Difference between revisions of "Module:Citation/CS1/Identifiers"

From Shipbucket Wiki
Jump to: navigation, search
(asin() calls isbn() so asin() must follow isbn();)
m (1 revision imported)
Line 6: Line 6:
 
]]
 
]]
  
−
local is_set, in_array, set_error, select_one, add_maint_cat, substitute, make_wikilink; -- functions in Module:Citation/CS1/Utilities
+
local is_set, in_array, set_error, select_one, add_maint_cat; -- functions in Module:Citation/CS1/Utilities
  
 
local z; -- table of tables defined in Module:Citation/CS1/Utilities
 
local z; -- table of tables defined in Module:Citation/CS1/Utilities
Line 12: Line 12:
 
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
  
−
 
−
--============================<< H E L P E R  F U N C T I O N S >>============================================
 
  
 
--[[--------------------------< 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 >----------------------------------------------
Line 23: Line 21:
 
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;
 
−
 
 
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
−
+
return mw.ustring.format( '[[%s|%s]]%s[%s%s%s %s]',
−
ext_link = mw.ustring.format ('[%s%s%s %s]', options.prefix, url_string, options.suffix or "", mw.text.nowiki(options.id));
+
options.link, options.label, options.separator or "&nbsp;",
−
if is_set(options.access) then
+
options.prefix, url_string, options.suffix or "",
−
ext_link = substitute (cfg.presentation['access-signal'], {ext_link, cfg.presentation[options.access]}); -- add the free-to-read / paywall lock
+
mw.text.nowiki(options.id)
−
end
+
);
−
 
−
return table.concat ({
 
−
make_wikilink (options.link, options.label),
 
−
options.separator or '&nbsp;',
 
−
ext_link
 
−
});
 
 
end
 
end
  
Line 49: Line 39:
  
 
local function internal_link_id(options)
 
local function internal_link_id(options)
−
 
+
return mw.ustring.format( '[[%s|%s]]%s[[%s%s%s|%s]]',
−
return table.concat (
+
options.link, options.label, options.separator or "&nbsp;",
−
{
+
options.prefix, options.id, options.suffix or "",
−
make_wikilink (options.link, options.label),
+
mw.text.nowiki(options.id)
−
options.separator or '&nbsp;',
+
);
−
make_wikilink (
 
−
table.concat (
 
−
{
 
−
options.prefix,
 
−
options.id,
 
−
options.suffix or ''
 
−
}),
 
−
mw.text.nowiki (options.id)
 
−
);
 
−
});
 
−
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 |embargo= 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
 
−
|embargo= was not set in this cite.
 
−
 
 
−
]]
 
−
 
 
−
local function is_embargoed (embargo)
 
−
if is_set (embargo) then
 
−
local lang = mw.getContentLanguage();
 
−
local good1, embargo_date, good2, todays_date;
 
−
good1, embargo_date = pcall( lang.formatDate, lang, 'U', embargo );
 
−
good2, todays_date = pcall( lang.formatDate, lang, 'U' );
 
−
 
−
if good1 and good2 then -- if embargo date and today's date are good dates
 
−
if tonumber( embargo_date ) >= tonumber( todays_date ) then -- is embargo date is in the future?
 
−
return embargo; -- still embargoed
 
−
else
 
−
add_maint_cat ('embargo')
 
−
return ''; -- unset because embargo has expired
 
−
end
 
−
end
 
−
end
 
−
return ''; -- |embargo= not set return empty string
 
 
end
 
end
  
Line 98: Line 50:
  
 
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 check_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, issbn/issn must be checked for length
Line 107: Line 59:
 
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
Line 120: Line 72:
  
  
−
--[[--------------------------< 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.
Line 131: Line 83:
 
local temp=0;
 
local temp=0;
 
 
−
isxn_str = { isxn_str:byte(1, 13) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39
+
isxn_str = { isxn_str:byte(1, 13) }; -- make a table of byte values '0' → 0x30 .. '9' → 0x39
 
for i, v in ipairs( isxn_str ) do
 
for i, v in ipairs( isxn_str ) do
 
temp = temp + (3 - 2*(i % 2)) * tonumber( string.char(v) ); -- multiply odd index digits by 1, even index digits by 3 and sum; includes check digit
 
temp = temp + (3 - 2*(i % 2)) * tonumber( string.char(v) ); -- multiply odd index digits by 1, even index digits by 3 and sum; includes check digit
Line 139: Line 91:
  
  
−
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------
+
--[[--------------------------< C H E C K _ I S B 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 9108 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
 
−
<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 = '';
 
−
local 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
 
−
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 = ' ' .. set_error( 'bad_arxiv' ); -- set 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 w/ & w/o version
 
−
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
 
−
year = tonumber(year);
 
−
month = tonumber(month);
 
−
if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or -- is year invalid or is month invalid? (doesn't test for future years)
 
−
((7 == year) and (4 > month)) then --or -- when year is 07, is month invalid (before April)?
 
−
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- set error message
 
−
end
 
−
elseif id:match("^%d%d[01]%d%.%d%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%d%dv%d+$") then -- test for the 1501- format w/ & w/o 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 = ' ' .. set_error( 'bad_arxiv' ); -- set error message
 
−
end
 
−
else
 
−
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- arXiv id doesn't match any format
 
−
end
 
−
 
 
−
text = external_link_id({link = handler.link, label = handler.label,
 
−
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, access=handler.access}) .. err_cat;
 
−
 
 
−
if is_set (class) then
 
−
class = ' [[' .. '//arxiv.org/archive/' .. class .. ' ' .. class .. ']]'; -- external link within square brackets, not wikilink
 
−
else
 
−
class = ''; -- empty string for concatenation
 
−
end
 
−
 
−
return text .. class;
 
−
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 must be letter, 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
 
−
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,
 
−
prefix=handler.prefix, id=id, separator=handler.separator, encode=handler.encode,
 
−
access=access});
 
−
 
−
if 19 ~= id:len() then
 
−
err_type = 'length';
 
−
else
 
−
year = id:match ("^(%d%d%d%d)[%a][%a&%.][%a&%.%d][%a&%.%d][%a%d%.]+[%a%.]$") --
 
−
if not year then -- if nil then no pattern match
 
−
err_type = '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 = 'year'; -- year out of bounds
 
−
end
 
−
if id:find('&%.') then
 
−
err_type = 'journal'; -- journal abbreviation must not have '&.' (if it does its missing a letter)
 
−
end
 
−
end
 
−
end
 
−
 
 
−
if is_set (err_type) then -- if there was an error detected
 
−
text = text .. ' ' .. set_error( 'bad_bibcode', {err_type});
 
−
end
 
−
return text;
 
−
end
 
−
 
 
−
 
 
−
--[[--------------------------< B I O R X I V >-----------------------------------------------------------------
 
−
 
 
−
Format bioRxiv id and do simple error checking.  BiorXiv ids are exactly 6 digits.
 
