() or id:match ("^9790%d*$" ) == nil then -- ISMN must be 13 digits and begin with 9790
valid_ismn = false;
else
else
text = external_link_id({link = handler.link, label = handler.label, -- no embargo date or embargo has expired, ok to link to article
valid_ismn=is_valid_isxn_13 (id); -- validate ISMN
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, free=handler.free}) .. err_cat;
end
end
-- text = internal_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect, -- use this (or external version) when there is some place to link to
-- prefix = handler.prefix, id = id_copy, separator = handler.separator, encode = handler.encode})
text = table.concat ( -- because no place to link to yet
{
make_wikilink (link_label_make (handler), handler.label),
handler.separator,
id_copy
});
if false == valid_ismn then
options.coins_list_t['ISMN'] = nil; -- when error, unset so not included in COinS; not really necessary here because ismn not made part of COinS
text = text .. ' ' .. set_message ('err_bad_ismn' ) -- add an error message if the ISMN is invalid
end
return text;
return text;
end
end




--[[--------------------------< D O I >------------------------------------------------------------------------
--[[--------------------------< I S S N >----------------------------------------------------------------------
 
Validate and format an ISSN.  This code fixes the case where an editor has included an ISSN in the citation but
has separated the two groups of four digits with a space.  When that condition occurred, the resulting link looked
like this:
 
|issn=0819 4327 gives: [http://www.worldcat.org/issn/0819 4327 0819 4327] -- can't have spaces in an external link
This code now prevents that by inserting a hyphen at the ISSN midpoint.  It also validates the ISSN for length
and makes sure that the checkdigit agrees with the calculated value.  Incorrect length (8 digits), characters
other than 0-9 and X, or checkdigit / calculated value mismatch will all cause a check ISSN error message.  The
ISSN is always displayed with a hyphen, even if the ISSN was given as a single group of 8 digits.


Formats a DOI and checks for DOI errors.
]]


DOI names contain two parts: prefix and suffix separated by a forward slash.
local function issn (options)
Prefix: directory indicator '10.' followed by a registrant code
local id = options.id;
Suffix: character string of any length chosen by the registrant
local handler = options.handler;
local ignore_invalid = options.accept;


This function checks a DOI name for: prefix/suffix.  If the doi name contains spaces or endashes, or, if it ends
local issn_copy = id; -- save a copy of unadulterated ISSN; use this version for display if ISSN does not validate
with a period or a comma, this function will emit a bad_doi error message.
local text;
local valid_issn = true;


DOI names are case-insensitive and can incorporate any printable Unicode characters so the test for spaces, endash,
id = id:gsub ('[%s-]', ''); -- remove hyphens and whitespace
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in doi names.


]]
if 8 ~= id:len() or nil == id:match ("^%d*X?$" ) then -- validate the ISSN: 8 digits long, containing only 0-9 or X in the last position
valid_issn = false; -- wrong length or improper character
else
valid_issn = is_valid_isxn (id, 8); -- validate ISSN
end


local function doi(id, inactive)
if true == valid_issn then
local cat = ""
id = string.sub (id, 1, 4 ) .. "-" .. string.sub (id, 5 ); -- if valid, display correctly formatted version
local handler = cfg.id_handlers['DOI'];
else
id = issn_copy; -- if not valid, show the invalid ISSN with error message
local text;
if is_set(inactive) then
local inactive_year = inactive:match("%d%d%d%d") or ''; -- try to get the year portion from the inactive date
text = "[[" .. handler.link .. "|" .. handler.label .. "]]:" .. id;
if is_set(inactive_year) then
table.insert( z.error_categories, "Pages with DOIs inactive since " .. inactive_year );
else
table.insert( z.error_categories, "Pages with inactive DOIs" ); -- when inactive doesn't contain a recognizable year
end
inactive = " (" .. cfg.messages['inactive'] .. " " .. inactive .. ")"
else  
text = external_link_id({link = handler.link, label = handler.label,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
inactive = ""
end
end


if nil == id:match("^10%.[^%s–]-/[^%s–]-[^%.,]$") then -- doi must begin with '10.', must contain a fwd slash, must not contain spaces or endashes, and must not end with period or comma
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
cat = ' ' .. set_error( 'bad_doi' );
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
 
if ignore_invalid then
set_message ('maint_issn_ignore');
else
if false == valid_issn then
options.coins_list_t['ISSN'] = nil; -- when error, unset so not included in COinS
text = text .. ' ' .. set_message ('err_bad_issn', (options.hkey == 'EISSN') and 'e' or ''); -- add an error message if the ISSN is invalid
end
end
end
return text .. inactive .. cat
return text
end
end




--[[--------------------------< H D L >------------------------------------------------------------------------
--[[--------------------------< J F M >-----------------------------------------------------------------------
 
A numerical identifier in the form nn.nnnn.nn
 
]]
 
local function jfm (options)
local id = options.id;
local handler = options.handler;
local id_num;
local err_cat = '';
id_num = id:match ('^[Jj][Ff][Mm](.*)$'); -- identifier with jfm prefix; extract identifier
 
if is_set (id_num) then
set_message ('maint_jfm_format');
else -- plain number without JFM prefix
id_num = id; -- if here id does not have prefix
end


