Changes
no edit summary
]]
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=
--============================<< H E L P E R F U N C T I O N S >>============================================
--[[--------------------------< 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 Wikidata.
returns :<lang code>:<article title> when <q> has an <article title> for <lang code>; nil else
for identifiers that do not have q, returns nil
for wikis that do not have mw.wikibase installed, returns nil
]]
local function wikidata_article_name_get (q)
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
local wd_article;
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
if wd_article then
wd_article = table.concat ({':', this_wiki_code, ':', wd_article}); -- interwiki-style link without brackets if taken from WD; leading colon required
end
return wd_article; -- article title from WD; 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
--[[--------------------------< 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/suffixstring, 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; -- wikipedia Wikipedia subdomain; 'en' for en.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
if is_set(options.access) then
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 table.concat ({
make_wikilink (wd_article or link_label_make (options.link), options.label), -- wikidata redirect, Wikidata link , or locally specifiedlink (in that order)
options.separator or ' ',
ext_link
--[[--------------------------< I N T E R N A L _ L I N K _ I D >----------------------------------------------
Formats a wiki -style internal link
]]
return table.concat (
{
make_wikilink (link_label_make (options.link), options.label), -- wiki-link the identifier label options.separator or ' ', -- add the separator
make_wikilink (
table.concat (
{
options.prefix,
id, -- translated to western Western digits
options.suffix or ''
}),
substitute (cfg.presentation['bdi'], {'', mw.text.nowiki (options.id)}) -- bdi tags to prevent Latn Latin script identifiers from being reversed at rtl RTL language wikis
); -- nowiki because MediaWiki still has magic links for ISBN and the like; TODO: is it really required?
});
--[[--------------------------< 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 isin the future, returns the content of |pmc-embargo-date=; otherwise, returns and an empty string because the embargo has expired or because|pmc-embargo-date= was not set in this cite.
]]
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 and today's is a good date are good dates if tonumber( embargo_date ) >= tonumber( todays_date ) then -- is embargo date is in the future?
return embargo; -- still embargoed
else
return ''; -- unset because embargo has expired
end
end
end
return ''; -- |pmc-embargo-date= not set return empty stringend --[=[-------------------------< 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 UTCtoday 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 This function does not work if it is fed month names for languages other than English. Wikimedia #time: parserapparently doesn't understand non-English date month names. This function will always return false when the datecontains a non-English month name because good1 is false after the call to lang.formatDate(). To get around thatcall this function with YYYY-MM-DD format dates. ]=] 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(); good1, biorxiv_ts = pcall (lang_object.formatDate, lang_object, 'U', biorxiv_date ); -- convert biorxiv_date value to Unix timestamp 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 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 Unix timestamp end 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 >-----------------------------------------------------
ISBN-10 and ISSN validator code calculates checksum across all isbnISBN/issn ISSN 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/issn ISSN must be checked for lengthand stripped of dashes, spaces and other non-isxn 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
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 isbnISBN/ismn ISMN digits including the check digit.If the number is valid, the result will be 0. Before calling this function, isbnISBN-13/ismn ISMN must be checked for lengthand stripped of dashes, spaces and other non-isxnISxN-13 characters.
]]
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/ismn ISMN is correct
end
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------
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.