−
The bioRxiv id is the number following the last slash in the bioRxiv-issued DOI:
 
−
https://doi.org/10.1101/078733 -> 078733
 
−
 
 
−
]]
 
−
 
 
−
local function biorxiv(id)
 
−
local handler = cfg.id_handlers['BIORXIV'];
 
−
local err_cat = ''; -- presume that bioRxiv id is valid
 
−
 
−
if nil == id:match("^%d%d%d%d%d%d$") then -- if bioRxiv id has anything but six digits
 
−
err_cat = ' ' .. set_error( 'bad_biorxiv'); -- set an error message
 
−
end
 
−
 
−
return external_link_id({link = handler.link, label = handler.label,
 
−
prefix=handler.prefix,id=id,separator=handler.separator,
 
−
encode=handler.encode, access=handler.access}) .. err_cat;
 
−
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,
 
−
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_error( '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)
 
−
local 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
 
−
if is_set(inactive_year) then
 
−
table.insert( z.error_categories, "Pages with DOIs inactive since " .. inactive_year );
 
−
else
 
−
table.insert( z.error_categories, "Pages with inactive DOIs" ); -- when inactive doesn't contain a recognizable year
 
−
end
 
−
inactive = " (" .. cfg.messages['inactive'] .. " " .. inactive .. ")"
 
−
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
 
−
cat = ' ' .. set_error( 'bad_doi' );
 
−
end
 
−
 
 
−
return text .. cat
 
−
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.
 
−
 
 
−
]]
 
−
 
 
−
local function hdl(id, access)
 
−
local handler = cfg.id_handlers['HDL'];
 
−
 
−
local text = external_link_id({link = handler.link, label = handler.label,
 
−
prefix=handler.prefix,id=id,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
 
−
text = text .. ' ' .. set_error( 'bad_hdl' );
 
−
end
 
−
return text;
 
−
end
 
−
 
 
−
 
 
−
--[[--------------------------< I S B N >----------------------------------------------------------------------
 
  
 
Determines whether an ISBN string is valid
 
Determines whether an ISBN string is valid
Line 425: Line 97:
 
]]
 
]]
  
−
local function isbn( isbn_str )
+
local function check_isbn( isbn_str )
 
if nil ~= isbn_str:match("[^%s-0-9X]") then
 
if nil ~= isbn_str:match("[^%s-0-9X]") then
 
return false, 'invalid character'; -- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
 
return false, 'invalid character'; -- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
Line 450: Line 122:
 
return is_valid_isxn_13 (isbn_str), 'checksum';
 
return is_valid_isxn_13 (isbn_str), 'checksum';
 
end
 
end
−
end
 
−
 
−
 
−
--[[--------------------------< A M A Z O 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 isbn10, if mixed and first character is a digit.
 
−
 
−
This function is positioned here because it calls isbn()
 
−
 
−
]]
 
−
 
−
local function asin(id, domain)
 
−
local err_cat = ""
 
−
 
−
if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
 
−
err_cat = ' ' .. set_error ('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 isbn10
 
−
add_maint_cat ('ASIN');
 
−
elseif not is_set (err_cat) then
 
−
err_cat = ' ' .. set_error ('bad_asin'); -- asin is not isbn10
 
−
end
 
−
elseif not id:match("^%u[%d%u]+$") then
 
−
err_cat = ' ' .. set_error ('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, prefix=handler.prefix .. domain .. "/dp/",
 
−
id=id, encode=handler.encode, separator = handler.separator}) .. err_cat;
 
 
end
 
end
  
Line 506: Line 137:
 
local text;
 
local text;
 
local valid_ismn = true;
 
local valid_ismn = true;
−
local id_copy;
 
  
−
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-–]", "" ); -- strip spaces, hyphens, and endashes from the ismn
  
Line 520: Line 149:
 
-- prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
 
-- prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
 
   
 
   
−
text = table.concat (
+
text="[[" .. handler.link .. "|" .. handler.label .. "]]" .. handler.separator .. id; -- because no place to link to yet
−
{
 
−
make_wikilink (handler.link, handler.label),
 
−
handler.separator,
 
−
id_copy
 
−
}); -- because no place to link to yet
 
  
 
if false == valid_ismn then
 
if false == valid_ismn then
Line 541: Line 165:
 
like this:
 
like this:
  
−
|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
Line 551: Line 175:
  
 
local function issn(id, e)
 
local function issn(id, e)
−
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;
Line 562: Line 186:
 
end
 
end
  
−
id=id:gsub( "[%s-–]", "" ); -- strip spaces, hyphens, and endashes from the issn
+
id=id:gsub( "[%s-–]", "" ); -- strip spaces, hyphens, and endashes from the issn
  
−
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
  
 
if true == valid_issn then
 
if true == valid_issn then
−
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, use the show the invalid issn with error message
 
end
 
end
 
 
Line 587: Line 211:
  
  
−
--[[--------------------------< J F M >-----------------------------------------------------------------------
+
--[[--------------------------< A M A Z O N >------------------------------------------------------------------
  
−
A numerical identifier in the form nn.nnnn.nn
+
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 isbn10, if mixed and first character is a digit.
  
 
]]
 
]]
  
−
local function jfm (id)
+
local function amazon(id, domain)
−
local handler = cfg.id_handlers['JFM'];
+
local err_cat = ""
−
local id_num;
+
 
 +
if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
 +
err_cat =  ' ' .. set_error ('bad_asin'); -- asin is not a mix of 10 uppercase alpha and numeric characters
 +
else
 +
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
 +
if check_isbn( id ) then -- see if asin value is isbn10
 +
add_maint_cat ('ASIN');
 +
elseif not is_set (err_cat) then
 +
err_cat =  ' ' .. set_error ('bad_asin'); -- asin is not isbn10
 +
end
 +
elseif not id:match("^%u[%d%u]+$") then
 +
err_cat =  ' ' .. set_error ('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, prefix=handler.prefix .. domain .. "/dp/",
 +
id=id, encode=handler.encode, separator = handler.separator}) .. err_cat;
 +
end
 +
 
 +
 
 +
--[[--------------------------< 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 9108 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
 +
<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 = '';
 
local err_cat = '';
 +
local text;
 