Formats an HDL with minor error checking.
if id_num and id_num:match('^%d%d%.%d%d%d%d%.%d%d$') then
id = id_num; -- jfm matches pattern
else
err_cat = ' ' .. set_message ('err_bad_jfm' ); -- set an error message
options.coins_list_t['JFM'] = nil; -- when error, unset so not included in COinS
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end


HDL names contain two parts: prefix and suffix separated by a forward slash.
Prefix: character string using any character in the UCS-2 character set except '/'
Suffix: character string of any length using any character in the UCS-2 character set chosen by the registrant


This function checks a HDL name for: prefix/suffix.  If the HDL name contains spaces, endashes, or, if it ends
--[[--------------------------< J S T O R >--------------------------------------------------------------------
with a period or a comma, this function will emit a bad_hdl error message.


HDL names are case-insensitive and can incorporate any printable Unicode characters so the test for endashes and
Format a JSTOR with some error checking
terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in HDLs.


]]
]]


local function hdl(id)
local function jstor (options)
local handler = cfg.id_handlers['HDL'];
local id = options.id;
local access = options.access;
local text = external_link_id({link = handler.link, label = handler.label,
local handler = options.handler;
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
local err_msg = '';


if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then -- hdl must contain a fwd slash, must not contain spaces, endashes, and must not end with period or comma
if id:find ('[Jj][Ss][Tt][Oo][Rr]') or id:find ('^https?://') or id:find ('%s') then
text = text .. ' ' .. set_error( 'bad_hdl' );
err_msg = ' ' .. set_message ('err_bad_jstor'); -- set an error message
options.coins_list_t['JSTOR'] = nil; -- when error, unset so not included in COinS
end
end
return text;
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access}) .. err_msg;
end
end




--[[--------------------------< O P E N L I B R A R Y >--------------------------------------------------------
--[[--------------------------< L C C N >----------------------------------------------------------------------
 
Format LCCN link and do simple error checking.  LCCN is a character string 8-12 characters long. The length of
the LCCN dictates the character type of the first 1-3 characters; the rightmost eight are always digits.
http://info-uri.info/registry/OAIHandler?verb=GetRecord&metadataPrefix=reg&identifier=info:lccn/


Formats an OpenLibrary link, and checks for associated errors.
length = 8 then all digits
length = 9 then lccn[1] is lowercase alpha
length = 10 then lccn[1] and lccn[2] are both lowercase alpha or both digits
length = 11 then lccn[1] is lower case alpha, lccn[2] and lccn[3] are both lowercase alpha or both digits
length = 12 then lccn[1] and lccn[2] are both lowercase alpha


]]
]]


local function openlibrary(id)
local function lccn (options)
local code = id:match("^%d+([AMW])$"); -- only digits followed by 'A', 'M', or 'W'
local lccn = options.id;
local handler = cfg.id_handlers['OL'];
local handler = options.handler;
local err_cat = ''; -- presume that LCCN is valid
local id = lccn; -- local copy of the LCCN


if ( code == "A" ) then
id = normalize_lccn (id); -- get canonical form (no whitespace, hyphens, forward slashes)
return external_link_id({link=handler.link, label=handler.label,
local len = id:len(); -- get the length of the LCCN
prefix=handler.prefix .. 'authors/OL',
 
id=id, separator=handler.separator, encode = handler.encode})
if 8 == len then
elseif ( code == "M" ) then
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits)
return external_link_id({link=handler.link, label=handler.label,
err_cat = ' ' .. set_message ('err_bad_lccn'); -- set an error message
prefix=handler.prefix .. 'books/OL',
end
id=id, separator=handler.separator, encode = handler.encode})
elseif 9 == len then -- LCCN should be adddddddd
elseif ( code == "W" ) then
if nil == id:match("%l%d%d%d%d%d%d%d%d") then -- does it match our pattern?
return external_link_id({link=handler.link, label=handler.label,
err_cat = ' ' .. set_message ('err_bad_lccn'); -- set an error message
prefix=handler.prefix .. 'works/OL',
end
id=id, separator=handler.separator, encode = handler.encode})
elseif 10 == len then -- LCCN should be aadddddddd or dddddddddd
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits) ...
if nil == id:match("^%l%l%d%d%d%d%d%d%d%d") then -- ... see if it matches our pattern
err_cat = ' ' .. set_message ('err_bad_lccn'); -- no match, set an error message
end
end
elseif 11 == len then -- LCCN should be aaadddddddd or adddddddddd
if not (id:match("^%l%l%l%d%d%d%d%d%d%d%d") or id:match("^%l%d%d%d%d%d%d%d%d%d%d")) then -- see if it matches one of our patterns
err_cat = ' ' .. set_message ('err_bad_lccn'); -- no match, set an error message
end
elseif 12 == len then -- LCCN should be aadddddddddd
if not id:match("^%l%l%d%d%d%d%d%d%d%d%d%d") then -- see if it matches our pattern
err_cat = ' ' .. set_message ('err_bad_lccn'); -- no match, set an error message
end
else
else
return external_link_id({link=handler.link, label=handler.label,
err_cat = ' ' .. set_message ('err_bad_lccn'); -- wrong length, set an error message
prefix=handler.prefix .. 'OL',
id=id, separator=handler.separator, encode = handler.encode}) .. ' ' .. set_error( 'bad_ol' );
end
end
if not is_set (err_cat) and nil ~= lccn:find ('%s') then
err_cat = ' ' .. set_message ('err_bad_lccn'); -- lccn contains a space, set an error message
end
if is_set (err_cat) then
options.coins_list_t['LCCN'] = nil; -- when error, unset so not included in COinS
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = lccn, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
end




