]]
local is_set, in_array, set_error, select_one, add_maint_cat, substitute, make_wikilink; -- 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
--============================<< 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 >----------------------------------------------
end
return mwtable.ustring.formatconcat ({ make_wikilink ( '[[%s|%s]]%s%s', options.link, options.label), options.separator or "' "', ext_link });
end
local function internal_link_id(options)
return mwtable.ustring.formatconcat ( '[[%s|%s]]%s[[%s%s%s|%s]]', { make_wikilink (options.link, options.label), options.separator or "' "', 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 isin 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
ISBN-10 and ISSN validator code calculates checksum across all isbn/issn digits including the check digit.
ISBN-13 is checked in check_isbnisbn().
If the number is valid the result will be 0. Before calling this function, issbn/issn must be checked for length
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
--[[--------------------------< 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.
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
--[[--------------------------< C H N O R M A L I Z E C K _ I S B L C C N >------------------------------------------------------------
Determines whether an ISBN 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 validsix.
Returns a normalized lccn for lccn() to validate. There is no error checking (step 3.b.1) performed in this function.
]]
local function check_isbnnormalize_lccn ( isbn_str lccn) if nil ~lccn = isbn_strlccn:matchgsub ("[^%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 end1. strip whitespace
if len nil ~== 10 then if isbn_str:matchstring.find ( "^%d*X?$" ) == nil then -- fail if isbn_str has 'X' anywhere but last position return falselccn, 'invalid form/'; end return is_valid_isxn(isbn_str, 10), 'checksum'; elsethen if isbn_strlccn = lccn:match( "^%d+$(.-)/" ) == nil then return false, 'invalid character'; -- fail if isbn13 is not 2. remove forward slash and all digits end if isbn_str:match( "^97[89]%d*$" ) == nil then return false, 'invalid prefix'; -- fail when isbn13 does not begin with 978 or 979 end return is_valid_isxn_13 (isbn_str), 'checksum';character to the right of it
end
end
local prefix
local suffix
prefix, suffix = lccn:match ("(.+)%-(.+)"); -- 3.a remove hyphen by splitting the string into prefix and suffix
--[[--------------------------< I S M N >---------------------------------------------------------------------- Determines whether an ISMN string is valid. Similar to isbn-13, ismn is 13 digits begining 979-0-... and uses thesame check digit calculations. See http://www.ismn-international.org/download/Web_ISMN_Users_Manual_2008-6.pdfsection 2, pages 9–12. ]] local function ismn (id) local handler if nil ~= cfg.id_handlers['ISMN']; local text; local valid_ismn = true; local id_copy; id_copy = id; suffix then -- save if there was a copy because this testing is destructivehyphen id suffix=id:gsubstring.rep( "[%s-–]0", "" ); 6-- strip spaces, hyphens, and endashes from the ismn if 13 ~= id:string.len(suffix) or id:match( "^9790%d*$" ) == nil then .. suffix; -- ismn must be 13 digits and begin 97903.b.2 left fill the suffix with 0s if suffix length less than 6 valid_ismn lccn= false; else valid_ismn=is_valid_isxn_13 (id)prefix..suffix; -- validate ismnreassemble the lccn
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_copy; -- 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 buthas separated the two groups of four digits with a space. When that condition occurred, the resulting link lookedlike 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 lengthand makes sure that the checkdigit agrees with the calculated value. Incorrect length (8 digits), charactersother than 0-9 and X, or checkdigit / calculated value mismatch will all cause a check issn error message. Theissn 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']lccn;
end