 
−
id_num = id:match ('^[Jj][Ff][Mm](.*)$'); -- identifier with jfm prefix; extract identifier
+
if id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%d$") or id:match("^%a[%a%.%-]+/[90]%d[01]%d%d%d%dv%d+$") then -- test for the 9108-0703 format w/ & w/o version
 +
year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
 +
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 = ' ' .. set_error( 'bad_arxiv' ); -- set 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 w/ & w/o version
 +
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
 +
year = tonumber(year);
 +
month = tonumber(month);
 +
if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or -- is year invalid or is month invalid? (doesn't test for future years)
 +
((7 == year) and (4 > month)) then --or -- when year is 07, is month invalid (before April)?
 +
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- set error message
 +
end
 +
elseif id:match("^%d%d[01]%d%.%d%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%d%dv%d+$") then -- test for the 1501- format w/ & w/o version
 +
year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d%d[v%d]*$");
 +
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 = ' ' .. set_error( 'bad_arxiv' ); -- set error message
 +
end
 +
else
 +
err_cat = ' ' .. set_error( 'bad_arxiv' ); -- arXiv id doesn't match any format
 +
end
  
−
if is_set (id_num) then
+
text = external_link_id({link = handler.link, label = handler.label,
−
add_maint_cat ('jfm_format');
+
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode}) .. err_cat;
−
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
+
if is_set (class) then
−
id = id_num; -- jfm matches pattern
+
class = ' [[' .. '//arxiv.org/archive/' .. class .. ' ' .. class .. ']]'; -- external link within square brackets, not wikilink
 
else
 
else
−
err_cat = ' ' .. set_error( 'bad_jfm' ); -- set an error message
+
class = ''; -- empty string for concatenation
 
end
 
end
 
 
−
return external_link_id({link = handler.link, label = handler.label,
+
return text .. class;
−
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode}) .. err_cat;
 
 
end
 
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
  
  
Line 633: Line 375:
 
local function lccn(lccn)
 
local function lccn(lccn)
 
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_error( '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_error( '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_error( '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_error( '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_error( '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_error( '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_error( 'bad_lccn' ); -- lccn contains a space, set an error message
 
end
 
end
  
Line 674: Line 416:
  
  
−
--[[--------------------------< M R >--------------------------------------------------------------------------
+
--[[--------------------------< P M I D >----------------------------------------------------------------------
  
−
A seven digit number; if not seven digits, zero-fill leading digits to make seven digits.
+
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 mr (id)
+
local function pmid(id)
−
local handler = cfg.id_handlers['MR'];
+
local test_limit = 30000000; -- update this value as PMIDs approach
−
local id_num;
+
local handler = cfg.id_handlers['PMID'];
−
local id_len;
+
local err_cat = ''; -- presume that PMID is valid
−
local err_cat = '';
 
 
 
−
id_num = id:match ('^[Mm][Rr](%d+)$'); -- identifier with mr prefix
+
if id:match("[^%d]") then -- if PMID has anything but digits
−
 
+
err_cat = ' ' .. set_error( 'bad_pmid' ); -- set an error message
−
if is_set (id_num) then
+
else -- PMID is only digits
−
add_maint_cat ('mr_format');
+
local id_num = tonumber(id); -- convert id to a number for range testing
−
else -- plain number without mr prefix
+
if 1 > id_num or test_limit < id_num then -- if PMID is outside test limit boundaries
−
id_num = id:match ('^%d+$'); -- if here id is all digits
+
err_cat = ' ' .. set_error( 'bad_pmid' ); -- set an error message
−
end
+
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_error( 'bad_mr' ); -- set an error message
 
 
end
 
end
 
 
Line 706: Line 443:
  
  
−
--[[--------------------------< O C L C >----------------------------------------------------------------------
+
--[[--------------------------< I S _ E M B A R G O E D >------------------------------------------------------
  
−
Validate and format an oclc id.  https://www.oclc.org/batchload/controlnumber.en.html
+
Determines if a PMC identifier's online version is embargoed. Compares the date in |embargo= 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
 +
|embargo= was not set in this cite.
  
 
]]
 
]]
  
−
local function oclc (id)
+
local function is_embargoed (embargo)
−
local handler = cfg.id_handlers['OCLC'];
+
if is_set (embargo) then
−
local number;
+
local lang = mw.getContentLanguage();
−
local err_msg = ''; -- empty string for concatenation
+
local good1, embargo_date, good2, todays_date;
 +
good1, embargo_date = pcall( lang.formatDate, lang, 'U', embargo );
 +
good2, todays_date = pcall( lang.formatDate, lang, 'U' );
 
 
−
if id:match('^ocm%d%d%d%d%d%d%d%d$') then -- ocm prefix and 8 digits; 001 field (12 characters)
+
if good1 and good2 then -- if embargo date and today's date are good dates
−
number = id:match('ocm(%d+)'); -- get the number
+
if tonumber( embargo_date ) >= tonumber( todays_date ) then -- is embargo date is in the future?
−
elseif id:match('^ocn%d%d%d%d%d%d%d%d%d$') then -- ocn prefix and 9 digits; 001 field (12 characters)
+
return embargo; -- still embargoed
−
number = id:match('ocn(%d+)'); -- get the number
+
else
−
elseif id:match('^on%d%d%d%d%d%d%d%d%d%d+$') then -- on prefix and 10 or more digits; 001 field (12 characters)
+
add_maint_cat ('embargo')
−
number = id:match('^on(%d%d%d%d%d%d%d%d%d%d+)$'); -- get the number
+
return ''; -- unset because embargo has expired
−
elseif id:match('^%(OCoLC%)[1-9]%d*$') then -- (OCoLC) prefix and variable number digits; no leading zeros; 035 field
+
end
−
number = id:match('%(OCoLC%)([1-9]%d*)'); -- get the number
 
−
if 9 < number:len() then
 
−
number = nil; -- contrain 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; -- contrain to 1 to 10 digits; change this when oclc issues 11-digit numbers
 
 
end
 
end
 
end
 
end
−
 
+
return ''; -- |embargo= not set return empty string
−
if number then -- proper format
 
−
id = number; -- exclude prefix, if any, from external link
 
−
else
 
−
err_msg = ' ' .. set_error( 'bad_oclc' ) -- add an error message if the id is malformed
 
−
end
 
−
 
−
local text = external_link_id({link=handler.link, label=handler.label,
 
−
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 code;
 
−
local handler = cfg.id_handlers['OL'];
 