--[[--------------------------< M E S S A G E _ I D >----------------------------------------------------------
--[[--------------------------< M R >--------------------------------------------------------------------------


Validate and format a usenet message id.  Simple error checking, looks for 'id-left@id-right' not enclosed in
A seven digit number; if not seven digits, zero-fill leading digits to make seven digits.
'<' and/or '>' angle brackets.


]]
]]


local function message_id (id)
local function mr (options)
local handler = cfg.id_handlers['USENETID'];
local id = options.id;
local handler = options.handler;
local id_num;
local id_len;
local err_cat = '';
id_num = id:match ('^[Mm][Rr](%d+)$'); -- identifier with mr prefix
 
if is_set (id_num) then
set_message ('maint_mr_format'); -- add maint cat
else -- plain number without mr prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
end


local text = external_link_id({link = handler.link, label = handler.label,
id_len = id_num and id_num:len() or 0;
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
if (7 >= id_len) and (0 ~= id_len) then
id = string.rep ('0', 7-id_len) .. id_num; -- zero-fill leading digits
if not id:match('^.+@.+$') or not id:match('^[^<].*[^>]$')then -- doesn't have '@' or has one or first or last character is '< or '>'
else
text = text .. ' ' .. set_error( 'bad_message_id' ) -- add an error message if the message id is invalid
err_cat = ' ' .. set_message ('err_bad_mr'); -- set an error message
end  
options.coins_list_t['MR'] = nil; -- when error, unset so not included in COinS
end
return text
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
end


Line 651: Line 1,037:
--[[--------------------------< O C L C >----------------------------------------------------------------------
--[[--------------------------< O C L C >----------------------------------------------------------------------


Validate and format an oclc id. https://www.oclc.org/batchload/controlnumber.en.html
Validate and format an OCLC ID.  https://www.oclc.org/batchload/controlnumber.en.html {{dead link}}
archived at: https://web.archive.org/web/20161228233804/https://www.oclc.org/batchload/controlnumber.en.html


]]
]]


local function oclc (id)
local function oclc (options)
local handler = cfg.id_handlers['OCLC'];
local id = options.id;
local handler = options.handler;
local number;
local number;
local err_msg = ''; -- empty string for concatenation
local err_msg = ''; -- empty string for concatenation
Line 669: Line 1,057:
number = id:match('%(OCoLC%)([1-9]%d*)'); -- get the number
number = id:match('%(OCoLC%)([1-9]%d*)'); -- get the number
if 9 < number:len() then
if 9 < number:len() then
number = nil; -- contrain to 1 to 9 digits; change this when oclc issues 10-digit numbers
number = nil; -- constrain to 1 to 9 digits; change this when OCLC issues 10-digit numbers
end
end
elseif id:match('^%d+$') then -- no prefix
elseif id:match('^%d+$') then -- no prefix
number = id; -- get the number
number = id; -- get the number
if 10 < number:len() then
if 10 < number:len() then
number = nil; -- contrain to 1 to 10 digits; change this when oclc issues 11-digit numbers
number = nil; -- constrain to 1 to 10 digits; change this when OCLC issues 11-digit numbers
end
end
end
end
Line 681: Line 1,069:
id = number; -- exclude prefix, if any, from external link
id = number; -- exclude prefix, if any, from external link
else
else
err_msg = ' ' .. set_error( 'bad_oclc' ) -- add an error message if the id is malformed
err_msg = ' ' .. set_message ('err_bad_oclc') -- add an error message if the id is malformed
options.coins_list_t['OCLC'] = nil; -- when error, unset so not included in COinS
end
end
local text = external_link_id({link=handler.link, label=handler.label,
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix=handler.prefix, id=id, separator=handler.separator, encode=handler.encode}) .. err_msg;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_msg;


return text;
return text;
Line 691: Line 1,080:




--[[--------------------------< B I B C O D E >--------------------------------------------------------------------
--[[--------------------------< O P E N L I B R A R Y >--------------------------------------------------------
 
Formats an OpenLibrary link, and checks for associated errors.
 
]]
 
local function openlibrary (options)
local id = options.id;
local access = options.access;
local handler = options.handler;
local ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- strip optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W';
local error_msg = '';
local prefix = { -- these are appended to the handler.prefix according to code
['A']='authors/OL',
['M']='books/OL',
['W']='works/OL',
['X']='OL' -- not a code; spoof when 'code' in id is invalid
};
 
if not ident then
code = 'X'; -- no code or id completely invalid
ident = id; -- copy id to ident so that we display the flawed identifier
error_msg = ' ' .. set_message ('err_bad_ol');
end
 
if not is_set (error_msg) then
options.coins_list_t['OL'] = handler.prefix .. prefix[code] .. ident; -- experiment for ol coins
else
options.coins_list_t['OL'] = nil; -- when error, unset so not included in COinS
end
 
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix .. prefix[code],
id = ident, separator = handler.separator, encode = handler.encode,
access = access}) .. error_msg;
end
 
 
--[[--------------------------< O S T I >----------------------------------------------------------------------
 
Format OSTI and do simple error checking. OSTIs are sequential numbers beginning at 1 and counting up.  This
code checks the OSTI to see that it contains only digits and is less than test_limit specified in the configuration;
the value in test_limit will need to be updated periodically as more OSTIs are issued.
 
