Module:Ċitazzjoni/CS1/Identifiers: Differenza bejn il-verżjonijiet

Content deleted Content added
Paġna ġdida: --[[--------------------------< 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, s...
 
Update
 
Linja 3:
]]
 
local has_accept_as_written, is_set, in_array, set_errorset_message, select_one, add_maint_cat, substitute, make_wikilink; -- functions in Module:Citation/CS1/Utilities
substitute, make_wikilink;
 
local z; -- table of tables defined in Module:Citation/CS1/Utilities
 
local cfg; -- table of configuration tables that are defined in Module:Citation/CS1/Configuration
 
 
--[[--------------------------< P A G E S C O P E V A R I A B L E S >--------------------------------------
 
declare variables here that have page-wide scope that are not brought in from other modules; that are created here and used here
 
]]
 
local auto_link_urls = {}; -- holds identifier URLs for those identifiers that can auto-link |title=
 
 
Line 14 ⟶ 24:
--[[--------------------------< W I K I D A T A _ A R T I C L E _ N A M E _ G E T >----------------------------
 
as an aid to internationalizing identifier-label wikilinks, gets identifier article names from wikidataWikidata.
 
returns :<lang code>:<article title> when <q> has an <article title> for <lang code>; nil else
Line 21 ⟶ 31:
 
for wikis that do not have mw.wikibase installed, returns nil
 
The call to mw.wikibase.getEntity() bumps the expensive parser function count
 
]]
Line 32 ⟶ 40:
 
local wd_article;
local this_wiki_code = cfg.this_wiki_code; -- wikipediaWikipedia subdomain; 'mten' for mten.wikipedia.org
 
wd_article = mw.wikibase.getEntitygetSitelink (q):getSitelink, (this_wiki_code .. 'wiki'); -- fetch article title from wdWD; nil when no title available at this wiki; bumps expensive parser function count
 
if wd_article then
wd_article = table.concat ({':', this_wiki_code, ':', wd_article}); -- interwiki-style link without brackets if taken from wdWD; leading colon required
end
 
return wd_article; -- article title from wdWD; nil else
end
 
 
--[[--------------------------< L I N K _ L A B E L _ M A K E >------------------------------------------------
 
common function to create identifier link label from handler table or from Wikidata
 
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 link_label_make (handler)
local wd_article;
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
wd_article = wikidata_article_name_get (handler.q); -- if Wikidata has an article title for this wiki, get it;
end
return (cfg.use_identifier_redirects and is_set (handler.redirect) and handler.redirect) or wd_article or handler.link;
end
 
Line 46 ⟶ 76:
--[[--------------------------< E X T E R N A L _ L I N K _ I D >----------------------------------------------
 
Formats a wiki -style external link
 
NB. Wikidata P1630 has a formatter URL with $1 placeholder for the ID which could be worked into our prefix/id/suffix
string, either overriding local definitions (auto-update) or as fallback for identifiers without local definitions.
But is expensive and could be risky if WD gets vandalized.
 
See Template_talk:Authority_control/Archive_8#Use_Wikidata_as_the_source_for_the_external_link
 
]]
 
local function external_link_id (options)
local url_string = options.id;
local ext_link;
local this_wiki_code = cfg.this_wiki_code; -- wikipediaWikipedia subdomain; 'mten' for mten.wikipedia.org
local wd_article; -- article title from wikidataWikidata
if options.encode == true or options.encode == nil then
url_string = mw.uri.encode( url_string );
end
 
if options.auto_link and is_set (options.access) then
auto_link_urls[options.auto_link] = table.concat ({options.prefix, url_string, options.suffix});
end
 
Line 64 ⟶ 104:
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
 
if not (cfg.use_identifier_redirects and is_set (options.redirect)) then -- redirect has priority so if enabled and available don't fetch from wikidata because expensive
wd_article = wikidata_article_name_get (options.q); -- if wikidata has an article title for this wiki, get it;
end
local label_link = (cfg.use_identifier_redirects and is_set (options.redirect) and options.redirect) or wd_article or options.link;
 
return table.concat ({
make_wikilink (label_linklink_label_make (options), options.label), -- redirect, wikidataWikidata link, or locally specified link (in that order)
options.separator or '&nbsp;',
ext_link
Line 81 ⟶ 115:
--[[--------------------------< I N T E R N A L _ L I N K _ I D >----------------------------------------------
 