id=id:gsub( "[%s-–]", "" ); -- strip spaces, hyphens, and endashes from the issn if 8 ~= id:len() or nil == id:match( "^%d*X?$" ) then -- validate the issn: 8 digits long, containing only 0-9 or X in the last position valid_issn=false; -- wrong length or improper character else valid_issn=is_valid_isxn(id, 8); -- validate issn end if true == valid_issn then id = string.sub( id, 1, 4 ) .. "-" .. string.sub( id, 5 ); -- if valid, display correctly formatted version else id = issn_copy; -- if not valid, use the show the invalid issn with error message end text = external_link_id({link = handler.link, label = handler.label, prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode}) if false == valid_issn then text = text .. ' ' .. set_error( 'bad_issn', e and 'e' or '' ) -- add an error message if the issn is invalid end return textend --[[--------------------------=========< A M A Z < I D E N T I F I E R F U N C T I O N S >>------------------------------------------------------------------ Formats a link to Amazon. Do simple error checking: asin must be mix of 10 numeric or uppercase alphacharacters. If a mix, first character must be uppercase alpha; if all numeric, asins must be 10-digitisbn. If 10-digit isbn, add a maintenance category so a bot or awb script can replace |asin= with |isbn=.Error message if not 10 characters, if not isbn10, if mixed and first character is a digit. ]] local function amazon(id, domain) local err_cat = "" if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") 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 >--------------------------------------------------------------------
--[[--------------------------< B I B C O R X I V D E >--------------------------------------------------------------------
Format bioRxiv id Validates (sort of) and do simple error checking. BiorXiv ids are exactly 6 digits.The bioRxiv formats a bibcode id is the number following the last slash in the bioRxiv-issued DOI:https://doi.org/10.1101/078733 -> 078733
]]Format for bibcodes is specified here: http://adsabs.harvard.edu/abs_doc/help_pages/data.html#bibcodes
local function biorxiv(id)But, this: 2015arXiv151206696F is apparently valid so apparently, the only things that really matter are length, 19 characters local handler = cfgand first four digits must be a year.id_handlers['BIORXIV']; This function makes these tests: local err_cat = ''; -- presume that bioRxiv id is validlength must be 19 characters characters in position if nil == id:match("^%d%d%d%d%d%d$") then -- if bioRxiv id has anything but six 1–4 must be digitsand must represent a year in the range of 1000 – next year err_cat = ' ' .. set_error5 must be a letter 6 must be letter, ampersand, or dot ( 'bad_biorxiv')ampersand cannot directly precede a dot; -- set an error message end return external_link_id({link = handler&.link, label = handler.label,) prefix=handler.prefix 7–8 must be letter,id=iddigit,separator=handler.separatorampersand, encode=handleror dot (ampersand cannot directly precede a dot; &.encode, access=handler.access}) .. err_cat;end 9–18 must be letter, digit, or dot 19 must be a letter or dot
--[[--------------------------< 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 bibcode (lccnid, access) lccn local handler = lccn:gsub ("%s", ""); -- 1cfg. strip whitespaceid_handlers['BIBCODE']; local err_type; local year;
if nil ~local text = external_link_id({link= stringhandler.find (lccnlink,'/') then lccn label= lccn:match ("(handler.-)/"); -- 2. remove forward slash and all character to the right of itlabel, end local prefix local suffix prefix, suffix = lccn:match ("(.+)%-(.+)"); -- 3handler.a remove hyphen by splitting the string into prefix and suffix if nil ~, id= suffix then -- if there was a hyphen suffixid, separator=stringhandler.rep("0"separator, 6-stringencode=handler.len (suffix)) .. suffix; -- 3.b.2 left fill the suffix with 0s if suffix length less than 6encode, lccnaccess=prefix..suffixaccess}); -- 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 ofthe 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 digitslength = 9 then lccn[1] is lower case alphalength = 10 then lccn[1] and lccn[2] are both lower case alpha or both digitslength = 11 then lccn[1] is lower case alpha, lccn[2] and lccn[3] are both lower case alpha or both digitslength = 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 if 19 ~= 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 err_type = ' length' .. set_error( 'bad_lccn' ); -- set an error message end elseif 9 == len then -- LCCN should be addddddddelse if nil =year = id:match("%l^(%d%d%d%d)[%a][%a&%.][%a&%.%d][%a&%.%d][%a%d%d.]+[%a%.]$") then -- if not year then -- does it match our if nil then no pattern?match err_cat err_type = ' value' ; -- so value error else local next_year = tonumber(os.. set_errordate ( 'bad_lccn%Y' ))+1; -- set an error messageget the current year as a number and add one for next year end elseif 10 year == len then tonumber (year); -- LCCN should be aadddddddd or ddddddddddconvert year portion of bibcode to a number if id:match("[^%d]"1000 > year) or (year > next_year) then err_type = 'year'; -- if LCCN has anything but digits (nil if only digits) ...year out of bounds end if nil == id:matchfind("^'&%l%l%d%d%d%d%d%d%d%d".') then -- ... see if it matches our pattern err_cat err_type = ' journal' ; -- journal abbreviation must not have '&.. set_error' ( 'bad_lccn' if it does its missing a letter); -- 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'err_type) then -- if there was an error detected err_cat text = text .. ' ' .. set_error( 'bad_lccnbad_bibcode' , {err_type}); -- 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_cattext;