NB. 1018 is the lowest OSTI number found in the wild (so far) and resolving OK on the OSTI site
 
]]
 
local function osti (options)
local id = options.id;
local access = options.access;
local handler = options.handler;
local err_cat = ''; -- presume that OSTI is valid
if id:match("[^%d]") then -- if OSTI has anything but digits
err_cat = ' ' .. set_message ('err_bad_osti'); -- set an error message
options.coins_list_t['OSTI'] = nil; -- when error, unset so not included in COinS
else -- OSTI is only digits
local id_num = tonumber (id); -- convert id to a number for range testing
if 1018 > id_num or handler.id_limit < id_num then -- if OSTI is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_osti'); -- set an error message
options.coins_list_t['OSTI'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access}) .. err_cat;
end
 
 
--[[--------------------------< P M C >------------------------------------------------------------------------
 
Format a PMC, do simple error checking, and check for embargoed articles.


Validates (sort of) and formats a bibcode id.
The embargo parameter takes a date for a value. If the embargo date is in the future the PMC identifier will not
be linked to the article.  If the embargo date is today or in the past, or if it is empty or omitted, then the
PMC identifier is linked to the article through the link at cfg.id_handlers['PMC'].prefix.


Format for bibcodes is specified here: http://adsabs.harvard.edu/abs_doc/help_pages/data.html#bibcodes
PMC embargo date testing is done in function is_embargoed () which is called earlier because when the citation
has |pmc=<value> but does not have a |url= then |title= is linked with the PMC link. Function is_embargoed ()
returns the embargo date if the PMC article is still embargoed, otherwise it returns an empty string.


But, this: 2015arXiv151206696F is apparently valid so apparently, the only things that really matter are length, 19 characters
PMCs are sequential numbers beginning at 1 and counting up.  This code checks the PMC to see that it contains only digits and is less
and first four digits must be a year.  This function makes these tests:
than test_limit; the value in local variable test_limit will need to be updated periodically as more PMCs are issued.
length must be 19 characters
characters in position
1–4 must be digits and must represent a year in the range of 1000 – next year
5 must be a letter
6 must be letter, ampersand, or dot (ampersand cannot directly precede a dot; &. )
7–8 must be letter, digit, ampersand, or dot (ampersand cannot directly precede a dot; &. )
9–18 must be letter, digit, or dot
19 must be a letter or dot


]]
]]


local function bibcode (id)
local function pmc (options)
local handler = cfg.id_handlers['BIBCODE'];
local id = options.id;
local err_type;
local embargo = options.Embargo; -- TODO: lowercase?
local year;
local handler = options.handler;
local err_cat;
local id_num;
local text;
 
id_num = id:match ('^[Pp][Mm][Cc](%d+)$'); -- identifier with PMC prefix
 
if is_set (id_num) then
set_message ('maint_pmc_format');
else -- plain number without PMC prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
end


local text = external_link_id({link=handler.link, label=handler.label,
if is_set (id_num) then -- id_num has a value so test it
prefix=handler.prefix, id=id, separator=handler.separator, encode=handler.encode});
id_num = tonumber (id_num); -- convert id_num to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if PMC is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_pmc'); -- set an error message
else
id = tostring (id_num); -- make sure id is a string
end
else -- when id format incorrect
err_cat = ' ' .. set_message ('err_bad_pmc'); -- set an error message
end
if 19 ~= id:len() then
if is_set (embargo) and is_set (is_embargoed (embargo)) then -- is PMC is still embargoed?
err_type = 'length';
text = table.concat ( -- still embargoed so no external link
{
make_wikilink (link_label_make (handler), handler.label),
handler.separator,
id,
(err_cat and err_cat or '') -- parens required
});
else
else
year = id:match ("^(%d%d%d%d)[%a][%a&%.][%a&%.%d][%a&%.%d][%a%d%.]+[%a%.]$") --  
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect, -- no embargo date or embargo has expired, ok to link to article
if not year then -- if nil then no pattern match
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access,
err_type = 'value'; -- so value error
auto_link = not err_cat and 'pmc' or nil -- do not auto-link when PMC has error
else
}) .. (err_cat and err_cat or ''); -- parentheses required
local next_year = tonumber(os.date ('%Y'))+1; -- get the current year as a number and add one for next year
year = tonumber (year); -- convert year portion of bibcode to a number
if (1000 > year) or (year > next_year) then
err_type = 'year'; -- year out of bounds
end
if id:find('&%.') then
err_type = 'journal'; -- journal abbreviation must not have '&.' (if it does its missing a letter)
end
end
end
end


if is_set (err_type) then -- if there was an error detected
if err_cat then
text = text .. ' ' .. set_error( 'bad_bibcode', {err_type});
options.coins_list_t['PMC'] = nil; -- when error, unset so not included in COinS
end
end
return text;
return text;
end
end




--[[--------------------------< B U I L D _ I D _ L I S T >--------------------------------------------------------
--[[--------------------------< P M I D >----------------------------------------------------------------------
 
Format PMID and do simple error checking.  PMIDs are sequential numbers beginning at 1 and counting up.  This
code checks the PMID to see that it contains only digits and is less than test_limit; the value in local variable
test_limit will need to be updated periodically as more PMIDs are issued.
 
