「モジュール:Citation/CS1/Identifiers」の版間の差分

提供:脳科学辞典
ナビゲーションに移動 検索に移動
(1版)
 
bsd>Trappist the monk
編集の要約なし
1行目: 1行目:
local identifiers = {};


--[[--------------------------< F O R W A R D  D E C L A R A T I O N S >--------------------------------------
--[[--------------------------< F O R W A R D  D E C L A R A T I O N S >--------------------------------------
]]
]]


local is_set, in_array, set_error, select_one, add_maint_cat, 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


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
--[[--------------------------< 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
 
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 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 = cfg.this_wiki_code; -- Wikipedia subdomain; 'en' for en.wikipedia.org
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 );
end
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
 
ext_link = mw.ustring.format ('[%s%s%s %s]', options.prefix, url_string, options.suffix or "", mw.text.nowiki(options.id));
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 is_set(options.access) then
ext_link = substitute (cfg.presentation['access-signal'], {ext_link, cfg.presentation[options.access]}); -- add the free-to-read / paywall lock
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
 
return table.concat ({
return table.concat ({
make_wikilink (options.link, options.label),
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
44行目: 115行目:
--[[--------------------------< 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


]]
]]


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


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,
options.id,
id, -- translated to Western digits
options.suffix or ''
options.suffix or ''
}),
}),
mw.text.nowiki (options.id)
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?
});
});
end
end
69行目: 141行目:
--[[--------------------------< 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 today's date.  If embargo date is
in the future, returns the content of |embargo=; otherwise, returns and empty string because the embargo has expired or because
in the future, returns the content of |pmc-embargo-date=; otherwise, returns an empty string because the embargo has expired or because
|embargo= was not set in this cite.
|pmc-embargo-date= was not set in this cite.


]]
]]
78行目: 150行目:
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


97行目: 208行目:
--[[--------------------------< 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.


]]
]]
107行目: 218行目:
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


122行目: 233行目:
--[[--------------------------< 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.


]]
]]
135行目: 246行目:
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


141行目: 252行目:
--[[--------------------------< 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.
150行目: 261行目:
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 >>====================================
179行目: 291行目:


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:
<archive> is a string of alpha characters - may be hyphenated; no other punctuation
<archive> is a string of alpha characters - may be hyphenated; no other punctuation
<class> is a string of alpha characters - may be hyphenated; no other punctuation
<class> is a string of alpha characters - may be hyphenated; no other punctuation; not the same as |class= parameter which is not supported in this form
<date code> is four digits in the form YYMM where YY is the last two digits of the four-digit year and MM is the month number January = 01
<date code> is four digits in the form YYMM where YY is the last two digits of the four-digit year and MM is the month number January = 01
first digit of YY for this form can only 9 and 0
first digit of YY for this form can only 9 and 0
206行目: 318行目:
local handler = cfg.id_handlers['ARXIV'];
local handler = cfg.id_handlers['ARXIV'];
local year, month, version;
local year, month, version;
local err_cat = '';
local err_cat = false; -- assume no error message
local 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);
215行目: 327行目:
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?
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- set error message
err_cat = true; -- flag for error message
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 = ' ' .. set_error( 'bad_arxiv' ); -- set 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 = ' ' .. set_error( 'bad_arxiv' ); -- set error message
err_cat = true; -- flag for error message
end
end
else
else
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- arXiv id doesn't match any format
err_cat = true; -- not a recognized format; flag for error message
end
end


text = external_link_id({link = handler.link, label = handler.label,
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
class = ' [[' .. '//arxiv.org/archive/' .. class .. ' ' .. class .. ']]'; -- external link within square brackets, not wikilink
if id:match ('^%d+') then
else
text = table.concat ({text, ' [[//arxiv.org/archive/', class, ' ', class, ']]'}); -- external link within square brackets, not wikilink
class = ''; -- empty string for concatenation
else
text = table.concat ({text, ' ', set_message ('err_class_ignored')});
end
end
end
 
return text .. class;
return text;
end
end


251行目: 370行目:
--[[--------------------------< 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
261行目: 380行目:
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
273行目: 391行目:
local year;
local year;


local text = external_link_id({link=handler.link, label=handler.label,
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
296行目: 414行目:


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});
end
end
return text;
return text;
304行目: 422行目:
--[[--------------------------< 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


]]
]]
312行目: 437行目:
local function biorxiv(id)
local function biorxiv(id)
local handler = cfg.id_handlers['BIORXIV'];
local handler = cfg.id_handlers['BIORXIV'];
local err_cat = ''; -- presume that bioRxiv id is valid
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)
end
'^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)
}
return external_link_id({link = handler.link, label = handler.label,
for _, pattern in ipairs (patterns) do -- spin through the patterns looking for a match
prefix=handler.prefix,id=id,separator=handler.separator,
if id:match (pattern) then
encode=handler.encode, access=handler.access}) .. err_cat;
local y, m, d = id:match (pattern); -- found a match, attempt to get year, month and date from the identifier
 
if m then -- m is nil when id is the six-digit form
if not is_valid_biorxiv_date (y .. '-' .. m .. '-' .. d) then -- validate the encoded date; TODO: don't ignore leap-year and actual month lengths ({{#time:}} is a poor date validator)
break; -- date fail; break out early so we don't unset the error message
end
end
err_cat = nil; -- we found a match so unset the error message
break; -- and done
end
end -- err_cat remains set here when no match
 
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator,
encode = handler.encode, access = handler.access}) .. (err_cat and (' ' .. set_message( 'err_bad_biorxiv')) or '');
end
end


335行目: 476行目:
local matched;
local matched;
local text = external_link_id({link=handler.link, label=handler.label,
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' );
end
end
return text;
return text;
355行目: 496行目:
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 (id, inactive, access, ignore_invalid)
local cat = ""
local err_cat;
local handler = cfg.id_handlers['DOI'];
local handler = cfg.id_handlers['DOI'];
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 since " .. 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
else
table.insert( z.error_categories, "Pages with inactive DOIs" ); -- when inactive doesn't contain a recognizable year
inactive_year = nil; -- |doi-broken-date= has something but it isn't a date
end
end
inactive = " (" .. cfg.messages['inactive'] .. " " .. inactive .. ")"
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
end
text = external_link_id({link = handler.link, label = handler.label,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, access=access}) .. (inactive or '')


if nil == id:match("^10%.[^%s–]-/[^%s–]-[^%.,]$") then -- doi must begin with '10.', must contain a fwd slash, must not contain spaces or endashes, and must not end with period or comma
local registrant = id:match ('^10%.([^/]+)/[^%s–]-[^%.,]$'); -- registrant set when DOI has the proper basic form
cat = ' ' .. set_error( 'bad_doi' );
local registrant_err_patterns = { -- these patterns are for code ranges that are not supported
'^[^1-3]%d%d%d%d%.%d%d*$', -- 5 digits with subcode (0xxxx, 40000+); accepts: 10000–39999
'^[^1-4]%d%d%d%d$', -- 5 digits without subcode (0xxxx, 40000+); accepts: 10000–49999
'^[^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
'%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
end


return text .. 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 = 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
end


403行目: 587行目:
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.


