Module:Citation/CS1/Identifiers: Difference between revisions

m
(sync from sandbox;)
 
(15 intermediate revisions by 4 users not shown)
Line 3: Line 3:
]]
]]


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


local z; -- table of tables defined in Module:Citation/CS1/Utilities
local z; -- table of tables defined in Module:Citation/CS1/Utilities
Line 9: Line 10:
local cfg; -- table of configuration tables that are defined in Module:Citation/CS1/Configuration
local cfg; -- table of configuration tables that are defined in Module:Citation/CS1/Configuration


local wd_int_lang = (mw.site.server:match ('wikidata') and mw.getCurrentFrame():preprocess('{{int:lang}}')) or '';
 
--[[--------------------------< 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 >>============================================
--============================<< 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 >----------------------------------------------
--[[--------------------------< E X T E R N A L _ L I N K _ I D >----------------------------------------------


Formats a wiki style external link
Formats a wiki-style external link


]]
]]


local function external_link_id(options)
local function external_link_id (options)
local url_string = options.id;
local url_string = options.id;
local ext_link;
local ext_link;
local this_wiki_code; -- wikipedia subdomain; 'en' for en.wikipedia.org
local this_wiki_code = cfg.this_wiki_code; -- Wikipedia subdomain; 'en' for en.wikipedia.org
local wd_article; -- article title from wikidata
local wd_article; -- article title from Wikidata
if options.encode == true or options.encode == nil then
if options.encode == true or options.encode == nil then
url_string = mw.uri.encode( url_string );
url_string = mw.uri.encode (url_string, 'PATH');
end
end


ext_link = mw.ustring.format ('[%s%s%s %s]', options.prefix, url_string, options.suffix or "", mw.text.nowiki(options.id));
if options.auto_link and is_set (options.access) then
if is_set(options.access) then
auto_link_urls[options.auto_link] = table.concat ({options.prefix, url_string, options.suffix});
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
end


this_wiki_code = mw.language.getContentLanguage():getCode(); -- get this wikipedia's language code (subdomain)
ext_link = mw.ustring.format ('[%s%s%s %s]', options.prefix, url_string, options.suffix or "", mw.text.nowiki (options.id));
if is_set (options.access) then
if string.match (mw.site.server, 'wikidata') 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
this_wiki_code = mw.getCurrentFrame():preprocess('{{int:lang}}'); -- on wikidata so use interface language setting instead
end
if is_set (options.q) and mw.wikibase then -- wikibase test here avoids script errors in third party wikis that aren't using mw.wikibase extension
wd_article = mw.wikibase.getEntity (options.q):getSitelink (this_wiki_code .. 'wiki'); -- fetch article title from wd
if wd_article then
wd_article = table.concat ({':', this_wiki_code, ':', wd_article}); -- make interwiki link if taken from wd; leading colon required
end
end
end


return table.concat ({
return table.concat ({
make_wikilink (wd_article or options.link, options.label), -- wikidata link or locally specified
make_wikilink (link_label_make (options), options.label), -- redirect, Wikidata link, or locally specified link (in that order)
options.separator or '&nbsp;',
options.separator or '&nbsp;',
ext_link
ext_link
Line 58: Line 109:
--[[--------------------------< I N T E R N A L _ L I N K _ I D >----------------------------------------------
--[[--------------------------< I N T E R N A L _ L I N K _ I D >----------------------------------------------


Formats a wiki style internal link
Formats a wiki-style internal link
 
TODO: Does not currently need to support options.access, options.encode, auto-linking and COinS (as in external_link_id),
but may be needed in the future for :m:Interwiki_map custom-prefixes like :arxiv:, :bibcode:, :DOI:, :hdl:, :ISSN:,
:JSTOR:, :Openlibrary:, :PMID:, :RFC:.


]]
]]


local function internal_link_id(options)
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
local id = mw.ustring.gsub (options.id, '%d', cfg.date_names.local_digits); -- translate 'local' digits to Western 0-9


return table.concat (
return table.concat (
{
{
make_wikilink (options.link, options.label),
make_wikilink (link_label_make (options), options.label), -- wiki-link the identifier label
options.separator or '&nbsp;',
options.separator or '&nbsp;', -- add the separator
make_wikilink (
make_wikilink (
table.concat (
table.concat (
{
{
options.prefix,
options.prefix,
id, -- translated to western digits
id, -- translated to Western digits
options.suffix or ''
options.suffix or ''
}),
}),
substitute (cfg.presentation['bdi'], {'', mw.text.nowiki (options.id)}) -- bdi tags to prevent Latn script identifiers from being reversed at rtl language wikis
substitute (cfg.presentation['bdi'], {'', mw.text.nowiki (options.id)}) -- bdi tags to prevent Latin script identifiers from being reversed at RTL language wikis
); -- nowiki because MediaWiki still has magic links for ISBN and the like; TODO: is it really required?
); -- nowiki because MediaWiki still has magic links for ISBN and the like; TODO: is it really required?
});
});
Line 84: Line 139:
--[[--------------------------< I S _ E M B A R G O E D >------------------------------------------------------
--[[--------------------------< 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 |embargo= against today's date.  If embargo date is
Determines if a PMC identifier's online version is embargoed. Compares the date in |pmc-embargo-date= against
in the future, returns the content of |embargo=; otherwise, returns and empty string because the embargo has expired or because
today's date.  If embargo date is in the future, returns the content of |pmc-embargo-date=; otherwise, returns
|embargo= was not set in this cite.
an empty string because the embargo has expired or because |pmc-embargo-date= was not set in this cite.


]]
]]
Line 93: Line 148:
if is_set (embargo) then
if is_set (embargo) then
local lang = mw.getContentLanguage();
local lang = mw.getContentLanguage();
local good1, embargo_date, good2, todays_date;
local good1, embargo_date, todays_date;
good1, embargo_date = pcall( lang.formatDate, lang, 'U', embargo );
good1, embargo_date = pcall (lang.formatDate, lang, 'U', embargo);
good2, todays_date = pcall( lang.formatDate, lang, 'U' );
todays_date = lang:formatDate ('U');
if good1 and good2 then -- if embargo date and today's date are good dates
if good1 then -- if embargo date is a good date
if tonumber( embargo_date ) >= tonumber( todays_date ) then -- is embargo date is in the future?
if tonumber (embargo_date) >= tonumber (todays_date) then -- is embargo date is in the future?
return embargo; -- still embargoed
return embargo; -- still embargoed
else
else
add_maint_cat ('embargo')
set_message ('maint_pmc_embargo'); -- embargo has expired; add main cat
return ''; -- unset because embargo has expired
return ''; -- unset because embargo has expired
end
end
end
end
end
end
return ''; -- |embargo= not set return empty string
return ''; -- |pmc-embargo-date= not set return empty string
end
 
 
--[=[-------------------------< I S _ V A L I D _ B I O R X I V _ D A T E >------------------------------------
 
returns true if:
2019-12-11T00:00Z <= biorxiv_date < today + 2 days
The dated form of biorxiv identifier has a start date of 2019-12-11.  The Unix timestamp for that date is {{#time:U|2019-12-11}} = 1576022400
 
biorxiv_date is the date provided in those |biorxiv= parameter values that are dated at time 00:00:00 UTC
today is the current date at time 00:00:00 UTC plus 48 hours
if today is 2015-01-01T00:00:00 then
adding 24 hours gives 2015-01-02T00:00:00 – one second more than today
adding 24 hours gives 2015-01-03T00:00:00 – one second more than tomorrow
 
This function does not work if it is fed month names for languages other than English.  Wikimedia #time: parser
apparently doesn't understand non-English 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.
 
]=]
 
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
end


Line 112: Line 206:
--[[--------------------------< IS _ V A L I D _ I S X N >-----------------------------------------------------
--[[--------------------------< IS _ V A L I D _ I S X N >-----------------------------------------------------


ISBN-10 and ISSN validator code calculates checksum across all isbn/issn digits including the check digit.
ISBN-10 and ISSN validator code calculates checksum across all ISBN/ISSN digits including the check digit.
ISBN-13 is checked in isbn().
ISBN-13 is checked in isbn().


If the number is valid the result will be 0. Before calling this function, issbn/issn must be checked for length
If the number is valid the result will be 0. Before calling this function, ISBN/ISSN must be checked for length
and stripped of dashes, spaces and other non-isxn characters.
and stripped of dashes, spaces and other non-ISxN characters.


]]
]]
Line 122: Line 216:
local function is_valid_isxn (isxn_str, len)
local function is_valid_isxn (isxn_str, len)
local temp = 0;
local temp = 0;
isxn_str = { isxn_str:byte(1, len) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39, 'X' → 0x58
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
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
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)
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
temp = temp + 10 * (len - i); -- it represents 10 decimal
else
else
temp = temp + tonumber( string.char(v) )*(len-i);
temp = temp + tonumber (string.char (v) )*(len-i);
end
end
end
end
return temp % 11 == 0; -- returns true if calculation result is zero
return temp % 11 == 0; -- returns true if calculation result is zero
end
end


Line 137: Line 231:
--[[--------------------------< IS _ V A L I D _ I S X N _ 1 3 >-----------------------------------------------
--[[--------------------------< IS _ V A L I D _ I S X N _ 1 3 >-----------------------------------------------


ISBN-13 and ISMN validator code calculates checksum across all 13 isbn/ismn digits including the check digit.
ISBN-13 and ISMN validator code calculates checksum across all 13 ISBN/ISMN digits including the check digit.
If the number is valid, the result will be 0. Before calling this function, isbn-13/ismn must be checked for length
If the number is valid, the result will be 0. Before calling this function, ISBN-13/ISMN must be checked for length
and stripped of dashes, spaces and other non-isxn-13 characters.
and stripped of dashes, spaces and other non-ISxN-13 characters.