Formats a wiki -style internal link
 
]]
Line 87 ⟶ 121:
local function internal_link_id(options)
local id = mw.ustring.gsub (options.id, '%d', cfg.date_names.local_digits); -- translate 'local' digits to Western 0-9
 
if not (cfg.use_identifier_redirects and is_set (options.redirect)) then -- redirect has priority so if enabled and available don't fetch from wikidata because expensive
wd_article = wikidata_article_name_get (options.q); -- if wikidata has an article title for this wiki, get it;
end
 
local label_link = (cfg.use_identifier_redirects and is_set (options.redirect) and options.redirect) or wd_article or options.link;
 
return table.concat (
{
make_wikilink (label_linklink_label_make (options), options.label), -- wiki -link the identifier label
options.separator or '&nbsp;', -- add the separator
make_wikilink (
Line 102 ⟶ 130:
{
options.prefix,
id, -- translated to westernWestern digits
options.suffix or ''
}),
substitute (cfg.presentation['bdi'], {'', mw.text.nowiki (options.id)}) -- bdi tags to prevent LatnLatin script identifiers from being reversed at rtlRTL language wikis
); -- nowiki because MediaWiki still has magic links for ISBN and the like; TODO: is it really required?
});
Line 113 ⟶ 141:
--[[--------------------------< I S _ E M B A R G O E D >------------------------------------------------------
 
Determines if a PMC identifier's online version is embargoed. Compares the date in |pmc-embargo-date= against today's date. If embargo date is
in the future, returns the content of |pmc-embargo-date=; otherwise, returns andan empty string because the embargo has expired or because
|pmc-embargo-date= was not set in this cite.
 
]]
Line 122 ⟶ 150:
if is_set (embargo) then
local lang = mw.getContentLanguage();
local good1, embargo_date, good2, todays_date;
good1, embargo_date = pcall ( lang.formatDate, lang, 'U', embargo );
good2, todays_date = pcall( lang.:formatDate, lang, ('U' );
if good1 and good2 then -- if embargo date andis today's date area good datesdate
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
add_maint_cat ('embargo')
return ''; -- unset because embargo has expired
end
end
end
return ''; -- |pmc-embargo-date= not set return empty string
end
 
Line 144 ⟶ 172:
2019-12-11T00:00Z <= biorxiv_date < today + 2 days
The dated form of biorxiv identifier has a start date of 2019-12-11. The unixUnix 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
Line 153 ⟶ 181:
 
This function does not work if it is fed month names for languages other than English. Wikimedia #time: parser
apparently doesn't understand non-EngishEnglish date month names. This function will always return false when the date
contains a non-English month name because good1 is false after the call to lang.formatDate(). To get around that
call this function with YYYY-MM-DD format dates.
Line 161 ⟶ 189:
local function is_valid_biorxiv_date (biorxiv_date)
local good1, good2;
local biorxiv_ts, tomorrow_ts; -- to hold unixUnix time stampstimestamps representing the dates
local lang_object = mw.getContentLanguage();
 
good1, biorxiv_ts = pcall (lang_object.formatDate, lang_object, 'U', biorxiv_date ); -- convert biorxiv_date value to unixUnix timesatmptimestamp
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
if good1 and good2 then -- lang.formatDate() returns a timestamp in the local script which which tonumber() may not understand
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
return false; -- one or both failed to convert to unixUnix time stamptimestamp
end
 
Line 180 ⟶ 208:
--[[--------------------------< IS _ V A L I D _ I S X N >-----------------------------------------------------
 
ISBN-10 and ISSN validator code calculates checksum across all isbnISBN/issnISSN digits including the check digit.
ISBN-13 is checked in isbn().
 
If the number is valid the result will be 0. Before calling this function, issbnISBN/issnISSN must be checked for length
and stripped of dashes, spaces and other non-isxnISxN characters.
 
]]
Line 191 ⟶ 219:
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)
Line 205 ⟶ 233:
--[[--------------------------< IS _ V A L I D _ I S X N _ 1 3 >-----------------------------------------------
 
ISBN-13 and ISMN validator code calculates checksum across all 13 isbnISBN/ismnISMN digits including the check digit.
If the number is valid, the result will be 0. Before calling this function, isbnISBN-13/ismnISMN must be checked for length
and stripped of dashes, spaces and other non-isxnISxN-13 characters.
 
]]
Line 218 ⟶ 246:
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
return temp % 10 == 0; -- sum modulo 10 is zero when isbnISBN-13/ismnISMN is correct
end
 
Line 224 ⟶ 252:
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------
 
lccnLCCN normalization (http://www.loc.gov/marc/lccn-namespace.html#normalization)
1. Remove all blanks.
2. If there is a forward slash (/) in the string, remove it, and remove all characters to the right of the forward slash.
Line 233 ⟶ 261:
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 lccnLCCN for lccn() to validate. There is no error checking (step 3.b.1) performed in this function.
]]
 
local function normalize_lccn (lccn)
lccn = lccn:gsub ("%s", ""); -- 1. strip whitespace
 
if nil ~= string.find (lccn, '/') then
lccn = lccn:match ("(.-)/"); -- 2. remove forward slash and all character to the right of it
end
 
local prefix
local suffix
prefix, suffix = lccn:match ("(.+)%-(.+)"); -- 3.a remove hyphen by splitting the string into prefix and suffix
 
if nil ~= suffix then -- if there was a hyphen
suffix = string.rep("0", 6-string.len (suffix)) .. suffix; -- 3.b.2 left fill the suffix with 0s if suffix length less than 6
lccn = prefix..suffix; -- reassemble the lccnLCCN
end
return lccn;
end
 
