Module:Citation/CS1/Identifiers: Difference between revisions
Content deleted Content added
sync from sandbox; |
Johnrdorazio (talk | contribs) m 1 revision imported |
||
(4 intermediate revisions by 2 users not shown) | |||
Line 1:
--[[--------------------------< F O R W A R D D E C L A R A T I O N S >--------------------------------------
]]
Line 180 ⟶ 179:
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
around that call this function with date parts and create a YYYY-MM-DD format
]=]
local function is_valid_biorxiv_date (
local biorxiv_date = table.concat ({y, m, d}, '-'); -- make ymd date
local good1, good2;
local biorxiv_ts, tomorrow_ts; -- to hold Unix timestamps representing the dates
Line 250:
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------
LCCN normalization (
1. Remove all blanks.
2. If there is a forward slash (/) in the string, remove it, and remove all characters to the right of the forward slash.
Line 260:
Returns a normalized LCCN for lccn() to validate. There is no error checking (step 3.b.1) performed in this function.
]]
Line 286 ⟶ 287:
--[[--------------------------< A R X I V >--------------------------------------------------------------------
See:
format and error check arXiv identifier. There are three valid forms of the identifier:
Line 311 ⟶ 312:
<date code> and <version> are as defined for 0704-1412
<number> is a five-digit number
]]
Line 318 ⟶ 320:
local handler = options.handler;
local year, month, version;
local
local text; -- output text
Line 327 ⟶ 329:
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?
end
Line 336 ⟶ 338:
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 -- when year is 07, is month invalid (before April)?
end
Line 344 ⟶ 346:
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)
end
else
end
if
options.coins_list_t['ARXIV'] = nil; -- when error, unset so not included in COinS
end
set_message ('err_bad_arxiv');
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})
if is_set (class) then
Line 364 ⟶ 369:
text = table.concat ({text, ' [[//arxiv.org/archive/', class, ' ', class, ']]'}); -- external link within square brackets, not wikilink
else
end
end
return text;
end
Line 376 ⟶ 381:
Validates (sort of) and formats a bibcode ID.
Format for bibcodes is specified here:
But, this: 2015arXiv151206696F is apparently valid so apparently, the only things that really matter are length, 19 characters
Line 395 ⟶ 400:
local handler = options.handler;
local err_type;
local year;
Line 420 ⟶ 426:
if is_set (err_type) then -- if there was an error detected
options.coins_list_t['BIBCODE'] = nil; -- when error, unset so not included in COinS
end
return text;
end
Line 446 ⟶ 453:
local id = options.id;
local handler = options.handler;
local
local patterns = {
Line 459 ⟶ 466:
if m then -- m is nil when id is the six-digit form
if not is_valid_biorxiv_date (y
break; -- date fail; break out early so we don't unset the error message
end
end
break; -- and done
end
end -- err_cat remains set here when no match
if
options.coins_list_t['BIORXIV'] = nil; -- when error, unset so not included in COinS
set_message ('err_bad_biorxiv'); -- and set the error message
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator,
encode = handler.encode, access = handler.access}
end
Line 483 ⟶ 491:
The description of the structure of this identifier can be found at Help_talk:Citation_Style_1/Archive_26#CiteSeerX_id_structure
]]
Line 489 ⟶ 498:
local handler = options.handler;
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,
Line 496 ⟶ 505:
matched = id:match ("^10%.1%.1%.[1-9]%d?%d?%d?%.[1-9]%d?%d?%d?$");
if not matched then
options.coins_list_t['CITESEERX'] = nil; -- when error, unset so not included in COinS
end
return text;
end
Line 517 ⟶ 527:
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in DOI names.