end
--[[--------------------------< P M B I D O R X I V >----------------------------------------------------------------------
Format PMID bioRxiv id and do simple error checking. PMIDs BiorXiv ids are sequential numbers beginning at 1 and counting upexactly 6 digits. Thiscode checks the PMID to see that it contains only digits and The bioRxiv id is less than test_limit; the value number following the last slash in local variablethe bioRxiv-issued DOI:test_limit will need to be updated periodically as more PMIDs are issuedhttps://doi.org/10.1101/078733 -> 078733
]]
local function pmidbiorxiv(id) local test_limit = 30000000; -- update this value as PMIDs approach local handler = cfg.id_handlers['PMIDBIORXIV']; local err_cat = ''; -- presume that PMID bioRxiv id is valid
if nil == id:match("[^%d]%d%d%d%d%d$") then -- if PMID bioRxiv id has anything but six digits err_cat = ' ' .. set_error( 'bad_pmidbad_biorxiv' ); -- 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, access=handler.access}) .. err_cat;
end
--[[--------------------------< C I T E S _ E M B A R G O E D R X >------------------------------------------------------------
Determines if a PMC identifier's online version is embargoed. Compares CiteSeerX use their own notion of "doi" (not to be confused with the date in |embargo= against today's dateidentifiers resolved via doi. If embargo date isin 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 citeorg).
The description of the structure of this identifier can be found at Help_talk:Citation_Style_1/Archive_26#CiteSeerX_id_structure
]]
local function is_embargoed citeseerx (embargoid) if is_set (embargo) then local lang handler = mwcfg.getContentLanguage(); local good1, embargo_date, good2, todays_date; good1, embargo_date = pcall( lang.formatDate, lang, id_handlers['UCITESEERX', embargo )]; good2, todays_date = pcall( lang.formatDate, lang, 'U' ) local matched;
if good1 and good2 then -- if embargo date and today's date are good dates if tonumber( embargo_date ) > local text = tonumberexternal_link_id( 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{link= not set return empty stringend --[[--------------------------< P M C >------------------------------------------------------------------------ Format a PMChandler.link, do simple error checking, and check for embargoed articleslabel=handler. The embargo parameter takes a date for a value. If the embargo date is in the future the PMC identifier will notbe linked to the article. If the embargo date is today or in the past, or if it is empty or omittedlabel, then thePMC identifier is linked to the article through the link at cfg.id_handlers['PMC'] prefix=handler.prefix. PMC embargo date testing is done in function is_embargoed () which is called earlier because when the citationhas |pmc, id=<value> but does not have a |urlid, separator= then |title= is linked with the PMC linkhandler. Function is_embargoed ()returns the embargo date if the PMC article is still embargoedseparator, otherwise it returns an empty stringencode=handler. PMCs are sequential numbers beginning at 1 and counting up. This code checks the PMC to see that it contains only digits and is lessthan test_limit; the value in local variable test_limit will need to be updated periodically as more PMCs are issued. ]] local function pmc(idencode, embargo) local test_limit access= 6000000; -- update this value as PMCs approach local handler = cfg.id_handlers['PMC']; local err_cat = ''; -- presume that PMC is valid local id_num; local textaccess});
id_num matched = id:match ('"^10%.1%.1%.[Pp1-9]%d?%d?%d?%.[Mm1-9][Cc](%d+)$'); -- identifier with pmc prefix if is_set (id_num) then add_maint_cat ('pmc_format'); else -- plain number without pmc prefix id_num = id:match ('^?%d+?%d?$'"); -- if here id is all digits end if is_set (id_num) not matched then -- id_num has a value so test it id_num text = tonumber(id_num); -- convert id_num to a number for range testing if 1 > id_num or test_limit < id_num then -- if PMC is outside test limit boundaries err_cat = ' ' text .. set_error( 'bad_pmc' ); -- set an error message else id = tostring (id_num); -- make sure id is a string end else -- when id format incorrect err_cat = ' ' .. set_error( 'bad_pmcbad_citeseerx' ); -- set an error message end if is_set (embargo) then -- is PMC is still embargoed? text="[[" .. handler.link .. "|" .. handler.label .. "]]" .. handler.separator .. id .. err_cat; -- still embargoed so no external link else text = external_link_id({link = handler.link, label = handler.label, -- no embargo date or embargo has expired, ok to link to article prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, access=handler.access}) .. err_cat;