 
--============================<< I D E N T I F I E R F U N C T I O N S >>====================================
Line 262 ⟶ 291:
 
format and error check arXiv identifier. There are three valid forms of the identifier:
the first form, valid only between date codes 91089107 and 0703, is:
arXiv:<archive>.<class>/<date code><number><version>
where:
Line 292 ⟶ 321:
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 91089107-0703 format w/with &or w/owithout version
year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
year = tonumber(year);
Line 301 ⟶ 330:
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/with &or w/owithout version
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
year = tonumber(year);
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)
((7 == year) and (4 > month)) then --or -- when year is 07, is month invalid (before April)?
err_cat = true; -- flag for error message
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/with &or w/owithout version
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d%d[v%d]*$");
year = tonumber(year);
Line 322 ⟶ 351:
end
 
err_cat = err_cat and table.concat ({' ', set_errorset_message ('bad_arxiverr_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
Line 331 ⟶ 360:
text = table.concat ({text, ' [[//arxiv.org/archive/', class, ' ', class, ']]'}); -- external link within square brackets, not wikilink
else
text = table.concat ({text, ' ', set_errorset_message ('class_ignorederr_class_ignored')});
end
end
Line 341 ⟶ 370:
--[[--------------------------< B I B C O D E >--------------------------------------------------------------------
 
Validates (sort of) and formats a bibcode idID.
 
Format for bibcodes is specified here: http://adsabs.harvard.edu/abs_doc/help_pages/data.html#bibcodes
Line 362 ⟶ 391:
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
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 itsit's missing a letter)
end
end
Line 385 ⟶ 414:
 
if is_set (err_type) then -- if there was an error detected
text = text .. ' ' .. set_errorset_message( 'bad_bibcodeerr_bad_bibcode', {err_type});
end
return text;
Line 393 ⟶ 422:
--[[--------------------------< B I O R X I V >-----------------------------------------------------------------
 
Format bioRxiv idID and do simple error checking. Before 2019-12-11, biorXiv idsIDs were 10.1101/ followed by exactly
6 digits. After 2019-12-11, biorXiv idsIDs retained the six-digit identifier but prefixed that with a yyyy.mm.dd.
date and suffixed with an optional version identifier.
 
The bioRxiv idID is the string of characters:
https://doi.org/10.1101/078733 -> 10.1101/078733
or a date followed by a six-digit number followed by an optional version indicator 'v' and one or more digits:
Line 414 ⟶ 443:
'^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
Line 421 ⟶ 450:
 
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 leapyearleap-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
Line 430 ⟶ 459:
end -- err_cat remains set here when no match
 
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_errorset_message( 'bad_biorxiverr_bad_biorxiv')) or '');
end
 
Line 447 ⟶ 476:
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});
matched = id:match ("^10%.1%.1%.[1-9]%d?%d?%d?%.[1-9]%d?%d?%d?$");
if not matched then
text = text .. ' ' .. set_errorset_message( 'bad_citeseerxerr_bad_citeseerx' );
end
return text;
Line 467 ⟶ 496:
Suffix: character string of any length chosen by the registrant
 
This function checks a DOI name for: prefix/suffix. If the doiDOI name contains spaces or endashes, or, if it ends
with a period or a comma, this function will emit a bad_doi error message.
 
DOI names are case-insensitive and can incorporate any printable Unicode characters so the test for spaces, endash,
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in doiDOI names.
 
]]
 
local function doi (id, inactive, access, ignore_invalid)
local cat = ""err_cat;
local handler = cfg.id_handlers['DOI'];
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
local inactive_month, good;
Line 494 ⟶ 523:
end
else
inactive_year = nil; -- |doi-broken-date= has something but it isn't a date
end
if is_set(inactive_year) and is_set (inactive_month) then
set_message ('maint_doi_inactive_dated', {inactive_year, inactive_month, ' '});
table.insert( z.error_categories, 'Pages with DOIs inactive as of ' .. inactive_year .. ' ' .. inactive_month); -- use inactive month in category name
elseif is_set(inactive_year) then
set_message ('maint_doi_inactive_dated', {inactive_year, '', ''});
table.insert( z.error_categories, 'Pages with DOIs inactive as of ' .. inactive_year);
else
set_message ('maint_doi_inactive');
table.insert( z.error_categories, 'Pages with inactive DOIs'); -- when inactive doesn't contain a recognizable date
end
inactive = " (" .. cfg.messages['inactive'] .. ' ' .. inactive .. ')';
end
 
local registrant = id:match ('^10%.([^/]+)/[^%s–]-[^%.,]$'); -- registrant set when DOI has the proper basic form
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}) .. (inactive or '')
 