]]
]]
408行目: 596行目:
local function hdl(id, access)
local function hdl(id, access)
local handler = cfg.id_handlers['HDL'];
local handler = cfg.id_handlers['HDL'];
local query_params = { -- list of known query parameters from http://www.handle.net/proxy_servlet.html
'noredirect',
'ignore_aliases',
'auth',
'cert',
'index',
'type',
'urlappend',
'locatt',
'action',
}
local text = external_link_id({link = handler.link, label = handler.label,
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 hdl then -- when there are query strings, this is the handle identifier portion
for _, q in ipairs (query_params) do -- spin through the list of query parameters
if param:match ('^' .. q) then -- if the query string begins with one of the parameters
found = true; -- announce a find
break; -- and stop looking
end
end
end
 
if found then
id = hdl; -- found so replace id with the handle portion; this will be URL-encoded, suffix will not
else
suffix = ''; -- make sure suffix is empty string for concatenation else
end
 
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, suffix = suffix, separator = handler.separator, encode = handler.encode, access = access})


if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then -- hdl must contain a fwd slash, must not contain spaces, endashes, and must not end with period or comma
if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then -- HDL must contain a forward slash, must not contain spaces, endashes, and must not end with period or comma
text = text .. ' ' .. set_error( 'bad_hdl' );
text = text .. ' ' .. set_message( 'err_bad_hdl' );
end
end
return text;
return text;
425行目: 642行目:
]]
]]


local function isbn( isbn_str )
local function isbn (isbn_str, ignore_invalid)
if nil ~= isbn_str:match("[^%s-0-9X]") then
local handler = cfg.id_handlers['ISBN'];
return false, 'invalid character'; -- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
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
 
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
end
isbn_str = isbn_str:gsub( "-", "" ):gsub( " ", "" ); -- remove hyphens and spaces
 
local len = isbn_str:len();
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
end
return is_valid_isxn(isbn_str, 10), 'checksum';
return return_result (is_valid_isxn(id, 10), cfg.err_msg_supl.check);
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
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
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=.
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.
Error message if not 10 characters, if not ISBN-10, if mixed and first character is a digit.
 
|asin=630....... is (apparently) not a legitimate ISBN though it checksums as one; these do not cause this
function to emit the maint_asin message


This function is positioned here because it calls isbn()
This function is positioned here because it calls isbn()
464行目: 703行目:
]]
]]


local function asin(id, domain)
local function asin (id, domain)
local err_cat = ""
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 isbn (id) then -- see if ASIN value is or validates as ISBN-10
add_maint_cat ('ASIN');
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
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) then  
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, {'au', 'br', 'mx'}) then -- Australia, Brazil, Mexico
domain = "com." .. domain;
domain = "com." .. domain;
end
end
local handler = cfg.id_handlers['ASIN'];
local handler = cfg.id_handlers['ASIN'];
return external_link_id({link=handler.link,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
label=handler.label, prefix=handler.prefix .. domain .. "/dp/",
prefix = handler.prefix .. domain .. "/dp/",
id=id, encode=handler.encode, separator = handler.separator}) .. err_cat;
id = id, encode = handler.encode, separator = handler.separator}) .. err_cat;
end
end


496行目: 737行目:
--[[--------------------------< 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.
509行目: 750行目:


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
text = text .. ' ' .. set_message( 'err_bad_ismn' ) -- add an error message if the ISMN is invalid
end  
end  
537行目: 778行目:
--[[--------------------------< 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:
543行目: 784行目:
|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 (id, e, ignore_invalid)
local issn_copy = id; -- save a copy of unadulterated issn; use this version for display if issn does not validate
local issn_copy = id; -- save a copy of unadulterated ISSN; use this version for display if ISSN does not validate
local handler;
local handler;
local text;
local text;
local valid_issn = true;
local valid_issn = true;
if e then
handler = cfg.id_handlers[e and 'EISSN' or 'ISSN'];
handler = cfg.id_handlers['EISSN'];
else
handler = cfg.id_handlers['ISSN'];
end


id=id:gsub( "[%s-]", "" ); -- strip spaces, hyphens, and endashes from the issn
id = id:gsub ('[%s-]', ''); -- remove hyphens and whitespace


if 8 ~= id:len() or nil == id:match( "^%d*X?$" ) then -- validate the issn: 8 digits long, containing only 0-9 or X in the last position
if 8 ~= id:len() or nil == id:match( "^%d*X?$" ) 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
valid_issn = false; -- wrong length or improper character
else
else
valid_issn=is_valid_isxn(id, 8); -- validate issn
valid_issn = is_valid_isxn(id, 8); -- validate ISSN
end
end


573行目: 810行目:
id = string.sub( id, 1, 4 ) .. "-" .. string.sub( id, 5 ); -- if valid, display correctly formatted version
id = string.sub( id, 1, 4 ) .. "-" .. string.sub( id, 5 ); -- if valid, display correctly formatted version
else
else
id = issn_copy; -- if not valid, use the show the invalid issn with error message
id = issn_copy; -- if not valid, show the invalid ISSN with error message
end
 
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
 
if ignore_invalid then
set_message ('maint_issn_ignore');
else
if false == valid_issn then
text = text .. ' ' .. set_message ('err_bad_issn', e and 'e' or ''); -- add an error message if the ISSN is invalid
end
end
end
text = external_link_id({link = handler.link, label = handler.label,
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
601行目: 842行目:


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 mr prefix
id_num = id; -- if here id does not have prefix
id_num = id; -- if here id does not have prefix
609行目: 850行目:
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
end
end
return external_link_id({link = handler.link, label = handler.label,
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


624行目: 865行目:


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


]]
]]
634行目: 875行目:
local handler = cfg.id_handlers['LCCN'];
local handler = cfg.id_handlers['LCCN'];
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
end


return external_link_id({link = handler.link, label = handler.label,
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


689行目: 930行目:


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
698行目: 939行目:
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
end
end
return external_link_id({link = handler.link, label = handler.label,
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


708行目: 949行目:
--[[--------------------------< O C L C >----------------------------------------------------------------------
--[[--------------------------< O C L C >----------------------------------------------------------------------


Validate and format an oclc id. https://www.oclc.org/batchload/controlnumber.en.html
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


]]
]]
726行目: 968行目:
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
738行目: 980行目:
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
end
end
local text = external_link_id({link=handler.link, label=handler.label,
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;
755行目: 997行目:


local function openlibrary(id, access)
local function openlibrary(id, access)
local code;
local handler = cfg.id_handlers['OL'];
local handler = cfg.id_handlers['OL'];
local ident;
local ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- strip optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W';
local error_msg = '';
ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W'; remove OL prefix
local 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 is_set (ident) then -- if malformed return an error
if not ident then
return external_link_id({link=handler.link, label=handler.label,
code = 'X'; -- no code or id completely invalid
prefix=handler.prefix .. 'OL',
ident = id; -- copy id to ident so that we display the flawed identifier
id=id, separator=handler.separator, encode = handler.encode,
error_msg = ' ' .. set_message ('err_bad_ol');
access = access}) .. ' ' .. set_error( 'bad_ol' );
end
id = ident; -- use ident without the optional OL prefix (it has been removed)
if ( code == "A" ) then
return external_link_id({link=handler.link, label=handler.label,
prefix=handler.prefix .. 'authors/OL',
id=id, separator=handler.separator, encode = handler.encode,
access = access})
end
if ( code == "M" ) then
return external_link_id({link=handler.link, label=handler.label,
prefix=handler.prefix .. 'books/OL',
id=id, separator=handler.separator, encode = handler.encode,
access = access})
end
end


if ( code == "W" ) 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,
prefix = handler.prefix .. prefix[code],
prefix=handler.prefix .. 'works/OL',
id = ident, separator = handler.separator, encode = handler.encode,
id=id, separator=handler.separator, encode = handler.encode,
access = access}) .. error_msg;
access = access})
end
end
end