]]
]]
Line 147: Line 241:
isxn_str = { isxn_str:byte(1, 13) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39
isxn_str = { isxn_str:byte(1, 13) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39
for i, v in ipairs( isxn_str ) do
for i, v in ipairs (isxn_str) do
temp = temp + (3 - 2*(i % 2)) * tonumber( string.char(v) ); -- multiply odd index digits by 1, even index digits by 3 and sum; includes check digit
temp = temp + (3 - 2*(i % 2)) * tonumber (string.char (v) ); -- multiply odd index digits by 1, even index digits by 3 and sum; includes check digit
end
end
return temp % 10 == 0; -- sum modulo 10 is zero when isbn-13/ismn is correct
return temp % 10 == 0; -- sum modulo 10 is zero when ISBN-13/ISMN is correct
end
end


Line 156: Line 250:
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------


lccn normalization (http://www.loc.gov/marc/lccn-namespace.html#normalization)
LCCN normalization (http://www.loc.gov/marc/lccn-namespace.html#normalization)
1. Remove all blanks.
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 there is a forward slash (/) in the string, remove it, and remove all characters to the right of the forward slash.
Line 165: Line 259:
2. If the length of the substring is less than 6, left-fill the substring with zeroes until the length is six.
2. If the length of the substring is less than 6, left-fill the substring with zeroes until the length is six.


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


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


if nil ~= string.find (lccn,'/') then
if nil ~= string.find (lccn, '/') then
lccn = lccn:match ("(.-)/"); -- 2. remove forward slash and all character to the right of it
lccn = lccn:match ("(.-)/"); -- 2. remove forward slash and all character to the right of it
end
end


local prefix
local prefix
local suffix
local suffix
prefix, suffix = lccn:match ("(.+)%-(.+)"); -- 3.a remove hyphen by splitting the string into prefix and 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
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
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 lccn
lccn = prefix..suffix; -- reassemble the LCCN
end
end
return lccn;
return lccn;
end
end


--============================<< I D E N T I F I E R  F U N C T I O N S >>====================================
--============================<< I D E N T I F I E R  F U N C T I O N S >>====================================
Line 194: Line 289:


format and error check arXiv identifier.  There are three valid forms of the identifier:
format and error check arXiv identifier.  There are three valid forms of the identifier:
the first form, valid only between date codes 9108 and 0703 is:
the first form, valid only between date codes 9107 and 0703, is:
arXiv:<archive>.<class>/<date code><number><version>
arXiv:<archive>.<class>/<date code><number><version>
where:
where:
Line 218: Line 313:
]]
]]


local function arxiv (id, class)
local function arxiv (options)
local handler = cfg.id_handlers['ARXIV'];
local id = options.id;
local class = options.Class; -- TODO: lowercase?
local handler = options.handler;
local year, month, version;
local year, month, version;
local err_cat = false; -- assume no error message
local err_cat = false; -- assume no error message
local text; -- output text
local text; -- output text
if id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%d$") or id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%dv%d+$") then -- test for the 9108-0703 format w/ & w/o version
if id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%d$") or id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%dv%d+$") then -- test for the 9107-0703 format with or without version
year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
year = tonumber(year);
year = tonumber (year);
month = tonumber(month);
month = tonumber (month);
if ((not (90 < year or 8 > year)) or (1 > month or 12 < month)) or -- if invalid year or invalid month
if ((not (90 < year or 8 > year)) or (1 > month or 12 < month)) or -- if invalid year or invalid month
((91 == year and 7 > month) or (7 == year and 3 < month)) then -- if years ok, are starting and ending months ok?
((91 == year and 7 > month) or (7 == year and 3 < month)) then -- if years ok, are starting and ending months ok?
Line 233: Line 330:
end
end


elseif id:match("^%d%d[01]%d%.%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%dv%d+$") then -- test for the 0704-1412 w/ & w/o version
elseif id:match("^%d%d[01]%d%.%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%dv%d+$") then -- test for the 0704-1412 with or without version
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
year = tonumber(year);
year = tonumber (year);
month = tonumber(month);
month = tonumber (month);
if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or -- is year invalid or is month invalid? (doesn't test for future years)
if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or -- is year invalid or is month invalid? (doesn't test for future years)
((7 == year) and (4 > month)) then --or -- when year is 07, is month invalid (before April)?
((7 == year) and (4 > month)) then -- when year is 07, is month invalid (before April)?
err_cat = true; -- flag for error message
err_cat = true; -- flag for error message
end
end


elseif id:match("^%d%d[01]%d%.%d%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%d%dv%d+$") then -- test for the 1501- format w/ & w/o version
elseif id:match("^%d%d[01]%d%.%d%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%d%dv%d+$") then -- test for the 1501- format with or without version
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d%d[v%d]*$");
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d%d[v%d]*$");
year = tonumber(year);
year = tonumber (year);
month = tonumber(month);
month = tonumber (month);
if ((15 > year) or (1 > month or 12 < month)) then -- is year invalid or is month invalid? (doesn't test for future years)
if ((15 > year) or (1 > month or 12 < month)) then -- is year invalid or is month invalid? (doesn't test for future years)
err_cat = true; -- flag for error message
err_cat = true; -- flag for error message
Line 254: Line 351:
end
end


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


if is_set (class) then
if is_set (class) then
Line 263: Line 364:
text = table.concat ({text, ' [[//arxiv.org/archive/', class, ' ', class, ']]'}); -- external link within square brackets, not wikilink
text = table.concat ({text, ' [[//arxiv.org/archive/', class, ' ', class, ']]'}); -- external link within square brackets, not wikilink
else
else
text = table.concat ({text, ' ', set_error ('class_ignored')});
text = table.concat ({text, ' ', set_message ('err_class_ignored')});
end
end
end
end
Line 273: Line 374:
--[[--------------------------< B I B C O D E >--------------------------------------------------------------------
--[[--------------------------< B I B C O D E >--------------------------------------------------------------------


Validates (sort of) and formats a bibcode id.
Validates (sort of) and formats a bibcode ID.


Format for bibcodes is specified here: http://adsabs.harvard.edu/abs_doc/help_pages/data.html#bibcodes
Format for bibcodes is specified here: http://adsabs.harvard.edu/abs_doc/help_pages/data.html#bibcodes
Line 283: Line 384:
1–4 must be digits and must represent a year in the range of 1000 – next year
1–4 must be digits and must represent a year in the range of 1000 – next year
5 must be a letter
5 must be a letter
6 must be letter, ampersand, or dot (ampersand cannot directly precede a dot; &. )
6–8 must be letter, digit, ampersand, or dot (ampersand cannot directly precede a dot; &. )
7–8 must be letter, digit, ampersand, or dot (ampersand cannot directly precede a dot; &. )
9–18 must be letter, digit, or dot
9–18 must be letter, digit, or dot
19 must be a letter or dot
19 must be a letter or dot
Line 290: Line 390:
]]
]]


local function bibcode (id, access)
local function bibcode (options)
local handler = cfg.id_handlers['BIBCODE'];
local id = options.id;
local access = options.access;
local handler = options.handler;
local err_type;
local err_type;
local year;
local year;


local text = external_link_id({link=handler.link, label=handler.label, q = handler.q,
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,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode,
access=access});
access = access});
if 19 ~= id:len() then
if 19 ~= id:len() then
err_type = 'length';
err_type = cfg.err_msg_supl.length;
else
else
year = id:match ("^(%d%d%d%d)[%a][%a&%.][%a&%.%d][%a&%.%d][%a%d%.]+[%a%.]$") --
year = id:match ("^(%d%d%d%d)[%a][%w&%.][%w&%.][%w&%.][%w.]+[%a%.]$");
if not year then -- if nil then no pattern match
if not year then -- if nil then no pattern match
err_type = 'value'; -- so value error
err_type = cfg.err_msg_supl.value; -- so value error
else
else
local next_year = tonumber(os.date ('%Y'))+1; -- get the current year as a number and add one for next year
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
year = tonumber (year); -- convert year portion of bibcode to a number
if (1000 > year) or (year > next_year) then
if (1000 > year) or (year > next_year) then
err_type = 'year'; -- year out of bounds
err_type = cfg.err_msg_supl.year; -- year out of bounds
end
end
if id:find('&%.') then
if id:find('&%.') then
err_type = 'journal'; -- journal abbreviation must not have '&.' (if it does its missing a letter)
err_type = cfg.err_msg_supl.journal; -- journal abbreviation must not have '&.' (if it does it's missing a letter)
end
end
end
end
Line 318: Line 420:


if is_set (err_type) then -- if there was an error detected
if is_set (err_type) then -- if there was an error detected
text = text .. ' ' .. set_error( 'bad_bibcode', {err_type});
text = text .. ' ' .. set_message ('err_bad_bibcode', {err_type});
options.coins_list_t['BIBCODE'] = nil; -- when error, unset so not included in COinS
 
end
end
return text;
return text;
Line 326: Line 430:
--[[--------------------------< B I O R X I V >-----------------------------------------------------------------
--[[--------------------------< B I O R X I V >-----------------------------------------------------------------


Format bioRxiv id and do simple error checking.  BiorXiv ids are exactly 6 digits.
Format bioRxiv ID and do simple error checking.  Before 2019-12-11, biorXiv IDs were 10.1101/ followed by exactly
The bioRxiv id is the number following the last slash in the bioRxiv-issued DOI:
6 digits. After 2019-12-11, biorXiv IDs retained the six-digit identifier but prefixed that with a yyyy.mm.dd.
https://doi.org/10.1101/078733 -> 078733
date and suffixed with an optional version identifier.
 