−
local ident;
 
−
 
−
ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W'; remove OL prefix
 
−
 
 
−
if not is_set (ident) then -- if malformed return an error
 
−
return external_link_id({link=handler.link, label=handler.label,
 
−
prefix=handler.prefix .. 'OL',
 
−
id=id, separator=handler.separator, encode = handler.encode,
 
−
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
 
−
 
 
−
if ( code == "W" ) then
 
−
return external_link_id({link=handler.link, label=handler.label,
 
−
prefix=handler.prefix .. 'works/OL',
 
−
id=id, separator=handler.separator, encode = handler.encode,
 
−
access = access})
 
−
end
 
 
end
 
end
  
Line 811: Line 489:
  
 
local function pmc(id, embargo)
 
local function pmc(id, embargo)
−
local test_limit = 6000000; -- update this value as PMCs approach
+
local test_limit = 5000000; -- 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 = ''; -- presume that PMC is valid
−
local id_num;
+
 
local text;
 
local text;
−
 
−
id_num = id:match ('^[Pp][Mm][Cc](%d+)$'); -- identifier with pmc prefix
 
  
−
if is_set (id_num) then
+
if id:match("[^%d]") then -- if PMC has anything but digits
−
add_maint_cat ('pmc_format');
+
err_cat = ' ' .. set_error( 'bad_pmc' ); -- set an error message
−
else -- plain number without pmc prefix
+
else -- PMC is only digits
−
id_num = id:match ('^%d+$'); -- if here id is all digits
+
local id_num = tonumber(id); -- convert id to a number for range testing
−
end
+
if 1 > id_num or test_limit < id_num then -- if PMC is outside test limit boundaries
−
 
+
err_cat = ' ' .. set_error( 'bad_pmc' ); -- set an error message
−
if is_set (id_num) then -- id_num has a value so test it
 
−
id_num = tonumber(id_num); -- convert id_num to a number for range testing
 
−
if 1 > id_num or test_limit < id_num then -- if PMC is outside test limit boundaries
 
−
err_cat = ' ' .. set_error( 'bad_pmc' ); -- set an error message
 
−
else
 
−
id = tostring (id_num); -- make sure id is a string
 
 
end
 
end
−
else -- when id format incorrect
 
−
err_cat = ' ' .. set_error( 'bad_pmc' ); -- set an error message
 
 
end
 
end
 
 
 
if is_set (embargo) then -- is PMC is still embargoed?
 
if is_set (embargo) then -- is PMC is still embargoed?
−
text = table.concat ( -- still embargoed so no external link
+
text="[[" .. handler.link .. "|" .. handler.label .. "]]:" .. handler.separator .. id .. err_cat; -- still embargoed so no external link
−
{
 
−
make_wikilink (handler.link, handler.label),
 
−
handler.separator,
 
−
id,
 
−
err_cat
 
−
});
 
 
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, -- 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}) .. err_cat;
 
end
 
end
 
return text;
 
return text;
Line 852: Line 514:
  
  
−
--[[--------------------------< P M I D >----------------------------------------------------------------------
+
--[[--------------------------< 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.
  
−
Format PMID and do simple error checking.  PMIDs are sequential numbers beginning at 1 and counting up.  This
+
DOI names are case-insensitive and can incorporate any printable Unicode characters so the test for spaces, endash,
−
code checks the PMID to see that it contains only digits and is less than test_limit; the value in local variable
+
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
−
test_limit will need to be updated periodically as more PMIDs are issued.
+
if ever used in doi names.
  
 
]]
 
]]
  
−
local function pmid(id)
+
local function doi(id, inactive)
−
local test_limit = 32000000; -- update this value as PMIDs approach
+
local cat = ""
−
local handler = cfg.id_handlers['PMID'];
+
local handler = cfg.id_handlers['DOI'];
−
local err_cat = ''; -- presume that PMID is valid
 
 
 
−
if id:match("[^%d]") then -- if PMID has anything but digits
+
local text;
−
err_cat = ' ' .. set_error( 'bad_pmid' ); -- set an error message
+
if is_set(inactive) then
−
else -- PMID is only digits
+
local inactive_year = inactive:match("%d%d%d%d") or ''; -- try to get the year portion from the inactive date
−
local id_num = tonumber(id); -- convert id to a number for range testing
+
text = "[[" .. handler.link .. "|" .. handler.label .. "]]:" .. id;
−
if 1 > id_num or test_limit < id_num then -- if PMID is outside test limit boundaries
+
if is_set(inactive_year) then
−
err_cat = ' ' .. set_error( 'bad_pmid' ); -- set an error message
+
table.insert( z.error_categories, "Pages with DOIs inactive since " .. inactive_year );
 +
else
 +
table.insert( z.error_categories, "Pages with inactive DOIs" ); -- when inactive doesn't contain a recognizable year
 
end
 
end
 +
inactive = " (" .. cfg.messages['inactive'] .. " " .. inactive .. ")"
 +
else
 +
text = external_link_id({link = handler.link, label = handler.label,
 +
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
 +
inactive = ""
 
end
 
end
−
+
 
−
return external_link_id({link = handler.link, label = handler.label,
+
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
−
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode}) .. err_cat;
+
cat = ' ' .. set_error( 'bad_doi' );
 +
end
 +
return text .. inactive .. cat
 
end
 
end
  
  
−
--[[--------------------------< S S R N >----------------------------------------------------------------------
+
--[[--------------------------< H D L >------------------------------------------------------------------------
  
−
Format an ssrn, do simple error checking
+
Formats an HDL with minor error checking.
  
−
SSRNs are sequential numbers beginning at 100? and counting up. This code checks the ssrn to see that it is
+
HDL names contain two parts: prefix and suffix separated by a forward slash.
−
only digits and is greater than 99 and less than test_limit; the value in local variable test_limit will need
+
Prefix: character string using any character in the UCS-2 character set except '/'
−
to be updated periodically as more SSRNs are issued.
+
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.
  