810行目: 1,037行目:
]]
]]


local function pmc(id, embargo)
local function pmc (id, embargo)
local test_limit = 6000000; -- update this value as PMCs approach
local handler = cfg.id_handlers['PMC'];
local handler = cfg.id_handlers['PMC'];
local err_cat = ''; -- presume that PMC is valid
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
827行目: 1,053行目:
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, -- 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
return text;
return text;
861行目: 1,089行目:


local function pmid(id)
local function pmid(id)
local test_limit = 32000000; -- update this value as PMIDs approach
local handler = cfg.id_handlers['PMID'];
local handler = cfg.id_handlers['PMID'];
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
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
end
end
end
end
return external_link_id({link = handler.link, label = handler.label,
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
 
 
--[[--------------------------< 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 (id, access)
local handler = cfg.id_handlers['S2CID'];
local err_cat = ''; -- presume that S2CID is valid
local id_num;
local text;
id_num = id:match ('^[1-9]%d*$'); -- id must be all digits; must not begin with 0; no open access flag
 
if is_set (id_num) then -- id_num has a value so test it
id_num = tonumber(id_num); -- convert id_num to a number for range testing
if handler.id_limit < id_num then -- if S2CID is outside test limit boundaries
err_cat = ' ' .. set_message ('bad_s2cid_err'); -- set an error message
end
 
else -- when id format incorrect
err_cat = ' ' .. set_message ('bad_s2cid_err'); -- set an error message
end
 
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id:gsub ('%.%a%a', ''), separator = handler.separator, encode = handler.encode, access = access}) .. err_cat;
 
return text;
end
 
 
--[[--------------------------< S B N >------------------------------------------------------------------------
 
9-digit form of ISBN-10; uses same check-digit validation when SBN is prefixed with an additional '0' to make 10 digits
 
]]
 
local function sbn (id, ignore_invalid)
local handler = cfg.id_handlers['SBN'];
local function return_result (check, err_type) -- local function to handle the various returns
local SBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator});
if not ignore_invalid then -- if not ignoring SBN errors
if not check then
return SBN .. set_message ('err_bad_sbn', {err_type}, false, ' '); -- display an error message
end
else
set_message ('maint_isbn_ignore'); -- add a maint category even when there is no error (ToDo: Possibly switch to separate message for SBNs only)
end
return SBN;
end
 
if id:match ('[^%s-0-9X]') then
return return_result (false, cfg.err_msg_supl.char); -- fail if SBN contains anything but digits, hyphens, or the uppercase X
end
 
local ident = id:gsub ('[%s-]', ''); -- remove hyphens and whitespace; they interfere with the rest of the tests
 
if  9 ~= ident:len() then
return return_result (false, cfg.err_msg_supl.length); -- fail if incorrect length
end
 
if ident:match ('^%d*X?$') == nil then
return return_result (false, cfg.err_msg_supl.form); -- fail if SBN has 'X' anywhere but last position
end
 
return return_result (is_valid_isxn ('0' .. ident, 10), cfg.err_msg_supl.check);
end
end


881行目: 1,182行目:
--[[--------------------------< 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.
890行目: 1,191行目:


local function ssrn (id)
local function ssrn (id)
local test_limit = 3500000; -- update this value as SSRNs approach
local handler = cfg.id_handlers['SSRN'];
local handler = cfg.id_handlers['SSRN'];
local err_cat = ''; -- presume that SSRN is valid
local err_cat = ''; -- presume that SSRN is valid
900行目: 1,200行目:
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
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
end
end
text = external_link_id({link = handler.link, label = handler.label,
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;
924行目: 1,224行目:
local handler = cfg.id_handlers['USENETID'];
local handler = cfg.id_handlers['USENETID'];


local text = external_link_id({link = handler.link, label = handler.label,
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
end  
end  
938行目: 1,238行目:


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


]]
]]
943行目: 1,247行目:
local function zbl (id)
local function zbl (id)
local handler = cfg.id_handlers['ZBL'];
local handler = cfg.id_handlers['ZBL'];
local id_num;
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
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,
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


979行目: 1,274行目:
local function build_id_list( id_list, options )
local function build_id_list( id_list, options )
local new_list, handler = {};
local new_list, handler = {};
local accept;


local function fallback(k) return { __index = function(t,i) return cfg.id_handlers[k][i] end } end;
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
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
-- fallback to read-only cfg
handler = setmetatable( { ['id'] = v, ['access'] = options.IdAccessLevels[k] }, fallback(k) );
handler = setmetatable( { ['id'] = v, ['access'] = options.IdAccessLevels[k] }, fallback(k) );
1,003行目: 1,302行目:
table.insert( new_list, {handler.label, citeseerx( v ) } );
table.insert( new_list, {handler.label, citeseerx( v ) } );
elseif k == 'DOI' then
elseif k == 'DOI' then
table.insert( new_list, {handler.label, doi( v, options.DoiBroken, handler.access ) } );
table.insert( new_list, {handler.label, doi( v, options.DoiBroken, handler.access, accept) } );
elseif k == 'EISSN' then
elseif k == 'EISSN' then
table.insert( new_list, {handler.label, issn( v, true ) } ); -- true distinguishes eissn from issn
table.insert( new_list, {handler.label, issn( v, true, accept) } ); -- true distinguishes EISSN from ISSN
elseif k == 'HDL' then
elseif k == 'HDL' then
table.insert( new_list, {handler.label, hdl( v, handler.access ) } );
table.insert( new_list, {handler.label, hdl( v, handler.access ) } );
elseif k == 'ISBN' then
elseif k == 'ISBN' then
local ISBN = internal_link_id( handler );
table.insert( new_list, {handler.label, isbn (v, (accept or options.IgnoreISBN)) } );
local check;
local err_type = '';
check, err_type = isbn( v );
if not check then
if is_set(options.IgnoreISBN) then -- ISBN is invalid; if |ignore-isbn-error= set
add_maint_cat ('ignore_isbn_err'); -- ad a maint category
else
ISBN = ISBN .. set_error( 'bad_isbn', {err_type}, false, " ", "" ); -- else display an error message
end
elseif is_set(options.IgnoreISBN) then -- ISBN is OK; if |ignore-isbn-error= set
add_maint_cat ('ignore_isbn_err'); -- because |ignore-isbn-error= unnecessary
end
table.insert( new_list, {handler.label, ISBN } );
elseif k == 'ISMN' then
elseif k == 'ISMN' then
table.insert( new_list, {handler.label, ismn( v ) } );
table.insert( new_list, {handler.label, ismn( v ) } );
elseif k == 'ISSN' then
elseif k == 'ISSN' then
table.insert( new_list, {handler.label, issn( v ) } );
table.insert( new_list, {handler.label, issn( v, false, accept) } );
elseif k == 'JFM' then
elseif k == 'JFM' then
table.insert( new_list, {handler.label, jfm( v ) } );
table.insert( new_list, {handler.label, jfm( v ) } );
1,041行目: 1,327行目:
elseif k == 'PMID' then
elseif k == 'PMID' then
table.insert( new_list, {handler.label, pmid( v ) } );
table.insert( new_list, {handler.label, pmid( v ) } );
elseif k == 'S2CID' then
table.insert( new_list, {handler.label, s2cid( v, handler.access ) } );
elseif k == 'SBN' then
table.insert( new_list, {handler.label, sbn (v, accept) } );
elseif k == 'SSRN' then
elseif k == 'SSRN' then
table.insert( new_list, {handler.label, ssrn( v ) } );
table.insert( new_list, {handler.label, ssrn( v ) } );
1,069行目: 1,359行目:
Populates ID table from arguments using configuration settings. Loops through cfg.id_handlers and searches args for
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
any of the parameters listed in each cfg.id_handlers['...'].parameters.  If found, adds the parameter and value to
the identifier list.  Emits redundant error message is more than one alias exists in args
the identifier list.  Emits redundant error message if more than one alias exists in args