The bioRxiv ID 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:
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 function biorxiv (options)
local handler = cfg.id_handlers['BIORXIV'];
local id = options.id;
local err_cat = ''; -- presume that bioRxiv id is valid
local handler = options.handler;
local err_cat = true; -- flag; assume that there will be an error
if nil == id:match("^%d%d%d%d%d%d$") then -- if bioRxiv id has anything but six digits
local patterns = {
err_cat = ' ' .. set_error( 'bad_biorxiv'); -- set an error message
'^10.1101/%d%d%d%d%d%d$', -- simple 6-digit identifier (before 2019-12-11)
'^10.1101/(20[1-9]%d)%.([01]%d)%.([0-3]%d)%.%d%d%d%d%d%dv%d+$', -- y.m.d. date + 6-digit identifier + version (after 2019-12-11)
'^10.1101/(20[1-9]%d)%.([01]%d)%.([0-3]%d)%.%d%d%d%d%d%d$', -- y.m.d. date + 6-digit identifier (after 2019-12-11)
}
for _, pattern in ipairs (patterns) do -- spin through the patterns looking for a match
if id:match (pattern) then
local y, m, d = id:match (pattern); -- found a match, attempt to get year, month and date from the identifier
 
if m then -- m is nil when id is the six-digit form
if not is_valid_biorxiv_date (y .. '-' .. m .. '-' .. d) then -- validate the encoded date; TODO: don't ignore leap-year and actual month lengths ({{#time:}} is a poor date validator)
break; -- date fail; break out early so we don't unset the error message
end
end
err_cat = nil; -- we found a match so unset the error message
break; -- and done
end
end -- err_cat remains set here when no match
 
if err_cat then
options.coins_list_t['BIORXIV'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id({link = handler.link, label = handler.label, q = handler.q,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix=handler.prefix,id=id,separator=handler.separator,
prefix = handler.prefix, id = id, separator = handler.separator,
encode=handler.encode, access=handler.access}) .. err_cat;
encode = handler.encode, access = handler.access}) .. (err_cat and (' ' .. set_message ('err_bad_biorxiv')) or '');
end
end


Line 353: Line 485:
]]
]]


local function citeseerx (id)
local function citeseerx (options)
local handler = cfg.id_handlers['CITESEERX'];
local id = options.id;
local handler = options.handler;
local matched;
local matched;
local text = external_link_id({link=handler.link, label=handler.label, q = handler.q,
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,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode,
access=handler.access});
access = handler.access});
matched = id:match ("^10%.1%.1%.[1-9]%d?%d?%d?%.[1-9]%d?%d?%d?$");
matched = id:match ("^10%.1%.1%.[1-9]%d?%d?%d?%.[1-9]%d?%d?%d?$");
if not matched then
if not matched then
text = text .. ' ' .. set_error( 'bad_citeseerx' );
text = text .. ' ' .. set_message ('err_bad_citeseerx' );
options.coins_list_t['CITESEERX'] = nil; -- when error, unset so not included in COinS
end
end
return text;
return text;
Line 377: Line 511:
Suffix: character string of any length chosen by the registrant
Suffix: character string of any length chosen by the registrant


This function checks a DOI name for: prefix/suffix.  If the doi name contains spaces or endashes, or, if it ends
This function checks a DOI name for: prefix/suffix.  If the 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.
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,
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
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in doi names.
if ever used in DOI names.


]]
]]


local function doi(id, inactive, access)
local function doi (options)
local cat = ""
local id = options.id;
local handler = cfg.id_handlers['DOI'];
local inactive = options.DoiBroken
local access = options.access;
local ignore_invalid = options.accept;
local handler = options.handler;
local err_cat;
 
local text;
local text;
if is_set(inactive) then
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_year = inactive:match("%d%d%d%d") or ''; -- try to get the year portion from the inactive date
if is_set(inactive_year) then
local inactive_month, good;
table.insert( z.error_categories, "Pages with DOIs inactive as of " .. inactive_year );
 
if is_set (inactive_year) then
if 4 < inactive:len() then -- inactive date has more than just a year (could be anything)
local lang_obj = mw.getContentLanguage(); -- get a language object for this wiki
good, inactive_month = pcall (lang_obj.formatDate, lang_obj, 'F', inactive); -- try to get the month name from the inactive date
if not good then
inactive_month = nil; -- something went wrong so make sure this is unset
end
end
else
inactive_year = nil; -- |doi-broken-date= has something but it isn't a date
end
if is_set (inactive_year) and is_set (inactive_month) then
set_message ('maint_doi_inactive_dated', {inactive_year, inactive_month, ' '});
elseif is_set (inactive_year) then
set_message ('maint_doi_inactive_dated', {inactive_year, '', ''});
else
set_message ('maint_doi_inactive');
end
inactive = " (" .. cfg.messages['inactive'] .. ' ' .. inactive .. ')';
end
 
local registrant = id:match ('^10%.([^/]+)/[^%s–]-[^%.,]$'); -- registrant set when DOI has the proper basic form
local registrant_err_patterns = { -- these patterns are for code ranges that are not supported
'^[^1-3]%d%d%d%d%.%d%d*$', -- 5 digits with subcode (0xxxx, 40000+); accepts: 10000–39999
'^[^1-5]%d%d%d%d$', -- 5 digits without subcode (0xxxx, 60000+); accepts: 10000–59999
'^[^1-9]%d%d%d%.%d%d*$', -- 4 digits with subcode (0xxx); accepts: 1000–9999
'^[^1-9]%d%d%d$', -- 4 digits without subcode (0xxx); accepts: 1000–9999
'^%d%d%d%d%d%d+', -- 6 or more digits
'^%d%d?%d?$', -- less than 4 digits without subcode (with subcode is legitimate)
'^5555$', -- test registrant will never resolve
'[^%d%.]', -- any character that isn't a digit or a dot
}
 
if not ignore_invalid then
if registrant then -- when DOI has proper form
for i, pattern in ipairs (registrant_err_patterns) do -- spin through error patterns
if registrant:match (pattern) then -- to validate registrant codes
err_cat = ' ' .. set_message ('err_bad_doi'); -- when found, mark this DOI as bad
break; -- and done
end
end
else
else
table.insert( z.error_categories, "Pages with inactive DOIs" ); -- when inactive doesn't contain a recognizable year
err_cat = ' ' .. set_message ('err_bad_doi'); -- invalid directory or malformed
end
end
inactive = " (" .. cfg.messages['inactive'] .. " " .. inactive .. ")"
else
set_message ('maint_doi_ignore');
end
end
text = external_link_id({link = handler.link, label = handler.label, q = handler.q,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, access=access}) .. (inactive or '')


if nil == id:match("^10%.[^%s–]-/[^%s–]-[^%.,]$") then -- doi must begin with '10.', must contain a fwd slash, must not contain spaces or endashes, and must not end with period or comma
if err_cat then
cat = ' ' .. set_error( 'bad_doi' );
options.coins_list_t['DOI'] = nil; -- when error, unset so not included in COinS
end
end
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access,
auto_link = not (err_cat or is_set (inactive) or ignore_invalid) and 'doi' or nil -- do not auto-link when |doi-broken-date= has a value or when there is a DOI error or (to play it safe, after all, auto-linking is not essential) when invalid DOIs are ignored
}) .. (inactive or '');


return text .. cat
return text .. (err_cat and err_cat or ''); -- parentheses required
end
end


Line 425: Line 610:
terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in HDLs.
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 function hdl (options)
local handler = cfg.id_handlers['HDL'];
local id = options.id;
local access = options.access;
local handler = options.handler;
local query_params = { -- list of known query parameters from http://www.handle.net/proxy_servlet.html
'noredirect',
'ignore_aliases',
'auth',
'cert',
'index',
'type',
'urlappend',
'locatt',
'action',
}
local text = external_link_id({link = handler.link, label = handler.label, q = handler.q,
local hdl, suffix, param = id:match ('(.-)(%?(%a+).+)$'); -- look for query string
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, access=access})
local found;


if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then -- hdl must contain a fwd slash, must not contain spaces, endashes, and must not end with period or comma
if hdl then -- when there are query strings, this is the handle identifier portion
text = text .. ' ' .. set_error( 'bad_hdl' );
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 must contain a forward slash, must not contain spaces, endashes, and must not end with period or comma
text = text .. ' ' .. set_message ('err_bad_hdl' );
options.coins_list_t['HDL'] = nil; -- when error, unset so not included in COinS
end
end
return text;
return text;
Line 447: Line 668:
]]
]]


local function isbn( isbn_str )
local function isbn (options)
if nil ~= isbn_str:match("[^%s-0-9X]") then
local isbn_str = options.id;
return false, 'invalid character'; -- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
local ignore_invalid = options.accept;
local handler = options.handler;
 
local function return_result (check, err_type) -- local function to handle the various returns
local ISBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
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
options.coins_list_t['ISBN'] = nil; -- when error, unset so not included in COinS
return ISBN .. set_message ('err_bad_isbn', {err_type}, false, ' '); -- display an error message
end
end
return ISBN;
end
end
isbn_str = isbn_str:gsub( "-", "" ):gsub( " ", "" ); -- remove hyphens and spaces
 
local len = isbn_str:len();
if nil ~= isbn_str:match ('[^%s-0-9X]') then
return return_result (false, cfg.err_msg_supl.char); -- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
end
 
local id = isbn_str:gsub ('[%s-]', ''); -- remove hyphens and whitespace
 
local len = id:len();
   
   
if len ~= 10 and len ~= 13 then
if len ~= 10 and len ~= 13 then
return false, 'length'; -- fail if incorrect length
return return_result (false, cfg.err_msg_supl.length); -- fail if incorrect length
end
end