]]
 
local function pmid (options)
local id = options.id;
local handler = options.handler;
local err_cat = ''; -- presume that PMID is valid
if id:match("[^%d]") then -- if PMID has anything but digits
err_cat = ' ' .. set_message ('err_bad_pmid'); -- set an error message
options.coins_list_t['PMID'] = nil; -- when error, unset so not included in COinS
else -- PMID is only digits
local id_num = tonumber (id); -- convert id to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if PMID is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_pmid'); -- set an error message
options.coins_list_t['PMID'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
 
 
--[[--------------------------< R F C >------------------------------------------------------------------------


Takes a table of IDs created by extract_ids() and turns it into a table of formatted ID outputs.
Format RFC and do simple error checking. RFCs are sequential numbers beginning at 1 and counting up.  This
code checks the RFC to see that it contains only digits and is less than test_limit specified in the configuration;
the value in test_limit will need to be updated periodically as more RFCs are issued.


inputs:
An index of all RFCs is here: https://tools.ietf.org/rfc/
id_list – table of identifiers built by extract_ids()
options – table of various template parameter values used to modify some manually handled identifiers


]]
]]


local function build_id_list( id_list, options )
local function rfc (options)
local new_list, handler = {};
local id = options.id;
local handler = options.handler;
local err_cat = ''; -- presume that RFC is valid
if id:match("[^%d]") then -- if RFC has anything but digits
err_cat = ' ' .. set_message ('err_bad_rfc'); -- set an error message
options.coins_list_t['RFC'] = nil; -- when error, unset so not included in COinS
else -- RFC is only digits
local id_num = tonumber (id); -- convert id to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if RFC is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_rfc'); -- set an error message
options.coins_list_t['RFC'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;
end
 
 
--[[--------------------------< S 2 C I D >--------------------------------------------------------------------
 
Format an S2CID, do simple error checking
 
S2CIDs are sequential numbers beginning at 1 and counting up.  This code checks the S2CID to see that it is only
digits and is less than test_limit; the value in local variable test_limit will need to be updated periodically
as more S2CIDs are issued.


local function fallback(k) return { __index = function(t,i) return cfg.id_handlers[k][i] end } end;
]]
 
local function s2cid (options)
local id = options.id;
local access = options.access;
local handler = options.handler;
local err_cat = ''; -- presume that S2CID is valid
local id_num;
local text;
for k, v in pairs( id_list ) do -- k is uc identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
id_num = id:match ('^[1-9]%d*$'); -- id must be all digits; must not begin with 0; no open access flag
-- fallback to read-only cfg
handler = setmetatable( { ['id'] = v }, fallback(k) );


if handler.mode == 'external' then
if is_set (id_num) then -- id_num has a value so test it
table.insert( new_list, {handler.label, external_link_id( handler ) } );
id_num = tonumber (id_num); -- convert id_num to a number for range testing
elseif handler.mode == 'internal' then
if handler.id_limit < id_num then -- if S2CID is outside test limit boundaries
table.insert( new_list, {handler.label, internal_link_id( handler ) } );
err_cat = ' ' .. set_message ('err_bad_s2cid'); -- set an error message
elseif handler.mode ~= 'manual' then
options.coins_list_t['S2CID'] = nil; -- when error, unset so not included in COinS
error( cfg.messages['unknown_ID_mode'] );
end
elseif k == 'BIBCODE' then
 
table.insert( new_list, {handler.label, bibcode( v ) } );
else -- when id format incorrect
elseif k == 'DOI' then
err_cat = ' ' .. set_message ('err_bad_s2cid'); -- set an error message
table.insert( new_list, {handler.label, doi( v, options.DoiBroken ) } );
options.coins_list_t['S2CID'] = nil; -- when error, unset so not included in COinS
elseif k == 'HDL' then
end
table.insert( new_list, {handler.label, hdl( v ) } );
 
elseif k == 'ARXIV' then
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
table.insert( new_list, {handler.label, arxiv( v, options.Class ) } );
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access}) .. err_cat;
elseif k == 'ASIN' then
 
table.insert( new_list, {handler.label, amazon( v, options.ASINTLD ) } );
return text;
elseif k == 'LCCN' then
end
table.insert( new_list, {handler.label, lccn( v ) } );
 
elseif k == 'OL' or k == 'OLA' then
 
table.insert( new_list, {handler.label, openlibrary( v ) } );
--[[--------------------------< S B N >------------------------------------------------------------------------
elseif k == 'PMC' then
 
