Bharatpedia

Module:Citation/CS1/Identifiers: Difference between revisions

Module Talk

Module:Citation/CS1/Identifiers (view source)

Revision as of 14:52, 30 September 2023

8,459 bytes added ,  30 September 2023
Update
m (1 revision imported)
(Update)
 
Line 1: Line 1:
--[[--------------------------< F O R W A R D  D E C L A R A T I O N S >--------------------------------------
--[[--------------------------< F O R W A R D  D E C L A R A T I O N S >--------------------------------------
]]
]]
Line 87: Line 86:
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, 'PATH');
end
end


Line 165: Line 164:




--[=[-------------------------< I S _ V A L I D _ B I O R X I V _ D A T E >------------------------------------
--[=[-------------------------< I S _ V A L I D _ R X I V _ D A T E >------------------------------------------


returns true if:
for biorxiv, returns true if:
2019-12-11T00:00Z <= biorxiv_date < today + 2 days
2019-12-11T00:00Z <= biorxiv_date < today + 2 days
for medrxiv, returns true if:
2020-01-01T00:00Z <= medrxiv_date < today + 2 days
The dated form of biorxiv identifier has a start date of 2019-12-11.  The Unix timestamp for that date is {{#time:U|2019-12-11}} = 1576022400
The dated form of biorxiv identifier has a start date of 2019-12-11.  The Unix timestamp for that date is {{#time:U|2019-12-11}} = 1576022400
The medrxiv identifier has a start date of 2020-01-01.  The Unix timestamp for that date is {{#time:U|2020-01-01}} = 1577836800


biorxiv_date is the date provided in those |biorxiv= parameter values that are dated at time 00:00:00 UTC
<rxiv_date> is the date provided in those |biorxiv= parameter values that are dated and in |medrxiv= parameter values at time 00:00:00 UTC
today is the current date at time 00:00:00 UTC plus 48 hours
<today> is the current date at time 00:00:00 UTC plus 48 hours
if today is 2015-01-01T00:00:00 then
if today's date is 2023-01-01T00:00:00 then
adding 24 hours gives 2015-01-02T00:00:00 – one second more than today
adding 24 hours gives 2023-01-02T00:00:00 – one second more than today
adding 24 hours gives 2015-01-03T00:00:00 – one second more than tomorrow
adding 24 hours gives 2023-01-03T00:00:00 – one second more than tomorrow


This function does not work if it is fed month names for languages other than English.  Wikimedia #time: parser
inputs:
apparently doesn't understand non-English date month names. This function will always return false when the date
<y>, <m>, <d> – year, month, day parts of the date from the birxiv or medrxiv identifier
contains a non-English month name because good1 is false after the call to lang.formatDate().  To get around that
<select> 'b' for biorxiv, 'm' for medrxiv; defaults to 'b'
call this function with YYYY-MM-DD format dates.


]=]
]=]


local function is_valid_biorxiv_date (biorxiv_date)
local function is_valid_rxiv_date (y, m, d, select)
if 0 == tonumber (m) and 12 < tonumber (m) then -- <m> must be a number 1–12
return false;
end
if 0 == tonumber (d) and 31 < tonumber (d) then -- <d> must be a number 1–31; TODO: account for month length and leap yer?
return false;
end
local rxiv_date = table.concat ({y, m, d}, '-'); -- make ymd date string
local good1, good2;
local good1, good2;
local biorxiv_ts, tomorrow_ts; -- to hold Unix timestamps representing the dates
local rxiv_ts, tomorrow_ts; -- to hold Unix timestamps representing the dates
local lang_object = mw.getContentLanguage();
local lang_object = mw.getContentLanguage();


good1, biorxiv_ts = pcall (lang_object.formatDate, lang_object, 'U', biorxiv_date); -- convert biorxiv_date value to Unix timestamp  
good1, rxiv_ts = pcall (lang_object.formatDate, lang_object, 'U', rxiv_date); -- convert rxiv_date value to Unix timestamp  
good2, tomorrow_ts = pcall (lang_object.formatDate, lang_object, 'U', 'today + 2 days' ); -- today midnight + 2 days is one second more than all day tomorrow
good2, tomorrow_ts = pcall (lang_object.formatDate, lang_object, 'U', 'today + 2 days' ); -- today midnight + 2 days is one second more than all day tomorrow
if good1 and good2 then -- lang.formatDate() returns a timestamp in the local script which tonumber() may not understand
if good1 and good2 then -- lang.formatDate() returns a timestamp in the local script which tonumber() may not understand
biorxiv_ts = tonumber (biorxiv_ts) or lang_object:parseFormattedNumber (biorxiv_ts); -- convert to numbers for the comparison;
rxiv_ts = tonumber (rxiv_ts) or lang_object:parseFormattedNumber (rxiv_ts); -- convert to numbers for the comparison;
tomorrow_ts = tonumber (tomorrow_ts) or lang_object:parseFormattedNumber (tomorrow_ts);
tomorrow_ts = tonumber (tomorrow_ts) or lang_object:parseFormattedNumber (tomorrow_ts);
else
else
Line 200: Line 209:
end
end


return ((1576022400 <= biorxiv_ts) and (biorxiv_ts < tomorrow_ts)) -- 2012-12-11T00:00Z <= biorxiv_date < tomorrow's date
local limit_ts = ((select and ('m' == select)) and 1577836800) or 1576022400; -- choose the appropriate limit timesatmp
 
return ((limit_ts <= rxiv_ts) and (rxiv_ts < tomorrow_ts)) -- limit_ts <= rxiv_date < tomorrow's date
end
end