end
return text;
local text;
if is_set(inactive) then
local inactive_year = inactive:match("%d%d%d%d") or ''; -- try to get the year portion from the inactive date text = "[[" .. handler.link .. "|" .. handler.label .. "]]:" .. id;
if is_set(inactive_year) then
table.insert( z.error_categories, "Pages with DOIs inactive since " .. inactive_year );
else
table.insert( z.error_categories, "Pages with inactive DOIs" ); -- when inactive doesn't contain a recognizable year
end
inactive = " (" .. cfg.messages['inactive'] .. " " .. inactive .. ")"
else
text = external_link_id({link = handler.link, label = handler.label,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, access=access})
inactive = ""
end
text = external_link_id({link = handler.link, label = handler.label,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode, access=access}) .. (inactive or '')
if nil == id:match("^10%.[^%s–]-/[^%s–]-[^%.,]$") 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 .. inactive .. cat
end
--[[--------------------------< O P E N L I S B R A R Y N >----------------------------------------------------------------------
Formats Determines whether an OpenLibrary link, and checks for associated errors.ISBN string is valid
]]
local function openlibraryisbn(id, accessisbn_str ) local code if nil ~= idisbn_str:match("[^%d+([AMWs-0-9X])$")then return false, 'invalid character'; -- only fail if isbn_str contains anything but digits followed by 'A', 'M'hyphens, or 'W'the uppercase X end isbn_str = isbn_str:gsub( "-", "" ):gsub( " ", "" ); -- remove hyphens and spaces local handler len = cfg.id_handlers[isbn_str:len(); if len ~= 10 and len ~= 13 then return false, 'OLlength']; -- fail if incorrect length end
if ( code len == "A" ) 10 then return external_link_idif isbn_str:match({link"^%d*X?$" ) =handler.link, label=handler.label, prefix=handler.prefix .. nil then -- fail if isbn_str has 'authors/OLX',anywhere but last position id=id, separator=handler.separator, encode = handler.encode, access = access}) elseif ( code == "M" ) then return external_link_id({link=handler.linkfalse, label=handler.label, prefix=handler.prefix .. 'books/OLinvalid form',; id=id, separator=handler.separator, encode = handler.encode, access = access}) elseif ( code == "W" ) then end return external_link_idis_valid_isxn({link=handler.linkisbn_str, label=handler.label10), prefix=handler.prefix .. 'works/OLchecksum', id=id, separator=handler.separator, encode = handler.encode, access = access});
else
return external_link_idif isbn_str:match({link"^%d+$" ) =handler.link, label=handler.label,nil then prefix=handler.prefix .. return false, 'OLinvalid character',; -- fail if isbn13 is not all digits end id if isbn_str:match( "^97[89]%d*$" ) =id, separator=handler.separator, encode = handler.encode,nil then access = access}) .. return false, ' invalid prefix' .. set_error; -- fail when isbn13 does not begin with 978 or 979 end return is_valid_isxn_13 ( isbn_str), 'bad_olchecksum' );
end
end
--[[--------------------------< A M E S S A G E _ I D Z O N >------------------------------------------------------------------
Validate and format Formats a usenet message idlink to Amazon. Simple Do simple error checking: asin must be mix of 10 numeric or uppercase alphacharacters. If a mix, looks for 'idfirst character must be uppercase alpha; if all numeric, asins must be 10-left@iddigitisbn. If 10-right' not enclosed indigit 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/or '>' angle bracketsfirst character is a digit. This function is positioned here because it calls isbn()
]]
local function message_id asin(id, domain) local handler err_cat = cfg.id_handlers['USENETID'];""