]]
]]
1,075行目: 1,365行目:
local function extract_ids( args )
local function extract_ids( args )
local id_list = {}; -- list of identifiers found in args
local id_list = {}; -- list of identifiers found in args
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
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, 'redundant_parameters' ); -- v.parameters is a table of aliases for k; here we pick one from args if present
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
if is_set(v) then id_list[k] = v; end -- if found in args, add identifier to our list
end
end
1,088行目: 1,378行目:
Parameters which have a predefined access level (e.g. arxiv) do not use this
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.
function as they are directly rendered as free without using an additional parameter.
access-level values must match the case used in cfg.keywords_lists['id-access'] (lowercase unless there is some special reason for something else)


]]
]]
1,094行目: 1,386行目:
local id_accesses_list = {};
local id_accesses_list = {};
for k, v in pairs( cfg.id_handlers ) do
for k, v in pairs( cfg.id_handlers ) do
local access_param = v.custom_access;
local access_param = v.custom_access; -- name of identifier's access-level parameter
local k_lower = string.lower(k);
if is_set(access_param) then
if is_set(access_param) then
local access_level = args[access_param];
local access_level = args[access_param]; -- get the assigned value if there is one
if is_set(access_level) then
if is_set (access_level) then
if not in_array (access_level:lower(), cfg.keywords['id-access']) then
if not in_array (access_level, cfg.keywords_lists['id-access']) then -- exact match required
table.insert( z.message_tail, { set_error( 'invalid_param_val', {access_param, access_level}, true ) } );
table.insert( z.message_tail, { set_message( 'invalid_param_val', {access_param, access_level}, true ) } );
access_level = nil;
access_level = nil; -- invalid so unset
end
end
if not is_set(id_list[k]) then
if not is_set(id_list[k]) then -- identifier access-level must have a matching identifier
table.insert( z.message_tail, { set_error( 'param_access_requires_param', {k_lower}, true ) } );
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
end
end
if is_set(access_level) then
id_accesses_list[k] = cfg.keywords_xlate[access_level]; -- get translated keyword
access_level = access_level:lower();
end
id_accesses_list[k] = access_level;
end
end
end
end
1,126行目: 1,414行目:
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;
1,137行目: 1,425行目:
end
end


--[[--------------------------< E X P O R T E D  F U N C T I O N S >------------------------------------------
]]


return {
return {
auto_link_urls = auto_link_urls, -- table of identifier URLs to be used when auto-linking |title=
build_id_list = build_id_list,
build_id_list = build_id_list,
extract_ids = extract_ids,
extract_ids = extract_ids,

2020年10月13日 (火) 20:15時点における版

このモジュールについての説明文ページを モジュール:Citation/CS1/Identifiers/doc に作成できます

--[[--------------------------< F O R W A R D   D E C L A R A T I O N S >--------------------------------------
]]

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 cfg;																		-- table of configuration tables that are defined in Module:Citation/CS1/Configuration


--[[--------------------------< P A G E   S C O P E   V A R I A B L E S >--------------------------------------

declare variables here that have page-wide scope that are not brought in from other modules; that are created here and used here

]]

local auto_link_urls = {};														-- holds identifier URLs for those identifiers that can auto-link |title=


--============================<< H E L P E R   F U N C T I O N S >>============================================

--[[--------------------------< W I K I D A T A _ A R T I C L E _ N A M E _ G E T >----------------------------

as an aid to internationalizing identifier-label wikilinks, gets identifier article names from Wikidata.

returns :<lang code>:<article title> when <q> has an <article title> for <lang code>; nil else

for identifiers that do not have q, returns nil

for wikis that do not have mw.wikibase installed, returns nil

]]

local function wikidata_article_name_get (q)
	if not is_set (q) or (q and not mw.wikibase) then							-- when no q number or when a q number but mw.wikibase not installed on this wiki
		return nil;																-- abandon
	end

	local wd_article;
	local this_wiki_code = cfg.this_wiki_code;									-- Wikipedia subdomain; 'en' for en.wikipedia.org

	wd_article = mw.wikibase.getSitelink (q, this_wiki_code .. 'wiki');			-- fetch article title from WD; nil when no title available at this wiki

	if wd_article then
		wd_article = table.concat ({':', this_wiki_code, ':', wd_article});		-- interwiki-style link without brackets if taken from WD; leading colon required
	end

	return wd_article;															-- article title from WD; nil else
end


--[[--------------------------< L I N K _ L A B E L _ M A K E >------------------------------------------------

common function to create identifier link label from handler table or from Wikidata

returns the first available of
	1. redirect from local wiki's handler table (if enabled)
	2. Wikidata (if there is a Wikidata entry for this identifier in the local wiki's language)
	3. label specified in the local wiki's handler table
	
]]

local function link_label_make (handler)
	local wd_article;
	
	if not (cfg.use_identifier_redirects and is_set (handler.redirect)) then	-- redirect has priority so if enabled and available don't fetch from Wikidata because expensive
		wd_article = wikidata_article_name_get (handler.q);						-- if Wikidata has an article title for this wiki, get it;
	end
	
	return (cfg.use_identifier_redirects and is_set (handler.redirect) and handler.redirect) or wd_article or handler.link;
end


--[[--------------------------< E X T E R N A L _ L I N K _ I D >----------------------------------------------

Formats a wiki-style external link

NB. Wikidata P1630 has a formatter URL with $1 placeholder for the ID which could be worked into our prefix/id/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;									-- Wikipedia subdomain; 'en' for en.wikipedia.org
	local wd_article;															-- article title from Wikidata
	
	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

	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
		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 (link_label_make (options), options.label),				-- redirect, Wikidata link, or locally specified link (in that order)
		options.separator or '&nbsp;',
		ext_link
		});
end


--[[--------------------------< I N T E R N A L _ L I N K _ I D >----------------------------------------------

Formats a wiki-style internal link

]]

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 (link_label_make (options), options.label),				-- wiki-link the identifier label
		options.separator or '&nbsp;',											-- add the separator
		make_wikilink (
			table.concat (
				{
				options.prefix,
				id,																-- translated to Western digits
				options.suffix or ''
				}),
			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?
		});
end


--[[--------------------------< 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 an empty string because the embargo has expired or because
|pmc-embargo-date= was not set in this cite.

]]

local function is_embargoed (embargo)
	if is_set (embargo) then
		local lang = mw.getContentLanguage();
		local good1, embargo_date, todays_date;
		good1, embargo_date = pcall (lang.formatDate, lang, 'U', embargo);
		todays_date = lang:formatDate ('U');
	
		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?
				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


--[=[-------------------------< 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


--[[--------------------------< 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-13 is checked in isbn().

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.