Line 242: Line 253:
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
end
end
return temp % 10 == 0; -- sum modulo 10 is zero when ISBN-13/ISMN is correct
return temp % 10 == 0; -- sum modulo 10 is zero when ISBN-13/ISMN is correct
Line 250: Line 261:
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------
--[[--------------------------< N O R M A L I Z E _ L C C N >--------------------------------------------------


LCCN normalization (http://www.loc.gov/marc/lccn-namespace.html#normalization)
LCCN normalization (https://www.loc.gov/marc/lccn-namespace.html#normalization)
1. Remove all blanks.
1. Remove all blanks.
2. If there is a forward slash (/) in the string, remove it, and remove all characters to the right of the forward slash.
2. If there is a forward slash (/) in the string, remove it, and remove all characters to the right of the forward slash.
Line 260: Line 271:


Returns a normalized LCCN for lccn() to validate.  There is no error checking (step 3.b.1) performed in this function.
Returns a normalized LCCN for lccn() to validate.  There is no error checking (step 3.b.1) performed in this function.
]]
]]


Line 286: Line 298:
--[[--------------------------< A R X I V >--------------------------------------------------------------------
--[[--------------------------< A R X I V >--------------------------------------------------------------------


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


format and error check arXiv identifier.  There are three valid forms of the identifier:
format and error check arXiv identifier.  There are three valid forms of the identifier:
Line 311: Line 323:
<date code> and <version> are as defined for 0704-1412
<date code> and <version> are as defined for 0704-1412
<number> is a five-digit number
<number> is a five-digit number
]]
]]


Line 318: Line 331:
local handler = options.handler;
local handler = options.handler;
local year, month, version;
local year, month, version;
local err_cat = false; -- assume no error message
local err_msg = false; -- assume no error message
local text; -- output text
local text; -- output text
Line 327: Line 340:
if ((not (90 < year or 8 > year)) or (1 > month or 12 < month)) or -- if invalid year or invalid month
if ((not (90 < year or 8 > year)) or (1 > month or 12 < month)) or -- if invalid year or invalid month
((91 == year and 7 > month) or (7 == year and 3 < month)) then -- if years ok, are starting and ending months ok?
((91 == year and 7 > month) or (7 == year and 3 < month)) then -- if years ok, are starting and ending months ok?
err_cat = true; -- flag for error message
err_msg = true; -- flag for error message
end
end


Line 336: Line 349:
if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or -- is year invalid or is month invalid? (doesn't test for future years)
if ((7 > year) or (14 < year) or (1 > month or 12 < month)) or -- is year invalid or is month invalid? (doesn't test for future years)
((7 == year) and (4 > month)) then -- when year is 07, is month invalid (before April)?
((7 == year) and (4 > month)) then -- when year is 07, is month invalid (before April)?
err_cat = true; -- flag for error message
err_msg = true; -- flag for error message
end
end


Line 344: Line 357:
month = tonumber (month);
month = tonumber (month);
if ((15 > year) or (1 > month or 12 < month)) then -- is year invalid or is month invalid? (doesn't test for future years)
if ((15 > year) or (1 > month or 12 < month)) then -- is year invalid or is month invalid? (doesn't test for future years)
err_cat = true; -- flag for error message
err_msg = true; -- flag for error message
end
end


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


err_cat = err_cat and table.concat ({' ', set_message ('err_bad_arxiv')}) or ''; -- set error message if flag is true
if err_msg then
options.coins_list_t['ARXIV'] = nil; -- when error, unset so not included in COinS
end
local err_msg_t = {};
if err_msg then
set_message ('err_bad_arxiv');
end
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access});


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


return text;
return text;
end
end


Line 372: Line 392:
Validates (sort of) and formats a bibcode ID.
Validates (sort of) and formats a bibcode ID.


Format for bibcodes is specified here: http://adsabs.harvard.edu/abs_doc/help_pages/data.html#bibcodes
Format for bibcodes is specified here: https://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
But, this: 2015arXiv151206696F is apparently valid so apparently, the only things that really matter are length, 19 characters
Line 390: Line 410:
local access = options.access;
local access = options.access;
local handler = options.handler;
local handler = options.handler;
local ignore_invalid = options.accept;
local err_type;
local err_type;
local err_msg = '';
local year;
local year;


Line 404: Line 426:
err_type = cfg.err_msg_supl.value; -- so value error
err_type = cfg.err_msg_supl.value; -- so value error
else
else
local next_year = tonumber (os.date ('%Y')) + 1; -- get the current year as a number and add one for next year
local next_year = tonumber (os.date ('%Y')) + 1; -- get the current year as a number and add one for next year
year = tonumber (year); -- convert year portion of bibcode to a number
year = tonumber (year); -- convert year portion of bibcode to a number
if (1000 > year) or (year > next_year) then
if (1000 > year) or (year > next_year) then
Line 411: Line 433:
if id:find('&%.') then
if id:find('&%.') then
err_type = cfg.err_msg_supl.journal; -- journal abbreviation must not have '&.' (if it does it's missing a letter)
err_type = cfg.err_msg_supl.journal; -- journal abbreviation must not have '&.' (if it does it's missing a letter)
end
if id:match ('.........%.tmp%.') then -- temporary bibcodes when positions 10–14 are '.tmp.'
set_message ('maint_bibcode');
end
end
end
end
end
end