table.insert( new_list, {handler.label, pmc( v, options.Embargo ) } );
9-digit form of ISBN-10; uses same check-digit validation when SBN is prefixed with an additional '0' to make 10 digits
elseif k == 'PMID' then
 
table.insert( new_list, {handler.label, pmid( v ) } );
sbn value not made part of COinS metadata because we don't have a url or isn't a COinS-defined identifier (rft.xxx)
elseif k == 'OCLC' then
or an identifier registered at info-uri.info (info:)
table.insert( new_list, {handler.label, oclc( v ) } );
 
elseif k == 'ISMN' then
]]
table.insert( new_list, {handler.label, ismn( v ) } );
 
elseif k == 'ISSN' then
local function sbn (options)
table.insert( new_list, {handler.label, issn( v ) } );
local id = options.id;
elseif k == 'EISSN' then
local ignore_invalid = options.accept;
table.insert( new_list, {handler.label, issn( v, true ) } ); -- true distinguishes eissn from issn
local handler = options.handler;
elseif k == 'ISBN' then
local function return_result (check, err_type) -- local function to handle the various returns
local ISBN = internal_link_id( handler );
local SBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
local check;
prefix = handler.prefix, id = id, separator = handler.separator});
local err_type = '';
if not ignore_invalid then -- if not ignoring SBN errors
-- if not check_isbn( v ) and not is_set(options.IgnoreISBN) then
-- ISBN = ISBN .. set_error( 'bad_isbn', {}, false, " ", "" );
-- end
check, err_type = check_isbn( v );
if not check then
if not check then
if is_set(options.IgnoreISBN) then -- ISBN is invalid; if |ignore-isbn-error= set
options.coins_list_t['SBN'] = nil; -- when error, unset so not included in COinS; not really necessary here because sbn not made part of COinS
add_maint_cat ('ignore_isbn_err'); -- ad a maint category
return SBN .. set_message ('err_bad_sbn', {err_type}, false, ' '); -- display an error message
else
ISBN = ISBN .. set_error( 'bad_isbn', {err_type}, false, " ", "" ); -- else display an error message
end
elseif is_set(options.IgnoreISBN) then -- ISBN is OK; if |ignore-isbn-error= set
add_maint_cat ('ignore_isbn_err'); -- because |ignore-isbn-error= unnecessary
end
end
table.insert( new_list, {handler.label, ISBN } );
elseif k == 'USENETID' then
table.insert( new_list, {handler.label, message_id( v ) } );
else
else
error( cfg.messages['unknown_manual_ID'] );
set_message ('maint_isbn_ignore'); -- add a maint category even when there is no error (ToDo: Possibly switch to separate message for SBNs only)
end
end
return SBN;
end
if id:match ('[^%s-0-9X]') then
return return_result (false, cfg.err_msg_supl.char); -- fail if SBN contains anything but digits, hyphens, or the uppercase X
end
local ident = id:gsub ('[%s-]', ''); -- remove hyphens and whitespace; they interfere with the rest of the tests
if  9 ~= ident:len() then
return return_result (false, cfg.err_msg_supl.length); -- fail if incorrect length
end
end
if ident:match ('^%d*X?$') == nil then
return return_result (false, cfg.err_msg_supl.form); -- fail if SBN has 'X' anywhere but last position
end
return return_result (is_valid_isxn ('0' .. ident, 10), cfg.err_msg_supl.check);
end
--[[--------------------------< S S R N >----------------------------------------------------------------------
Format an SSRN, do simple error checking
SSRNs are sequential numbers beginning at 100? and counting up.  This code checks the SSRN to see that it is
only digits and is greater than 99 and less than test_limit; the value in local variable test_limit will need
to be updated periodically as more SSRNs are issued.
]]
local function ssrn (options)
local id = options.id;
local handler = options.handler;
local err_cat = ''; -- presume that SSRN is valid
local id_num;
local text;
local function comp( a, b ) -- used in following table.sort()
id_num = id:match ('^%d+$'); -- id must be all digits
return a[1] < b[1];
 
if is_set (id_num) then -- id_num has a value so test it
id_num = tonumber (id_num); -- convert id_num to a number for range testing
if 100 > id_num or handler.id_limit < id_num then -- if SSRN is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_ssrn'); -- set an error message
options.coins_list_t['SSRN'] = nil; -- when error, unset so not included in COinS
end
else -- when id format incorrect
err_cat = ' ' .. set_message ('err_bad_ssrn'); -- set an error message
options.coins_list_t['SSRN'] = nil; -- when error, unset so not included in COinS
end
end
table.sort( new_list, comp );
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
for k, v in ipairs( new_list ) do
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;
new_list[k] = v[2];
 
return text;
end
 
 
--[[--------------------------< U S E N E T _ I D >------------------------------------------------------------
 
Validate and format a usenet message id.  Simple error checking, looks for 'id-left@id-right' not enclosed in
'<' and/or '>' angle brackets.
 