]]

local function is_valid_isxn (isxn_str, len)
	local temp = 0;
	isxn_str = { isxn_str:byte(1, len) };										-- make a table of byte values '0' → 0x30 .. '9' → 0x39, 'X' → 0x58
	len = len + 1;																-- adjust to be a loop counter
	for i, v in ipairs( isxn_str ) do											-- loop through all of the bytes and calculate the checksum
		if v == string.byte( "X" ) then											-- if checkdigit is X (compares the byte value of 'X' which is 0x58)
			temp = temp + 10*( len - i );										-- it represents 10 decimal
		else
			temp = temp + tonumber( string.char(v) )*(len-i);
		end
	end
	return temp % 11 == 0;														-- returns true if calculation result is zero
end


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

ISBN-13 and ISMN validator code calculates checksum across all 13 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
and stripped of dashes, spaces and other non-ISxN-13 characters.

]]

local function is_valid_isxn_13 (isxn_str)
	local temp=0;
	
	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
		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 ISBN-13/ISMN is correct
end


--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------

LCCN normalization (http://www.loc.gov/marc/lccn-namespace.html#normalization)
1. Remove all blanks.
2. If there is a forward slash (/) in the string, remove it, and remove all characters to the right of the forward slash.
3. If there is a hyphen in the string:
	a. Remove it.
	b. Inspect the substring following (to the right of) the (removed) hyphen. Then (and assuming that steps 1 and 2 have been carried out):
		1. All these characters should be digits, and there should be six or less. (not done in this function)
		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.
]]

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 LCCN
	end
	
	return lccn;
	end


--============================<< I D E N T I F I E R   F U N C T I O N S >>====================================

--[[--------------------------< A R X I V >--------------------------------------------------------------------

See: http://arxiv.org/help/arxiv_identifier

format and error check arXiv identifier.  There are three valid forms of the identifier:
the first form, valid only between date codes 9107 and 0703, is:
	arXiv:<archive>.<class>/<date code><number><version>
where:
	<archive> is a string of alpha characters - may be hyphenated; no other punctuation
	<class> is a string of alpha characters - may be hyphenated; no other punctuation; not the same as |class= parameter which is not supported in this form
	<date code> is four digits in the form YYMM where YY is the last two digits of the four-digit year and MM is the month number January = 01
		first digit of YY for this form can only 9 and 0
	<number> is a three-digit number
	<version> is a 1 or more digit number preceded with a lowercase v; no spaces (undocumented)
	
the second form, valid from April 2007 through December 2014 is:
	arXiv:<date code>.<number><version>
where:
	<date code> is four digits in the form YYMM where YY is the last two digits of the four-digit year and MM is the month number January = 01
	<number> is a four-digit number
	<version> is a 1 or more digit number preceded with a lowercase v; no spaces

the third form, valid from January 2015 is:
	arXiv:<date code>.<number><version>
where:
	<date code> and <version> are as defined for 0704-1412
	<number> is a five-digit number
]]

local function arxiv (id, class)
	local handler = cfg.id_handlers['ARXIV'];
	local year, month, version;
	local err_cat = false;														-- assume no error message
	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 9107-0703 format with or without version
		year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
		year = tonumber(year);
		month = tonumber(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?
				err_cat = true;													-- flag for error message
		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 with or without version
		year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
		year = tonumber(year);
		month = tonumber(month);
		if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or			-- is year invalid or is month invalid? (doesn't test for future years)
			((7 == year) and (4 > month)) then									-- when year is 07, is month invalid (before April)?
				err_cat = true;													-- flag for error message
		end

	elseif id:match("^%d%d[01]%d%.%d%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%d%dv%d+$") then	-- test for the 1501- format with or without version
		year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d%d[v%d]*$");
		year = tonumber(year);
		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)
				err_cat = true;													-- flag for error message
		end

	else
		err_cat = true;															-- not a recognized format; flag for error message
	end

	err_cat = err_cat and table.concat ({' ', set_message ('err_bad_arxiv')}) or '';	-- set error message if flag is true
	
	text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
			prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;

	if is_set (class) then
		if id:match ('^%d+') then
			text = table.concat ({text, ' [[//arxiv.org/archive/', class, ' ', class, ']]'});	-- external link within square brackets, not wikilink
		else
			text = table.concat ({text, ' ', set_message ('err_class_ignored')});		
		end
	end

	return text;	
end


--[[--------------------------< B I B C O D E >--------------------------------------------------------------------

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

But, this: 2015arXiv151206696F is apparently valid so apparently, the only things that really matter are length, 19 characters
and first four digits must be a year.  This function makes these tests:
	length must be 19 characters
	characters in position
		1–4 must be digits and must represent a year in the range of 1000 – next year
		5 must be a letter
		6–8 must be letter, digit, ampersand, or dot (ampersand cannot directly precede a dot; &. )
		9–18 must be letter, digit, or dot
		19 must be a letter or dot

]]

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

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

	if is_set (err_type) then													-- if there was an error detected
		text = text .. ' ' .. set_message( 'err_bad_bibcode', {err_type});
	end
	return text;
end


--[[--------------------------< B I O R X I V >-----------------------------------------------------------------

Format bioRxiv ID and do simple error checking.  Before 2019-12-11, biorXiv IDs were 10.1101/ followed by exactly
6 digits.  After 2019-12-11, biorXiv IDs retained the six-digit identifier but prefixed that with a yyyy.mm.dd. 
date and suffixed with an optional version identifier.

The bioRxiv ID 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 handler = cfg.id_handlers['BIORXIV'];
	local err_cat = true;														-- flag; assume that there will be an error
	
	local patterns = {
		'^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

	return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
			prefix = handler.prefix, id = id, separator = handler.separator,
			encode = handler.encode, access = handler.access}) .. (err_cat and (' ' .. set_message( 'err_bad_biorxiv')) or '');
end


--[[--------------------------< C I T E S E E R X >------------------------------------------------------------

CiteSeerX use their own notion of "doi" (not to be confused with the identifiers resolved via doi.org).

The description of the structure of this identifier can be found at Help_talk:Citation_Style_1/Archive_26#CiteSeerX_id_structure
]]

local function citeseerx (id)
	local handler = cfg.id_handlers['CITESEERX'];
	local matched;
	
	local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
		prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode,
		access = handler.access});
	
	matched = id:match ("^10%.1%.1%.[1-9]%d?%d?%d?%.[1-9]%d?%d?%d?$");
	if not matched then
		text = text .. ' ' .. set_message( 'err_bad_citeseerx' );
	end
	return text;
end


--[[--------------------------< D O I >------------------------------------------------------------------------

Formats a DOI and checks for DOI errors.

DOI names contain two parts: prefix and suffix separated by a forward slash.
	Prefix: directory indicator '10.' followed by a registrant code
	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
with a period or a comma, this function will emit a bad_doi error message.

DOI names are case-insensitive and can incorporate any printable Unicode characters so the test for spaces, endash,
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in DOI names.

]]

local function doi (id, inactive, access, ignore_invalid)
	local err_cat;
	local handler = cfg.id_handlers['DOI'];
	
	local text;
	if is_set (inactive) then
		local inactive_year = inactive:match("%d%d%d%d") or '';					-- try to get the year portion from the inactive date
		local inactive_month, good;

		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*$',												-- 5 digits with subcode (0xxxx, 40000+); accepts: 10000–39999
		'^[^1-4]%d%d%d%d$',														-- 5 digits without subcode (0xxxx, 40000+); accepts: 10000–49999
		'^[^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
		'%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


--[[--------------------------< H D L >------------------------------------------------------------------------

Formats an HDL with minor error checking.