 
]]
 
]]
  
−
local function ssrn (id)
+
local function hdl(id)
−
local test_limit = 3500000; -- update this value as SSRNs approach
+
local handler = cfg.id_handlers['HDL'];
−
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
+
local text = external_link_id({link = handler.link, label = handler.label,
 +
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
  
−
if is_set (id_num) then -- id_num has a value so test it
+
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
−
id_num = tonumber(id_num); -- convert id_num to a number for range testing
+
text = text .. ' ' .. set_error( 'bad_hdl' );
−
if 100 > id_num or test_limit < id_num then -- if SSRN is outside test limit boundaries
 
−
err_cat = ' ' .. set_error( 'bad_ssrn' ); -- set an error message
 
−
end
 
−
else -- when id format incorrect
 
−
err_cat = ' ' .. set_error( 'bad_ssrn' ); -- set an error message
 
 
end
 
end
−
+
return text;
−
text = external_link_id({link = handler.link, label = handler.label,
+
end
−
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, access=handler.access}) .. err_cat;
+
 
 +
 
 +
--[[--------------------------< O P E N L I B R A R Y >--------------------------------------------------------
 +
 
 +
Formats an OpenLibrary link, and checks for associated errors.
 +
 
 +
]]
 +
 
 +
local function openlibrary(id)
 +
local code = id:match("^%d+([AMW])$"); -- only digits followed by 'A', 'M', or 'W'
 +
local handler = cfg.id_handlers['OL'];
  
−
return text;
+
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})
 +
elseif ( 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})
 +
elseif ( code == "W" ) then
 +
return external_link_id({link=handler.link, label=handler.label,
 +
prefix=handler.prefix .. 'works/OL',
 +
id=id, separator=handler.separator, encode = handler.encode})
 +
else
 +
return external_link_id({link=handler.link, label=handler.label,
 +
prefix=handler.prefix .. 'OL',
 +
id=id, separator=handler.separator, encode = handler.encode}) .. ' ' .. set_error( 'bad_ol' );
 +
end
 
end
 
end
  
  
−
--[[--------------------------< U S E N E T _ I D >------------------------------------------------------------
+
--[[--------------------------< M E S S A G E _ I D >----------------------------------------------------------
  
 
Validate and format a usenet message id.  Simple error checking, looks for 'id-left@id-right' not enclosed in
 
Validate and format a usenet message id.  Simple error checking, looks for 'id-left@id-right' not enclosed in
Line 921: Line 625:
 
]]
 
]]
  
−
local function usenet_id (id)
+
local function message_id (id)
 
local handler = cfg.id_handlers['USENETID'];
 
local handler = cfg.id_handlers['USENETID'];
  
Line 928: Line 632:
 
   
 
   
 
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_error( 'bad_message_id' ) -- add an error message if the message id is invalid
 
end  
 
end  
 
 
Line 935: Line 639:
  
  
−
--[[--------------------------< Z B L >-----------------------------------------------------------------------
+
--[[--------------------------< O C L C >----------------------------------------------------------------------
  
−
A numerical identifier in the form nnnn.nnnnn - leading zeros in the first quartet optional
+
Validate and format an oclc id. https://www.oclc.org/batchload/controlnumber.en.html
  
 
]]
 
]]
  
−
local function zbl (id)
+
local function oclc (id)
−
local handler = cfg.id_handlers['ZBL'];
+
local handler = cfg.id_handlers['OCLC'];
−
local id_num;
+
local number;
−
local err_cat = '';
+
local err_msg = ''; -- empty string for concatenation
 
 
−
id_num = id:match ('^[Zz][Bb][Ll](.*)$'); -- identifier with zbl prefix; extract identifier
+
if id:match('^ocm%d%d%d%d%d%d%d%d$') then -- ocm prefix and 8 digits; 001 field (12 characters)
−
 
+
number = id:match('ocm(%d+)'); -- get the number
−
if is_set (id_num) then
+
elseif id:match('^ocn%d%d%d%d%d%d%d%d%d$') then -- ocn prefix and 9 digits; 001 field (12 characters)
−
add_maint_cat ('zbl_format');
+
number = id:match('ocn(%d+)'); -- get the number
−
else -- plain number without zbl prefix
+
elseif id:match('^on%d%d%d%d%d%d%d%d%d%d+$') then -- on prefix and 10 or more digits; 001 field (12 characters)
−
id_num = id; -- if here id does not have prefix
+
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; -- contrain 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; -- contrain to 1 to 10 digits; change this when oclc issues 11-digit numbers
 +
end
 
end
 
end
  
−
if id_num:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then
+
if number then -- proper format
−
id = id_num; -- id matches pattern
+
id = number; -- exclude prefix, if any, from external link
 
else
 
else
−
err_cat = ' ' .. set_error( 'bad_zbl' ); -- set an error message
+
err_msg = ' ' .. set_error( 'bad_oclc' ) -- add an error message if the id is malformed
 
end
 
end
 
 
−
return external_link_id({link = handler.link, label = handler.label,
+
local text = external_link_id({link=handler.link, label=handler.label,
−
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_msg;
 +
 
 +
return text;
 
end
 
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 >--------------------------------------------------------
 
--[[--------------------------< B U I L D _ I D _ L I S T >--------------------------------------------------------
Line 984: Line 698:
 
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 -- k is uc identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
 
-- fallback to read-only cfg
 
-- fallback to read-only cfg
−
handler = setmetatable( { ['id'] = v, ['access'] = options.IdAccessLevels[k] }, fallback(k) );
+
handler = setmetatable( { ['id'] = v }, fallback(k) );
−
 
+
 
if handler.mode == 'external' then
 
if handler.mode == 'external' then
 
table.insert( new_list, {handler.label, external_link_id( handler ) } );
 
table.insert( new_list, {handler.label, external_link_id( handler ) } );
Line 992: Line 706:
 
elseif handler.mode ~= 'manual' then
 
elseif handler.mode ~= 'manual' then
 
error( cfg.messages['unknown_ID_mode'] );
 
error( cfg.messages['unknown_ID_mode'] );
 +
elseif k == 'DOI' then
 +
table.insert( new_list, {handler.label, doi( v, options.DoiBroken ) } );
 +
elseif k == 'HDL' then
 +
table.insert( new_list, {handler.label, hdl( v ) } );
 
elseif k == 'ARXIV' then
 
elseif k == 'ARXIV' then
 
table.insert( new_list, {handler.label, arxiv( v, options.Class ) } );  
 
table.insert( new_list, {handler.label, arxiv( v, options.Class ) } );  
 