]]
 
local function usenet_id (options)
local id = options.id;
local handler = options.handler;
 
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
if not id:match('^.+@.+$') or not id:match('^[^<].*[^>]$') then -- doesn't have '@' or has one or first or last character is '< or '>'
text = text .. ' ' .. set_message ('err_bad_usenet_id') -- add an error message if the message id is invalid
options.coins_list_t['USENETID'] = nil; -- when error, unset so not included in COinS
end
return text
end
 
 
--[[--------------------------< Z B L >-----------------------------------------------------------------------
 
A numerical identifier in the form nnnn.nnnnn - leading zeros in the first quartet optional
 
format described here: http://emis.mi.sanu.ac.rs/ZMATH/zmath/en/help/search/
 
temporary format is apparently eight digits.  Anything else is an error
 
]]
 
local function zbl (options)
local id = options.id;
local handler = options.handler;
local err_cat = '';
if id:match('^%d%d%d%d%d%d%d%d$') then -- is this identifier using temporary format?
set_message ('maint_zbl'); -- yes, add maint cat
elseif not id:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then -- not temporary, is it normal format?
err_cat = ' ' .. set_message ('err_bad_zbl'); -- no, set an error message
options.coins_list_t['ZBL'] = nil; -- when error, unset so not included in COinS
end
end
return new_list;
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
end


--============================<< I N T E R F A C E  F U N C T I O N S >>==========================================


--[[--------------------------< E X T R A C T _ I D S >------------------------------------------------------------
--[[--------------------------< E X T R A C T _ I D S >------------------------------------------------------------
Line 836: Line 1,454:
Populates ID table from arguments using configuration settings. Loops through cfg.id_handlers and searches args for
Populates ID table from arguments using configuration settings. Loops through cfg.id_handlers and searches args for
any of the parameters listed in each cfg.id_handlers['...'].parameters.  If found, adds the parameter and value to
any of the parameters listed in each cfg.id_handlers['...'].parameters.  If found, adds the parameter and value to
the identifier list.  Emits redundant error message is more than one alias exists in args
the identifier list.  Emits redundant error message if more than one alias exists in args


]]
]]