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+); accecptsaccepts: 10000–39999
'^[^1-34]%d%d%d%d$', -- 5 digits without subcode (0xxxx, 40000+); accecptsaccepts: 10000–3999910000–49999
'^[^1-9]%d%d%d%.%d%d*$', -- 4 digits with subcode (0xxx); accecptsaccepts: 1000–9999
'^[^1-9]%d%d%d$', -- 4 digits without subcode (0xxx); accecptsaccepts: 1000–9999
'^%d%d%d%d%d%d+', -- 6 or more digits
'^%d%d?%d?$', -- less than 4 digits without subcode (with subcode is legitimate)
Line 522 ⟶ 548:
'%s', -- any space character in registrant
}
 
if not ignore_invalid then
if registrant then -- when doi has proper form
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
cat err_cat = ' ' .. set_errorset_message ('bad_doierr_bad_doi'); -- when found, mark this doiDOI as bad
break; -- and done
end
end
else
err_cat = ' ' .. set_message ('err_bad_doi'); -- invalid directory or malformed
end
else
set_message ('maint_doi_ignore');
cat = ' ' .. set_error ('bad_doi'); -- invalid directory or malformed
end
 
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return text .. cat
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 text .. (err_cat and err_cat or ''); -- parentheses required
end
 
Line 554 ⟶ 589:
 
Query string parameters are named here: http://www.handle.net/proxy_servlet.html. query strings are not displayed
but since '?' is anallowedan allowed character in an hdlHDL, '?' followed by one of the query parameters is the only way we
have to detect the query string so that it isn't url URL-encoded with the rest of the identifier.
 
]]
Line 586 ⟶ 621:
 
if found then
id = hdl; -- found so replace id with the handle portion; this will be url 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 -- hdlHDL must contain a fwdforward slash, must not contain spaces, endashes, and must not end with period or comma
text = text .. ' ' .. set_errorset_message( 'bad_hdlerr_bad_hdl' );
end
return text;
Line 607 ⟶ 642:
]]
 
local function isbn( (isbn_str, ignore_invalid)
local handler = cfg.id_handlers['ISBN'];
if nil ~= isbn_str:match("[^%s-0-9X]") then
local function return_result (check, err_type) -- local function to handle the various returns
return false, cfg.err_msg_supl.char; -- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
local ISBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
prefix = handler.prefix, id = isbn_str, separator = handler.separator});
if ignore_invalid then -- if ignoring ISBN errors
set_message ('maint_isbn_ignore'); -- add a maint category even when there is no error
else -- here when not ignoring
if not check then -- and there is an error
return ISBN .. set_message ('err_bad_isbn', {err_type}, false, ' '); -- display an error message
end
end
return ISBN;
end
 
isbn_str = isbn_str:gsub( "-", "" ):gsub( " ", "" ); -- remove hyphens and spaces
localif lennil ~= isbn_str:lenmatch ('[^%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 isbn_strid:match ( "'^%d*X?$" ') == nil then -- fail if isbn_str has 'X' anywhere but last position
return return_result (false, cfg.err_msg_supl.form);
end
return return_result (is_valid_isxn(isbn_strid, 10), cfg.err_msg_supl.check);
else
if isbn_strid:match ( "'^%d+$" ') == nil then
return return_result (false, cfg.err_msg_supl.char); -- fail if isbn13ISBN-13 is not all digits
end
if isbn_strid:match ( "'^97[89]%d*$" ') == nil then
return return_result (false, cfg.err_msg_supl.prefix); -- fail when isbn13ISBN-13 does not begin with 978 or 979
end
if isbn_strid:match ('^9790') then
return return_result (false, cfg.err_msg_supl.group); -- group identifier '0' is reserved to ismnISMN
end
return return_result (is_valid_isxn_13 (isbn_strid), cfg.err_msg_supl.check);
end
end
 
 
--[[--------------------------< A MS A Z OI N >----------------------------------------------------------------------
 
Formats a link to Amazon. Do simple error checking: asinASIN must be mix of 10 numeric or uppercase alpha
characters. If a mix, first character must be uppercase alpha; if all numeric, asinsASINs must be 10-digit
isbnISBN. If 10-digit isbnISBN, add a maintenance category so a bot or awbAWB script can replace |asin= with |isbn=.
Error message if not 10 characters, if not isbn10ISBN-10, if mixed and first character is a digit.
 
|asin=630....... is (apparently) not a legitimate ISBN though it checksums as one; these do not cause this
function to emit the maint_asin message
 