https://www.doi.org/doi_handbook/2_Numbering.html -- 2.2 Syntax of a DOI name
https://www.doi.org/doi_handbook/2_Numbering.html#2.2.2 -- 2.2.2 DOI prefix
]]
Line 526 ⟶ 539:
local ignore_invalid = options.accept;
local handler = options.handler;
local
local text;
Line 555 ⟶ 568:
end
local registrant =
local registrant_err_patterns = { -- these patterns are for code ranges that are not supported
'^[^1-3]%d%d%d%d%.%d%d*$', -- 5 digits with subcode (0xxxx, 40000+); accepts: 10000–39999
Line 563 ⟶ 576:
'^[^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 (3 digits with subcode is legitimate)
'^%d%d?%.[%d%.]+', -- 1 or 2 digits with subcode
'^5555$', -- test registrant will never resolve
'[^%d%.]', -- any character that isn't a digit or a dot
Line 572 ⟶ 586:
for i, pattern in ipairs (registrant_err_patterns) do -- spin through error patterns
if registrant:match (pattern) then -- to validate registrant codes
break; -- and done
end
end
else
end
else
Line 583 ⟶ 597:
end
if
options.coins_list_t['DOI'] = nil; -- when error, unset so not included in COinS
end
Line 589 ⟶ 603:
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access,
auto_link = not (
}) .. (inactive or '');
return text;▼
end
Line 611 ⟶ 625:
if ever used in HDLs.
Query string parameters are named here:
but since '?' is an allowed character in an HDL, '?' followed by one of the query parameters is the only way we
have to detect the query string so that it isn't URL-encoded with the rest of the identifier.
Line 621 ⟶ 635:
local access = options.access;
local handler = options.handler;
local query_params = { -- list of known query parameters from
'noredirect',
'ignore_aliases',
Line 655 ⟶ 669:
if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then -- HDL must contain a forward slash, must not contain spaces, endashes, and must not end with period or comma
options.coins_list_t['HDL'] = nil; -- when error, unset so not included in COinS
end
return text;
end
Line 681 ⟶ 696:
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; -- return id text
end
end
return ISBN; -- return id text
end
Line 743 ⟶ 759:
local domain = options.ASINTLD;
local
if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
else
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
if is_valid_isxn (id, 10) then -- see if ASIN value is or validates as ISBN-10
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
end
elseif not is_set (
end
elseif not id:match("^%u[%d%u]+$") then
end
end
Line 769 ⟶ 785:
domain = "com." .. domain;
elseif not in_array (domain, {'ae', 'ca', 'cn', 'de', 'es', 'fr', 'in', 'it', 'nl', 'pl', 'sa', 'se', 'co.jp', 'co.uk', 'com', 'com.au', 'com.br', 'com.mx', 'com.sg', 'com.tr'}) then -- Arabic Emirates, Canada, China, Germany, Spain, France, Indonesia, Italy, Netherlands, Poland, Saudi Arabia, Sweden (as of 2021-03 Austria (.at), Liechtenstein (.li) and Switzerland (.ch) still redirect to the German site (.de) with special settings, so don't maintain local ASINs for them)
end
local handler = options.handler;
if not is_set (
options.coins_list_t['ASIN'] = handler.prefix .. domain .. "/dp/" .. id; --
else
options.coins_list_t['ASIN'] = nil; -- when error, unset so not included in COinS
Line 781 ⟶ 797:
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix .. domain .. "/dp/",
id = id, encode = handler.encode, separator = handler.separator})
end
Line 788 ⟶ 804:
Determines whether an ISMN string is valid. Similar to ISBN-13, ISMN is 13 digits beginning 979-0-... and uses the
same check digit calculations. See
section 2, pages 9–12.