local function extract_ids( args )
local function extract_ids (args)
local id_list = {}; -- list of identifiers found in args
local id_list = {}; -- list of identifiers found in args
for k, v in pairs( cfg.id_handlers ) do -- k is uc identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
for k, v in pairs (cfg.id_handlers) do -- k is uppercase identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
v = select_one( args, v.parameters, 'redundant_parameters' ); -- v.parameters is a table of aliases for k; here we pick one from args if present
v = select_one (args, v.parameters, 'err_redundant_parameters' ); -- v.parameters is a table of aliases for k; here we pick one from args if present
if is_set(v) then id_list[k] = v; end -- if found in args, add identifier to our list
if is_set (v) then id_list[k] = v; end -- if found in args, add identifier to our list
end
end
return id_list;
return id_list;
end
--[[--------------------------< E X T R A C T _ I D _ A C C E S S _ L E V E L S >--------------------------------------
Fetches custom id access levels from arguments using configuration settings. Parameters which have a predefined access
level (e.g. arxiv) do not use this function as they are directly rendered as free without using an additional parameter.
returns a table of k/v pairs where k is same as the identifier's key in cfg.id_handlers and v is the assigned (valid) keyword
access-level values must match the case used in cfg.keywords_lists['id-access'] (lowercase unless there is some special reason for something else)
]]
local function extract_id_access_levels (args, id_list)
local id_accesses_list = {};
for k, v in pairs (cfg.id_handlers) do
local access_param = v.custom_access; -- name of identifier's access-level parameter
if is_set (access_param) then
local access_level = args[access_param]; -- get the assigned value if there is one
if is_set (access_level) then
if not in_array (access_level, cfg.keywords_lists['id-access']) then -- exact match required
table.insert (z.message_tail, { set_message ('err_invalid_param_val', {access_param, access_level}, true) } );
access_level = nil; -- invalid so unset
end
if not is_set (id_list[k]) then -- identifier access-level must have a matching identifier
table.insert (z.message_tail, { set_message ('err_param_access_requires_param', {k:lower()}, true) } ); -- parameter name is uppercase in cfg.id_handlers (k); lowercase for error message
end
id_accesses_list[k] = cfg.keywords_xlate[access_level]; -- get translated keyword
end
end
end
return id_accesses_list;
end
--[[--------------------------< B U I L D _ I D _ L I S T >----------------------------------------------------
render the identifiers into a sorted sequence table
<ID_list_coins_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value
<options_t> is a table of various k/v option pairs provided in the call to new_build_id_list();
modified by this function and passed to all identifier rendering functions
<access_levels_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value (if valid)
returns a sequence table of sorted (by hkey - 'handler' key) rendered identifier strings
]]
local function build_id_list (ID_list_coins_t, options_t, access_levels_t)
local ID_list_t = {};
local accept;
local func_map = { --function map points to functions associated with hkey identifier
['ARXIV'] = arxiv,
['ASIN'] = asin,
['BIBCODE'] = bibcode,
['BIORXIV'] = biorxiv,
['CITESEERX'] = citeseerx,
['DOI'] = doi,
['EISSN'] = issn,
['HDL'] = hdl,
['ISBN'] = isbn,
['ISMN'] = ismn,
['ISSN'] = issn,
['JFM'] = jfm,
['JSTOR'] = jstor,
['LCCN'] = lccn,
['MR'] = mr,
['OCLC'] = oclc,
['OL'] = openlibrary,
['OSTI'] = osti,
['PMC'] = pmc,
['PMID'] = pmid,
['RFC']  = rfc,
['S2CID'] = s2cid,
['SBN'] = sbn,
['SSRN'] = ssrn,
['USENETID'] = usenet_id,
['ZBL'] = zbl,
}
for hkey, v in pairs (ID_list_coins_t) do
v, accept = has_accept_as_written (v); -- remove accept-as-written markup if present; accept is boolean true when markup removed; false else
-- every function gets the options table with value v and accept boolean
options_t.hkey = hkey; -- ~/Configuration handler key
options_t.id = v; -- add that identifier value to the options table
options_t.accept = accept; -- add the accept boolean flag
options_t.access = access_levels_t[hkey]; -- add the access level for those that have an |<identifier-access= parameter
options_t.handler = cfg.id_handlers[hkey];
options_t.coins_list_t = ID_list_coins_t; -- pointer to ID_list_coins_t; for |asin= and |ol=; also to keep erroneous values out of the citation's metadata
if func_map[hkey] then
table.insert (ID_list_t, {hkey, func_map[hkey] (options_t)}); -- call the function and add the results to the output sequence table
else
error (cfg.messages['unknown_ID_key'] .. ' ' .. hkey); -- here when func_map doesn't have a function for hkey
end
end
local function comp (a, b) -- used by following table.sort()
return a[1]:lower() < b[1]:lower(); -- sort by hkey
end
table.sort (ID_list_t, comp); -- sequence table of tables sort
for k, v in ipairs (ID_list_t) do -- convert sequence table of tables to simple sequence table of strings
ID_list_t[k] = v[2]; -- v[2] is the identifier rendering from the call to the various functions in func_map{}
end
return ID_list_t;
end
--[[--------------------------< O P T I O N S _ C H E C K >----------------------------------------------------
check that certain option parameters have their associated identifier parameters with values
<ID_list_coins_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value
<ID_support_t> is a sequence table of tables created in citation0() where each subtable has four elements:
[1] is the support parameter's assigned value; empty string if not set
[2] is a text string same as key in cfg.id_handlers
[3] is cfg.error_conditions key used to create error message
[4] is original ID support parameter name used to create error message
returns nothing; on error emits an appropriate error message
]]
local function options_check (ID_list_coins_t, ID_support_t)
for _, v in ipairs (ID_support_t) do
if is_set (v[1]) and not ID_list_coins_t[v[2]] then -- when support parameter has a value but matching identifier parameter is missing or empty
table.insert (z.message_tail, {set_message (v[3], (v[4]))}); -- emit the appropriate error message
end
end
end
--[[--------------------------< I D E N T I F I E R _ L I S T S _ G E T >--------------------------------------
Creates two identifier lists: a k/v table of identifiers and their values to be used locally and for use in the
COinS metadata, and a sequence table of the rendered identifier strings that will be included in the rendered
citation.
]]
local function identifier_lists_get (args, options_t, ID_support_t)
local ID_list_coins_t = extract_ids (args); -- get a table of identifiers and their values for use locally and for use in COinS
options_check (ID_list_coins_t, ID_support_t); -- ID support parameters must have matching identifier parameters
local ID_access_levels_t = extract_id_access_levels (args, ID_list_coins_t); -- get a table of identifier access levels
local ID_list_t = build_id_list (ID_list_coins_t, options_t, ID_access_levels_t); -- get a sequence table of rendered identifier strings
return ID_list_t, ID_list_coins_t; -- return the tables
end
end


Line 859: Line 1,627:
cfg = cfg_table_ptr;
cfg = cfg_table_ptr;


is_set = utilities_page_ptr.is_set; -- import functions from select Module:Citation/CS1/Utilities module
has_accept_as_written = utilities_page_ptr.has_accept_as_written; -- import functions from select Module:Citation/CS1/Utilities module
is_set = utilities_page_ptr.is_set;
in_array = utilities_page_ptr.in_array;
in_array = utilities_page_ptr.in_array;
set_error = utilities_page_ptr.set_error;
set_message = utilities_page_ptr.set_message;
select_one = utilities_page_ptr.select_one;
select_one = utilities_page_ptr.select_one;
add_maint_cat = utilities_page_ptr.add_maint_cat;
substitute = utilities_page_ptr.substitute;
substitute = utilities_page_ptr.substitute;
make_wikilink = utilities_page_ptr.make_wikilink;


z = utilities_page_ptr.z; -- table of tables in Module:Citation/CS1/Utilities
z = utilities_page_ptr.z; -- table of tables in Module:Citation/CS1/Utilities
Line 870: Line 1,639:




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


return {
return {
build_id_list = build_id_list,
auto_link_urls = auto_link_urls, -- table of identifier URLs to be used when auto-linking |title=
extract_ids = extract_ids,
identifier_lists_get = identifier_lists_get, -- experiment to replace individual calls to build_id_list, extract_ids, extract_id_access_levels
is_embargoed = is_embargoed;
is_embargoed = is_embargoed;
set_selected_modules = set_selected_modules;
set_selected_modules = set_selected_modules;
}
}
Retrieved from "https://nonbinary.wiki/wiki/Special:MobileDiff/17927...17955"