This function is positioned here because it calls isbn()
Line 649 ⟶ 703:
]]
 
local function asin (id, domain)
local err_cat = ""
 
if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
err_cat = ' ' .. set_errorset_message ('bad_asinerr_bad_asin'); -- asinASIN is not a mix of 10 uppercase alpha and numeric characters
else
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
if isbn ( id ) then -- see if asinASIN value is isbn10or validates as ISBN-10
if not id:find ('^630') then -- 630xxxxxxx is (apparently) not a valid isbn prefix but is used by amazon as a numeric identifier
add_maint_cat ('ASIN');
set_message ('maint_asin'); -- begins with something other than 630 so possibly an isbn
end
elseif not is_set (err_cat) then
err_cat = ' ' .. set_errorset_message ('bad_asinerr_bad_asin'); -- asinASIN is not isbn10ISBN-10
end
elseif not id:match("^%u[%d%u]+$") then
err_cat = ' ' .. set_errorset_message ('bad_asinerr_bad_asin'); -- asin doesn't begin with uppercase alpha
end
end
if not is_set(domain) then
domain = "com";
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
local handler = cfg.id_handlers['ASIN'];
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix .. domain .. "/dp/",
id = id, encode = handler.encode, separator = handler.separator}) .. err_cat;
end
 
Line 681 ⟶ 737:
--[[--------------------------< I S M N >----------------------------------------------------------------------
 
Determines whether an ISMN string is valid. Similar to isbnISBN-13, ismnISMN is 13 digits beginingbeginning 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.
Line 694 ⟶ 750:
 
id_copy = id; -- save a copy because this testing is destructive
id = id:gsub( "('[%s-]"', "" ''); -- strip spaces,remove hyphens, and endashes from thewhite ismnspace
 
if 13 ~= id:len() or id:match( "^9790%d*$" ) == nil then -- ismnISMN must be 13 digits and begin with 9790
valid_ismn = false;
else
valid_ismn=is_valid_isxn_13 (id); -- validate ismnISMN
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.link), handler.label),
handler.separator,
id_copy
});
}); -- because no place to link to yet
 
if false == valid_ismn then
text = text .. ' ' .. set_errorset_message( 'bad_ismnerr_bad_ismn' ) -- add an error message if the ismnISMN is invalid
end
Line 722 ⟶ 778:
--[[--------------------------< I S S N >----------------------------------------------------------------------
 
Validate and format an issnISSN. 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:
Line 728 ⟶ 784:
|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 issnISSN midpoint. It also validates the issnISSN 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 issnISSN error message. The
issnISSN is always displayed with a hyphen, even if the issnISSN was given as a single group of 8 digits.
 
]]
 
local function issn (id, e, ignore_invalid)
local issn_copy = id; -- save a copy of unadulterated issnISSN; use this version for display if issnISSN does not validate
local handler;
local text;
local valid_issn = true;
handler = cfg.id_handlers[e and 'EISSN' or 'ISSN'];
if e then
handler = cfg.id_handlers['EISSN'];
else
handler = cfg.id_handlers['ISSN'];
end
 
id = id:gsub( "('[%s-]"', "" ''); -- strip spaces,remove hyphens, and endashes from the issnwhitespace
 
if 8 ~= id:len() or nil == id:match( "^%d*X?$" ) then -- validate the issnISSN: 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 issnISSN
end
 
Line 758 ⟶ 810:
id = string.sub( id, 1, 4 ) .. "-" .. string.sub( id, 5 ); -- if valid, display correctly formatted version
else
id = issn_copy; -- if not valid, use the show the invalid issnISSN with error message
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})
 
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
if false == valid_issn then
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
text = text .. ' ' .. set_error( 'bad_issn', e and 'e' or '' ) -- add an error message if the issn is invalid
 
end
if ignore_invalid then
set_message ('maint_issn_ignore');
else
if false == valid_issn then
text = text .. ' ' .. set_message ('err_bad_issn', e and 'e' or ''); -- add an error message if the ISSN is invalid
end
end
return text
Line 786 ⟶ 842:
 
if is_set (id_num) then
add_maint_catset_message ('jfm_formatmaint_jfm_format');
else -- plain number without mr prefix
id_num = id; -- if here id does not have prefix
Line 794 ⟶ 850:
id = id_num; -- jfm matches pattern
else
err_cat = ' ' .. set_errorset_message( 'bad_jfmerr_bad_jfm' ); -- set an error message
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
 
Line 809 ⟶ 865:
 
length = 8 then all digits
length = 9 then lccn[1] is lower caselowercase alpha
length = 10 then lccn[1] and lccn[2] are both lower caselowercase alpha or both digits
length = 11 then lccn[1] is lower case alpha, lccn[2] and lccn[3] are both lower caselowercase alpha or both digits
length = 12 then lccn[1] and lccn[2] are both lower caselowercase alpha
 
