Module:Citation/CS1/Identifiers: Difference between revisions
Content deleted Content added
Johnrdorazio (talk | contribs) m 1 revision imported |
No edit summary |
||
Line 3:
]]
local has_accept_as_written, is_set, in_array,
substitute, make_wikilink;
local z; -- table of tables defined in Module:Citation/CS1/Utilities
local cfg; -- table of configuration tables that are defined in Module:Citation/CS1/Configuration
--[[--------------------------< P A G E S C O P E V A R I A B L E S >--------------------------------------
declare variables here that have page-wide scope that are not brought in from other modules; that are created here and used here
]]
local auto_link_urls = {}; -- holds identifier URLs for those identifiers that can auto-link |title=
Line 14 ⟶ 24:
--[[--------------------------< W I K I D A T A _ A R T I C L E _ N A M E _ G E T >----------------------------
as an aid to internationalizing identifier-label wikilinks, gets identifier article names from
returns :<lang code>:<article title> when <q> has an <article title> for <lang code>; nil else
Line 30 ⟶ 40:
local wd_article;
local this_wiki_code = cfg.this_wiki_code; --
wd_article = mw.wikibase.getSitelink (q, this_wiki_code .. 'wiki'); -- fetch article title from
if wd_article then
wd_article = table.concat ({':', this_wiki_code, ':', wd_article}); -- interwiki-style link without brackets if taken from
end
return wd_article; -- article title from
end
--[[--------------------------< L I N K _ L A B E L _ M A K E >------------------------------------------------
common function to create identifier link label from handler table or from Wikidata
returns the first available of
1. redirect from local wiki's handler table (if enabled)
2. Wikidata (if there is a Wikidata entry for this identifier in the local wiki's language)
3. label specified in the local wiki's handler table
]]
local function link_label_make (handler)
local wd_article;
if not (cfg.use_identifier_redirects and is_set (handler.redirect)) then -- redirect has priority so if enabled and available don't fetch from Wikidata because expensive
wd_article = wikidata_article_name_get (handler.q); -- if Wikidata has an article title for this wiki, get it;
end
return (cfg.use_identifier_redirects and is_set (handler.redirect) and handler.redirect) or wd_article or handler.link;
end
Line 44 ⟶ 76:
--[[--------------------------< E X T E R N A L _ L I N K _ I D >----------------------------------------------
Formats a wiki
NB. Wikidata P1630 has a formatter URL with $1 placeholder for the ID which could be worked into our prefix/id/suffix
string, either overriding local definitions (auto-update) or as fallback for identifiers without local definitions.
But is expensive and could be risky if WD gets vandalized.
See Template_talk:Authority_control/Archive_8#Use_Wikidata_as_the_source_for_the_external_link
]]
local function external_link_id (options)
local url_string = options.id;
local ext_link;
local this_wiki_code = cfg.this_wiki_code; --
local wd_article; -- article title from
if options.encode == true or options.encode == nil then
url_string = mw.uri.encode( url_string );
end
if options.auto_link and is_set (options.access) then
auto_link_urls[options.auto_link] = table.concat ({options.prefix, url_string, options.suffix});
end
Line 62 ⟶ 104:
ext_link = substitute (cfg.presentation['ext-link-access-signal'], {cfg.presentation[options.access].class, cfg.presentation[options.access].title, ext_link}); -- add the free-to-read / paywall lock
end
return table.concat ({
make_wikilink (
options.separator or ' ',
ext_link
Line 79 ⟶ 115:
--[[--------------------------< I N T E R N A L _ L I N K _ I D >----------------------------------------------
Formats a wiki
]]
Line 85 ⟶ 121:
local function internal_link_id(options)
local id = mw.ustring.gsub (options.id, '%d', cfg.date_names.local_digits); -- translate 'local' digits to Western 0-9
return table.concat (
{
make_wikilink (
options.separator or ' ', -- add the separator
make_wikilink (
Line 100 ⟶ 130:
{
options.prefix,
id, -- translated to
options.suffix or ''
}),
substitute (cfg.presentation['bdi'], {'', mw.text.nowiki (options.id)}) -- bdi tags to prevent
); -- nowiki because MediaWiki still has magic links for ISBN and the like; TODO: is it really required?