HDL names contain two parts: prefix and suffix separated by a forward slash.
	Prefix: character string using any character in the UCS-2 character set except '/'
	Suffix: character string of any length using any character in the UCS-2 character set chosen by the registrant

This function checks a HDL name for: prefix/suffix.  If the HDL name contains spaces, endashes, or, if it ends
with a period or a comma, this function will emit a bad_hdl error message.

HDL names are case-insensitive and can incorporate any printable Unicode characters so the test for endashes and
terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in HDLs.

Query string parameters are named here: http://www.handle.net/proxy_servlet.html.  query strings are not displayed
but since '?' is an allowed character in an HDL, '?' followed by one of the query parameters is the only way we
have to detect the query string so that it isn't URL-encoded with the rest of the identifier.

]]

local function hdl(id, access)
	local handler = cfg.id_handlers['HDL'];
	local query_params = {														-- list of known query parameters from http://www.handle.net/proxy_servlet.html
		'noredirect',
		'ignore_aliases',
		'auth',
		'cert',
		'index',
		'type',
		'urlappend',
		'locatt',
		'action',
		}
	
	local hdl, suffix, param = id:match ('(.-)(%?(%a+).+)$');					-- look for query string
	local found;

	if hdl then																	-- when there are query strings, this is the handle identifier portion
		for _, q in ipairs (query_params) do									-- spin through the list of query parameters
			if param:match ('^' .. q) then										-- if the query string begins with one of the parameters
				found = true;													-- announce a find
				break;															-- and stop looking
			end
		end
	end

	if found then
		id = hdl;																-- found so replace id with the handle portion; this will be URL-encoded, suffix will not
	else
		suffix = '';															-- make sure suffix is empty string for concatenation else
	end

	local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
			prefix = handler.prefix, id = id, suffix = suffix, separator = handler.separator, encode = handler.encode, access = access})

	if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then							-- HDL 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' );
	end
	return text;
end


--[[--------------------------< I S B N >----------------------------------------------------------------------

Determines whether an ISBN string is valid

]]

local function isbn (isbn_str, ignore_invalid)
	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

	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
		return return_result (false, cfg.err_msg_supl.length);					-- fail if incorrect length
	end

	if len == 10 then
		if id:match ('^%d*X?$') == nil then										-- fail if isbn_str has 'X' anywhere but last position
			return return_result (false, cfg.err_msg_supl.form);									
		end
		return return_result (is_valid_isxn(id, 10), cfg.err_msg_supl.check);
	else
		if id:match ('^%d+$') == nil then
			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
		if id:match ('^9790') then
			return return_result (false, cfg.err_msg_supl.group);				-- group identifier '0' is reserved to ISMN
		end
		return return_result (is_valid_isxn_13 (id), cfg.err_msg_supl.check);
	end
end


--[[--------------------------< 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.

|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()

]]

local function asin (id, domain)
	local err_cat = ""

	if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
		err_cat = ' ' .. set_message ('err_bad_asin');							-- ASIN is not a mix of 10 uppercase alpha and numeric characters
	else
		if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then							-- if 10-digit numeric (or 9 digits with terminal X)
			if isbn (id) then													-- see if ASIN value is or validates as ISBN-10
				if not id:find ('^630') then									-- 630xxxxxxx is (apparently) not a valid isbn prefix but is used by amazon as a numeric identifier
					set_message ('maint_asin');									-- begins with something other than 630 so possibly an isbn
				end
			elseif not is_set (err_cat) then
				err_cat = ' ' .. set_message ('err_bad_asin');					-- ASIN is not ISBN-10
			end
		elseif not id:match("^%u[%d%u]+$") then
			err_cat = ' ' .. set_message ('err_bad_asin');						-- asin doesn't begin with uppercase alpha
		end
	end
	if not is_set(domain) then 
		domain = "com";
	elseif in_array (domain, {'jp', 'uk'}) then									-- Japan, United Kingdom
		domain = "co." .. domain;
	elseif in_array (domain, {'au', 'br', 'mx'}) then							-- Australia, Brazil, Mexico
		domain = "com." .. domain;
	end
	local handler = cfg.id_handlers['ASIN'];
	return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
		prefix = handler.prefix .. domain .. "/dp/",
		id = id, encode = handler.encode, separator = handler.separator}) .. err_cat;
end


--[[--------------------------< I S M N >----------------------------------------------------------------------

Determines whether an ISMN string is valid.  Similar to 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
section 2, pages 9–12.

]]

local function ismn (id)
	local handler = cfg.id_handlers['ISMN'];
	local text;
	local valid_ismn = true;
	local id_copy;

	id_copy = id;																-- save a copy because this testing is destructive
	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 with 9790
		valid_ismn = false;
	else
		valid_ismn=is_valid_isxn_13 (id);										-- validate ISMN
	end

	--	text = internal_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,		-- use this (or external version) when there is some place to link to
	--		prefix = handler.prefix, id = id_copy, separator = handler.separator, encode = handler.encode})

	text = table.concat (														-- because no place to link to yet
		{
		make_wikilink (link_label_make (handler), handler.label),
		handler.separator,
		id_copy
		});

	if false == valid_ismn then
		text = text .. ' ' .. set_message( 'err_bad_ismn' )						-- add an error message if the ISMN is invalid
	end 
	
	return text;
end


--[[--------------------------< 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
has separated the two groups of four digits with a space.  When that condition occurred, the resulting link looked
like this:

	|issn=0819 4327 gives: [http://www.worldcat.org/issn/0819 4327 0819 4327]	-- can't have spaces in an external link
	
This code now prevents that by inserting a hyphen at the ISSN midpoint.  It also validates the ISSN for length
and makes sure that the checkdigit agrees with the calculated value.  Incorrect length (8 digits), characters
other than 0-9 and X, or checkdigit / calculated value mismatch will all cause a check ISSN error message.  The
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, ignore_invalid)
	local issn_copy = id;														-- save a copy of unadulterated ISSN; use this version for display if ISSN does not validate
	local handler;
	local text;
	local valid_issn = true;
	
	handler = cfg.id_handlers[e and 'EISSN' or 'ISSN'];

	id = id:gsub ('[%s-]', '');													-- remove hyphens and whitespace

	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
		valid_issn = is_valid_isxn(id, 8);										-- validate ISSN
	end

	if true == valid_issn then
		id = string.sub( id, 1, 4 ) .. "-" .. string.sub( id, 5 );				-- if valid, display correctly formatted version
	else
		id = issn_copy;															-- if not valid, show the invalid ISSN with error message
	end

	text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
		prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})

	if ignore_invalid then
		set_message ('maint_issn_ignore');
	else
		if false == valid_issn then
			text = text .. ' ' .. set_message ('err_bad_issn', e and 'e' or '');	-- add an error message if the ISSN is invalid
		end 
	end
	
	return text
end


--[[--------------------------< J F M >-----------------------------------------------------------------------

A numerical identifier in the form nn.nnnn.nn

]]

local function jfm (id)
	local handler = cfg.id_handlers['JFM'];
	local id_num;
	local err_cat = '';
	
	id_num = id:match ('^[Jj][Ff][Mm](.*)$');									-- identifier with jfm prefix; extract identifier

	if is_set (id_num) then
		set_message ('maint_jfm_format');
	else																		-- plain number without mr prefix
		id_num = id;															-- if here id does not have prefix
	end

	if id_num and id_num:match('^%d%d%.%d%d%d%d%.%d%d$') then
		id = id_num;															-- jfm matches pattern
	else
		err_cat = ' ' .. set_message( 'err_bad_jfm' );							-- set an error message
	end
	
	return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
			prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end


--[[--------------------------< L C C N >----------------------------------------------------------------------

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.
http://info-uri.info/registry/OAIHandler?verb=GetRecord&metadataPrefix=reg&identifier=info:lccn/

length = 8 then all digits
length = 9 then lccn[1] is lowercase alpha
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 lowercase alpha or both digits
length = 12 then lccn[1] and lccn[2] are both lowercase alpha

]]

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

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

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

	if not is_set (err_cat) and nil ~= lccn:find ('%s') then
		err_cat = ' ' .. set_message ('err_bad_lccn');							-- lccn contains a space, set an error message
	end

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


--[[--------------------------< M R >--------------------------------------------------------------------------

A seven digit number; if not seven digits, zero-fill leading digits to make seven digits.