]]
Line 819 ⟶ 875:
local handler = cfg.id_handlers['LCCN'];
local err_cat = ''; -- presume that LCCN is valid
local id = lccn; -- local copy of the lccnLCCN
 
id = normalize_lccn (id); -- get canonical form (no whitespace, hyphens, forward slashes)
local len = id:len(); -- get the length of the lccnLCCN
 
if 8 == len then
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits)
err_cat = ' ' .. set_errorset_message ( 'bad_lccnerr_bad_lccn' ); -- set an error message
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?
err_cat = ' ' .. set_errorset_message ( 'bad_lccnerr_bad_lccn' ); -- set an error message
end
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_errorset_message ( 'bad_lccnerr_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_errorset_message ( 'bad_lccnerr_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_errorset_message ( 'bad_lccnerr_bad_lccn' ); -- no match, set an error message
end
else
err_cat = ' ' .. set_errorset_message ( 'bad_lccnerr_bad_lccn' ); -- wrong length, set an error message
end
 
if not is_set (err_cat) and nil ~= lccn:find ('%s') then
err_cat = ' ' .. set_errorset_message ( 'bad_lccnerr_bad_lccn' ); -- lccn contains a space, set an error message
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
 
Line 874 ⟶ 930:
 
if is_set (id_num) then
set_message ('maint_mr_format'); -- add maint cat
add_maint_cat ('mr_format');
else -- plain number without mr prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
Line 883 ⟶ 939:
id = string.rep ('0', 7-id_len ) .. id_num; -- zero-fill leading digits
else
err_cat = ' ' .. set_errorset_message ( 'bad_mrerr_bad_mr' ); -- set an error message
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
 
Line 893 ⟶ 949:
--[[--------------------------< O C L C >----------------------------------------------------------------------
 
Validate and format an oclcOCLC idID. 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
 
Line 912 ⟶ 968:
number = id:match('%(OCoLC%)([1-9]%d*)'); -- get the number
if 9 < number:len() then
number = nil; -- contrainconstrain to 1 to 9 digits; change this when oclcOCLC issues 10-digit numbers
end
elseif id:match('^%d+$') then -- no prefix
number = id; -- get the number
if 10 < number:len() then
number = nil; -- contrainconstrain to 1 to 10 digits; change this when oclcOCLC issues 11-digit numbers
end
end
Line 924 ⟶ 980:
id = number; -- exclude prefix, if any, from external link
else
err_msg = ' ' .. set_errorset_message ( 'bad_oclcerr_bad_oclc' ) -- add an error message if the id is malformed
end
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;
 
return text;
Line 941 ⟶ 997:
 
local function openlibrary(id, access)
local code;
local handler = cfg.id_handlers['OL'];
local ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- strip optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W';
local ident;
local error_msg = '';
local prefix = { -- these are appended to the handler.prefix according to code
ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W'; remove OL prefix
['A']='authors/OL',
['M']='books/OL',
['W']='works/OL',
['X']='OL' -- not a code; spoof when 'code' in id is invalid
};
 
if not is_set (ident) then -- if malformed return an error
code = 'X'; -- no code or id completely invalid
return external_link_id ({link=handler.link, label=handler.label, q=handler.q, redirect=handler.redirect,
ident = id; -- copy id to ident so that we display the flawed identifier
prefix=handler.prefix .. 'OL',
error_msg = ' ' .. set_message ('err_bad_ol');
id=id, separator=handler.separator, encode = handler.encode,
access = access}) .. ' ' .. set_error( 'bad_ol' );
end
id = ident; -- use ident without the optional OL prefix (it has been removed)
if ( code == "A" ) then
return external_link_id ({link=handler.link, label=handler.label, q=handler.q, redirect=handler.redirect,
prefix=handler.prefix .. 'authors/OL',
id=id, separator=handler.separator, encode = handler.encode,
access = access})
end
if ( code == "M" ) then
return external_link_id ({link=handler.link, label=handler.label, q=handler.q, redirect=handler.redirect,
prefix=handler.prefix .. 'books/OL',
id=id, separator=handler.separator, encode = handler.encode,
access = access})
end
 
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
if ( code == "W" ) then
prefix = handler.prefix .. prefix[code],
return external_link_id ({link=handler.link, label=handler.label, q=handler.q, redirect=handler.redirect,
id = ident, separator = handler.separator, encode = handler.encode,
prefix=handler.prefix .. 'works/OL',
access = access}) .. error_msg;
id=id, separator=handler.separator, encode = handler.encode,
access = access})
end
end
 