});
Line 111 ⟶ 141:
--[[--------------------------< I S _ E M B A R G O E D >------------------------------------------------------
Determines if a PMC identifier's online version is embargoed. Compares the date in |pmc-embargo-date= against today's date. If embargo date is
in the future, returns the content of |pmc-embargo-date=; otherwise, returns
|pmc-embargo-date= was not set in this cite.
]]
Line 120 ⟶ 150:
if is_set (embargo) then
local lang = mw.getContentLanguage();
local good1, embargo_date
good1, embargo_date = pcall (
if good1
if tonumber (
return embargo; -- still embargoed
else
set_message ('maint_pmc_embargo'); -- embargo has expired; add main cat
return ''; -- unset because embargo has expired
end
end
end
return ''; -- |pmc-embargo-date= not set return empty string
end
Line 142 ⟶ 172:
2019-12-11T00:00Z <= biorxiv_date < today + 2 days
The dated form of biorxiv identifier has a start date of 2019-12-11. The
biorxiv_date is the date provided in those |biorxiv= parameter values that are dated at time 00:00:00 UTC
Line 151 ⟶ 181:
This function does not work if it is fed month names for languages other than English. Wikimedia #time: parser
apparently doesn't understand non-
contains a non-English month name because good1 is false after the call to lang.formatDate(). To get around that
call this function with YYYY-MM-DD format dates.
Line 159 ⟶ 189:
local function is_valid_biorxiv_date (biorxiv_date)
local good1, good2;
local biorxiv_ts, tomorrow_ts; -- to hold
local lang_object = mw.getContentLanguage();
good1, biorxiv_ts = pcall (lang_object.formatDate, lang_object, 'U', biorxiv_date ); -- convert biorxiv_date value to
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
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
end
Line 178 ⟶ 208:
--[[--------------------------< IS _ V A L I D _ I S X N >-----------------------------------------------------
ISBN-10 and ISSN validator code calculates checksum across all
ISBN-13 is checked in isbn().
If the number is valid the result will be 0. Before calling this function,
and stripped of dashes, spaces and other non-
]]
Line 189 ⟶ 219:
local temp = 0;
isxn_str = { isxn_str:byte(1, len) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39, 'X' → 0x58
len = len + 1; -- adjust to be a loop counter
for i, v in ipairs( isxn_str ) do -- loop through all of the bytes and calculate the checksum
if v == string.byte( "X" ) then -- if checkdigit is X (compares the byte value of 'X' which is 0x58)
Line 203 ⟶ 233:
--[[--------------------------< IS _ V A L I D _ I S X N _ 1 3 >-----------------------------------------------
ISBN-13 and ISMN validator code calculates checksum across all 13
If the number is valid, the result will be 0. Before calling this function,
and stripped of dashes, spaces and other non-
]]
Line 216 ⟶ 246:
temp = temp + (3 - 2*(i % 2)) * tonumber( string.char(v) ); -- multiply odd index digits by 1, even index digits by 3 and sum; includes check digit
end
return temp % 10 == 0; -- sum modulo 10 is zero when
end
Line 222 ⟶ 252:
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------
1. Remove all blanks.
2. If there is a forward slash (/) in the string, remove it, and remove all characters to the right of the forward slash.