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]+$") then err_cat = ' ' .. set_error ('bad_asin'); -- asin doesn't begin with uppercase alpha end end if not is_set(domain) then domain = "com"; elseif in_array (domain, {'jp', 'uk'}) then -- Japan, United Kingdom domain = "co." .. domain; elseif in_array (domain, {'au', 'br', 'mx'}) then -- Australia, Brazil, Mexico domain = "com." .. domain; end local handler = cfg.id_handlers['ASIN']; return external_link_id({link=handler.link, label=handler.label, prefix=handler.prefix .. domain .. "/dp/", id=id, encode=handler.encode, separator = handler.separator}) .. err_cat;end --[[--------------------------< I S M N >---------------------------------------------------------------------- Determines whether an ISMN string is valid. Similar to isbn-13, ismn is 13 digits begining 979-0-... and uses thesame check digit calculations. See http://www.ismn-international.org/download/Web_ISMN_Users_Manual_2008-6.pdfsection 2, pages 9–12. ]] local function ismn (id) local handler = cfg.id_handlers['ISMN']; local text; local valid_ismn = true; local id_copy; id_copy = id; -- save a copy because this testing is destructive id=id:gsub( "[%s-–]", "" ); -- strip spaces, hyphens, and endashes from the ismn if 13 ~= id:len() or id:match( "^9790%d*$" ) == 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 = table.concat ( { make_wikilink (handler.link, handler.label), handler.separator, id_copy }); -- 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 buthas separated the two groups of four digits with a space. When that condition occurred, the resulting link lookedlike 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 lengthand makes sure that the checkdigit agrees with the calculated value. Incorrect length (8 digits), charactersother than 0-9 and X, or checkdigit / calculated value mismatch will all cause a check issn error message. Theissn 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?$" ) then -- validate the issn: 8 digits long, containing only 0-9 or X in the last position valid_issn=false; -- wrong length or improper character else valid_issn=is_valid_isxn(id, 8); -- validate issn end if true == valid_issn then id = string.sub( id, 1, 4 ) .. "-" .. string.sub( id, 5 ); -- if valid, display correctly formatted version else id = issn_copy; -- if not valid, use the show the invalid issn with error message end text = external_link_id({link = handler.link, label = handler.label,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
if not id:match('^.+@.+$') or not id:match('^[^<].*[^>]$')false == valid_issn then -- doesn't have '@' or has one or first or last character is '< or '>' text = text .. ' ' .. set_error( 'bad_message_idbad_issn', e and 'e' or '' ) -- add an error message if the message id issn is invalid
end
return text
end
--[[--------------------------< J F M >-----------------------------------------------------------------------
A numerical identifier in the form nn.nnnn.nn
]]
local function jfm (id)
local handler = cfg.id_handlers['JFM'];
local id_num;
local err_cat = '';
id_num = id:match ('^[Jj][Ff][Mm](.*)$'); -- identifier with jfm prefix; extract identifier
if is_set (id_num) then
add_maint_cat ('jfm_format');
else -- plain number without mr prefix
id_num = id; -- if here id does not have prefix
end
if id_num and id_num:match('^%d%d%.%d%d%d%d%.%d%d$') then
id = id_num; -- jfm matches pattern
else
err_cat = ' ' .. set_error( 'bad_jfm' ); -- 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}) .. err_cat;
end
--[[--------------------------< L C C N >----------------------------------------------------------------------
Format LCCN link and do simple error checking. LCCN is a character string 8-12 characters long. The length of
the LCCN dictates the character type of the first 1-3 characters; the rightmost eight are always digits.
http://info-uri.info/registry/OAIHandler?verb=GetRecord&metadataPrefix=reg&identifier=info:lccn/
length = 8 then all digits
length = 9 then lccn[1] is 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
--[[--------------------------< M R >--------------------------------------------------------------------------
A seven digit number; if not seven digits, zero-fill leading digits to make seven digits.
]]
local function mr (id)
local handler = cfg.id_handlers['MR'];
local id_num;
local id_len;
local err_cat = '';
id_num = id:match ('^[Mm][Rr](%d+)$'); -- identifier with mr prefix
if is_set (id_num) then
add_maint_cat ('mr_format');
else -- plain number without mr prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
end
id_len = id_num and id_num:len() or 0;
if (7 >= id_len) and (0 ~= id_len) then
id = string.rep ('0', 7-id_len ) .. id_num; -- zero-fill leading digits
else
err_cat = ' ' .. set_error( 'bad_mr' ); -- set an error message
end
return external_link_id({link = handler.link, label = handler.label,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode}) .. err_cat;
end
--[[--------------------------< B O P E N L I B C O D E R A R Y >--------------------------------------------------------------------
Validates (sort of) Formats an OpenLibrary link, and formats a bibcode idchecks for associated errors.