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 LCCN 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 >>====================================
format and error check arXiv identifier. There are three valid forms of the identifier:
the first form, valid only between date codes 9108 9107 and 0703 , is:
arXiv:<archive>.<class>/<date code><number><version>
where:
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+$") then -- test for the 91089107-0703 format w/ & w/o with or without version
year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
year = tonumber(year);
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 with or without 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+$") then -- test for the 1501- format w/ & w/o with or without version
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d%d[v%d]*$");
year = tonumber(year);
end
err_cat = err_cat and table.concat ({' ', set_error set_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
text = table.concat ({text, ' [[//arxiv.org/archive/', class, ' ', class, ']]'}); -- external link within square brackets, not wikilink
else
text = table.concat ({text, ' ', set_error set_message ('class_ignorederr_class_ignored')});
end
end
--[[--------------------------< 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
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 its it's missing a letter)
end
end
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;
--[[--------------------------< B I O R X I V >-----------------------------------------------------------------
Format bioRxiv id ID and do simple error checking. BiorXiv ids are 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. The bioRxiv id ID is the number following the last slash in the bioRxiv-issued DOIstring of characters: https://doi.org/10.1101/078733 -> 10.1101/078733or a date followed by a six-digit number followed by an optional version indicator 'v' and one or more digits: https://www.biorxiv.org/content/10.1101/2019.12.11.123456v2 -> 10.1101/2019.12.11.123456v2 see https://www.biorxiv.org/about-biorxiv
]]
local function biorxiv(id)
local handler = cfg.id_handlers['BIORXIV'];
local err_cat = ''true; -- presume flag; assume that bioRxiv id is validthere will be an error
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 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_catand (' ' .. set_message( 'err_bad_biorxiv')) or '');
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});
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;
Suffix: character string of any length chosen by the registrant
This function checks a DOI name for: prefix/suffix. If the doi DOI 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 doi DOI 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;
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
elseif is_set(inactive_year) then
else
end
inactive = " (" .. cfg.messages['inactive'] .. ' ' .. inactive .. ')';
end
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?$', -- less than 4 digits without subcode (with subcode is legitimate)
'^5555$', -- test registrant will never resolve
'%s', -- any space character in registrant
}
if not ignore_invalid then if registrant then -- when doi 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_error set_message ('bad_doierr_bad_doi'); -- when found, mark this doi DOI as bad break; -- and done end
end
else
err_cat = ' ' .. set_message ('err_bad_doi'); -- invalid directory or malformed
end
else
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 text .. cat (err_cat and err_cat or ''); -- parentheses required
end
terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in HDLs.
Query string parameters are named here: http://www.handle.net/proxy_servlet.html. query strings are not displayed
but since '?' is an allowed character in an HDL, '?' followed by one of the query parameters is the only way we
have to detect the query string so that it isn't URL-encoded with the rest of the identifier.
]]
local function hdl(id, access)
local handler = cfg.id_handlers['HDL'];
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',
}
local hdl, suffix, param = id:match ('(.-)(%?(%a+).+)$'); -- look for query string local found; if hdl then -- when there are query strings, this is the handle identifier portion for _, q in ipairs (query_params) do -- spin through the list of query parameters if param:match ('^' .. q) then -- if the query string begins with one of the parameters found = true; -- announce a find break; -- and stop looking end end end if found then id = hdl; -- found so replace id with the handle portion; this will be URL-encoded, suffix will not else suffix = ''; -- make sure suffix is empty string for concatenation else end local text = external_link_id({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect, prefix=handler.prefix,id=id, suffix = suffix,separator=handler.separator, encode=handler.encode, access=access})
if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then -- hdl HDL must contain a fwd forward 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;
]]
local function isbn( isbn_str , ignore_invalid) if nil ~local handler = isbn_str:matchcfg.id_handlers['ISBN']; local function return_result ("[^%scheck, err_type) -0-9X]"local function to handle the various returns 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, 'invalid character'); -- fail if isbn_str contains anything but digits, hyphens, or the uppercase Xdisplay an error message end end return ISBN;
end