Line 231 ⟶ 261:
2. If the length of the substring is less than 6, left-fill the substring with zeroes until the length is six.
Returns a normalized
]]
local function normalize_lccn (lccn)
lccn = lccn:gsub ("%s", ""); -- 1. strip whitespace
if nil ~= string.find (lccn, '/') then
lccn = lccn:match ("(.-)/"); -- 2. remove forward slash and all character to the right of it
end
local prefix
local suffix
prefix, suffix = lccn:match ("(.+)%-(.+)"); -- 3.a remove hyphen by splitting the string into prefix and suffix
if nil ~= suffix then -- if there was a hyphen
suffix = string.rep("0", 6-string.len (suffix)) .. suffix; -- 3.b.2 left fill the suffix with 0s if suffix length less than 6
lccn = prefix..suffix; -- reassemble the
end
return lccn;
end
--============================<< I D E N T I F I E R F U N C T I O N S >>====================================
Line 260 ⟶ 291:
format and error check arXiv identifier. There are three valid forms of the identifier:
the first form, valid only between date codes
arXiv:<archive>.<class>/<date code><number><version>
where:
Line 290 ⟶ 321:
local text; -- output text
if id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%d$") or id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%dv%d+$") then -- test for the
year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
year = tonumber(year);
Line 299 ⟶ 330:
end
elseif id:match("^%d%d[01]%d%.%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%dv%d+$") then -- test for the 0704-1412
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
year = tonumber(year);
month = tonumber(month);
if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or -- is year invalid or is month invalid? (doesn't test for future years)
((7 == year) and (4 > month)) then
err_cat = true; -- flag for error message
end
elseif id:match("^%d%d[01]%d%.%d%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%d%dv%d+$") then -- test for the 1501- format
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d%d[v%d]*$");
year = tonumber(year);
Line 320 ⟶ 351:
end
err_cat = err_cat and table.concat ({' ',
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;
if is_set (class) then
Line 329 ⟶ 360:
text = table.concat ({text, ' [[//arxiv.org/archive/', class, ' ', class, ']]'}); -- external link within square brackets, not wikilink
else
text = table.concat ({text, ' ',
end
end
Line 339 ⟶ 370:
--[[--------------------------< B I B C O D E >--------------------------------------------------------------------
Validates (sort of) and formats a bibcode
Format for bibcodes is specified here: http://adsabs.harvard.edu/abs_doc/help_pages/data.html#bibcodes
Line 360 ⟶ 391:
local year;
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode,
access = access});
if 19 ~= id:len() then
err_type = cfg.err_msg_supl.length;
else
year = id:match ("^(%d%d%d%d)[%a][%w&%.][%w&%.][%w&%.][%w.]+[%a%.]$")
if not year then -- if nil then no pattern match
err_type = cfg.err_msg_supl.value;
else
local next_year = tonumber(os.date ('%Y')) + 1; -- get the current year as a number and add one for next year
year = tonumber (year); -- convert year portion of bibcode to a number
if (1000 > year) or (year > next_year) then
err_type = cfg.err_msg_supl.year;
end
if id:find('&%.') then
err_type = cfg.err_msg_supl.journal;
end
end
Line 383 ⟶ 414:
if is_set (err_type) then -- if there was an error detected
text = text .. ' ' ..
end
return text;
Line 391 ⟶ 422:
--[[--------------------------< B I O R X I V >-----------------------------------------------------------------
Format bioRxiv
6 digits. After 2019-12-11, biorXiv
date and suffixed with an optional version identifier.
The bioRxiv
https://doi.org/10.1101/078733 -> 10.1101/078733
or a date followed by a six-digit number followed by an optional version indicator 'v' and one or more digits:
Line 412 ⟶ 443:
'^10.1101/(20[1-9]%d)%.([01]%d)%.([0-3]%d)%.%d%d%d%d%d%dv%d+$', -- y.m.d. date + 6-digit identifier + version (after 2019-12-11)
'^10.1101/(20[1-9]%d)%.([01]%d)%.([0-3]%d)%.%d%d%d%d%d%d$', -- y.m.d. date + 6-digit identifier (after 2019-12-11)
for _, pattern in ipairs (patterns) do -- spin through the patterns looking for a match
Line 419 ⟶ 450:
if m then -- m is nil when id is the six-digit form
if not is_valid_biorxiv_date (y .. '-' .. m .. '-' .. d) then -- validate the encoded date; TODO: don't ignore
break; -- date fail; break out early so we don't unset the error message
end
Line 428 ⟶ 459:
end -- err_cat remains set here when no match
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator,
encode = handler.encode, access = handler.access}) .. (err_cat and (' ' ..