if is_set (err_type) then -- if there was an error detected
if is_set (err_type) and not ignore_invalid then -- if there was an error detected and accept-as-written markup not used
text = text .. ' ' .. set_message ('err_bad_bibcode', {err_type});
set_message ('err_bad_bibcode', {err_type});
options.coins_list_t['BIBCODE'] = nil; -- when error, unset so not included in COinS
end
end
return text;
return text;
end
end
Line 440: Line 467:
local id = options.id;
local id = options.id;
local handler = options.handler;
local handler = options.handler;
local err_cat = true; -- flag; assume that there will be an error
local err_msg = true; -- flag; assume that there will be an error
local patterns = {
local patterns = {
'^10.1101/%d%d%d%d%d%d$', -- simple 6-digit identifier (before 2019-12-11)
'^10%.1101/%d%d%d%d%d%d$', -- simple 6-digit identifier (before 2019-12-11)
'^10.1101/(20[1-9]%d)%.([01]%d)%.([0-3]%d)%.%d%d%d%d%d%dv%d+$', -- y.m.d. date + 6-digit identifier + version (after 2019-12-11)
'^10%.1101/(20%d%d)%.(%d%d)%.(%d%d)%.%d%d%d%d%d%dv%d+$', -- y.m.d. date + 6-digit identifier + version (after 2019-12-11)
'^10.1101/(20[1-9]%d)%.([01]%d)%.([0-3]%d)%.%d%d%d%d%d%d$', -- y.m.d. date + 6-digit identifier (after 2019-12-11)
'^10%.1101/(20%d%d)%.(%d%d)%.(%d%d)%.%d%d%d%d%d%d$', -- y.m.d. date + 6-digit identifier (after 2019-12-11)
}
}
Line 453: Line 480:


if m then -- m is nil when id is the six-digit form
if m then -- m is nil when id is the six-digit form
if not is_valid_biorxiv_date (y .. '-' .. m .. '-' .. d) then -- validate the encoded date; TODO: don't ignore leap-year and actual month lengths ({{#time:}} is a poor date validator)
if not is_valid_rxiv_date (y, m, d, 'b') then -- validate the encoded date; 'b' for biorxiv limit
break; -- date fail; break out early so we don't unset the error message
break; -- date fail; break out early so we don't unset the error message
end
end
end
end
err_cat = nil; -- we found a match so unset the error message
err_msg = nil; -- we found a match so unset the error message
break; -- and done
break; -- and done
end
end
end -- err_cat remains set here when no match
end -- err_cat remains set here when no match


if err_msg then
options.coins_list_t['BIORXIV'] = nil; -- when error, unset so not included in COinS
set_message ('err_bad_biorxiv'); -- and set the error message
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator,
prefix = handler.prefix, id = id, separator = handler.separator,
encode = handler.encode, access = handler.access}) .. (err_cat and (' ' .. set_message ('err_bad_biorxiv')) or '');
encode = handler.encode, access = handler.access});
end
end


Line 473: Line 505:


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


Line 479: Line 512:
local handler = options.handler;
local handler = options.handler;
local matched;
local matched;
 
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode,
Line 486: Line 519:
matched = id:match ("^10%.1%.1%.[1-9]%d?%d?%d?%.[1-9]%d?%d?%d?$");
matched = id:match ("^10%.1%.1%.[1-9]%d?%d?%d?%.[1-9]%d?%d?%d?$");
if not matched then
if not matched then
text = text .. ' ' .. set_message ('err_bad_citeseerx' );
set_message ('err_bad_citeseerx' );
options.coins_list_t['CITESEERX'] = nil; -- when error, unset so not included in COinS
end
end
return text;
return text;
end
end
Line 506: Line 541:
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
and terminal punctuation may not be technically correct but it appears, that in practice these characters are rarely
if ever used in DOI names.
if ever used in DOI names.
https://www.doi.org/doi_handbook/2_Numbering.html -- 2.2 Syntax of a DOI name
https://www.doi.org/doi_handbook/2_Numbering.html#2.2.2 -- 2.2.2 DOI prefix


]]
]]
Line 515: Line 553:
local ignore_invalid = options.accept;
local ignore_invalid = options.accept;
local handler = options.handler;
local handler = options.handler;
local err_cat;
local err_flag;


local text;
local text;
if is_set (inactive) then
if is_set (inactive) then
local inactive_year = inactive:match("%d%d%d%d") or ''; -- try to get the year portion from the inactive date
local inactive_year = inactive:match("%d%d%d%d"); -- try to get the year portion from the inactive date
local inactive_month, good;
local inactive_month, good;


Line 530: Line 568:
end
end
end
end
else
end -- otherwise, |doi-broken-date= has something but it isn't a date
inactive_year = nil; -- |doi-broken-date= has something but it isn't a date
end
if is_set (inactive_year) and is_set (inactive_month) then
if is_set (inactive_year) and is_set (inactive_month) then
Line 544: Line 580:
end
end


local registrant = id:match ('^10%.([^/]+)/[^%s–]-[^%.,]$'); -- registrant set when DOI has the proper basic form
local registrant = mw.ustring.match (id, '^10%.([^/]+)/[^%s–]-[^%.,]$'); -- registrant set when DOI has the proper basic form
 
local registrant_err_patterns = { -- these patterns are for code ranges that are not supported  
local registrant_err_patterns = { -- these patterns are for code ranges that are not supported  
'^[^1-3]%d%d%d%d%.%d%d*$', -- 5 digits with subcode (0xxxx, 40000+); accepts: 10000–39999
'^[^1-3]%d%d%d%d%.%d+$', -- 5 digits with subcode (0xxxx, 40000+); accepts: 10000–39999
'^[^1-4]%d%d%d%d$', -- 5 digits without subcode (0xxxx, 40000+); accepts: 10000–49999
'^[^1-6]%d%d%d%d$', -- 5 digits without subcode (0xxxx, 60000+); accepts: 10000–69999
'^[^1-9]%d%d%d%.%d%d*$', -- 4 digits with subcode (0xxx); accepts: 1000–9999
'^[^1-9]%d%d%d%.%d+$', -- 4 digits with subcode (0xxx); accepts: 1000–9999
'^[^1-9]%d%d%d$', -- 4 digits without subcode (0xxx); accepts: 1000–9999
'^[^1-9]%d%d%d$', -- 4 digits without subcode (0xxx); accepts: 1000–9999
'^%d%d%d%d%d%d+', -- 6 or more digits
'^%d%d%d%d%d%d+', -- 6 or more digits
'^%d%d?%d?$', -- less than 4 digits without subcode (with subcode is legitimate)
'^%d%d?%d?$', -- less than 4 digits without subcode (3 digits with subcode is legitimate)
'^%d%d?%.[%d%.]+', -- 1 or 2 digits with subcode
'^5555$', -- test registrant will never resolve
'^5555$', -- test registrant will never resolve
'[^%d%.]', -- any character that isn't a digit or a dot
'[^%d%.]', -- any character that isn't a digit or a dot
Line 558: Line 595:


if not ignore_invalid then
if not ignore_invalid then
if registrant then -- when DOI has proper form
if registrant then -- when DOI has proper form
for i, pattern in ipairs (registrant_err_patterns) do -- spin through error patterns
for i, pattern in ipairs (registrant_err_patterns) do -- spin through error patterns
if registrant:match (pattern) then -- to validate registrant codes
if registrant:match (pattern) then -- to validate registrant codes
err_cat = ' ' .. set_message ('err_bad_doi'); -- when found, mark this DOI as bad
err_flag = set_message ('err_bad_doi'); -- when found, mark this DOI as bad
break; -- and done
break; -- and done
end
end
end
end
else
else
err_cat = ' ' .. set_message ('err_bad_doi'); -- invalid directory or malformed
err_flag = set_message ('err_bad_doi'); -- invalid directory or malformed
end
end
else
else
Line 572: Line 609:
end
end


if err_flag then
options.coins_list_t['DOI'] = nil; -- when error, unset so not included in COinS
end
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access,
auto_link = not (err_cat or is_set (inactive) or ignore_invalid) and 'doi' or nil -- do not auto-link when |doi-broken-date= has a value or when there is a DOI error or (to play it safe, after all, auto-linking is not essential) when invalid DOIs are ignored
auto_link = not (err_flag or is_set (inactive) or ignore_invalid) and 'doi' or nil -- do not auto-link when |doi-broken-date= has a value or when there is a DOI error or (to play it safe, after all, auto-linking is not essential) when invalid DOIs are ignored
}) .. (inactive or '');
}) .. (inactive or '');


return text .. (err_cat and err_cat or ''); -- parentheses required
return text;
end
end


Line 596: Line 637:
if ever used in HDLs.
if ever used in HDLs.


Query string parameters are named here: http://www.handle.net/proxy_servlet.html.  query strings are not displayed
Query string parameters are named here: https://www.handle.net/proxy_servlet.html.  query strings are not displayed
but since '?' is an allowed character in an HDL, '?' followed by one of the query parameters is the only way we
but since '?' is an allowed character in an HDL, '?' followed by one of the query parameters is the only way we
have to detect the query string so that it isn't URL-encoded with the rest of the identifier.
have to detect the query string so that it isn't URL-encoded with the rest of the identifier.
Line 606: Line 647:
local access = options.access;
local access = options.access;
local handler = options.handler;
local handler = options.handler;
local query_params = { -- list of known query parameters from http://www.handle.net/proxy_servlet.html
local query_params = { -- list of known query parameters from https://www.handle.net/proxy_servlet.html
'noredirect',
'noredirect',
'ignore_aliases',
'ignore_aliases',
Line 640: Line 681:


if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then -- HDL must contain a forward slash, must not contain spaces, endashes, and must not end with period or comma
if nil == id:match("^[^%s–]-/[^%s–]-[^%.,]$") then -- HDL must contain a forward slash, must not contain spaces, endashes, and must not end with period or comma
text = text .. ' ' .. set_message ('err_bad_hdl' );
set_message ('err_bad_hdl' );
options.coins_list_t['HDL'] = nil; -- when error, unset so not included in COinS
end
end
return text;
return text;
end
end
Line 664: Line 707:
else -- here when not ignoring
else -- here when not ignoring
if not check then -- and there is an error
if not check then -- and there is an error
return ISBN .. set_message ('err_bad_isbn', {err_type}, false, ' '); -- display an error message
options.coins_list_t['ISBN'] = nil; -- when error, unset so not included in COinS
set_message ('err_bad_isbn', err_type); -- set an error message
return ISBN; -- return id text
end
end
end
end
return ISBN;
return ISBN; -- return id text
end
end


Line 686: Line 731:
return return_result (false, cfg.err_msg_supl.form);
return return_result (false, cfg.err_msg_supl.form);
end
end
return return_result (is_valid_isxn (id, 10), cfg.err_msg_supl.check);
if not is_valid_isxn (id, 10) then -- test isbn-10 for numerical validity
return return_result (false, cfg.err_msg_supl.check); -- fail if isbn-10 is not numerically valid
end
if id:find ('^63[01]') then -- 630xxxxxxx and 631xxxxxxx are (apparently) not valid isbn group ids but are used by amazon as numeric identifiers (asin)
return return_result (false, cfg.err_msg_supl.group); -- fail if isbn-10 begins with 630/1
end
return return_result (true, cfg.err_msg_supl.check); -- pass if isbn-10 is numerically valid
else
else
if id:match ('^%d+$') == nil then
if id:match ('^%d+$') == nil then
Line 709: Line 760:
Error message if not 10 characters, if not ISBN-10, if mixed and first character is a digit.
Error message if not 10 characters, if not ISBN-10, if mixed and first character is a digit.