if len == 10 then
if len == 10 then
if isbn_str:match( "^%d*X?$" ) == nil then -- fail if isbn_str has 'X' anywhere but last position
if id:match ('^%d*X?$') == nil then -- fail if isbn_str has 'X' anywhere but last position
return false, 'invalid form';
return return_result (false, cfg.err_msg_supl.form);
end
if not is_valid_isxn (id, 10) then -- test isbn-10 for numerical validity
return return_result (false, cfg.err_msg_supl.check); -- fail if isbn-10 is not numerically valid
end
if id:find ('^63[01]') then -- 630xxxxxxx and 631xxxxxxx are (apparently) not valid isbn group ids but are used by amazon as numeric identifiers (asin)
return return_result (false, cfg.err_msg_supl.group); -- fail if isbn-10 begins with 630/1
end
end
return is_valid_isxn(isbn_str, 10), 'checksum';
return return_result (true, cfg.err_msg_supl.check); -- pass if isbn-10 is numerically valid
else
else
if isbn_str:match( "^%d+$" ) == nil then
if id:match ('^%d+$') == nil then
return false, 'invalid character'; -- fail if isbn13 is not all digits
return return_result (false, cfg.err_msg_supl.char); -- fail if ISBN-13 is not all digits
end
if id:match ('^97[89]%d*$') == nil then
return return_result (false, cfg.err_msg_supl.prefix); -- fail when ISBN-13 does not begin with 978 or 979
end
end
if isbn_str:match( "^97[89]%d*$" ) == nil then
if id:match ('^9790') then
return false, 'invalid prefix'; -- fail when isbn13 does not begin with 978 or 979
return return_result (false, cfg.err_msg_supl.group); -- group identifier '0' is reserved to ISMN
end
end
return is_valid_isxn_13 (isbn_str), 'checksum';
return return_result (is_valid_isxn_13 (id), cfg.err_msg_supl.check);
end
end
end
end




--[[--------------------------< A M A Z O N >------------------------------------------------------------------
--[[--------------------------< A S I N >----------------------------------------------------------------------
 
Formats a link to Amazon.  Do simple error checking: ASIN must be mix of 10 numeric or uppercase alpha
characters.  If a mix, first character must be uppercase alpha; if all numeric, ASINs must be 10-digit
ISBN. If 10-digit ISBN, add a maintenance category so a bot or AWB script can replace |asin= with |isbn=.
Error message if not 10 characters, if not ISBN-10, if mixed and first character is a digit.


Formats a link to Amazon. Do simple error checking: asin must be mix of 10 numeric or uppercase alpha
|asin=630....... and |asin=631....... are (apparently) not a legitimate ISBN though it checksums as one; these
characters. If a mix, first character must be uppercase alpha; if all numeric, asins must be 10-digit
do not cause this function to emit the maint_asin message
isbn. If 10-digit isbn, add a maintenance category so a bot or awb script can replace |asin= with |isbn=.
Error message if not 10 characters, if not isbn10, if mixed and first character is a digit.


This function is positioned here because it calls isbn()
This function is positioned here because it calls isbn()
Line 486: Line 739:
]]
]]


local function asin(id, domain)
local function asin (options)
local err_cat = ""
local id = options.id;
local domain = options.ASINTLD;
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]$") then
if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
err_cat = ' ' .. set_error ('bad_asin'); -- asin is not a mix of 10 uppercase alpha and numeric characters
err_cat = ' ' .. set_message ('err_bad_asin'); -- ASIN is not a mix of 10 uppercase alpha and numeric characters
else
else
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
if isbn( id ) then -- see if asin value is isbn10
if is_valid_isxn (id, 10) then -- see if ASIN value is or validates as ISBN-10
add_maint_cat ('ASIN');
if not id:find ('^63[01]') then -- 630xxxxxxx and 631xxxxxxx are (apparently) not a valid isbn prefixes but are used by amazon as a numeric identifier
err_cat = ' ' .. set_message ('err_bad_asin'); -- ASIN has ISBN-10 form but begins with something other than 630/1 so probably an isbn
end
elseif not is_set (err_cat) then
elseif not is_set (err_cat) then
err_cat = ' ' .. set_error ('bad_asin'); -- asin is not isbn10
err_cat = ' ' .. set_message ('err_bad_asin'); -- ASIN is not ISBN-10
end
end
elseif not id:match("^%u[%d%u]+$") then
elseif not id:match("^%u[%d%u]+$") then
err_cat = ' ' .. set_error ('bad_asin'); -- asin doesn't begin with uppercase alpha
err_cat = ' ' .. set_message ('err_bad_asin'); -- asin doesn't begin with uppercase alpha
end
end
end
end
if not is_set(domain) then  
if (not is_set (domain)) or in_array (domain, {'us'}) then -- default: United States
domain = "com";
domain = "com";
elseif in_array (domain, {'jp', 'uk'}) then -- Japan, United Kingdom
elseif in_array (domain, {'jp', 'uk'}) then -- Japan, United Kingdom
domain = "co." .. domain;
domain = "co." .. domain;
elseif in_array (domain, {'au', 'br', 'mx'}) then -- Australia, Brazil, Mexico
elseif in_array (domain, {'z.cn'}) then -- China
domain = "cn";
elseif in_array (domain, {'au', 'br', 'mx', 'sg', 'tr'}) then -- Australia, Brazil, Mexico, Singapore, Turkey
domain = "com." .. domain;
domain = "com." .. domain;
elseif not in_array (domain, {'ae', 'ca', 'cn', 'de', 'es', 'fr', 'in', 'it', 'nl', 'pl', 'sa', 'se', 'co.jp', 'co.uk', 'com', 'com.au', 'com.br', 'com.mx', 'com.sg', 'com.tr'}) then -- Arabic Emirates, Canada, China, Germany, Spain, France, Indonesia, Italy, Netherlands, Poland, Saudi Arabia, Sweden (as of 2021-03 Austria (.at), Liechtenstein (.li) and Switzerland (.ch) still redirect to the German site (.de) with special settings, so don't maintain local ASINs for them)
err_cat = ' ' .. set_message ('err_bad_asin_tld'); -- unsupported asin-tld value
end
end
local handler = cfg.id_handlers['ASIN'];
local handler = options.handler;
return external_link_id({link=handler.link,
 
label=handler.label, q = handler.q, prefix=handler.prefix .. domain .. "/dp/",
if not is_set (err_cat) then
id=id, encode=handler.encode, separator = handler.separator}) .. err_cat;
options.coins_list_t['ASIN'] = handler.prefix .. domain .. "/dp/" .. id; -- experiment for asin coins
else
options.coins_list_t['ASIN'] = nil; -- when error, unset so not included in COinS
end
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
end


Line 518: Line 787:
--[[--------------------------< I S M N >----------------------------------------------------------------------
--[[--------------------------< I S M N >----------------------------------------------------------------------


Determines whether an ISMN string is valid.  Similar to isbn-13, ismn is 13 digits begining 979-0-... and uses the
Determines whether an ISMN string is valid.  Similar to ISBN-13, ISMN is 13 digits beginning 979-0-... and uses the
same check digit calculations.  See http://www.ismn-international.org/download/Web_ISMN_Users_Manual_2008-6.pdf
same check digit calculations.  See http://www.ismn-international.org/download/Web_ISMN_Users_Manual_2008-6.pdf
section 2, pages 9–12.
section 2, pages 9–12.
ismn value not made part of COinS metadata because we don't have a url or isn't a COinS-defined identifier (rft.xxx)
or an identifier registered at info-uri.info (info:)


]]
]]


local function ismn (id)
local function ismn (options)
local handler = cfg.id_handlers['ISMN'];
local id = options.id;
local handler = options.handler;
local text;
local text;
local valid_ismn = true;
local valid_ismn = true;
Line 531: Line 804:


id_copy = id; -- save a copy because this testing is destructive
id_copy = id; -- save a copy because this testing is destructive
id=id:gsub( "[%s-]", "" ); -- strip spaces, hyphens, and endashes from the ismn
id = id:gsub ('[%s-]', ''); -- remove hyphens and white space


if 13 ~= id:len() or id:match( "^9790%d*$" ) == nil then -- ismn must be 13 digits and begin 9790
if 13 ~= id:len() or id:match ("^9790%d*$" ) == nil then -- ISMN must be 13 digits and begin with 9790
valid_ismn = false;
valid_ismn = false;
else
else
valid_ismn=is_valid_isxn_13 (id); -- validate ismn
valid_ismn=is_valid_isxn_13 (id); -- validate ISMN
end
end


-- text = internal_link_id({link = handler.link, label = handler.label, -- use this (or external version) when there is some place to link to
-- 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,separator=handler.separator, encode=handler.encode})
-- prefix = handler.prefix, id = id_copy, separator = handler.separator, encode = handler.encode})
 
text = table.concat (
text = table.concat ( -- because no place to link to yet
{
{
make_wikilink (handler.link, handler.label),
make_wikilink (link_label_make (handler), handler.label),
handler.separator,
handler.separator,
id_copy
id_copy
}); -- because no place to link to yet
});


if false == valid_ismn then
if false == valid_ismn then
text = text .. ' ' .. set_error( 'bad_ismn' ) -- add an error message if the ismn is invalid
options.coins_list_t['ISMN'] = nil; -- when error, unset so not included in COinS; not really necessary here because ismn not made part of COinS
text = text .. ' ' .. set_message ('err_bad_ismn' ) -- add an error message if the ISMN is invalid
end  
end  
Line 559: Line 833:
--[[--------------------------< I S S N >----------------------------------------------------------------------
--[[--------------------------< I S S N >----------------------------------------------------------------------


Validate and format an issn.  This code fixes the case where an editor has included an ISSN in the citation but
Validate and format an ISSN.  This code fixes the case where an editor has included an ISSN in the citation but
has separated the two groups of four digits with a space.  When that condition occurred, the resulting link looked
has separated the two groups of four digits with a space.  When that condition occurred, the resulting link looked
like this:
like this:
Line 565: Line 839:
|issn=0819 4327 gives: [http://www.worldcat.org/issn/0819 4327 0819 4327] -- can't have spaces in an external link
|issn=0819 4327 gives: [http://www.worldcat.org/issn/0819 4327 0819 4327] -- can't have spaces in an external link
This code now prevents that by inserting a hyphen at the issn midpoint.  It also validates the issn for length
This code now prevents that by inserting a hyphen at the ISSN midpoint.  It also validates the ISSN for length
and makes sure that the checkdigit agrees with the calculated value.  Incorrect length (8 digits), characters
and makes sure that the checkdigit agrees with the calculated value.  Incorrect length (8 digits), characters
other than 0-9 and X, or checkdigit / calculated value mismatch will all cause a check issn error message.  The
other than 0-9 and X, or checkdigit / calculated value mismatch will all cause a check ISSN error message.  The
issn is always displayed with a hyphen, even if the issn was given as a single group of 8 digits.
ISSN is always displayed with a hyphen, even if the ISSN was given as a single group of 8 digits.