end
Line 445 ⟶ 476:
local matched;
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode,
access = handler.access});
matched = id:match ("^10%.1%.1%.[1-9]%d?%d?%d?%.[1-9]%d?%d?%d?$");
if not matched then
text = text .. ' ' ..
end
return text;
Line 465 ⟶ 496:
Suffix: character string of any length chosen by the registrant
This function checks a DOI name for: prefix/suffix. If the
with a period or a comma, this function will emit a bad_doi error message.
DOI names are case-insensitive and can incorporate any printable Unicode characters so the test for spaces, endash,
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in
]]
local function doi (id, inactive, access, ignore_invalid)
local
local handler = cfg.id_handlers['DOI'];
local text;
if is_set (inactive) then
local inactive_year = inactive:match("%d%d%d%d") or ''; -- try to get the year portion from the inactive date
local inactive_month, good;
Line 492 ⟶ 523:
end
else
inactive_year = nil; -- |doi-broken-date= has something but it isn't a date
end
if is_set(inactive_year) and is_set (inactive_month) then
set_message ('maint_doi_inactive_dated', {inactive_year, inactive_month, ' '});
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–]-[^%.,]$'); -- registrant set when DOI has the proper basic form
local registrant_err_patterns = {
'^[^1-3]%d%d%d%d%.%d%d*$', -- 5 digits with subcode (0xxxx, 40000+);
'^[^1-4]%d%d%d%d$', -- 5 digits without subcode (0xxxx, 40000+);
'^[^1-9]%d%d%d%.%d%d*$', -- 4 digits with subcode (0xxx);
'^[^1-9]%d%d%d$', -- 4 digits without subcode (0xxx);
'^%d%d%d%d%d%d+', -- 6 or more digits
'^%d%d?%d?$', -- less than 4 digits without subcode (with subcode is legitimate)
Line 520 ⟶ 548:
'%s', -- any space character in registrant
}
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
err_cat = ' ' .. set_message ('err_bad_doi'); -- invalid directory or malformed
end
else
set_message ('maint_doi_ignore');
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 .. (err_cat and err_cat or ''); -- parentheses required
end
Line 552 ⟶ 589:
Query string parameters are named here: http://www.handle.net/proxy_servlet.html. query strings are not displayed
but since '?' is
have to detect the query string so that it isn't
]]
Line 584 ⟶ 621:
if found then
id = hdl; -- found so replace id with the handle portion; this will be
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–]-[^%.,]$") then --
text = text .. ' ' ..
end
return text;
Line 605 ⟶ 642:
]]
local function isbn
local handler = cfg.id_handlers['ISBN'];
local function return_result (check, err_type) -- local function to handle the various returns
local ISBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
prefix = handler.prefix, id = isbn_str, separator = handler.separator});
if ignore_invalid then -- if ignoring ISBN errors
set_message ('maint_isbn_ignore'); -- add a maint category even when there is no error
else -- here when not ignoring
if not check then -- and there is an error
return ISBN .. set_message ('err_bad_isbn', {err_type}, false, ' '); -- display an error message
end
end
return ISBN;
end
return return_result (false, cfg.err_msg_supl.char); -- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
end
local id = isbn_str:gsub ('[%s-]', ''); -- remove hyphens and whitespace
local len = id:len();
if len ~= 10 and len ~= 13 then
return return_result (false, cfg.err_msg_supl.length);
end
if len == 10 then
if
return return_result (false, cfg.err_msg_supl.form);
end
return return_result (is_valid_isxn(
else
if
return return_result (false, cfg.err_msg_supl.char);
end
if
return return_result (false, cfg.err_msg_supl.prefix);
end
if
return return_result (false, cfg.err_msg_supl.group);
end
return return_result (is_valid_isxn_13 (
end
end
--[[--------------------------< A
Formats a link to Amazon. Do simple error checking:
characters. If a mix, first character must be uppercase alpha; if all numeric,
Error message if not 10 characters, if not
|asin=630....... is (apparently) not a legitimate ISBN though it checksums as one; these do not cause this
function to emit the maint_asin message
This function is positioned here because it calls isbn()
Line 647 ⟶ 703:
]]
local function asin (id, domain)
local err_cat = ""
if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
err_cat = ' ' ..