Line 824 ⟶ 840:
if false == valid_ismn then
options.coins_list_t['ISMN'] = nil; -- when error, unset so not included in COinS; not really necessary here because ismn not made part of COinS
end
Line 837 ⟶ 853:
like this:
|issn=0819 4327 gives: [
This code now prevents that by inserting a hyphen at the ISSN midpoint. It also validates the ISSN for length
Line 877 ⟶ 893:
if false == valid_issn then
options.coins_list_t['ISSN'] = nil; -- when error, unset so not included in COinS
end
end
return text;
end
Line 895 ⟶ 911:
local handler = options.handler;
local id_num;
▲ local err_cat = '';
id_num = id:match ('^[Jj][Ff][Mm](.*)$'); -- identifier with jfm prefix; extract identifier
Line 908 ⟶ 923:
id = id_num; -- jfm matches pattern
else
options.coins_list_t['JFM'] = nil; -- when error, unset so not included in COinS
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
end
Line 927 ⟶ 942:
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
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})
end
Line 943 ⟶ 957:
Format LCCN link and do simple error checking. LCCN is a character string 8-12 characters long. The length of
the LCCN dictates the character type of the first 1-3 characters; the rightmost eight are always digits.
https://oclc-research.github.io/infoURI-Frozen/info-uri.info/info:lccn/reg.html
length = 8 then all digits
Line 956 ⟶ 970:
local lccn = options.id;
local handler = options.handler;
local
local id = lccn; -- local copy of the LCCN
Line 964 ⟶ 978:
if 8 == len then
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits)
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?
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
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
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
end
else
end
if not is_set (
end
if is_set (
options.coins_list_t['LCCN'] = nil; -- when error, unset so not included in COinS
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = lccn, separator = handler.separator, encode = handler.encode})
end
Line 1,012 ⟶ 1,026:
local id_num;
local id_len;
▲ local err_cat = '';
id_num = id:match ('^[Mm][Rr](%d+)$'); -- identifier with mr prefix
Line 1,026 ⟶ 1,039:
id = string.rep ('0', 7-id_len) .. id_num; -- zero-fill leading digits
else
options.coins_list_t['MR'] = nil; -- when error, unset so not included in COinS
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
end
Line 1,046 ⟶ 1,059:
local handler = options.handler;
local number;
if id:match('^ocm%d%d%d%d%d%d%d%d$') then -- ocm prefix and 8 digits; 001 field (12 characters)
number = id:match('ocm(%d+)'); -- get the number
Line 1,069 ⟶ 1,081:
id = number; -- exclude prefix, if any, from external link
else
options.coins_list_t['OCLC'] = nil; -- when error, unset so not included in COinS
end
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
▲ return text;
end
Line 1,091 ⟶ 1,101:
local handler = options.handler;
local ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- strip optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W';
local
local prefix = { -- these are appended to the handler.prefix according to code
['A']='authors/OL',
Line 1,102 ⟶ 1,112:
code = 'X'; -- no code or id completely invalid
ident = id; -- copy id to ident so that we display the flawed identifier
end
if not is_set (
options.coins_list_t['OL'] = handler.prefix .. prefix[code] .. ident; -- experiment for ol coins
else
Line 1,114 ⟶ 1,124:
prefix = handler.prefix .. prefix[code],
id = ident, separator = handler.separator, encode = handler.encode,
access = access})
end
Line 1,132 ⟶ 1,142:
local access = options.access;
local handler = options.handler;
if id:match("[^%d]") then -- if OSTI has anything but digits
options.coins_list_t['OSTI'] = nil; -- when error, unset so not included in COinS
else -- OSTI is only digits
local id_num = tonumber (id); -- convert id to a number for range testing
if 1018 > id_num or handler.id_limit < id_num then -- if OSTI is outside test limit boundaries
options.coins_list_t['OSTI'] = nil; -- when error, unset so not included in COinS
end
Line 1,146 ⟶ 1,155:
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})
end
Line 1,171 ⟶ 1,180:
local embargo = options.Embargo; -- TODO: lowercase?