|asin=630....... is (apparently) not a legitimate ISBN though it checksums as one; these do not cause this
|asin=630....... and |asin=631....... are (apparently) not a legitimate ISBN though it checksums as one; these
function to emit the maint_asin message
do not cause this function to emit the maint_asin message


This function is positioned here because it calls isbn()
This function is positioned here because it calls isbn()
Line 720: Line 771:
local domain = options.ASINTLD;
local domain = options.ASINTLD;
local err_cat = ""
local err_flag;


if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
if not id:match("^[%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u][%d%u]$") then
err_cat = ' ' .. set_message ('err_bad_asin'); -- ASIN is not a mix of 10 uppercase alpha and numeric characters
err_flag = set_message ('err_bad_asin'); -- ASIN is not a mix of 10 uppercase alpha and numeric characters
else
else
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
if id:match("^%d%d%d%d%d%d%d%d%d[%dX]$") then -- if 10-digit numeric (or 9 digits with terminal X)
if isbn (options) then -- see if ASIN value is or validates as ISBN-10
if is_valid_isxn (id, 10) then -- see if ASIN value is or validates as ISBN-10
if not id:find ('^630') then -- 630xxxxxxx is (apparently) not a valid isbn prefix but is used by amazon as a numeric identifier
if not id:find ('^63[01]') then -- 630xxxxxxx and 631xxxxxxx are (apparently) not a valid isbn prefixes but are used by amazon as a numeric identifier
set_message ('maint_asin'); -- begins with something other than 630 so possibly an isbn
err_flag = set_message ('err_bad_asin'); -- ASIN has ISBN-10 form but begins with something other than 630/1 so probably an isbn  
end
end
elseif not is_set (err_cat) then
elseif not is_set (err_flag) then
err_cat = ' ' .. set_message ('err_bad_asin'); -- ASIN is not ISBN-10
err_flag = set_message ('err_bad_asin'); -- ASIN is not ISBN-10
end
end
elseif not id:match("^%u[%d%u]+$") then
elseif not id:match("^%u[%d%u]+$") then
err_cat = ' ' .. set_message ('err_bad_asin'); -- asin doesn't begin with uppercase alpha
err_flag = set_message ('err_bad_asin'); -- asin doesn't begin with uppercase alpha
end
end
end
end
if not is_set (domain) then  
if (not is_set (domain)) or in_array (domain, {'us'}) then -- default: United States
domain = "com";
domain = "com";
elseif in_array (domain, {'jp', 'uk'}) then -- Japan, United Kingdom
elseif in_array (domain, {'jp', 'uk'}) then -- Japan, United Kingdom
domain = "co." .. domain;
domain = "co." .. domain;
elseif in_array (domain, {'au', 'br', 'mx'}) then -- Australia, Brazil, Mexico
elseif in_array (domain, {'z.cn'}) then -- China
domain = "cn";
elseif in_array (domain, {'au', 'br', 'mx', 'sg', 'tr'}) then -- Australia, Brazil, Mexico, Singapore, Turkey
domain = "com." .. domain;
domain = "com." .. domain;
elseif not in_array (domain, {'ae', 'ca', 'cn', 'de', 'es', 'fr', 'in', 'it', 'nl', 'pl', 'sa', 'se', 'co.jp', 'co.uk', 'com', 'com.au', 'com.br', 'com.mx', 'com.sg', 'com.tr'}) then -- Arabic Emirates, Canada, China, Germany, Spain, France, Indonesia, Italy, Netherlands, Poland, Saudi Arabia, Sweden (as of 2021-03 Austria (.at), Liechtenstein (.li) and Switzerland (.ch) still redirect to the German site (.de) with special settings, so don't maintain local ASINs for them)
err_flag = set_message ('err_bad_asin_tld'); -- unsupported asin-tld value
end
end
local handler = options.handler;


local handler = options.handler;
if not is_set (err_flag) then
options.coins_list_t['ASIN'] = handler.prefix .. domain .. "/dp/" .. id; -- asin for coins
else
options.coins_list_t['ASIN'] = nil; -- when error, unset so not included in COinS
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix .. domain .. "/dp/",
prefix = handler.prefix .. domain .. "/dp/",
id = id, encode = handler.encode, separator = handler.separator}) .. err_cat;
id = id, encode = handler.encode, separator = handler.separator})
end
end


Line 755: Line 816:


Determines whether an ISMN string is valid.  Similar to ISBN-13, ISMN is 13 digits beginning 979-0-... and uses the
Determines whether an ISMN string is valid.  Similar to ISBN-13, ISMN is 13 digits beginning 979-0-... and uses the
same check digit calculations.  See http://www.ismn-international.org/download/Web_ISMN_Users_Manual_2008-6.pdf
same check digit calculations.  See https://www.ismn-international.org/download/Web_ISMN_Users_Manual_2008-6.pdf
section 2, pages 9–12.
section 2, pages 9–12.
ismn value not made part of COinS metadata because we don't have a url or isn't a COinS-defined identifier (rft.xxx)
or an identifier registered at info-uri.info (info:)


]]
]]
Line 787: Line 851:


if false == valid_ismn then
if false == valid_ismn then
text = text .. ' ' .. set_message ('err_bad_ismn' ) -- add an error message if the ISMN is invalid
options.coins_list_t['ISMN'] = nil; -- when error, unset so not included in COinS; not really necessary here because ismn not made part of COinS
set_message ('err_bad_ismn'); -- create an error message if the ISMN is invalid
end  
end  
Line 800: Line 865:
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: [https://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 839: Line 904:
else
else
if false == valid_issn then
if false == valid_issn then
text = text .. ' ' .. set_message ('err_bad_issn', (options.hkey == 'EISSN') and 'e' or ''); -- add an error message if the ISSN is invalid
options.coins_list_t['ISSN'] = nil; -- when error, unset so not included in COinS
set_message ('err_bad_issn', (options.hkey == 'EISSN') and 'e' or ''); -- create an error message if the ISSN is invalid
end  
end  
end
end
return text
return text;
end
end