else
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
if isbn (
if not id:find ('^630') then -- 630xxxxxxx is (apparently) not a valid isbn prefix but is used by amazon as a numeric identifier
set_message ('maint_asin'); -- begins with something other than 630 so possibly an isbn
end
elseif not is_set (err_cat) then
err_cat = ' ' ..
end
elseif not id:match("^%u[%d%u]+$") then
err_cat = ' ' ..
end
end
if not is_set(domain) then
domain = "com";
elseif in_array (domain, {'jp', 'uk'}) then -- Japan, United Kingdom
domain = "co." .. domain;
elseif in_array (domain, {'au', 'br', 'mx'}) then -- Australia, Brazil, Mexico
domain = "com." .. domain;
end
local handler = cfg.id_handlers['ASIN'];
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix .. domain .. "/dp/",
id = id, encode = handler.encode, separator = handler.separator}) .. err_cat;
end
Line 679 ⟶ 737:
--[[--------------------------< I S M N >----------------------------------------------------------------------
Determines whether an ISMN string is valid. Similar to
same check digit calculations. See http://www.ismn-international.org/download/Web_ISMN_Users_Manual_2008-6.pdf
section 2, pages 9–12.
Line 692 ⟶ 750:
id_copy = id; -- save a copy because this testing is destructive
id = id:gsub
if 13 ~= id:len() or id:match( "^9790%d*$" ) == nil then --
valid_ismn = false;
else
valid_ismn=is_valid_isxn_13 (id); -- validate
end
-- text = internal_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect, -- use this (or external version) when there is some place to link to
-- prefix = handler.prefix, id = id_copy, separator = handler.separator, encode = handler.encode})
text = table.concat ( -- because no place to link to yet
{
make_wikilink (
handler.separator,
id_copy
Line 713 ⟶ 769:
if false == valid_ismn then
text = text .. ' ' ..
end
Line 722 ⟶ 778:
--[[--------------------------< I S S N >----------------------------------------------------------------------
Validate and format an
has separated the two groups of four digits with a space. When that condition occurred, the resulting link looked
like this:
Line 728 ⟶ 784:
|issn=0819 4327 gives: [http://www.worldcat.org/issn/0819 4327 0819 4327] -- can't have spaces in an external link
This code now prevents that by inserting a hyphen at the
and makes sure that the checkdigit agrees with the calculated value. Incorrect length (8 digits), characters
other than 0-9 and X, or checkdigit / calculated value mismatch will all cause a check
]]
local function issn (id, e, ignore_invalid)
local issn_copy = id; -- save a copy of unadulterated
local handler;
local text;
local valid_issn = true;
handler = cfg.id_handlers[e and 'EISSN' or 'ISSN'];
id = id:gsub
if 8 ~= id:len() or nil == id:match( "^%d*X?$" ) then -- validate the
valid_issn = false; -- wrong length or improper character
else
valid_issn = is_valid_isxn(id, 8); -- validate
end
Line 758 ⟶ 810:
id = string.sub( id, 1, 4 ) .. "-" .. string.sub( id, 5 ); -- if valid, display correctly formatted version
else
id = issn_copy; -- if not valid,
end
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
if ignore_invalid then
set_message ('maint_issn_ignore');
else
if false == valid_issn then
text = text .. ' ' .. set_message ('err_bad_issn', e and 'e' or ''); -- add an error message if the ISSN is invalid
end
end
return text
Line 786 ⟶ 842:
if is_set (id_num) then
else -- plain number without mr prefix
id_num = id; -- if here id does not have prefix
Line 794 ⟶ 850:
id = id_num; -- jfm matches pattern
else
err_cat = ' ' ..