]]
]]


local function issn(id, e)
local function issn (options)
local issn_copy = id; -- save a copy of unadulterated issn; use this version for display if issn does not validate
local id = options.id;
local handler;
local handler = options.handler;
local ignore_invalid = options.accept;
 
local issn_copy = id; -- save a copy of unadulterated ISSN; use this version for display if ISSN does not validate
local text;
local text;
local valid_issn = true;
local valid_issn = true;
 
if e then
id = id:gsub ('[%s-]', ''); -- remove hyphens and whitespace
handler = cfg.id_handlers['EISSN'];
 
if 8 ~= id:len() or nil == id:match ("^%d*X?$" ) then -- validate the ISSN: 8 digits long, containing only 0-9 or X in the last position
valid_issn = false; -- wrong length or improper character
else
else
handler = cfg.id_handlers['ISSN'];
valid_issn = is_valid_isxn (id, 8); -- validate ISSN
end
end


id=id:gsub( "[%s-–]", "" ); -- strip spaces, hyphens, and endashes from the issn
if true == valid_issn then
 
id = string.sub (id, 1, 4 ) .. "-" .. string.sub (id, 5 ); -- if valid, display correctly formatted version
if 8 ~= id:len() or nil == id:match( "^%d*X?$" ) then -- validate the issn: 8 digits long, containing only 0-9 or X in the last position
valid_issn=false; -- wrong length or improper character
else
else
valid_issn=is_valid_isxn(id, 8); -- validate issn
id = issn_copy; -- if not valid, show the invalid ISSN with error message
end
end


if true == valid_issn then
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
id = string.sub( id, 1, 4 ) .. "-" .. string.sub( id, 5 ); -- if valid, display correctly formatted version
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
 
if ignore_invalid then
set_message ('maint_issn_ignore');
else
else
id = issn_copy; -- if not valid, use the show the invalid issn with error message
if false == valid_issn then
options.coins_list_t['ISSN'] = nil; -- when error, unset so not included in COinS
text = text .. ' ' .. set_message ('err_bad_issn', (options.hkey == 'EISSN') and 'e' or ''); -- add an error message if the ISSN is invalid
end
end
end
text = external_link_id({link = handler.link, label = handler.label, q = handler.q,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
if false == valid_issn then
text = text .. ' ' .. set_error( 'bad_issn', e and 'e' or '' ) -- add an error message if the issn is invalid
end
return text
return text
Line 615: Line 891:
]]
]]


local function jfm (id)
local function jfm (options)
local handler = cfg.id_handlers['JFM'];
local id = options.id;
local handler = options.handler;
local id_num;
local id_num;
local err_cat = '';
local err_cat = '';
Line 623: Line 900:


if is_set (id_num) then
if is_set (id_num) then
add_maint_cat ('jfm_format');
set_message ('maint_jfm_format');
else -- plain number without mr prefix
else -- plain number without JFM prefix
id_num = id; -- if here id does not have prefix
id_num = id; -- if here id does not have prefix
end
end
Line 631: Line 908:
id = id_num; -- jfm matches pattern
id = id_num; -- jfm matches pattern
else
else
err_cat = ' ' .. set_error( 'bad_jfm' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_jfm' ); -- set an error message
options.coins_list_t['JFM'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id({link = handler.link, label = handler.label, q = handler.q,
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;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
 
 
--[[--------------------------< J S T O R >--------------------------------------------------------------------
 
Format a JSTOR with some error checking
 
]]
 
local function jstor (options)
local id = options.id;
local access = options.access;
local handler = options.handler;
local err_msg = '';
 
if id:find ('[Jj][Ss][Tt][Oo][Rr]') or id:find ('^https?://') or id:find ('%s') then
err_msg = ' ' .. set_message ('err_bad_jstor'); -- set an error message
options.coins_list_t['JSTOR'] = nil; -- when error, unset so not included in COinS
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access}) .. err_msg;
end
end


Line 646: Line 946:


length = 8 then all digits
length = 8 then all digits
length = 9 then lccn[1] is lower case alpha
length = 9 then lccn[1] is lowercase alpha
length = 10 then lccn[1] and lccn[2] are both lower case alpha or both digits
length = 10 then lccn[1] and lccn[2] are both lowercase alpha or both digits
length = 11 then lccn[1] is lower case alpha, lccn[2] and lccn[3] are both lower case alpha or both digits
length = 11 then lccn[1] is lower case alpha, lccn[2] and lccn[3] are both lowercase alpha or both digits
length = 12 then lccn[1] and lccn[2] are both lower case alpha
length = 12 then lccn[1] and lccn[2] are both lowercase alpha


]]
]]


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


id = normalize_lccn (id); -- get canonical form (no whitespace, hyphens, forward slashes)
id = normalize_lccn (id); -- get canonical form (no whitespace, hyphens, forward slashes)
local len = id:len(); -- get the length of the lccn
local len = id:len(); -- get the length of the LCCN


if 8 == len then
if 8 == len then
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits)
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits)
err_cat = ' ' .. set_error( 'bad_lccn' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_lccn'); -- set an error message
end
end
elseif 9 == len then -- LCCN should be adddddddd
elseif 9 == len then -- LCCN should be adddddddd
if nil == id:match("%l%d%d%d%d%d%d%d%d") then -- does it match our pattern?
if nil == id:match("%l%d%d%d%d%d%d%d%d") then -- does it match our pattern?
err_cat = ' ' .. set_error( 'bad_lccn' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_lccn'); -- set an error message
end
end
elseif 10 == len then -- LCCN should be aadddddddd or dddddddddd
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 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
if nil == id:match("^%l%l%d%d%d%d%d%d%d%d") then -- ... see if it matches our pattern
err_cat = ' ' .. set_error( 'bad_lccn' ); -- no match, set an error message
err_cat = ' ' .. set_message ('err_bad_lccn'); -- no match, set an error message
end
end
end
end
elseif 11 == len then -- LCCN should be aaadddddddd or adddddddddd
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
if not (id:match("^%l%l%l%d%d%d%d%d%d%d%d") or id:match("^%l%d%d%d%d%d%d%d%d%d%d")) then -- see if it matches one of our patterns
err_cat = ' ' .. set_error( 'bad_lccn' ); -- no match, set an error message
err_cat = ' ' .. set_message ('err_bad_lccn'); -- no match, set an error message
end
end
elseif 12 == len then -- LCCN should be aadddddddddd
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
if not id:match("^%l%l%d%d%d%d%d%d%d%d%d%d") then -- see if it matches our pattern
err_cat = ' ' .. set_error( 'bad_lccn' ); -- no match, set an error message
err_cat = ' ' .. set_message ('err_bad_lccn'); -- no match, set an error message
end
end
else
else
err_cat = ' ' .. set_error( 'bad_lccn' ); -- wrong length, set an error message
err_cat = ' ' .. set_message ('err_bad_lccn'); -- wrong length, set an error message
end
end


if not is_set (err_cat) and nil ~= lccn:find ('%s') then
if not is_set (err_cat) and nil ~= lccn:find ('%s') then
err_cat = ' ' .. set_error( 'bad_lccn' ); -- lccn contains a space, set an error message
err_cat = ' ' .. set_message ('err_bad_lccn'); -- lccn contains a space, set an error message
end
 
if is_set (err_cat) then
options.coins_list_t['LCCN'] = nil; -- when error, unset so not included in COinS
end
end


return external_link_id({link = handler.link, label = handler.label, q = handler.q,
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;
prefix = handler.prefix, id = lccn, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
end


Line 702: Line 1,007:
]]
]]


local function mr (id)
local function mr (options)
local handler = cfg.id_handlers['MR'];
local id = options.id;
local handler = options.handler;
local id_num;
local id_num;
local id_len;
local id_len;
Line 711: Line 1,017:


if is_set (id_num) then
if is_set (id_num) then
add_maint_cat ('mr_format');
set_message ('maint_mr_format'); -- add maint cat
else -- plain number without mr prefix
else -- plain number without mr prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
id_num = id:match ('^%d+$'); -- if here id is all digits
Line 718: Line 1,024:
id_len = id_num and id_num:len() or 0;
id_len = id_num and id_num:len() or 0;
if (7 >= id_len) and (0 ~= id_len) then
if (7 >= id_len) and (0 ~= id_len) then
id = string.rep ('0', 7-id_len ) .. id_num; -- zero-fill leading digits
id = string.rep ('0', 7-id_len) .. id_num; -- zero-fill leading digits
else
else
err_cat = ' ' .. set_error( 'bad_mr' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_mr'); -- set an error message
options.coins_list_t['MR'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id({link = handler.link, label = handler.label, q = handler.q,
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;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
end