]]

local function mr (id)
	local handler = cfg.id_handlers['MR'];
	local id_num;
	local id_len;
	local err_cat = '';
	
	id_num = id:match ('^[Mm][Rr](%d+)$');										-- identifier with mr prefix

	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
	end

	id_len = id_num and id_num:len() or 0;
	if (7 >= id_len) and (0 ~= id_len) then
		id = string.rep ('0', 7-id_len ) .. id_num;								-- zero-fill leading digits
	else
		err_cat = ' ' .. set_message ('err_bad_mr');							-- set an error message
	end
	
	return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
			prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end


--[[--------------------------< O C L C >----------------------------------------------------------------------

Validate and format an OCLC 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

]]

local function oclc (id)
	local handler = cfg.id_handlers['OCLC'];
	local number;
	local err_msg = '';															-- empty string for concatenation
	
	if id:match('^ocm%d%d%d%d%d%d%d%d$') then									-- ocm prefix and 8 digits; 001 field (12 characters)
		number = id:match('ocm(%d+)');											-- get the number
	elseif id:match('^ocn%d%d%d%d%d%d%d%d%d$') then								-- ocn prefix and 9 digits; 001 field (12 characters)
		number = id:match('ocn(%d+)');											-- get the number
	elseif id:match('^on%d%d%d%d%d%d%d%d%d%d+$') then							-- on prefix and 10 or more digits; 001 field (12 characters)
		number = id:match('^on(%d%d%d%d%d%d%d%d%d%d+)$');						-- get the number
	elseif id:match('^%(OCoLC%)[1-9]%d*$') then									-- (OCoLC) prefix and variable number digits; no leading zeros; 035 field
		number = id:match('%(OCoLC%)([1-9]%d*)');								-- get the number
		if 9 < number:len() then
			number = nil;														-- constrain to 1 to 9 digits; change this when OCLC issues 10-digit numbers
		end
	elseif id:match('^%d+$') then												-- no prefix
		number = id;															-- get the number
		if 10 < number:len() then
			number = nil;														-- constrain to 1 to 10 digits; change this when OCLC issues 11-digit numbers
		end
	end

	if number then																-- proper format
		id = number;															-- exclude prefix, if any, from external link
	else
		err_msg = ' ' .. set_message ('err_bad_oclc')							-- add an error message if the id is malformed
	end
	
	local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
		prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_msg;

	return text;
end


--[[--------------------------< O P E N L I B R A R Y >--------------------------------------------------------

Formats an OpenLibrary link, and checks for associated errors.

]]

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 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

	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


--[[--------------------------< P M C >------------------------------------------------------------------------

Format a PMC, do simple error checking, and check for embargoed articles.

The embargo parameter takes a date for a value. If the embargo date is in the future the PMC identifier will not
be linked to the article.  If the embargo date is today or in the past, or if it is empty or omitted, then the
PMC identifier is linked to the article through the link at cfg.id_handlers['PMC'].prefix.

PMC embargo date testing is done in function is_embargoed () which is called earlier because when the citation
has |pmc=<value> but does not have a |url= then |title= is linked with the PMC link.  Function is_embargoed ()
returns the embargo date if the PMC article is still embargoed, otherwise it returns an empty string.

PMCs are sequential numbers beginning at 1 and counting up.  This code checks the PMC to see that it contains only digits and is less
than test_limit; the value in local variable test_limit will need to be updated periodically as more PMCs are issued.

]]

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 PMC prefix

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

	if is_set (id_num) then														-- id_num has a value so test it
		id_num = tonumber(id_num);												-- convert id_num to a number for range testing
		if 1 > id_num or handler.id_limit < id_num then							-- if PMC is outside test limit boundaries
			err_cat = ' ' .. set_message ('err_bad_pmc');						-- set an error message
		else
			id = tostring (id_num);												-- make sure id is a string
		end
	else																		-- when id format incorrect
		err_cat = ' ' .. set_message ('err_bad_pmc');							-- set an error message
	end
	
	if is_set (embargo) and is_set (is_embargoed (embargo)) then				-- is PMC is still embargoed?
		text = table.concat (													-- still embargoed so no external link
			{
			make_wikilink (link_label_make (handler), handler.label),
			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;
end


--[[--------------------------< P M I D >----------------------------------------------------------------------

Format PMID and do simple error checking.  PMIDs are sequential numbers beginning at 1 and counting up.  This
code checks the PMID to see that it contains only digits and is less than test_limit; the value in local variable
test_limit will need to be updated periodically as more PMIDs are issued.

]]

local function pmid(id)
	local handler = cfg.id_handlers['PMID'];
	local err_cat = '';															-- presume that PMID is valid
	
	if id:match("[^%d]") then													-- if PMID has anything but digits
		err_cat = ' ' .. set_message ('err_bad_pmid');							-- set an error message
	else																		-- PMID is only digits
		local id_num = tonumber(id);											-- convert id to a number for range testing
		if 1 > id_num or handler.id_limit < id_num then							-- if PMID is outside test limit boundaries
			err_cat = ' ' .. set_message ('err_bad_pmid');						-- set an error message
		end
	end
	
	return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
			prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end


--[[--------------------------< S 2 C I D >--------------------------------------------------------------------

Format an S2CID, do simple error checking

S2CIDs are sequential numbers beginning at 1 and counting up.  This code checks the S2CID to see that it is 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 (id, access)
	local handler = cfg.id_handlers['S2CID'];
	local err_cat = '';															-- presume that S2CID is valid
	local id_num;
	local text;
	
	id_num = id:match ('^[1-9]%d*$');											-- id must be all digits; must not begin with 0; no open access flag

 	if is_set (id_num) then														-- id_num has a value so test it
		id_num = tonumber(id_num);												-- convert id_num to a number for range testing
		if handler.id_limit < id_num then										-- if S2CID is outside test limit boundaries
			err_cat = ' ' .. set_message ('bad_s2cid_err');						-- set an error message
		end

	else																		-- when id format incorrect
		err_cat = ' ' .. set_message ('bad_s2cid_err');							-- set an error message
	end

	text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
		prefix = handler.prefix, id = id:gsub ('%.%a%a', ''), separator = handler.separator, encode = handler.encode, access = access}) .. err_cat;

	return text;
end


--[[--------------------------< S B N >------------------------------------------------------------------------

9-digit form of ISBN-10; uses same check-digit validation when SBN is prefixed with an additional '0' to make 10 digits