elseif k == 'ASIN' then
 
elseif k == 'ASIN' then
−
table.insert( new_list, {handler.label, asin( v, options.ASINTLD ) } );  
+
table.insert( new_list, {handler.label, amazon( v, options.ASINTLD ) } );  
−
elseif k == 'BIBCODE' then
+
elseif k == 'LCCN' then
−
table.insert( new_list, {handler.label, bibcode( v, handler.access ) } );
+
table.insert( new_list, {handler.label, lccn( v ) } );
−
elseif k == 'BIORXIV' then
+
elseif k == 'OL' or k == 'OLA' then
−
table.insert( new_list, {handler.label, biorxiv( v ) } );
+
table.insert( new_list, {handler.label, openlibrary( v ) } );
−
elseif k == 'CITESEERX' then
+
elseif k == 'PMC' then
−
table.insert( new_list, {handler.label, citeseerx( v ) } );
+
table.insert( new_list, {handler.label, pmc( v, options.Embargo ) } );
−
elseif k == 'DOI' then
+
elseif k == 'PMID' then
−
table.insert( new_list, {handler.label, doi( v, options.DoiBroken, handler.access ) } );
+
table.insert( new_list, {handler.label, pmid( v ) } );
 +
elseif k == 'OCLC' then
 +
table.insert( new_list, {handler.label, oclc( v ) } );
 +
elseif k == 'ISMN' then
 +
table.insert( new_list, {handler.label, ismn( v ) } );
 +
elseif k == 'ISSN' then
 +
table.insert( new_list, {handler.label, issn( v ) } );
 
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 ) } ); -- true distinguishes eissn from issn
−
elseif k == 'HDL' then
 
−
table.insert( new_list, {handler.label, hdl( v, handler.access ) } );
 
 
elseif k == 'ISBN' then
 
elseif k == 'ISBN' then
 
local ISBN = internal_link_id( handler );
 
local ISBN = internal_link_id( handler );
 
local check;
 
local check;
 
local err_type = '';
 
local err_type = '';
−
check, err_type = isbn( v );
+
-- if not check_isbn( v ) and not is_set(options.IgnoreISBN) then
 +
-- ISBN = ISBN .. set_error( 'bad_isbn', {}, false, " ", "" );
 +
-- end
 +
check, err_type = check_isbn( v );
 
if not check then
 
if not check then
 
if is_set(options.IgnoreISBN) then -- ISBN is invalid; if |ignore-isbn-error= set
 
if is_set(options.IgnoreISBN) then -- ISBN is invalid; if |ignore-isbn-error= set
Line 1,020: Line 745:
 
end
 
end
 
elseif is_set(options.IgnoreISBN) then -- ISBN is OK; if |ignore-isbn-error= set
 
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
+
add_maint_cat ('ignore_isbn_err'); -- because |ignore-isbn-error= unnecessary
 
end
 
end
 
table.insert( new_list, {handler.label, ISBN } );
 
table.insert( new_list, {handler.label, ISBN } );
−
elseif k == 'ISMN' then
 
−
table.insert( new_list, {handler.label, ismn( v ) } );
 
−
elseif k == 'ISSN' then
 
−
table.insert( new_list, {handler.label, issn( v ) } );
 
−
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 == 'SSRN' then
 
−
table.insert( new_list, {handler.label, ssrn( v ) } );
 
 
elseif k == 'USENETID' then
 
elseif k == 'USENETID' then
−
table.insert( new_list, {handler.label, usenet_id( v ) } );
+
table.insert( new_list, {handler.label, message_id( v ) } );
−
elseif k == 'ZBL' then
 
−
table.insert( new_list, {handler.label, zbl( v ) } );
 
 
else
 
else
 
error( cfg.messages['unknown_manual_ID'] );
 
error( cfg.messages['unknown_manual_ID'] );
Line 1,053: Line 756:
 
 
 
local function comp( a, b ) -- used in following table.sort()
 
local function comp( a, b ) -- used in following table.sort()
−
return a[1]:lower() < b[1]:lower();
+
return a[1] < b[1];
 
end
 
end
 
 
Line 1,080: Line 783:
 
end
 
end
 
return id_list;
 
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.
 
−
 
−
]]
 
−
 
−
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;
 
−
local k_lower = string.lower(k);
 
−
if is_set(access_param) then
 
−
local access_level = args[access_param];
 
−
if is_set(access_level) then
 
−
if not in_array (access_level:lower(), cfg.keywords['id-access']) then
 
−
table.insert( z.message_tail, { set_error( 'invalid_param_val', {access_param, access_level}, true ) } );
 
−
access_level = nil;
 
−
end
 
−
if not is_set(id_list[k]) then
 
−
table.insert( z.message_tail, { set_error( 'param_access_requires_param', {k_lower}, true ) } );
 
−
end
 
−
if is_set(access_level) then
 
−
access_level = access_level:lower();
 
−
end
 
−
id_accesses_list[k] = access_level;
 
−
end
 
−
end
 
−
end
 
−
return id_accesses_list;
 
 
end
 
end
  
Line 1,131: Line 800:
 
select_one = utilities_page_ptr.select_one;
 
select_one = utilities_page_ptr.select_one;
 
add_maint_cat = utilities_page_ptr.add_maint_cat;
 
add_maint_cat = utilities_page_ptr.add_maint_cat;
−
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
 
z = utilities_page_ptr.z; -- table of tables in Module:Citation/CS1/Utilities
 
end
 
end
 +
  
  
Line 1,141: Line 809:
 
build_id_list = build_id_list,
 
build_id_list = build_id_list,
 
extract_ids = extract_ids,
 
extract_ids = extract_ids,
−
extract_id_access_levels = extract_id_access_levels,
 
 
is_embargoed = is_embargoed;
 
is_embargoed = is_embargoed;
 
set_selected_modules = set_selected_modules;
 
set_selected_modules = set_selected_modules;
 
}
 
}

Revision as of 17:08, 31 January 2018

{{#lst:Module:Citation/CS1/doc|header}} This page renders and performs error checking on the various named identifiers supported by Module:Citation/CS1.

{{#lst:Module:Citation/CS1/doc|module_components_table}} 


local identifiers = {};


--[[--------------------------< 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;					-- functions 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


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

Formats a wiki style external link

]]

local function external_link_id(options)
	local url_string = options.id;
	if options.encode == true or options.encode == nil then
		url_string = mw.uri.encode( url_string );
	end
	return mw.ustring.format( '[[%s|%s]]%s[%s%s%s %s]',
		options.link, options.label, options.separator or "&nbsp;",
		options.prefix, url_string, options.suffix or "",
		mw.text.nowiki(options.id)
	);
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)
	return mw.ustring.format( '[[%s|%s]]%s[[%s%s%s|%s]]',
		options.link, options.label, options.separator or "&nbsp;",
		options.prefix, options.id, options.suffix or "",
		mw.text.nowiki(options.id)
	);
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 check_isbn().