Line 730: Line 1,037:
--[[--------------------------< O C L C >----------------------------------------------------------------------
--[[--------------------------< O C L C >----------------------------------------------------------------------


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


]]
]]


local function oclc (id)
local function oclc (options)
local handler = cfg.id_handlers['OCLC'];
local id = options.id;
local handler = options.handler;
local number;
local number;
local err_msg = ''; -- empty string for concatenation
local err_msg = ''; -- empty string for concatenation
Line 749: Line 1,057:
number = id:match('%(OCoLC%)([1-9]%d*)'); -- get the number
number = id:match('%(OCoLC%)([1-9]%d*)'); -- get the number
if 9 < number:len() then
if 9 < number:len() then
number = nil; -- contrain to 1 to 9 digits; change this when oclc issues 10-digit numbers
number = nil; -- constrain to 1 to 9 digits; change this when OCLC issues 10-digit numbers
end
end
elseif id:match('^%d+$') then -- no prefix
elseif id:match('^%d+$') then -- no prefix
number = id; -- get the number
number = id; -- get the number
if 10 < number:len() then
if 10 < number:len() then
number = nil; -- contrain to 1 to 10 digits; change this when oclc issues 11-digit numbers
number = nil; -- constrain to 1 to 10 digits; change this when OCLC issues 11-digit numbers
end
end
end
end
Line 761: Line 1,069:
id = number; -- exclude prefix, if any, from external link
id = number; -- exclude prefix, if any, from external link
else
else
err_msg = ' ' .. set_error( 'bad_oclc' ) -- add an error message if the id is malformed
err_msg = ' ' .. set_message ('err_bad_oclc') -- add an error message if the id is malformed
options.coins_list_t['OCLC'] = nil; -- when error, unset so not included in COinS
end
end
local text = external_link_id({link=handler.link, label=handler.label, q = handler.q,
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix=handler.prefix, id=id, separator=handler.separator, encode=handler.encode}) .. err_msg;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_msg;


return text;
return text;
Line 777: Line 1,086:
]]
]]


local function openlibrary(id, access)
local function openlibrary (options)
local code;
local id = options.id;
local handler = cfg.id_handlers['OL'];
local access = options.access;
local ident;
local handler = options.handler;
local ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- strip optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W';
ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W'; remove OL prefix
local error_msg = '';
local prefix = { -- these are appended to the handler.prefix according to code
['A']='authors/OL',
['M']='books/OL',
['W']='works/OL',
['X']='OL' -- not a code; spoof when 'code' in id is invalid
};
 
if not ident then
code = 'X'; -- no code or id completely invalid
ident = id; -- copy id to ident so that we display the flawed identifier
error_msg = ' ' .. set_message ('err_bad_ol');
end


if not is_set (ident) then -- if malformed return an error
if not is_set (error_msg) then
return external_link_id({link=handler.link, label=handler.label, q = handler.q,
options.coins_list_t['OL'] = handler.prefix .. prefix[code] .. ident; -- experiment for ol coins
prefix=handler.prefix .. 'OL',
else
id=id, separator=handler.separator, encode = handler.encode,
options.coins_list_t['OL'] = nil; -- when error, unset so not included in COinS
access = access}) .. ' ' .. set_error( 'bad_ol' );
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix .. prefix[code],
id = ident, separator = handler.separator, encode = handler.encode,
access = access}) .. error_msg;
end
--[[--------------------------< O S T I >----------------------------------------------------------------------
Format OSTI and do simple error checking. OSTIs are sequential numbers beginning at 1 and counting up.  This
code checks the OSTI to see that it contains only digits and is less than test_limit specified in the configuration;
the value in test_limit will need to be updated periodically as more OSTIs are issued.
NB. 1018 is the lowest OSTI number found in the wild (so far) and resolving OK on the OSTI site
]]
local function osti (options)
local id = options.id;
local access = options.access;
local handler = options.handler;
local err_cat = ''; -- presume that OSTI is valid
id = ident; -- use ident without the optional OL prefix (it has been removed)
if id:match("[^%d]") then -- if OSTI has anything but digits
err_cat = ' ' .. set_message ('err_bad_osti'); -- set an error message
if ( code == "A" ) then
options.coins_list_t['OSTI'] = nil; -- when error, unset so not included in COinS
return external_link_id({link=handler.link, label=handler.label, q = handler.q,
else -- OSTI is only digits
prefix=handler.prefix .. 'authors/OL',
local id_num = tonumber (id); -- convert id to a number for range testing
id=id, separator=handler.separator, encode = handler.encode,
if 1018 > id_num or handler.id_limit < id_num then -- if OSTI is outside test limit boundaries
access = access})
err_cat = ' ' .. set_message ('err_bad_osti'); -- set an error message
options.coins_list_t['OSTI'] = nil; -- when error, unset so not included in COinS
end
end
end
if ( code == "M" ) then
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id({link=handler.link, label=handler.label, q = handler.q,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access}) .. err_cat;
prefix=handler.prefix .. 'books/OL',
id=id, separator=handler.separator, encode = handler.encode,
access = access})
end
 
if ( code == "W" ) then
return external_link_id({link=handler.link, label=handler.label, q = handler.q,
prefix=handler.prefix .. 'works/OL',
id=id, separator=handler.separator, encode = handler.encode,
access = access})
end
end
end


Line 833: Line 1,167:
]]
]]


local function pmc(id, embargo)
local function pmc (options)
local test_limit = 6500000; -- update this value as PMCs approach
local id = options.id;
local handler = cfg.id_handlers['PMC'];
local embargo = options.Embargo; -- TODO: lowercase?
local err_cat = ''; -- presume that PMC is valid
local handler = options.handler;
local err_cat;
local id_num;
local id_num;
local text;
local text;
 
id_num = id:match ('^[Pp][Mm][Cc](%d+)$'); -- identifier with pmc prefix
id_num = id:match ('^[Pp][Mm][Cc](%d+)$'); -- identifier with PMC prefix


if is_set (id_num) then
if is_set (id_num) then
add_maint_cat ('pmc_format');
set_message ('maint_pmc_format');
else -- plain number without pmc prefix
else -- plain number without PMC prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
id_num = id:match ('^%d+$'); -- if here id is all digits
end
end


if is_set (id_num) then -- id_num has a value so test it
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
id_num = tonumber (id_num); -- convert id_num to a number for range testing
if 1 > id_num or test_limit < id_num then -- if PMC is outside test limit boundaries
if 1 > id_num or handler.id_limit < id_num then -- if PMC is outside test limit boundaries
err_cat = ' ' .. set_error( 'bad_pmc' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_pmc'); -- set an error message
else
else
id = tostring (id_num); -- make sure id is a string
id = tostring (id_num); -- make sure id is a string
end
end
else -- when id format incorrect
else -- when id format incorrect
err_cat = ' ' .. set_error( 'bad_pmc' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_pmc'); -- set an error message
end
end
if is_set (embargo) then -- is PMC is still embargoed?
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
text = table.concat ( -- still embargoed so no external link
{
{
make_wikilink (handler.link, handler.label),
make_wikilink (link_label_make (handler), handler.label),
handler.separator,
handler.separator,
id,
id,
err_cat
(err_cat and err_cat or '') -- parens required
});
});
else
else
text = external_link_id({link = handler.link, label = handler.label, q = handler.q, -- no embargo date or embargo has expired, ok to link to article
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;
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_cat and err_cat or ''); -- parentheses required
end
end
if err_cat then
options.coins_list_t['PMC'] = nil; -- when error, unset so not included in COinS
end
return text;
return text;
end
end
Line 883: Line 1,225:
]]
]]


local function pmid(id)
local function pmid (options)
local test_limit = 32000000; -- update this value as PMIDs approach
local id = options.id;
local handler = cfg.id_handlers['PMID'];
local handler = options.handler;
local err_cat = ''; -- presume that PMID is valid
local err_cat = ''; -- presume that PMID is valid
if id:match("[^%d]") then -- if PMID has anything but digits
if id:match("[^%d]") then -- if PMID has anything but digits
err_cat = ' ' .. set_error( 'bad_pmid' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_pmid'); -- set an error message
options.coins_list_t['PMID'] = nil; -- when error, unset so not included in COinS
else -- PMID is only digits
else -- PMID is only digits
local id_num = tonumber(id); -- convert id to a number for range testing
local id_num = tonumber (id); -- convert id to a number for range testing
if 1 > id_num or test_limit < id_num then -- if PMID is outside test limit boundaries
if 1 > id_num or handler.id_limit < id_num then -- if PMID is outside test limit boundaries
err_cat = ' ' .. set_error( 'bad_pmid' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_pmid'); -- set an error message
options.coins_list_t['PMID'] = nil; -- when error, unset so not included in COinS
end
end
end
end
return external_link_id({link = handler.link, label = handler.label, q = handler.q,
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;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
 
 
--[[--------------------------< R F C >------------------------------------------------------------------------
 
Format RFC and do simple error checking. RFCs are sequential numbers beginning at 1 and counting up.  This
code checks the RFC to see that it contains only digits and is less than test_limit specified in the configuration;
the value in test_limit will need to be updated periodically as more RFCs are issued.
 