]]

local function sbn (id, ignore_invalid)
	local handler = cfg.id_handlers['SBN'];
	local function return_result (check, err_type) -- local function to handle the various returns
		local SBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
						prefix = handler.prefix, id = id, separator = handler.separator});
		if not ignore_invalid then -- if not ignoring SBN errors
			if not check then
				return SBN .. set_message ('err_bad_sbn', {err_type}, false, ' '); -- display an error message
			end
		else
			set_message ('maint_isbn_ignore');									-- add a maint category even when there is no error (ToDo: Possibly switch to separate message for SBNs only)
		end
		return SBN;
	end

	if id:match ('[^%s-0-9X]') then
		return return_result (false, cfg.err_msg_supl.char);					-- fail if SBN contains anything but digits, hyphens, or the uppercase X
	end

	local ident = id:gsub ('[%s-]', '');											-- remove hyphens and whitespace; they interfere with the rest of the tests

	if  9 ~= ident:len() then
		return return_result (false, cfg.err_msg_supl.length);					-- fail if incorrect length
	end

	if ident:match ('^%d*X?$') == nil then
		return return_result (false, cfg.err_msg_supl.form);					-- fail if SBN has 'X' anywhere but last position
	end

	return return_result (is_valid_isxn ('0' .. ident, 10), cfg.err_msg_supl.check);
end


--[[--------------------------< S S R N >----------------------------------------------------------------------

Format an 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
only digits and is greater than 99 and less than test_limit; the value in local variable test_limit will need
to be updated periodically as more SSRNs are issued.

]]

local function ssrn (id)
	local handler = cfg.id_handlers['SSRN'];
	local err_cat = '';															-- presume that SSRN is valid
	local id_num;
	local text;
	
	id_num = id:match ('^%d+$');												-- id must be all digits

	if is_set (id_num) then														-- id_num has a value so test it
		id_num = tonumber(id_num);												-- convert id_num to a number for range testing
		if 100 > id_num or handler.id_limit < id_num then						-- if SSRN is outside test limit boundaries
			err_cat = ' ' .. set_message ('err_bad_ssrn');						-- set an error message
		end
	else																		-- when id format incorrect
		err_cat = ' ' .. set_message ('err_bad_ssrn');							-- set an error message
	end
	
	text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
		prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;

	return text;
end


--[[--------------------------< U S E N E T _ I D >------------------------------------------------------------

Validate and format a usenet message id.  Simple error checking, looks for 'id-left@id-right' not enclosed in
'<' and/or '>' angle brackets.

]]

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

	local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
		prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode})
 
	if not id:match('^.+@.+$') or not id:match('^[^<].*[^>]$') then				-- doesn't have '@' or has one or first or last character is '< or '>'
		text = text .. ' ' .. set_message ('err_bad_usenet_id')					-- add an error message if the message id is invalid
	end 
	
	return text
end


--[[--------------------------< Z B L >-----------------------------------------------------------------------

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 handler = cfg.id_handlers['ZBL'];
	local err_cat = '';
	
	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?
		err_cat = ' ' .. set_message ('err_bad_zbl');							-- no, set an error message
	end
	
	return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
			prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
end


--============================<< 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 >--------------------------------------------------------

Takes a table of IDs created by extract_ids() and turns it into a table of formatted ID outputs.

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 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) );

		if handler.mode == 'external' then
			table.insert( new_list, {handler.label, external_link_id( handler ) } );
		elseif handler.mode == 'internal' then
			table.insert( new_list, {handler.label, internal_link_id( handler ) } );
		elseif handler.mode ~= 'manual' then
			error( cfg.messages['unknown_ID_mode'] );
		elseif k == 'ARXIV' then
			table.insert( new_list, {handler.label, arxiv( v, options.Class ) } ); 
		elseif k == 'ASIN' then
			table.insert( new_list, {handler.label, asin( v, options.ASINTLD ) } ); 
		elseif k == 'BIBCODE' then
			table.insert( new_list, {handler.label, bibcode( v, handler.access ) } );
		elseif k == 'BIORXIV' then
			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, accept) } );
		elseif k == 'EISSN' then
			table.insert( new_list, {handler.label, issn( v, true, accept) } );	-- true distinguishes EISSN from ISSN
		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 ) } );
		elseif k == 'LCCN' then
			table.insert( new_list, {handler.label, lccn( v ) } );
		elseif k == 'MR' then
			table.insert( new_list, {handler.label, mr( v ) } );
		elseif k == 'OCLC' then
			table.insert( new_list, {handler.label, oclc( v ) } );
		elseif k == 'OL' or k == 'OLA' then
			table.insert( new_list, {handler.label, openlibrary( v, handler.access ) } );
		elseif k == 'PMC' then
			table.insert( new_list, {handler.label, pmc( v, options.Embargo ) } );
		elseif k == 'PMID' then
			table.insert( new_list, {handler.label, pmid( v ) } );
		elseif k == 'S2CID' then
			table.insert( new_list, {handler.label, s2cid( v, handler.access ) } );
		elseif k == 'SBN' then
			table.insert( new_list, {handler.label, sbn (v, accept) } );
		elseif k == 'SSRN' then
			table.insert( new_list, {handler.label, ssrn( v ) } );
		elseif k == 'USENETID' then
			table.insert( new_list, {handler.label, usenet_id( v ) } );
		elseif k == 'ZBL' then
			table.insert( new_list, {handler.label, zbl( v ) } );
		else
			error( cfg.messages['unknown_manual_ID'] );
		end
	end
	
	local function comp( a, b )	-- used in following table.sort()
		return a[1]:lower() < b[1]:lower();
	end
	
	table.sort( new_list, comp );
	for k, v in ipairs( new_list ) do
		new_list[k] = v[2];
	end
	
	return new_list;
end


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

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

]]

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.

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 function extract_id_access_levels( args, id_list )
	local id_accesses_list = {};
	for k, v in pairs( cfg.id_handlers ) do
		local access_param = v.custom_access;									-- name of identifier's access-level parameter
		if is_set(access_param) then
			local access_level = args[access_param];							-- get the assigned value if there is one
			if is_set (access_level) then
				if not in_array (access_level, cfg.keywords_lists['id-access']) then	-- exact match required
					table.insert( z.message_tail, { set_message( 'invalid_param_val', {access_param, access_level}, true ) } );	
					access_level = nil;											-- invalid so unset
				end
				if not is_set(id_list[k]) then									-- identifier access-level must have a matching identifier
					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
				end
				id_accesses_list[k] = cfg.keywords_xlate[access_level];			-- get translated keyword
			end
		end
	end
	return id_accesses_list;
end


--[[--------------------------< S E T _ S E L E C T E D _ M O D U L E S >--------------------------------------

Sets local cfg table and imported functions table to same (live or sandbox) as that used by the other modules.

]]

local function set_selected_modules (cfg_table_ptr, utilities_page_ptr)
	cfg = cfg_table_ptr;

	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;
	set_message = utilities_page_ptr.set_message;
	select_one = utilities_page_ptr.select_one;
	substitute = utilities_page_ptr.substitute;
	make_wikilink = utilities_page_ptr.make_wikilink;

	z = utilities_page_ptr.z;													-- table of tables in Module:Citation/CS1/Utilities
end


--[[--------------------------< E X P O R T E D   F U N C T I O N S >------------------------------------------
]]

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,
	extract_id_access_levels = extract_id_access_levels,
	is_embargoed = is_embargoed;
	set_selected_modules = set_selected_modules;
	}