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
Line 809 ⟶ 865:
length = 8 then all digits
length = 9 then lccn[1] is
length = 10 then lccn[1] and lccn[2] are both
length = 11 then lccn[1] is lower case alpha, lccn[2] and lccn[3] are both
length = 12 then lccn[1] and lccn[2] are both
]]
Line 819 ⟶ 875:
local handler = cfg.id_handlers['LCCN'];
local err_cat = ''; -- presume that LCCN is valid
local id = lccn; -- local copy of the
id = normalize_lccn (id); -- get canonical form (no whitespace, hyphens, forward slashes)
local len = id:len(); -- get the length of the
if 8 == len then
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits)
err_cat = ' ' ..
end
elseif 9 == len then -- LCCN should be adddddddd
if nil == id:match("%l%d%d%d%d%d%d%d%d") then -- does it match our pattern?
err_cat = ' ' ..
end
elseif 10 == len then -- LCCN should be aadddddddd or dddddddddd
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits) ...
if nil == id:match("^%l%l%d%d%d%d%d%d%d%d") then -- ... see if it matches our pattern
err_cat = ' ' ..
end
end
elseif 11 == len then -- LCCN should be aaadddddddd or adddddddddd
if not (id:match("^%l%l%l%d%d%d%d%d%d%d%d") or id:match("^%l%d%d%d%d%d%d%d%d%d%d")) then -- see if it matches one of our patterns
err_cat = ' ' ..
end
elseif 12 == len then -- LCCN should be aadddddddddd
if not id:match("^%l%l%d%d%d%d%d%d%d%d%d%d") then -- see if it matches our pattern
err_cat = ' ' ..
end
else
err_cat = ' ' ..
end
if not is_set (err_cat) and nil ~= lccn:find ('%s') then
err_cat = ' ' ..
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = lccn, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
Line 874 ⟶ 930:
if is_set (id_num) then
set_message ('maint_mr_format'); -- add maint cat
else -- plain number without mr prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
Line 883 ⟶ 939:
id = string.rep ('0', 7-id_len ) .. id_num; -- zero-fill leading digits
else
err_cat = ' ' ..
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
Line 893 ⟶ 949:
--[[--------------------------< O C L C >----------------------------------------------------------------------
Validate and format an
archived at: https://web.archive.org/web/20161228233804/https://www.oclc.org/batchload/controlnumber.en.html
Line 912 ⟶ 968:
number = id:match('%(OCoLC%)([1-9]%d*)'); -- get the number
if 9 < number:len() then
number = nil; --
end
elseif id:match('^%d+$') then -- no prefix
number = id; -- get the number
if 10 < number:len() then
number = nil; --
end
end
Line 924 ⟶ 980:
id = number; -- exclude prefix, if any, from external link
else
err_msg = ' ' ..