Line 996 ⟶ 1,037:
]]
 
local function pmc (id, embargo)
local test_limit = cfg.id_limits.pmc; -- update this value as PMCs approach
local handler = cfg.id_handlers['PMC'];
local err_cat = ''; -- presume that PMC is valid
local id_num;
local text;
 
id_num = id:match ('^[Pp][Mm][Cc](%d+)$'); -- identifier with pmcPMC prefix
 
if is_set (id_num) then
add_maint_catset_message ('pmc_formatmaint_pmc_format');
else -- plain number without pmcPMC prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
end
Line 1,013 ⟶ 1,053:
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 1 > id_num or test_limithandler.id_limit < id_num then -- if PMC is outside test limit boundaries
err_cat = ' ' .. set_errorset_message ( 'bad_pmcerr_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_errorset_message ( 'bad_pmcerr_bad_pmc' ); -- set an error message
end
if is_set (embargo) and is_set (is_embargoed (embargo)) then -- is PMC is still embargoed?
text = table.concat ( -- still embargoed so no external link
{
make_wikilink (link_label_make (handler.link), handler.label),
handler.separator,
id,
(err_cat and err_cat or '') -- parens required
err_cat
});
else
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
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;,
auto_link = not err_cat and 'pmc' or nil -- do not auto-link when PMC has error
}) .. (err_cat and err_cat or ''); -- parentheses required
end
return text;
Line 1,047 ⟶ 1,089:
 
local function pmid(id)
local test_limit = cfg.id_limits.pmid; -- update this value as PMIDs approach
local handler = cfg.id_handlers['PMID'];
local err_cat = ''; -- presume that PMID is valid
if id:match("[^%d]") then -- if PMID has anything but digits
err_cat = ' ' .. set_errorset_message ( 'bad_pmiderr_bad_pmid' ); -- set an error message
else -- PMID is only digits
local id_num = tonumber(id); -- convert id to a number for range testing
if 1 > id_num or test_limithandler.id_limit < id_num then -- if PMID is outside test limit boundaries
err_cat = ' ' .. set_errorset_message ( 'bad_pmiderr_bad_pmid' ); -- set an error message
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
 
Line 1,067 ⟶ 1,108:
--[[--------------------------< S 2 C I D >--------------------------------------------------------------------
 
Format an s2cidS2CID, do simple error checking
 
S2CIDs are sequential numbers beginning at 1 and counting up. This code checks the s2cidS2CID 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.
Line 1,076 ⟶ 1,117:
 
local function s2cid (id, access)
local test_limit = cfg.id_limits.s2cid; -- update this value as S2CIDs approach
local handler = cfg.id_handlers['S2CID'];
local err_cat = ''; -- presume that S2CID is valid
Line 1,082 ⟶ 1,122:
local text;
id_num = id:match ('^[1-9]%d+*$'); -- id must be all digits; must not begin with 0; no open access flag
 
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 test_limithandler.id_limit < id_num then -- if S2CID is outside test limit boundaries
err_cat = ' ' .. set_errorset_message ( 'bad_s2cidbad_s2cid_err' ); -- set an error message
end
 
else -- when id format incorrect
err_cat = ' ' .. set_errorset_message ( 'bad_s2cidbad_s2cid_err' ); -- set an error message
end
 
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id:gsub ('%.%a%a', ''), separator = handler.separator, encode = handler.encode, access = access}) .. err_cat;
 
return text;
Line 1,103 ⟶ 1,143:
--[[--------------------------< S B N >------------------------------------------------------------------------
 
9-digit form of isbn10ISBN-10; uses same check-digit validation when sbnSBN is prefixed with an additional '0' to make 10 digits
 
]]
 
local function sbn (id, ignore_invalid)
local handler = cfg.id_handlers['SBN'];
local check;
local function return_result (check, err_type) -- local function to handle the various returns
local err_type = '';
local SBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator});
if not ignore_invalid then -- if not ignoring SBN errors
if not check then
return SBN .. set_message ('err_bad_sbn', {err_type}, false, ' '); -- display an error message
end
else
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
return SBN;
end
 
if nil ~= id:match ("'[^%s-0-9X]"') then
return return_result (false, cfg.err_msg_supl.char); -- fail if sbnSBN contains anything but digits, hyphens, or the uppercase X
end
 
idlocal ident = id:gsub( "('[%s-]"', "" ''); -- stripremove spaceshyphens and hyphenswhitespace; they interfere with the rest fromof the sbntests
 
if 9 ~= idident:len() then
return return_result (false, cfg.err_msg_supl.length); -- fail if incorrect length
end
 
if idident:match ( "'^%d*X?$" ') == nil then -- fail if sbn has 'X' anywhere but last position
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' .. idident, 10), cfg.err_msg_supl.check); -- prefix sbn with '0' and validate as isbn10
end
 
Line 1,131 ⟶ 1,182:
--[[--------------------------< S S R N >----------------------------------------------------------------------
 
Format an ssrnSSRN, do simple error checking
 
SSRNs are sequential numbers beginning at 100? and counting up. This code checks the ssrnSSRN 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.
Line 1,140 ⟶ 1,191:
 
local function ssrn (id)
local test_limit = cfg.id_limits.ssrn; -- update this value as SSRNs approach
local handler = cfg.id_handlers['SSRN'];
local err_cat = ''; -- presume that SSRN is valid
Line 1,150 ⟶ 1,200:
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 test_limithandler.id_limit < id_num then -- if SSRN is outside test limit boundaries
err_cat = ' ' .. set_errorset_message ( 'bad_ssrnerr_bad_ssrn' ); -- set an error message
end
else -- when id format incorrect
err_cat = ' ' .. set_errorset_message ( 'bad_ssrnerr_bad_ssrn' ); -- set an error message
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 = handler.access}) .. err_cat;
 
return text;
Line 1,174 ⟶ 1,224:
local handler = cfg.id_handlers['USENETID'];
 