Line 857: Line 923:
local handler = options.handler;
local handler = options.handler;
local id_num;
local id_num;
local err_cat = '';
 
id_num = id:match ('^[Jj][Ff][Mm](.*)$'); -- identifier with jfm prefix; extract identifier
id_num = id:match ('^[Jj][Ff][Mm](.*)$'); -- identifier with jfm prefix; extract identifier


Line 870: Line 935:
id = id_num; -- jfm matches pattern
id = id_num; -- jfm matches pattern
else
else
err_cat = ' ' .. set_message ('err_bad_jfm' ); -- set an error message
set_message ('err_bad_jfm' ); -- set an error message
options.coins_list_t['JFM'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode});
end
end


Line 888: Line 954:
local access = options.access;
local access = options.access;
local handler = options.handler;
local handler = options.handler;
local err_msg = '';


if id:find ('[Jj][Ss][Tt][Oo][Rr]') or id:find ('^https?://') or id:find ('%s') then
if id:find ('[Jj][Ss][Tt][Oo][Rr]') or id:find ('^https?://') or id:find ('%s') then
err_msg = ' ' .. set_message ('err_bad_jstor'); -- set an error message
set_message ('err_bad_jstor'); -- set an error message
options.coins_list_t['JSTOR'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access}) .. err_msg;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access});
end
end


Line 903: Line 969:
Format LCCN link and do simple error checking.  LCCN is a character string 8-12 characters long. The length of
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.
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/
https://oclc-research.github.io/infoURI-Frozen/info-uri.info/info:lccn/reg.html


length = 8 then all digits
length = 8 then all digits
Line 916: Line 982:
local lccn = options.id;
local lccn = options.id;
local handler = options.handler;
local handler = options.handler;
local err_cat = ''; -- presume that LCCN is valid
local err_flag; -- presume that LCCN is valid
local id = lccn; -- local copy of the LCCN
local id = lccn; -- local copy of the LCCN


Line 924: Line 990:
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_message ('err_bad_lccn'); -- set an error message
err_flag = set_message ('err_bad_lccn'); -- set an error message
end
end
elseif 9 == len then -- LCCN should be adddddddd
elseif 9 == len then -- LCCN should be adddddddd
if nil == id:match("%l%d%d%d%d%d%d%d%d") then -- does it match our pattern?
if nil == id:match("%l%d%d%d%d%d%d%d%d") then -- does it match our pattern?
err_cat = ' ' .. set_message ('err_bad_lccn'); -- set an error message
err_flag = set_message ('err_bad_lccn'); -- set an error message
end
end
elseif 10 == len then -- LCCN should be aadddddddd or dddddddddd
elseif 10 == len then -- LCCN should be aadddddddd or dddddddddd
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits) ...
if id:match("[^%d]") then -- if LCCN has anything but digits (nil if only digits) ...
if nil == id:match("^%l%l%d%d%d%d%d%d%d%d") then -- ... see if it matches our pattern
if nil == id:match("^%l%l%d%d%d%d%d%d%d%d") then -- ... see if it matches our pattern
err_cat = ' ' .. set_message ('err_bad_lccn'); -- no match, set an error message
err_flag = set_message ('err_bad_lccn'); -- no match, set an error message
end
end
end
end
elseif 11 == len then -- LCCN should be aaadddddddd or adddddddddd
elseif 11 == len then -- LCCN should be aaadddddddd or adddddddddd
if not (id:match("^%l%l%l%d%d%d%d%d%d%d%d") or id:match("^%l%d%d%d%d%d%d%d%d%d%d")) then -- see if it matches one of our patterns
if not (id:match("^%l%l%l%d%d%d%d%d%d%d%d") or id:match("^%l%d%d%d%d%d%d%d%d%d%d")) then -- see if it matches one of our patterns
err_cat = ' ' .. set_message ('err_bad_lccn'); -- no match, set an error message
err_flag = set_message ('err_bad_lccn'); -- no match, set an error message
end
end
elseif 12 == len then -- LCCN should be aadddddddddd
elseif 12 == len then -- LCCN should be aadddddddddd
if not id:match("^%l%l%d%d%d%d%d%d%d%d%d%d") then -- see if it matches our pattern
if not id:match("^%l%l%d%d%d%d%d%d%d%d%d%d") then -- see if it matches our pattern
err_cat = ' ' .. set_message ('err_bad_lccn'); -- no match, set an error message
err_flag = set_message ('err_bad_lccn'); -- no match, set an error message
end
end
else
else
err_cat = ' ' .. set_message ('err_bad_lccn'); -- wrong length, set an error message
err_flag = set_message ('err_bad_lccn'); -- wrong length, set an error message
end
end


if not is_set (err_cat) and nil ~= lccn:find ('%s') then
if not is_set (err_flag) and nil ~= lccn:find ('%s') then
err_cat = ' ' .. set_message ('err_bad_lccn'); -- lccn contains a space, set an error message
err_flag = set_message ('err_bad_lccn'); -- lccn contains a space, set an error message
end
 
if is_set (err_flag) then
options.coins_list_t['LCCN'] = nil; -- when error, unset so not included in COinS
end
end


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




--[[--------------------------< M R >--------------------------------------------------------------------------
--[[--------------------------< M E D R X I V >-----------------------------------------------------------------
 
Format medRxiv ID and do simple error checking.  Similar to later bioRxiv IDs, medRxiv IDs are prefixed with a
yyyy.mm.dd. date and suffixed with an optional version identifier.  Ealiest date accepted is 2020.01.01