end
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_msg;
return text;
Line 941 ⟶ 997:
local function openlibrary(id, access)
local handler = cfg.id_handlers['OL'];
local ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- strip optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W';
local error_msg = '';
local prefix = { -- these are appended to the handler.prefix according to code
['A']='authors/OL',
['M']='books/OL',
['W']='works/OL',
['X']='OL' -- not a code; spoof when 'code' in id is invalid
};
if not
code = 'X'; -- no code or id completely invalid
ident = id; -- copy id to ident so that we display the flawed identifier
error_msg = ' ' .. set_message ('err_bad_ol');
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix .. prefix[code],
id = ident, separator = handler.separator, encode = handler.encode,
access = access}) .. error_msg;
end
Line 996 ⟶ 1,037:
]]
local function pmc (id, embargo)
local handler = cfg.id_handlers['PMC'];
local err_cat
local id_num;
local text;
id_num = id:match ('^[Pp][Mm][Cc](%d+)$'); -- identifier with
if is_set (id_num) then
else -- plain number without
id_num = id:match ('^%d+$'); -- if here id is all digits
end
Line 1,013 ⟶ 1,054:
id_num = tonumber(id_num); -- convert id_num to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if PMC is outside test limit boundaries
err_cat = ' ' ..
else
id = tostring (id_num); -- make sure id is a string
end
else -- when id format incorrect
err_cat = ' ' ..
end
if is_set (embargo) and is_set (is_embargoed (embargo)) then
text = table.concat ( -- still embargoed so no external link
{
make_wikilink (link_label_make (handler
handler.separator,
id,
(err_cat and err_cat or '') -- parens required
});
else
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect, -- no embargo date or embargo has expired, ok to link to article
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access
auto_link = not err_cat and 'pmc' or nil -- do not auto-link when PMC has error
}) .. (err_cat and err_cat or ''); -- parentheses required
end
return text;
Line 1,050 ⟶ 1,093:
if id:match("[^%d]") then -- if PMID has anything but digits
err_cat = ' ' ..
else -- PMID is only digits
local id_num = tonumber(id); -- convert id to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if PMID is outside test limit boundaries
err_cat = ' ' ..
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
Line 1,065 ⟶ 1,108:
--[[--------------------------< S 2 C I D >--------------------------------------------------------------------
Format an
S2CIDs are sequential numbers beginning at 1 and counting up. This code checks the
digits and is less than test_limit; the value in local variable test_limit will need to be updated periodically
as more S2CIDs are issued.
Line 1,084 ⟶ 1,127:
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 = ' ' ..
end
else -- when id format incorrect
err_cat = ' ' ..
end
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id:gsub ('%.%a%a', ''), separator = handler.separator, encode = handler.encode, access = access}) .. err_cat;
return text;
Line 1,100 ⟶ 1,143:
--[[--------------------------< S B N >------------------------------------------------------------------------
9-digit form of
]]
local function sbn (id, ignore_invalid)
local handler = cfg.id_handlers['SBN'];
local function return_result (check, err_type) -- local function to handle the various returns
local SBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator});
if not ignore_invalid then -- if not ignoring SBN errors
if not check then
return SBN .. set_message ('err_bad_sbn', {err_type}, false, ' '); -- display an error message
end
else
set_message ('maint_isbn_ignore'); -- add a maint category even when there is no error (ToDo: Possibly switch to separate message for SBNs only)
end
return SBN;
end
if
return return_result (false, cfg.err_msg_supl.char);
end
if 9 ~=
return return_result (false, cfg.err_msg_supl.length);
end
if
return return_result (false, cfg.err_msg_supl.form);
end
return return_result (is_valid_isxn ('0' ..
end
Line 1,128 ⟶ 1,182:
--[[--------------------------< S S R N >----------------------------------------------------------------------
Format an
SSRNs are sequential numbers beginning at 100? and counting up. This code checks the
only digits and is greater than 99 and less than test_limit; the value in local variable test_limit will need
to be updated periodically as more SSRNs are issued.