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_errorset_message ( 'bad_usenet_iderr_bad_usenet_id' ) -- add an error message if the message id is invalid
end
Line 1,200 ⟶ 1,250:
if id:match('^%d%d%d%d%d%d%d%d$') then -- is this identifier using temporary format?
add_maint_catset_message ('zblmaint_zbl'); -- yes, add maint cat
elseif not id:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then -- not temporary, is it normal format?
err_cat = ' ' .. set_errorset_message ( 'bad_zblerr_bad_zbl' ); -- no, set an error message
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
 
Line 1,224 ⟶ 1,274:
local function build_id_list( id_list, options )
local new_list, handler = {};
local accept;
 
local function fallback(k) return { __index = function(t,i) return cfg.id_handlers[k][i] end } end;
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
v, accept = has_accept_as_written (v); -- remove and note accept-as-written markup if present
 
-- k is uppercase identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
-- fallback to read-only cfg
handler = setmetatable( { ['id'] = v, ['access'] = options.IdAccessLevels[k] }, fallback(k) );
Line 1,248 ⟶ 1,302:
table.insert( new_list, {handler.label, citeseerx( v ) } );
elseif k == 'DOI' then
table.insert( new_list, {handler.label, doi( v, options.DoiBroken, handler.access, accept) } );
elseif k == 'EISSN' then
table.insert( new_list, {handler.label, issn( v, true, accept) } ); -- true distinguishes eissnEISSN from issnISSN
elseif k == 'HDL' then
table.insert( new_list, {handler.label, hdl( v, handler.access ) } );
elseif k == 'ISBN' then
table.insert( new_list, {handler.label, isbn (v, (accept or options.IgnoreISBN)) } );
local ISBN = internal_link_id( handler );
local check;
local err_type = '';
check, err_type = isbn( v );
if not check then
if is_set(options.IgnoreISBN) then -- ISBN is invalid; if |ignore-isbn-error= set
add_maint_cat ('ignore_isbn_err'); -- ad a maint category
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
table.insert( new_list, {handler.label, ISBN } );
elseif k == 'ISMN' then
table.insert( new_list, {handler.label, ismn( v ) } );
elseif k == 'ISSN' then
table.insert( new_list, {handler.label, issn( v, false, accept) } );
elseif k == 'JFM' then
table.insert( new_list, {handler.label, jfm( v ) } );
Line 1,289 ⟶ 1,330:
table.insert( new_list, {handler.label, s2cid( v, handler.access ) } );
elseif k == 'SBN' then
table.insert( new_list, {handler.label, sbn (v, accept) } );
local SBN = internal_link_id (handler);
local check; -- boolean validation result
local err_type = '';
check, err_type = sbn (v);
if not check then
SBN = SBN .. set_error( 'bad_sbn', {err_type}, false, " ", "" ); -- display an error message
end
table.insert( new_list, {handler.label, SBN } );
elseif k == 'SSRN' then
table.insert( new_list, {handler.label, ssrn( v ) } );
Line 1,325 ⟶ 1,359:
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
the identifier list. Emits redundant error message isif more than one alias exists in args
 
]]
Line 1,331 ⟶ 1,365:
local function extract_ids( args )
local id_list = {}; -- list of identifiers found in args
for k, v in pairs( cfg.id_handlers ) do -- k is ucuppercase 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_parameterserr_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
end
Line 1,357 ⟶ 1,391:
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_errorset_message( 'invalid_param_val', {access_param, access_level}, true ) } );
access_level = nil; -- invalid so unset
end
if not is_set(id_list[k]) then -- identiferidentifier access-level must have a matching identifier
table.insert( z.message_tail, { set_errorset_message( 'param_access_requires_paramerr_param_access_requires_param', {k:lower()}, true ) } ); -- paramparameter 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
Line 1,380 ⟶ 1,414:
cfg = cfg_table_ptr;
 
is_sethas_accept_as_written = utilities_page_ptr.is_sethas_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;
set_errorset_message = utilities_page_ptr.set_errorset_message;
select_one = utilities_page_ptr.select_one;
add_maint_cat = utilities_page_ptr.add_maint_cat;
substitute = utilities_page_ptr.substitute;
make_wikilink = utilities_page_ptr.make_wikilink;
Line 1,396 ⟶ 1,430:
 
return {
auto_link_urls = auto_link_urls, -- table of identifier URLs to be used when auto-linking |title=
build_id_list = build_id_list,
extract_ids = extract_ids,