If the number is valid the result will be 0. Before calling this function, issbn/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


--[[--------------------------< C H E C K _ I S B N >------------------------------------------------------------

Determines whether an ISBN string is valid

]]

local function check_isbn( isbn_str )
	if nil ~= isbn_str:match("[^%s-0-9X]") then
		return false, 'invalid character';										-- fail if isbn_str contains anything but digits, hyphens, or the uppercase X
	end
	isbn_str = isbn_str:gsub( "-", "" ):gsub( " ", "" );						-- remove hyphens and spaces
	local len = isbn_str:len();
 
	if len ~= 10 and len ~= 13 then
		return false, 'length';													-- fail if incorrect length
	end

	if len == 10 then
		if isbn_str:match( "^%d*X?$" ) == nil then								-- fail if isbn_str has 'X' anywhere but last position
			return false, 'invalid form';									
		end
		return is_valid_isxn(isbn_str, 10), 'checksum';
	else
		if isbn_str:match( "^%d+$" ) == nil then
			return false, 'invalid character';									-- fail if isbn13 is not all digits
		end
		if isbn_str:match( "^97[89]%d*$" ) == nil then
			return false, 'invalid prefix';										-- fail when isbn13 does not begin with 978 or 979
		end
		return is_valid_isxn_13 (isbn_str), 'checksum';
	end
end


--[[--------------------------< 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
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;

	id=id:gsub( "[%s-–]", "" );													-- strip spaces, hyphens, and endashes from the ismn

	if 13 ~= id:len() or id:match( "^9790%d*$" ) == nil then					-- ismn must be 13 digits and begin 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,		-- use this (or external version) when there is some place to link to
--		prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
 
	text="[[" .. handler.link .. "|" .. handler.label .. "]]" .. handler.separator .. id;		-- because no place to link to yet

	if false == valid_ismn then
		text = text .. ' ' .. set_error( '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)
	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;
	
	if e then
		 handler = cfg.id_handlers['EISSN'];
	else
		 handler = cfg.id_handlers['ISSN'];
	end

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

	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, use the show the invalid issn with error message
	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
end


--[[--------------------------< A M A Z O 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 isbn10, if mixed and first character is a digit.

]]

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

	if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
		err_cat =  ' ' .. set_error ('bad_asin');								-- asin is not a mix of 10 uppercase alpha and numeric characters
	else
		if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then								-- if 10-digit numeric (or 9 digits with terminal X)
			if check_isbn( id ) then												-- see if asin value is isbn10
				add_maint_cat ('ASIN');
			elseif not is_set (err_cat) then
				err_cat =  ' ' .. set_error ('bad_asin');						-- asin is not isbn10
			end
		elseif not id:match("^%u[%d%u]+$") then
			err_cat =  ' ' .. set_error ('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, prefix=handler.prefix .. domain .. "/dp/",
		id=id, encode=handler.encode, separator = handler.separator}) .. err_cat;
end


--[[--------------------------< 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 9108 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
	<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 = '';
	local 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
		year, month = id:match("^%a[%a%.%-]+/([90]%d)([01]%d)%d%d%d[v%d]*$");
		year = tonumber(year);
		month = tonumber(month);
		if ((not (90 < year or 8 > year)) or (1 > month or 12 < month)) or		-- if invalid year or invalid month
			((91 == year and 7 > month) or (7 == year and 3 < month)) then		-- if years ok, are starting and ending months ok?
				err_cat = ' ' .. set_error( 'bad_arxiv' );						-- set error message
		end
	elseif id:match("^%d%d[01]%d%.%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%dv%d+$") then	-- test for the 0704-1412 w/ & w/o version
		year, month = id:match("^(%d%d)([01]%d)%.%d%d%d%d[v%d]*$");
		year = tonumber(year);
		month = tonumber(month);
		if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or			-- is year invalid or is month invalid? (doesn't test for future years)
			((7 == year) and (4 > month)) then --or									-- when year is 07, is month invalid (before April)?
				err_cat = ' ' .. set_error( 'bad_arxiv' );						-- set error message
		end
	elseif id:match("^%d%d[01]%d%.%d%d%d%d%d$") or id:match("^%d%d[01]%d%.%d%d%d%d%dv%d+$") then	-- test for the 1501- format w/ & w/o 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 = ' ' .. set_error( 'bad_arxiv' );							-- set error message
		end
	else
		err_cat = ' ' .. set_error( 'bad_arxiv' );								-- arXiv id doesn't match any format
	end

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

	if is_set (class) then
		class = ' [[' .. '//arxiv.org/archive/' .. class .. ' ' .. class .. ']]';	-- external link within square brackets, not wikilink
	else
		class = '';																-- empty string for concatenation
	end
	
	return text .. class;
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


--[[--------------------------< 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 lower case alpha
length = 10 then lccn[1] and lccn[2] are both lower case 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 = 12 then lccn[1] and lccn[2] are both lower case 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_error( '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_error( '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_error( '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_error( '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_error( 'bad_lccn' );	-- no match, set an error message
		end
	else
		err_cat = ' ' .. set_error( 'bad_lccn' );		-- wrong length, set an error message
	end

	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
	end

	return external_link_id({link = handler.link, label = handler.label,
			prefix=handler.prefix,id=lccn,separator=handler.separator, encode=handler.encode}) .. err_cat;
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 test_limit = 30000000;						-- update this value as PMIDs approach
	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_error( 'bad_pmid' );		-- set an error message
	else												-- PMID is only digits
		local id_num = tonumber(id);					-- convert id to a number for range testing
		if 1 > id_num or test_limit < id_num then		-- if PMID is outside test limit boundaries
			err_cat = ' ' .. set_error( 'bad_pmid' );	-- set an error message
		end
	end
	
	return external_link_id({link = handler.link, label = handler.label,
			prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode}) .. err_cat;
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 |embargo= 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
|embargo= was not set in this cite.