An index of all RFCs is here: https://tools.ietf.org/rfc/
 
]]
 
local function rfc (options)
local id = options.id;
local handler = options.handler;
local err_cat = ''; -- presume that RFC is valid
if id:match("[^%d]") then -- if RFC has anything but digits
err_cat = ' ' .. set_message ('err_bad_rfc'); -- set an error message
options.coins_list_t['RFC'] = nil; -- when error, unset so not included in COinS
else -- RFC is only digits
local id_num = tonumber (id); -- convert id to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if RFC is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_rfc'); -- set an error message
options.coins_list_t['RFC'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;
end
 
 
--[[--------------------------< S 2 C I D >--------------------------------------------------------------------
 
Format an S2CID, do simple error checking
 
S2CIDs are sequential numbers beginning at 1 and counting up.  This code checks the S2CID to see that it is only
digits and is less than test_limit; the value in local variable test_limit will need to be updated periodically
as more S2CIDs are issued.
 
]]
 
local function s2cid (options)
local id = options.id;
local access = options.access;
local handler = options.handler;
local err_cat = ''; -- presume that S2CID is valid
local id_num;
local text;
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 ('err_bad_s2cid'); -- set an error message
options.coins_list_t['S2CID'] = nil; -- when error, unset so not included in COinS
end
 
else -- when id format incorrect
err_cat = ' ' .. set_message ('err_bad_s2cid'); -- set an error message
options.coins_list_t['S2CID'] = nil; -- when error, unset so not included in COinS
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}) .. 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
 
sbn value not made part of COinS metadata because we don't have a url or isn't a COinS-defined identifier (rft.xxx)
or an identifier registered at info-uri.info (info:)
 
]]
 
local function sbn (options)
local id = options.id;
local ignore_invalid = options.accept;
local handler = options.handler;
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
options.coins_list_t['SBN'] = nil; -- when error, unset so not included in COinS; not really necessary here because sbn not made part of COinS
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
end


Line 904: Line 1,363:
--[[--------------------------< S S R N >----------------------------------------------------------------------
--[[--------------------------< S S R N >----------------------------------------------------------------------


Format an ssrn, do simple error checking
Format an SSRN, do simple error checking


SSRNs are sequential numbers beginning at 100? and counting up.  This code checks the ssrn to see that it is
SSRNs are sequential numbers beginning at 100? and counting up.  This code checks the SSRN to see that it is
only digits and is greater than 99 and less than test_limit; the value in local variable test_limit will need
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.
to be updated periodically as more SSRNs are issued.
Line 912: Line 1,371:
]]
]]


local function ssrn (id)
local function ssrn (options)
local test_limit = 3500000; -- update this value as SSRNs approach
local id = options.id;
local handler = cfg.id_handlers['SSRN'];
local handler = options.handler;
local err_cat = ''; -- presume that SSRN is valid
local err_cat = ''; -- presume that SSRN is valid
local id_num;
local id_num;
Line 922: Line 1,381:


if is_set (id_num) then -- id_num has a value so test it
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
id_num = tonumber (id_num); -- convert id_num to a number for range testing
if 100 > id_num or test_limit < id_num then -- if SSRN is outside test limit boundaries
if 100 > id_num or handler.id_limit < id_num then -- if SSRN is outside test limit boundaries
err_cat = ' ' .. set_error( 'bad_ssrn' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_ssrn'); -- set an error message
options.coins_list_t['SSRN'] = nil; -- when error, unset so not included in COinS
end
end
else -- when id format incorrect
else -- when id format incorrect
err_cat = ' ' .. set_error( 'bad_ssrn' ); -- set an error message
err_cat = ' ' .. set_message ('err_bad_ssrn'); -- set an error message
options.coins_list_t['SSRN'] = nil; -- when error, unset so not included in COinS
end
end
text = external_link_id({link = handler.link, label = handler.label, q = handler.q,
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;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;


return text;
return text;
Line 944: Line 1,405:
]]
]]


local function usenet_id (id)
local function usenet_id (options)
local handler = cfg.id_handlers['USENETID'];
local id = options.id;
local handler = options.handler;


local text = external_link_id({link = handler.link, label = handler.label, q = handler.q,
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})
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 '>'
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_error( 'bad_usenet_id' ) -- add an error message if the message id is invalid
text = text .. ' ' .. set_message ('err_bad_usenet_id') -- add an error message if the message id is invalid
options.coins_list_t['USENETID'] = nil; -- when error, unset so not included in COinS
end  
end  
Line 961: Line 1,424:


A numerical identifier in the form nnnn.nnnnn - leading zeros in the first quartet optional
A numerical identifier in the form nnnn.nnnnn - leading zeros in the first quartet optional
format described here: http://emis.mi.sanu.ac.rs/ZMATH/zmath/en/help/search/
temporary format is apparently eight digits.  Anything else is an error


]]
]]


local function zbl (id)
local function zbl (options)
local handler = cfg.id_handlers['ZBL'];
local id = options.id;
local id_num;
local handler = options.handler;
local err_cat = '';
local err_cat = '';
id_num = id:match ('^[Zz][Bb][Ll](.*)$'); -- identifier with zbl prefix; extract identifier
if id:match('^%d%d%d%d%d%d%d%d$') then -- is this identifier using temporary format?
 
set_message ('maint_zbl'); -- yes, add maint cat
if is_set (id_num) then
elseif not id:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then -- not temporary, is it normal format?
add_maint_cat ('zbl_format');
err_cat = ' ' .. set_message ('err_bad_zbl'); -- no, set an error message
else -- plain number without zbl prefix
options.coins_list_t['ZBL'] = nil; -- when error, unset so not included in COinS
id_num = id; -- if here id does not have prefix
end
 
if id_num:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then
id = id_num; -- id matches pattern
else
err_cat = ' ' .. set_error( 'bad_zbl' ); -- set an error message
end
end
return external_link_id({link = handler.link, label = handler.label, q = handler.q,
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;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
end


Line 990: Line 1,450:
--============================<< I N T E R F A C E  F U N C T I O N S >>==========================================
--============================<< I N T E R F A C E  F U N C T I O N S >>==========================================


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


Takes a table of IDs created by extract_ids() and turns it into a table of formatted ID outputs.
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 if more than one alias exists in args


inputs:
]]
id_list – table of identifiers built by extract_ids()
options – table of various template parameter values used to modify some manually handled identifiers


]]
local function extract_ids (args)
local id_list = {}; -- list of identifiers found in args
for k, v in pairs (cfg.id_handlers) do -- k is uppercase identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
v = select_one (args, v.parameters, 'err_redundant_parameters' ); -- v.parameters is a table of aliases for k; here we pick one from args if present
if is_set (v) then id_list[k] = v; end -- if found in args, add identifier to our list
end
return id_list;
end
 
 
--[[--------------------------< E X T R A C T _ I D _ A C C E S S _ L E V E L S >--------------------------------------
 
Fetches custom id access levels from arguments using configuration settings. Parameters which have a predefined access
level (e.g. arxiv) do not use this function as they are directly rendered as free without using an additional parameter.
 
returns a table of k/v pairs where k is same as the identifier's key in cfg.id_handlers and v is the assigned (valid) keyword


local function build_id_list( id_list, options )
access-level values must match the case used in cfg.keywords_lists['id-access'] (lowercase unless there is some special reason for something else)
local new_list, handler = {};


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
-- fallback to read-only cfg
handler = setmetatable( { ['id'] = v, ['access'] = options.IdAccessLevels[k] }, fallback(k) );


if handler.mode == 'external' then
local function extract_id_access_levels (args, id_list)
table.insert( new_list, {handler.label, external_link_id( handler ) } );
local id_accesses_list = {};
elseif handler.mode == 'internal' then
for k, v in pairs (cfg.id_handlers) do
table.insert( new_list, {handler.label, internal_link_id( handler ) } );
local access_param = v.custom_access; -- name of identifier's access-level parameter
elseif handler.mode ~= 'manual' then
if is_set (access_param) then
error( cfg.messages['unknown_ID_mode'] );
local access_level = args[access_param]; -- get the assigned value if there is one
elseif k == 'ARXIV' then
if is_set (access_level) then
table.insert( new_list, {handler.label, arxiv( v, options.Class ) } );  
if not in_array (access_level, cfg.keywords_lists['id-access']) then -- exact match required
elseif k == 'ASIN' then
table.insert (z.message_tail, { set_message ('err_invalid_param_val', {access_param, access_level}, true) } );
table.insert( new_list, {handler.label, asin( v, options.ASINTLD ) } );
access_level = nil; -- invalid so unset
elseif k == 'BIBCODE' then
end
table.insert( new_list, {handler.label, bibcode( v, handler.access ) } );
if not is_set (id_list[k]) then -- identifier access-level must have a matching identifier
elseif k == 'BIORXIV' then
table.insert (z.message_tail, { set_message ('err_param_access_requires_param', {k:lower()}, true) } ); -- parameter name is uppercase in cfg.id_handlers (k); lowercase for error message
table.insert( new_list, {handler.label, biorxiv( v ) } );
elseif k == 'CITESEERX' then
table.insert( new_list, {handler.label, citeseerx( v ) } );
elseif k == 'DOI' then
table.insert( new_list, {handler.label, doi( v, options.DoiBroken, handler.access ) } );
elseif k == 'EISSN' then
table.insert( new_list, {handler.label, issn( v, true ) } ); -- true distinguishes eissn from issn
elseif k == 'HDL' then
table.insert( new_list, {handler.label, hdl( v, handler.access ) } );
elseif k == 'ISBN' then
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
end
elseif is_set(options.IgnoreISBN) then -- ISBN is OK; if |ignore-isbn-error= set
id_accesses_list[k] = cfg.keywords_xlate[access_level]; -- get translated keyword
add_maint_cat ('ignore_isbn_err'); -- because |ignore-isbn-error= unnecessary
end
end
table.insert( new_list, {handler.label, ISBN } );
end
elseif k == 'ISMN' then
end
table.insert( new_list, {handler.label, ismn( v ) } );
return id_accesses_list;
elseif k == 'ISSN' then
end
table.insert( new_list, {handler.label, issn( v ) } );
 