Line 1,147 ⟶ 1,201:
id_num = tonumber(id_num); -- convert id_num to a number for range testing
if 100 > id_num or handler.id_limit < id_num then -- if SSRN is outside test limit boundaries
err_cat = ' ' ..
end
else -- when id format incorrect
err_cat = ' ' ..
end
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;
return text;
Line 1,170 ⟶ 1,224:
local handler = cfg.id_handlers['USENETID'];
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
if not id:match('^.+@.+$') or not id:match('^[^<].*[^>]$') then -- doesn't have '@' or has one or first or last character is '< or '>'
text = text .. ' ' ..
end
Line 1,196 ⟶ 1,250:
if id:match('^%d%d%d%d%d%d%d%d$') then -- is this identifier using temporary format?
elseif not id:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then -- not temporary, is it normal format?
err_cat = ' ' ..
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end
Line 1,220 ⟶ 1,274:
local function build_id_list( id_list, options )
local new_list, handler = {};
local accept;
local function fallback(k) return { __index = function(t,i) return cfg.id_handlers[k][i] end } end;
for k, v in pairs( id_list ) do
v, accept = has_accept_as_written (v); -- remove and note accept-as-written markup if present
-- k is uppercase identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
-- fallback to read-only cfg
handler = setmetatable( { ['id'] = v, ['access'] = options.IdAccessLevels[k] }, fallback(k) );
Line 1,244 ⟶ 1,302:
table.insert( new_list, {handler.label, citeseerx( v ) } );
elseif k == 'DOI' then
table.insert( new_list, {handler.label, doi( v, options.DoiBroken, handler.access, accept) } );
elseif k == 'EISSN' then
table.insert( new_list, {handler.label, issn( v, true, accept) } );
elseif k == 'HDL' then
table.insert( new_list, {handler.label, hdl( v, handler.access ) } );
elseif k == 'ISBN' then
table.insert( new_list, {handler.label, isbn (v, (accept or options.IgnoreISBN)) } );
elseif k == 'ISMN' then
table.insert( new_list, {handler.label, ismn( v ) } );
elseif k == 'ISSN' then
table.insert( new_list, {handler.label, issn( v, false, accept) } );
elseif k == 'JFM' then
table.insert( new_list, {handler.label, jfm( v ) } );
Line 1,285 ⟶ 1,330:
table.insert( new_list, {handler.label, s2cid( v, handler.access ) } );
elseif k == 'SBN' then
table.insert( new_list, {handler.label, sbn (v, accept) } );
elseif k == 'SSRN' then
table.insert( new_list, {handler.label, ssrn( v ) } );
Line 1,321 ⟶ 1,359:
Populates ID table from arguments using configuration settings. Loops through cfg.id_handlers and searches args for
any of the parameters listed in each cfg.id_handlers['...'].parameters. If found, adds the parameter and value to
the identifier list. Emits redundant error message
]]
Line 1,327 ⟶ 1,365:
local function extract_ids( args )
local id_list = {}; -- list of identifiers found in args
for k, v in pairs( cfg.id_handlers ) do -- k is
v = select_one( args, v.parameters, '
if is_set(v) then id_list[k] = v; end -- if found in args, add identifier to our list
end
Line 1,353 ⟶ 1,391:
if is_set (access_level) then
if not in_array (access_level, cfg.keywords_lists['id-access']) then -- exact match required
table.insert( z.message_tail, {
access_level = nil; -- invalid so unset
end
if not is_set(id_list[k]) then --
table.insert( z.message_tail, {
end
id_accesses_list[k] = cfg.keywords_xlate[access_level]; -- get translated keyword
Line 1,376 ⟶ 1,414:
cfg = cfg_table_ptr;
is_set = utilities_page_ptr.is_set;
in_array = utilities_page_ptr.in_array;
select_one = utilities_page_ptr.select_one;
substitute = utilities_page_ptr.substitute;
make_wikilink = utilities_page_ptr.make_wikilink;
Line 1,392 ⟶ 1,430:
return {
auto_link_urls = auto_link_urls, -- table of identifier URLs to be used when auto-linking |title=
build_id_list = build_id_list,
extract_ids = extract_ids,
|