A seven digit number; if not seven digits, zero-fill leading digits to make seven digits.
The medRxiv ID is a date followed by an eight-digit number followed by an optional version indicator 'v' and one or more digits:
https://www.medrxiv.org/content/10.1101/2020.11.16.20232009v2 -> 10.1101/2020.11.16.20232009v2


]]
]]


local function mr (options)
local function medrxiv (options)
local id = options.id;
local id = options.id;
local handler = options.handler;
local handler = options.handler;
local id_num;
local err_msg_flag = true; -- flag; assume that there will be an error
local id_len;
local err_cat = '';
id_num = id:match ('^[Mm][Rr](%d+)$'); -- identifier with mr prefix


if is_set (id_num) then
local patterns = {
set_message ('maint_mr_format'); -- add maint cat
'%d%d%d%d%d%d%d%d$', -- simple 8-digit identifier; these should be relatively rare
else -- plain number without mr prefix
'^10%.1101/(20%d%d)%.(%d%d)%.(%d%d)%.%d%d%d%d%d%d%d%dv%d+$', -- y.m.d. date + 8-digit identifier + version (2020-01-01 and later)
id_num = id:match ('^%d+$'); -- if here id is all digits
'^10%.1101/(20%d%d)%.(%d%d)%.(%d%d)%.%d%d%d%d%d%d%d%d$', -- y.m.d. date + 8-digit identifier (2020-01-01 and later)
end
}
for _, pattern in ipairs (patterns) do -- spin through the patterns looking for a match
if id:match (pattern) then
local y, m, d = id:match (pattern); -- found a match, attempt to get year, month and date from the identifier
 
if m then -- m is nil when id is the 8-digit form
if not is_valid_rxiv_date (y, m, d, 'b') then -- validate the encoded date; 'b' for medrxiv limit
break; -- date fail; break out early so we don't unset the error message
end
end
err_msg_flag = nil; -- we found a match so unset the error message
break; -- and done
end
end -- <err_msg_flag> remains set here when no match


id_len = id_num and id_num:len() or 0;
if err_msg_flag then
if (7 >= id_len) and (0 ~= id_len) then
options.coins_list_t['MEDRXIV'] = nil; -- when error, unset so not included in COinS
id = string.rep ('0', 7-id_len) .. id_num; -- zero-fill leading digits
set_message ('err_bad_medrxiv'); -- and set the error message
else
err_cat = ' ' .. set_message ('err_bad_mr'); -- set an error message
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
prefix = handler.prefix, id = id, separator = handler.separator,
encode = handler.encode, access = handler.access});
end
end




--[[--------------------------< O C L C >----------------------------------------------------------------------
--[[--------------------------< M R >--------------------------------------------------------------------------


Validate and format an OCLC ID. https://www.oclc.org/batchload/controlnumber.en.html {{dead link}}
A seven digit number; if not seven digits, zero-fill leading digits to make seven digits.
archived at: https://web.archive.org/web/20161228233804/https://www.oclc.org/batchload/controlnumber.en.html


]]
]]


local function oclc (options)
local function mr (options)
local id = options.id;
local id = options.id;
local handler = options.handler;
local handler = options.handler;
local number;
local id_num;
local err_msg = ''; -- empty string for concatenation
local id_len;
 
id_num = id:match ('^[Mm][Rr](%d+)$'); -- identifier with mr prefix
 
if is_set (id_num) then
set_message ('maint_mr_format'); -- add maint cat
else -- plain number without mr prefix
id_num = id:match ('^%d+$'); -- if here id is all digits
end
 
id_len = id_num and id_num:len() or 0;
if (7 >= id_len) and (0 ~= id_len) then
id = string.rep ('0', 7-id_len) .. id_num; -- zero-fill leading digits
else
set_message ('err_bad_mr'); -- set an error message
options.coins_list_t['MR'] = nil; -- when error, unset so not included in COinS
end
if id:match('^ocm%d%d%d%d%d%d%d%d$') then -- ocm prefix and 8 digits; 001 field (12 characters)
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
number = id:match('ocm(%d+)'); -- get the number
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode});
elseif id:match('^ocn%d%d%d%d%d%d%d%d%d$') then -- ocn prefix and 9 digits; 001 field (12 characters)
end
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)
 
--[[--------------------------< O C L C >----------------------------------------------------------------------
 
Validate and format an OCLC ID.  https://www.oclc.org/batchload/controlnumber.en.html {{dead link}}
archived at: https://web.archive.org/web/20161228233804/https://www.oclc.org/batchload/controlnumber.en.html
 
]]
 
local function oclc (options)
local id = options.id;
local handler = options.handler;
local number;
 
if id:match('^ocm%d%d%d%d%d%d%d%d$') then -- ocm prefix and 8 digits; 001 field (12 characters)
number = id:match('ocm(%d+)'); -- get the number
elseif id:match('^ocn%d%d%d%d%d%d%d%d%d$') 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
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
elseif id:match('^%(OCoLC%)[1-9]%d*$') then -- (OCoLC) prefix and variable number digits; no leading zeros; 035 field
Line 1,016: Line 1,131:
elseif id:match('^%d+$') then -- no prefix
elseif id:match('^%d+$') then -- no prefix
number = id; -- get the number
number = id; -- get the number
if 10 < number:len() then
if tonumber (id) > handler.id_limit then
number = nil; -- constrain to 1 to 10 digits; change this when OCLC issues 11-digit numbers
number = nil; -- unset when id value exceeds the limit
end
end
end
end
Line 1,024: Line 1,139:
id = number; -- exclude prefix, if any, from external link
id = number; -- exclude prefix, if any, from external link
else
else
err_msg = ' ' .. set_message ('err_bad_oclc') -- add an error message if the id is malformed
set_message ('err_bad_oclc') -- add an error message if the id is malformed
options.coins_list_t['OCLC'] = nil; -- when error, unset so not included in COinS
end
end
local text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_msg;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode});
 