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?$" ') == nil then -- fail if isbn_str has 'X' anywhere but last position return return_result (false, 'invalid cfg.err_msg_supl.form');
end
return return_result (is_valid_isxn(isbn_strid, 10), 'checksum'cfg.err_msg_supl.check);
else
if isbn_strid:match( "'^%d+$" ') == nil then return return_result (false, 'invalid character'cfg.err_msg_supl.char); -- fail if isbn13 ISBN-13 is not all digits
end
if isbn_strid:match( "'^97[89]%d*$" ') == nil then return return_result (false, 'invalid cfg.err_msg_supl.prefix'); -- fail when isbn13 ISBN-13 does not begin with 978 or 979
end
if isbn_strid:match ('^9790') then return return_result (false, 'invalid cfg.err_msg_supl.group id'); -- group identifier '0' is reserved to ismnISMN
end
return return_result (is_valid_isxn_13 (isbn_strid), 'checksum'cfg.err_msg_supl.check);
end
end
--[[--------------------------< A M A Z O S I N >----------------------------------------------------------------------
Formats a link to Amazon. Do simple error checking: asin ASIN must be mix of 10 numeric or uppercase alphacharacters. If a mix, first character must be uppercase alpha; if all numeric, asins ASINs must be 10-digitisbnISBN. If 10-digit isbnISBN, add a maintenance category so a bot or awb AWB script can replace |asin= with |isbn=.Error message if not 10 characters, if not isbn10ISBN-10, if mixed and first character is a digit. |asin=630....... is (apparently) not a legitimate ISBN though it checksums as one; these do not cause thisfunction to emit the maint_asin message
This function is positioned here because it calls isbn()
]]
local function asin(id, domain)
local err_cat = ""
if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
err_cat = ' ' .. set_error set_message ('bad_asinerr_bad_asin'); -- asin ASIN 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]$") then -- if 10-digit numeric (or 9 digits with terminal X)
if isbn( id ) then -- see if asin ASIN value is isbn10or validates as ISBN-10 add_maint_cat if not id:find ('ASIN^630') then -- 630xxxxxxx is (apparently) not a valid isbn prefix but is used by amazon as a numeric identifier set_message ('maint_asin'); -- begins with something other than 630 so possibly an isbn end
elseif not is_set (err_cat) then
err_cat = ' ' .. set_error set_message ('bad_asinerr_bad_asin'); -- asin ASIN is not isbn10ISBN-10
end
elseif not id:match("^%u[%d%u]+$") then
err_cat = ' ' .. set_error set_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
--[[--------------------------< I S M N >----------------------------------------------------------------------
Determines whether an ISMN string is valid. Similar to isbnISBN-13, ismn ISMN is 13 digits begining 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.
id_copy = id; -- save a copy because this testing is destructive
id=id:gsub( "'[%s-–]"', "" ''); -- strip spaces, remove hyphens, and endashes from the ismnwhite space
if 13 ~= id:len() or id:match( "^9790%d*$" ) == nil then -- ismn ISMN 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=idid_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 ismn ISMN is invalid
end
--[[--------------------------< 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:
|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 ISSN midpoint. It also validates the issn 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 ISSN error message. Theissn ISSN is always displayed with a hyphen, even if the issn ISSN 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 issn ISSN does not validate
local handler;
local text;
local valid_issn = true;
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
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 issn ISSN 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}) 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
if is_set (id_num) then
else -- plain number without mr prefix
id_num = id; -- if here id does not have prefix
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
length = 8 then all digits
length = 9 then lccn[1] is lower case lowercase alphalength = 10 then lccn[1] and lccn[2] are both lower case lowercase alpha or both digitslength = 11 then lccn[1] is lower case alpha, lccn[2] and lccn[3] are both lower case lowercase alpha or both digitslength = 12 then lccn[1] and lccn[2] are both lower case lowercase alpha
]]
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
if is_set (id_num) then
else -- plain number without mr prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
id = string.rep ('0', 7-id_len ) .. id_num; -- zero-fill leading digits
else
err_cat = ' ' .. set_errorset_message ( 'bad_mrerr_bad_mr' ); -- set an error message
end
return external_link_id({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect, prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode}) .. err_cat;
end
--[[--------------------------< O C L C >----------------------------------------------------------------------
Validate and format an oclc idOCLC ID. https://www.oclc.org/batchload/controlnumber.en.html {{dead link}}
archived at: https://web.archive.org/web/20161228233804/https://www.oclc.org/batchload/controlnumber.en.html
number = id:match('%(OCoLC%)([1-9]%d*)'); -- get the number
if 9 < number:len() then
number = nil; -- contrain constrain to 1 to 9 digits; change this when oclc OCLC issues 10-digit numbers
end
elseif id:match('^%d+$') then -- no prefix
number = id; -- get the number
if 10 < number:len() then
number = nil; -- contrain constrain to 1 to 10 digits; change this when oclc OCLC issues 11-digit numbers
end
end
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;
local function openlibrary(id, access)