]]

local function is_embargoed (embargo)
	if is_set (embargo) then
		local lang = mw.getContentLanguage();
		local good1, embargo_date, good2, todays_date;
		good1, embargo_date = pcall( lang.formatDate, lang, 'U', embargo );
		good2, todays_date = pcall( lang.formatDate, lang, 'U' );
	
		if good1 and good2 then													-- if embargo date and today's date are good dates
			if tonumber( embargo_date ) >= tonumber( todays_date ) then			-- is embargo date is in the future?
				return embargo;													-- still embargoed
			else
				add_maint_cat ('embargo')
				return '';														-- unset because embargo has expired
			end
		end
	end
	return '';																	-- |embargo= not set return empty string
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 test_limit = 5000000;							-- update this value as PMCs approach
	local handler = cfg.id_handlers['PMC'];
	local err_cat =  '';								-- presume that PMC is valid
	
	local text;

	if id:match("[^%d]") then							-- if PMC has anything but digits
		err_cat = ' ' .. set_error( 'bad_pmc' );			-- set an error message
	else												-- PMC is only digits
		local id_num = tonumber(id);					-- convert id to a number for range testing
		if 1 > id_num or test_limit < id_num then		-- if PMC is outside test limit boundaries
			err_cat = ' ' .. set_error( 'bad_pmc' );		-- set an error message
		end
	end
	
	if is_set (embargo) then													-- is PMC is still embargoed?
		text="[[" .. handler.link .. "|" .. handler.label .. "]]:" .. handler.separator .. id .. err_cat;	-- still embargoed so no external link
	else
		text = external_link_id({link = handler.link, label = handler.label,			-- no embargo date or embargo has expired, ok to link to article
			prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode}) .. err_cat;
	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)
	local 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
		text = "[[" .. handler.link .. "|" .. handler.label .. "]]:" .. id;
		if is_set(inactive_year) then
			table.insert( z.error_categories, "Pages with DOIs inactive since " .. inactive_year );
		else
			table.insert( z.error_categories, "Pages with inactive DOIs" );	-- when inactive doesn't contain a recognizable year
		end
		inactive = " (" .. cfg.messages['inactive'] .. " " .. inactive .. ")" 
	else 
		text = external_link_id({link = handler.link, label = handler.label,
			prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
		inactive = "" 
	end

	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
		cat = ' ' .. set_error( 'bad_doi' );
	end
	return text .. inactive .. cat 
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.

]]

local function hdl(id)
	local handler = cfg.id_handlers['HDL'];
	
	local text = external_link_id({link = handler.link, label = handler.label,
			prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})

	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
		text = text .. ' ' .. set_error( 'bad_hdl' );
	end
	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)
	local code = id:match("^%d+([AMW])$");										-- only digits followed by 'A', 'M', or 'W'
	local handler = cfg.id_handlers['OL'];

	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})
	elseif ( 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})
	elseif ( code == "W" ) then
		return external_link_id({link=handler.link, label=handler.label,
			prefix=handler.prefix .. 'works/OL',
			id=id, separator=handler.separator,	encode = handler.encode})
	else
		return external_link_id({link=handler.link, label=handler.label,
			prefix=handler.prefix .. 'OL',
			id=id, separator=handler.separator,	encode = handler.encode}) .. ' ' .. set_error( 'bad_ol' );
	end
end


--[[--------------------------< M E S S A G E _ 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 message_id (id)
	local handler = cfg.id_handlers['USENETID'];

	local text = external_link_id({link = handler.link, label = handler.label,
		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_error( 'bad_message_id' )						-- add an error message if the message id is invalid
	end 
	
	return text
end


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

Validate and format an oclc id.  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*$') then									-- (OCoLC) prefix and variable number digits; no leading zeros; 035 field
		number = id:match('%(OCoLC%)([1-9]%d*)');								-- get the number
		if 9 < number:len() then
			number = nil;														-- contrain to 1 to 9 digits; change this when oclc issues 10-digit numbers
		end
	elseif id:match('^%d+$') then												-- no prefix
		number = id;															-- get the number
		if 10 < number:len() then
			number = nil;														-- contrain 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_error( 'bad_oclc' )								-- add an error message if the id is malformed
	end
	
	local text = external_link_id({link=handler.link, label=handler.label,
		prefix=handler.prefix, id=id, separator=handler.separator, encode=handler.encode}) .. err_msg;

	return text;
end


--[[--------------------------< 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 function fallback(k) return { __index = function(t,i) return cfg.id_handlers[k][i] end } end;
	
	for k, v in pairs( id_list ) do												-- k is uc identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
		-- fallback to read-only cfg
		handler = setmetatable( { ['id'] = v }, 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 == 'DOI' then
			table.insert( new_list, {handler.label, doi( v, options.DoiBroken ) } );
		elseif k == 'HDL' then
			table.insert( new_list, {handler.label, hdl( v ) } );
		elseif k == 'ARXIV' then
			table.insert( new_list, {handler.label, arxiv( v, options.Class ) } ); 
		elseif k == 'ASIN' then
			table.insert( new_list, {handler.label, amazon( v, options.ASINTLD ) } ); 
		elseif k == 'LCCN' then
			table.insert( new_list, {handler.label, lccn( v ) } );
		elseif k == 'OL' or k == 'OLA' then
			table.insert( new_list, {handler.label, openlibrary( v ) } );
		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 == 'OCLC' then
			table.insert( new_list, {handler.label, oclc( v ) } );
		elseif k == 'ISMN' then
			table.insert( new_list, {handler.label, ismn( v ) } );
		elseif k == 'ISSN' then
			table.insert( new_list, {handler.label, issn( v ) } );
		elseif k == 'EISSN' then
			table.insert( new_list, {handler.label, issn( v, true ) } );		-- true distinguishes eissn from issn
		elseif k == 'ISBN' then
			local ISBN = internal_link_id( handler );
			local check;
			local err_type = '';
--			if not check_isbn( v ) and not is_set(options.IgnoreISBN) then
--				ISBN = ISBN .. set_error( 'bad_isbn', {}, false, " ", "" );
--			end
			check, err_type = check_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 == 'USENETID' then
			table.insert( new_list, {handler.label, message_id( v ) } );
		else
			error( cfg.messages['unknown_manual_ID'] );
		end
	end
	
	local function comp( a, b )	-- used in following table.sort()
		return a[1] < b[1];
	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 is more than one alias exists in args

]]

local function extract_ids( args )
	local id_list = {};															-- list of identifiers found in args
	for k, v in pairs( cfg.id_handlers ) do										-- k is uc 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
		if is_set(v) then id_list[k] = v; end									-- if found in args, add identifier to our list
	end
	return id_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;

	is_set = utilities_page_ptr.is_set;											-- import functions from select Module:Citation/CS1/Utilities module
	in_array = utilities_page_ptr.in_array;
	set_error = utilities_page_ptr.set_error;
	select_one = utilities_page_ptr.select_one;
	add_maint_cat = utilities_page_ptr.add_maint_cat;

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



return {
	build_id_list = build_id_list,
	extract_ids = extract_ids,
	is_embargoed = is_embargoed;
	set_selected_modules = set_selected_modules;
	}
Retrieved from "http://shipbucket.com/wiki/index.php?title=Module:Citation/CS1/Identifiers&oldid=45"