Format for bibcodes is specified here: http://adsabs.harvard.edu/abs_doc/help_pages/data.html#bibcodes]]
Butlocal function openlibrary(id, this: 2015arXiv151206696F is apparently valid so apparently, the only things that really matter are length, 19 charactersaccess)and first four digits must be a year local code; local handler = cfg. This function makes these tests:id_handlers['OL']; length must be 19 characterslocal ident; characters in position 1–4 must be digits and must represent a year in the range of 1000 – next year 5 must be a letter 6 must be letter ident, ampersandcode = id:gsub('^OL', or dot ''):match("^(%d+(ampersand cannot directly precede a dot[AMW]))$"); &. ) 7–8 must be letter -- optional OL prefix followed immediately by digits followed by 'A', digit, ampersand'M', or dot (ampersand cannot directly precede a dot'W'; &. ) 9–18 must be letter, digit, or dot 19 must be a letter or dotremove 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
local function bibcode if (id, accesscode == "W" )then local return external_link_id({link=handler .link, label= cfghandler.label, prefix=handler.prefix ..id_handlers['BIBCODEworks/OL'];, local err_type; id=id, separator=handler.separator, encode = handler.encode, access = access}) local year;endend
local text = external_link_id({link=handler.link--[[--------------------------< P M C >------------------------------------------------------------------------ Format a PMC, label=handlerdo simple error checking, and check for embargoed articles.label The embargo parameter takes a date for a value. If the embargo date is in the future the PMC identifier will notbe 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 citationhas |pmc=handler<value> but does not have a |url= then |title= is linked with the PMC link.prefix Function is_embargoed ()returns the embargo date if the PMC article is still embargoed, id=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 lessthan test_limit; the value in local variable test_limit will need to be updated periodically as more PMCs are issued. ]] local function pmc(id, separatorembargo) local test_limit =6000000; -- update this value as PMCs approach local handler.separator, encode=handlercfg.encode,id_handlers['PMC']; access local err_cat =access})''; -- presume that PMC is valid local id_num; local text;
if 19 ~id_num = id:lenmatch ('^[Pp][Mm][Cc](%d+)$'); -- identifier with pmc prefix if is_set (id_num) then err_type = add_maint_cat ('lengthpmc_format'); else -- plain number without pmc prefix year id_num = id:match ("'^(%d%d%d%d)[%a][%a&%.][%a&%.%d][%a&%.%d][%a%d%.]+[%a%.]$"') ; -- if here id is all digits end if not year 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 nil 1 > id_num or test_limit < id_num then no pattern match -- if PMC is outside test limit boundaries err_type err_cat = 'value'.. set_error( 'bad_pmc' ); -- so value set an errormessage
else
local next_year id = tonumbertostring (os.date ('%Y'))+1; -- get the current year as a number and add one for next year year = tonumber (yearid_num); -- 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 make sure id:find('&%.') then err_type = 'journal'; -- journal abbreviation must not have '&.' (if it does its missing is a letter) endstring
end
else -- when id format incorrect
err_cat = ' ' .. set_error( 'bad_pmc' ); -- set an error message
end
if is_set (err_typeembargo) then -- if there was an error detectedis PMC is still embargoed? text = text table.concat ( -- still embargoed so no external link { make_wikilink (handler. ' ' link, handler.label), handler. set_error( 'bad_bibcode'separator, id, err_cat }); else text = external_link_id({err_typelink = 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;
end
return text;
--[[--------------------------< C P M I T E S E E R X D >----------------------------------------------------------------------
CiteSeerX use their own notion of "doi" (not Format PMID and do simple error checking. PMIDs are sequential numbers beginning at 1 and counting up. Thiscode checks the PMID to be confused with see that it contains only digits and is less than test_limit; the identifiers resolved via doi.org)value in local variabletest_limit will need to be updated periodically as more PMIDs are issued.