local handler = cfg.id_handlers['OL'];
local ident; ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- strip optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W'; remove local error_msg = ''; local prefix = { -- these are appended to the handler.prefix according to code ['A']='authors/OL prefix', ['M']='books/OL', ['W']='works/OL', ['X']='OL' -- not a code; spoof when 'code' in id is invalid };
if not is_set (ident) then -- if malformed return an error return external_link_id({link=handler.link, label=handler.label, q code = handler.q, prefix=handler.prefix .. 'OLX', id=; -- no code or id, separator=handler.separator, encode = handler.encode,completely invalid access ident = access}) .. ' ' .. set_error( 'bad_ol' ); end id = ident; -- use copy id to ident without so that we display the optional OL prefix (it has been removed) if ( code == "A" ) thenflawed identifier return external_link_id({link=handler.link, label=handler.label, q = handler.q, prefixerror_msg =handler.prefix .. 'authors/OL', id=id, separator=handler.separator, encode = handler.encode, access = access}) end if set_message ( code == "M" ) then return external_link_id({link=handler.link, label=handler.label, q = handler.q, prefix=handler.prefix .. 'books/OLerr_bad_ol', id=id, separator=handler.separator, encode = handler.encode, access = access});
end
end
]]
local function pmc(id, embargo) local test_limit = 8000000; -- update this value as PMCs approach
local handler = cfg.id_handlers['PMC'];
local err_cat = ''; -- presume that PMC is valid
local id_num;
local text;
if is_set (id_num) then
id_num = id:match ('^%d+$'); -- if here id is all digits
end
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_limit handler.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_catand err_cat or '') -- parens required
});
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, auto_link = not err_cat and 'pmc' or nil -- do not auto-link when PMC has error }) .. (err_catand err_cat or ''); -- parentheses required
end
return text;
local function pmid(id)
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_limit handler.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 --[[--------------------------< 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 onlydigits and is less than test_limit; the value in local variable test_limit will need to be updated periodicallyas more S2CIDs are issued. ]] local function s2cid (id, access) local handler = cfg.id_handlers['S2CID']; local err_cat = ''; -- presume that S2CID is valid local id_num; 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 handler.id_limit < id_num then -- if S2CID is outside test limit boundaries err_cat = ' ' .. set_message ('bad_s2cid_err'); -- set an error message end else -- when id format incorrect err_cat = ' ' .. set_message ('bad_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;end --[[--------------------------< S B N >------------------------------------------------------------------------ 9-digit form of ISBN-10; uses same check-digit validation when SBN is prefixed with an additional '0' to make 10 digits ]] local function sbn (id, ignore_invalid) local handler = cfg.id_handlers['SBN']; local function return_result (check, err_type) -- local function to handle the various returns 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 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 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 ssrnSSRN, do simple error checking
SSRNs are sequential numbers beginning at 100? and counting up. This code checks the ssrn 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 (id)
local handler = cfg.id_handlers['SSRN'];
local err_cat = ''; -- presume that SSRN is valid
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_limit handler.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;
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
if id:match('^%d%d%d%d%d%d%d%d$') then -- is this identifier using temporary format?
elseif not id:match('^%d?%d?%d?%d%.%d%d%d%d%d$') 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
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 v, accept = has_accept_as_written (v); -- remove and note accept-as-written markup if present -- k is uc 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) );
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 eissn EISSN from issnISSN
elseif k == 'HDL' then
table.insert( new_list, {handler.label, hdl( v, handler.access ) } );
elseif k == 'ISBN' then
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 ) } );
elseif k == 'PMID' then
table.insert( new_list, {handler.label, pmid( v ) } );
elseif k == 'S2CID' then
table.insert( new_list, {handler.label, s2cid( v, handler.access ) } );
elseif k == 'SBN' then
table.insert( new_list, {handler.label, sbn (v, accept) } );
elseif k == 'SSRN' then
table.insert( new_list, {handler.label, ssrn( v ) } );
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 is if more than one alias exists in args
]]
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 uc 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_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
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 -- identifer identifier 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 ) } ); -- param 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
cfg = cfg_table_ptr;
in_array = utilities_page_ptr.in_array;
select_one = utilities_page_ptr.select_one;
substitute = utilities_page_ptr.substitute;
make_wikilink = utilities_page_ptr.make_wikilink;
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,