elseif k == 'JFM' then
 
table.insert( new_list, {handler.label, jfm( v ) } );
--[[--------------------------< B U I L D _ I D _ L I S T >----------------------------------------------------
elseif k == 'LCCN' then
 
table.insert( new_list, {handler.label, lccn( v ) } );
render the identifiers into a sorted sequence table
elseif k == 'MR' then
 
table.insert( new_list, {handler.label, mr( v ) } );
<ID_list_coins_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value
elseif k == 'OCLC' then
<options_t> is a table of various k/v option pairs provided in the call to new_build_id_list();
table.insert( new_list, {handler.label, oclc( v ) } );
modified by this function and passed to all identifier rendering functions
elseif k == 'OL' or k == 'OLA' then
<access_levels_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value (if valid)
table.insert( new_list, {handler.label, openlibrary( v, handler.access ) } );
 
elseif k == 'PMC' then
returns a sequence table of sorted (by hkey - 'handler' key) rendered identifier strings
table.insert( new_list, {handler.label, pmc( v, options.Embargo ) } );
 
elseif k == 'PMID' then
]]
table.insert( new_list, {handler.label, pmid( v ) } );
 
elseif k == 'SSRN' then
local function build_id_list (ID_list_coins_t, options_t, access_levels_t)
table.insert( new_list, {handler.label, ssrn( v ) } );
local ID_list_t = {};
elseif k == 'USENETID' then
local accept;
table.insert( new_list, {handler.label, usenet_id( v ) } );
local func_map = { --function map points to functions associated with hkey identifier
elseif k == 'ZBL' then
['ARXIV'] = arxiv,
table.insert( new_list, {handler.label, zbl( v ) } );
['ASIN'] = asin,
['BIBCODE'] = bibcode,
['BIORXIV'] = biorxiv,
['CITESEERX'] = citeseerx,
['DOI'] = doi,
['EISSN'] = issn,
['HDL'] = hdl,
['ISBN'] = isbn,
['ISMN'] = ismn,
['ISSN'] = issn,
['JFM'] = jfm,
['JSTOR'] = jstor,
['LCCN'] = lccn,
['MR'] = mr,
['OCLC'] = oclc,
['OL'] = openlibrary,
['OSTI'] = osti,
['PMC'] = pmc,
['PMID'] = pmid,
['RFC']  = rfc,
['S2CID'] = s2cid,
['SBN'] = sbn,
['SSRN'] = ssrn,
['USENETID'] = usenet_id,
['ZBL'] = zbl,
}
 
for hkey, v in pairs (ID_list_coins_t) do
v, accept = has_accept_as_written (v); -- remove accept-as-written markup if present; accept is boolean true when markup removed; false else
-- every function gets the options table with value v and accept boolean
options_t.hkey = hkey; -- ~/Configuration handler key
options_t.id = v; -- add that identifier value to the options table
options_t.accept = accept; -- add the accept boolean flag
options_t.access = access_levels_t[hkey]; -- add the access level for those that have an |<identifier-access= parameter
options_t.handler = cfg.id_handlers[hkey];
options_t.coins_list_t = ID_list_coins_t; -- pointer to ID_list_coins_t; for |asin= and |ol=; also to keep erroneous values out of the citation's metadata
if func_map[hkey] then
table.insert (ID_list_t, {hkey, func_map[hkey] (options_t)}); -- call the function and add the results to the output sequence table
else
else
error( cfg.messages['unknown_manual_ID'] );
error (cfg.messages['unknown_ID_key'] .. ' ' .. hkey); -- here when func_map doesn't have a function for hkey
end
end
end
end
 
local function comp( a, b ) -- used in following table.sort()
local function comp (a, b) -- used by following table.sort()
return a[1]:lower() < b[1]:lower();
return a[1]:lower() < b[1]:lower(); -- sort by hkey
end
end
 
table.sort( new_list, comp );
table.sort (ID_list_t, comp); -- sequence table of tables sort
for k, v in ipairs( new_list ) do
for k, v in ipairs (ID_list_t) do -- convert sequence table of tables to simple sequence table of strings
new_list[k] = v[2];
ID_list_t[k] = v[2]; -- v[2] is the identifier rendering from the call to the various functions in func_map{}
end
end
return new_list;
return ID_list_t;
end
end




--[[--------------------------< E X T R A C T _ I D S >------------------------------------------------------------
--[[--------------------------< O P T I O N S _ C H E C K >----------------------------------------------------
 
check that certain option parameters have their associated identifier parameters with values


Populates ID table from arguments using configuration settings. Loops through cfg.id_handlers and searches args for
<ID_list_coins_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value
any of the parameters listed in each cfg.id_handlers['...'].parameters.  If found, adds the parameter and value to
<ID_support_t> is a sequence table of tables created in citation0() where each subtable has four elements:
the identifier list.  Emits redundant error message is more than one alias exists in args
[1] is the support parameter's assigned value; empty string if not set
[2] is a text string same as key in cfg.id_handlers
[3] is cfg.error_conditions key used to create error message
[4] is original ID support parameter name used to create error message
returns nothing; on error emits an appropriate error message


]]
]]


local function extract_ids( args )
local function options_check (ID_list_coins_t, ID_support_t)
local id_list = {}; -- list of identifiers found in args
for _, v in ipairs (ID_support_t) do
for k, v in pairs( cfg.id_handlers ) do -- k is uc identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
if is_set (v[1]) and not ID_list_coins_t[v[2]] then -- when support parameter has a value but matching identifier parameter is missing or empty
v = select_one( args, v.parameters, 'redundant_parameters' ); -- v.parameters is a table of aliases for k; here we pick one from args if present
table.insert (z.message_tail, {set_message (v[3], (v[4]))}); -- emit the appropriate error message
if is_set(v) then id_list[k] = v; end -- if found in args, add identifier to our list
end
end
end
return id_list;
end
end




--[[--------------------------< E X T R A C T _ I D _ A C C E S S _ L E V E L S >--------------------------------------
--[[--------------------------< I D E N T I F I E R _ L I S T S _ G E T >--------------------------------------


Fetches custom id access levels from arguments using configuration settings.
Creates two identifier lists: a k/v table of identifiers and their values to be used locally and for use in the
Parameters which have a predefined access level (e.g. arxiv) do not use this
COinS metadata, and a sequence table of the rendered identifier strings that will be included in the rendered
function as they are directly rendered as free without using an additional parameter.
citation.


]]
]]


local function extract_id_access_levels( args, id_list )
local function identifier_lists_get (args, options_t, ID_support_t)
local id_accesses_list = {};
local ID_list_coins_t = extract_ids (args); -- get a table of identifiers and their values for use locally and for use in COinS
for k, v in pairs( cfg.id_handlers ) do
options_check (ID_list_coins_t, ID_support_t); -- ID support parameters must have matching identifier parameters
local access_param = v.custom_access;
local ID_access_levels_t = extract_id_access_levels (args, ID_list_coins_t); -- get a table of identifier access levels
local k_lower = string.lower(k);
local ID_list_t = build_id_list (ID_list_coins_t, options_t, ID_access_levels_t); -- get a sequence table of rendered identifier strings
if is_set(access_param) then
 
local access_level = args[access_param];
return ID_list_t, ID_list_coins_t; -- return the tables
if is_set(access_level) then
if not in_array (access_level:lower(), cfg.keywords['id-access']) then
table.insert( z.message_tail, { set_error( 'invalid_param_val', {access_param, access_level}, true ) } );
access_level = nil;
end
if not is_set(id_list[k]) then
table.insert( z.message_tail, { set_error( 'param_access_requires_param', {k_lower}, true ) } );
end
if is_set(access_level) then
access_level = access_level:lower();
end
id_accesses_list[k] = access_level;
end
end
end
return id_accesses_list;
end
end


Line 1,149: Line 1,627:
cfg = cfg_table_ptr;
cfg = cfg_table_ptr;


is_set = utilities_page_ptr.is_set; -- import functions from select Module:Citation/CS1/Utilities module
has_accept_as_written = utilities_page_ptr.has_accept_as_written; -- import functions from select Module:Citation/CS1/Utilities module
is_set = utilities_page_ptr.is_set;
in_array = utilities_page_ptr.in_array;
in_array = utilities_page_ptr.in_array;
set_error = utilities_page_ptr.set_error;
set_message = utilities_page_ptr.set_message;
select_one = utilities_page_ptr.select_one;
select_one = utilities_page_ptr.select_one;
add_maint_cat = utilities_page_ptr.add_maint_cat;
substitute = utilities_page_ptr.substitute;
substitute = utilities_page_ptr.substitute;
make_wikilink = utilities_page_ptr.make_wikilink;
make_wikilink = utilities_page_ptr.make_wikilink;
Line 1,165: Line 1,643:


return {
return {
build_id_list = build_id_list,
auto_link_urls = auto_link_urls, -- table of identifier URLs to be used when auto-linking |title=
extract_ids = extract_ids,
extract_id_access_levels = extract_id_access_levels,
identifier_lists_get = identifier_lists_get, -- experiment to replace individual calls to build_id_list, extract_ids, extract_id_access_levels
is_embargoed = is_embargoed;
is_embargoed = is_embargoed;
set_selected_modules = set_selected_modules;
set_selected_modules = set_selected_modules;
}
}