local handler = options.handler;
local
local id_num;
local text;
Line 1,186 ⟶ 1,195:
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
else
id = tostring (id_num); -- make sure id is a string
end
else -- when id format incorrect
end
Line 1,200 ⟶ 1,209:
handler.separator,
id,
});
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
});
end
if
options.coins_list_t['PMC'] = nil; -- when error, unset so not included in COinS
end
Line 1,228 ⟶ 1,236:
local id = options.id;
local handler = options.handler;
if id:match("[^%d]") then -- if PMID has anything but digits
options.coins_list_t['PMID'] = nil; -- when error, unset so not included in COinS
else -- PMID is only digits
local id_num = tonumber (id); -- convert id to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if PMID is outside test limit boundaries
options.coins_list_t['PMID'] = nil; -- when error, unset so not included in COinS
end
Line 1,242 ⟶ 1,249:
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})
end
Line 1,259 ⟶ 1,266:
local id = options.id;
local handler = options.handler;
if id:match("[^%d]") then -- if RFC has anything but digits
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
options.coins_list_t['RFC'] = nil; -- when error, unset so not included in COinS
end
Line 1,273 ⟶ 1,279:
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})
end
Line 1,291 ⟶ 1,297:
local access = options.access;
local handler = options.handler;
local id_num;
local text;
Line 1,300 ⟶ 1,305:
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
options.coins_list_t['S2CID'] = nil; -- when error, unset so not included in COinS
end
else -- when id format incorrect
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})
return text;
Line 1,329 ⟶ 1,333:
local ignore_invalid = options.accept;
local handler = options.handler;
local function return_result (check, err_type)
local SBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator});
if not ignore_invalid then
if not 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;
end
else
Line 1,374 ⟶ 1,379:
local id = options.id;
local handler = options.handler;
local id_num;
local text;
Line 1,383 ⟶ 1,387:
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
options.coins_list_t['SSRN'] = nil; -- when error, unset so not included in COinS
end
else -- when id format incorrect
options.coins_list_t['SSRN'] = 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 =
return text;
Line 1,413 ⟶ 1,417:
if not id:match('^.+@.+$') or not id:match('^[^<].*[^>]$') then -- doesn't have '@' or has one or first or last character is '< or '>'
options.coins_list_t['USENETID'] = nil; -- when error, unset so not included in COinS
end
return text;
end
Line 1,434 ⟶ 1,438:
local id = options.id;
local handler = options.handler;
if id:match('^%d%d%d%d%d%d%d%d$') then -- is this identifier using temporary format?
set_message ('maint_zbl'); -- yes, add maint cat
elseif not id:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then -- not temporary, is it normal format?
options.coins_list_t['ZBL'] = 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})
end
Line 1,487 ⟶ 1,490:
if is_set (access_level) then
if not in_array (access_level, cfg.keywords_lists['id-access']) then -- exact match required
access_level = nil; -- invalid so unset
end
if not is_set (id_list[k]) then -- identifier access-level must have a matching identifier
end
id_accesses_list[k] = cfg.keywords_xlate[access_level]; -- get translated keyword
Line 1,555 ⟶ 1,558:
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
options_t.coins_list_t[hkey] = v; -- id value without accept-as-written markup for metadata
if options_t.handler.access and not in_array (options_t.handler.access, cfg.keywords_lists['id-access']) then
error (cfg.messages['unknown_ID_access'] .. options_t.handler.access); -- here when handler access key set to a value not listed in list of allowed id access keywords
end
if func_map[hkey] then
table.insert (ID_list_t, {hkey, id_text}); -- add identifier text to the output sequence table
else
error (cfg.messages['unknown_ID_key']
end
end
Line 1,594 ⟶ 1,603:
for _, v in ipairs (ID_support_t) do
if is_set (v[1]) and not ID_list_coins_t[v[2]] then -- when support parameter has a value but matching identifier parameter is missing or empty
end
end
Line 1,608 ⟶ 1,617:
]]
local function identifier_lists_get (
local ID_list_coins_t = extract_ids (
options_check (ID_list_coins_t, ID_support_t); -- ID support parameters must have matching identifier parameters
local ID_access_levels_t = extract_id_access_levels (
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
Line 1,645 ⟶ 1,654:
auto_link_urls = auto_link_urls, -- table of identifier URLs to be used when auto-linking |title=
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;
set_selected_modules = set_selected_modules;
|