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 pmid(id) local test_limit = 32000000; -- update this value as PMIDs approach local handler = cfg.id_handlers['CITESEERXPMID']; local matchederr_cat = ''; -- presume that PMID is valid
local text = external_link_idif id:match({link"[^%d]") then -- if PMID has anything but digits err_cat =handler' ' .link, label=handler.label,set_error( 'bad_pmid' ); -- set an error message else -- PMID is only digits prefixlocal id_num =handler.prefix, tonumber(id=); -- convert id, separator=handler.separator, encode=handler.encode,to a number for range testing accessif 1 > id_num or test_limit < id_num then -- if PMID is outside test limit boundaries err_cat =handler' ' .access}. set_error( 'bad_pmid' ); -- set an error message end end
matched return external_link_id({link = id:match ("^10%handler.1%link, label = handler.1%label, prefix=handler.[1-9]%d?%d?%d?%prefix,id=id,separator=handler.[1-9]%d?%d?%d?$"); if not matched then text separator, encode= text handler.encode}) . ' ' .. set_error( 'bad_citeseerx' ); end return texterr_cat;
end
local function ssrn (id)
local test_limit = 30000003500000; -- update this value as SSRNs approach
local handler = cfg.id_handlers['SSRN'];
local err_cat = ''; -- presume that SSRN is valid
local id_num;
local text;
end
--[[--------------------------< U S E N E T _ I D >------------------------------------------------------------
Validate and format a usenet message id. Simple error checking, looks for 'id-left@id-right' not enclosed in
'<' and/or '>' angle brackets.
]]
local function usenet_id (id)
local handler = cfg.id_handlers['USENETID'];
local text = external_link_id({link = handler.link, label = handler.label,
prefix=handler.prefix,id=id,separator=handler.separator, encode=handler.encode})
if not id:match('^.+@.+$') or not id:match('^[^<].*[^>]$')then -- doesn't have '@' or has one or first or last character is '< or '>'
text = text .. ' ' .. set_error( 'bad_usenet_id' ) -- add an error message if the message id is invalid
end
return text
end
--[[--------------------------< Z B L >-----------------------------------------------------------------------
A numerical identifier in the form nnnn.nnnnn - leading zeros in the first quartet optional
]]
local function zbl (id)
local handler = cfg.id_handlers['ZBL'];
local id_num;
local err_cat = '';
id_num = id:match ('^[Zz][Bb][Ll](.*)$'); -- identifier with zbl prefix; extract identifier
if is_set (id_num) then
add_maint_cat ('zbl_format');
else -- plain number without zbl prefix
id_num = id; -- if here id does not have prefix
end
if id_num:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then
id = id_num; -- id matches pattern
else
err_cat = ' ' .. set_error( 'bad_zbl' ); -- set an error message
end
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 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 >--------------------------------------------------------
for k, v in pairs( id_list ) do -- k is uc identifier name as index to cfg.id_handlers; e.g. cfg.id_handlers['ISBN'], v is a table
-- fallback to read-only cfg
handler = setmetatable( { ['id'] = v, ['access'] = options.IdAccessLevels[k] }, fallback(k) );
elseif handler.mode ~= 'manual' then
error( cfg.messages['unknown_ID_mode'] );
elseif k == 'ARXIV' then
table.insert( new_list, {handler.label, arxiv( v, options.Class ) } );
elseif k == 'ASIN' then
table.insert( new_list, {handler.label, asin( v, options.ASINTLD ) } );
elseif k == 'BIBCODE' then
table.insert( new_list, {handler.label, bibcode( v, handler.access ) } );
elseif k == 'DOI' then
table.insert( new_list, {handler.label, doi( v, options.DoiBroken, handler.access ) } );
elseif k == 'HDL' then
table.insert( new_list, {handler.label, hdl( v, handler.access ) } );
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, 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 == 'OCLC' then
table.insert( new_list, {handler.label, oclc( v ) } );
elseif k == 'SSRN' then
table.insert( new_list, {handler.label, ssrn( 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 == 'HDL' then
table.insert( new_list, {handler.label, hdl( v, handler.access ) } );
elseif k == 'ISBN' then
local ISBN = internal_link_id( handler );
local check;
local err_type = '';
check, err_type = check_isbnisbn( v );
if not check then
if is_set(options.IgnoreISBN) then -- ISBN is invalid; if |ignore-isbn-error= set
end
elseif is_set(options.IgnoreISBN) then -- ISBN is OK; if |ignore-isbn-error= set
add_maint_cat ('ignore_isbn_err'); -- because |ignore-isbn-error= unnecessary
end
table.insert( new_list, {handler.label, ISBN } );
elseif k == 'ISMN' then
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
table.insert( new_list, {handler.label, message_idusenet_id( v ) } ); elseif k == 'ZBL' then table.insert( new_list, {handler.label, zbl( v ) } );
else
error( cfg.messages['unknown_manual_ID'] );
local function comp( a, b ) -- used in following table.sort()
return a[1] :lower() < b[1]:lower();
end
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