return text;
end
end


Line 1,045: Line 1,159:
local handler = options.handler;
local handler = options.handler;
local ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- strip optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W';
local ident, code = id:gsub('^OL', ''):match("^(%d+([AMW]))$"); -- strip optional OL prefix followed immediately by digits followed by 'A', 'M', or 'W';
local error_msg = '';
local err_flag;
local prefix = { -- these are appended to the handler.prefix according to code
local prefix = { -- these are appended to the handler.prefix according to code
['A']='authors/OL',
['A']='authors/OL',
Line 1,056: Line 1,170:
code = 'X'; -- no code or id completely invalid
code = 'X'; -- no code or id completely invalid
ident = id; -- copy id to ident so that we display the flawed identifier
ident = id; -- copy id to ident so that we display the flawed identifier
error_msg = ' ' .. set_message ('err_bad_ol');
err_flag = set_message ('err_bad_ol');
end
end


return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
if not is_set (err_flag) then
options.coins_list_t['OL'] = handler.prefix .. prefix[code] .. ident; -- experiment for ol coins
else
options.coins_list_t['OL'] = nil; -- when error, unset so not included in COinS
end
 
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix .. prefix[code],
prefix = handler.prefix .. prefix[code],
id = ident, separator = handler.separator, encode = handler.encode,
id = ident, separator = handler.separator, encode = handler.encode,
access = access}) .. error_msg;
access = access});
end
end


Line 1,080: Line 1,200:
local access = options.access;
local access = options.access;
local handler = options.handler;
local handler = options.handler;
local err_cat = ''; -- presume that OSTI is valid
 
if id:match("[^%d]") then -- if OSTI has anything but digits
if id:match("[^%d]") then -- if OSTI has anything but digits
err_cat = ' ' .. set_message ('err_bad_osti'); -- set an error message
set_message ('err_bad_osti'); -- set an error message
options.coins_list_t['OSTI'] = nil; -- when error, unset so not included in COinS
else -- OSTI is only digits
else -- OSTI is only digits
local id_num = tonumber (id); -- convert id to a number for range testing
local id_num = tonumber (id); -- convert id to a number for range testing
if 1018 > id_num or handler.id_limit < id_num then -- if OSTI is outside test limit boundaries
if 1018 > id_num or handler.id_limit < id_num then -- if OSTI is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_osti'); -- set an error message
set_message ('err_bad_osti'); -- set an error message
options.coins_list_t['OSTI'] = nil; -- when error, unset so not included in COinS
end
end
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access}) .. err_cat;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = access});
end
end


Line 1,117: Line 1,238:
local embargo = options.Embargo; -- TODO: lowercase?
local embargo = options.Embargo; -- TODO: lowercase?
local handler = options.handler;
local handler = options.handler;
local err_cat;
local err_flag;
local id_num;
local id_num;
local text;
local text;
Line 1,132: Line 1,253:
id_num = tonumber (id_num); -- convert id_num to a number for range testing
id_num = tonumber (id_num); -- convert id_num to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if PMC is outside test limit boundaries
if 1 > id_num or handler.id_limit < id_num then -- if PMC is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_pmc'); -- set an error message
err_flag = set_message ('err_bad_pmc'); -- set an error message
else
else
id = tostring (id_num); -- make sure id is a string
id = tostring (id_num); -- make sure id is a string
end
end
else -- when id format incorrect
else -- when id format incorrect
err_cat = ' ' .. set_message ('err_bad_pmc'); -- set an error message
err_flag = set_message ('err_bad_pmc'); -- set an error message
end
end
Line 1,146: Line 1,267:
handler.separator,
handler.separator,
id,
id,
(err_cat and err_cat or '') -- parens required
});
});
else
else
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect, -- no embargo date or embargo has expired, ok to link to article
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect, -- no embargo date or embargo has expired, ok to link to article
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access,
auto_link = not err_cat and 'pmc' or nil -- do not auto-link when PMC has error
auto_link = not err_flag and 'pmc' or nil -- do not auto-link when PMC has error
}) .. (err_cat and err_cat or ''); -- parentheses required
});
end
end
if err_flag then
options.coins_list_t['PMC'] = nil; -- when error, unset so not included in COinS
end
return text;
return text;
end
end
Line 1,169: Line 1,294:
local id = options.id;
local id = options.id;
local handler = options.handler;
local handler = options.handler;
local err_cat = ''; -- presume that PMID is valid
 
if id:match("[^%d]") then -- if PMID has anything but digits
if id:match("[^%d]") then -- if PMID has anything but digits
err_cat = ' ' .. set_message ('err_bad_pmid'); -- set an error message
set_message ('err_bad_pmid'); -- set an error message
options.coins_list_t['PMID'] = nil; -- when error, unset so not included in COinS
else -- PMID is only digits
else -- PMID is only digits
local id_num = tonumber (id); -- convert id to a number for range testing
local id_num = tonumber (id); -- convert id to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if PMID is outside test limit boundaries
if 1 > id_num or handler.id_limit < id_num then -- if PMID is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_pmid'); -- set an error message
set_message ('err_bad_pmid'); -- set an error message
options.coins_list_t['PMID'] = nil; -- when error, unset so not included in COinS
end
end
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode});
end
end


Line 1,198: Line 1,324:
local id = options.id;
local id = options.id;
local handler = options.handler;
local handler = options.handler;
local err_cat = ''; -- presume that RFC is valid
 
if id:match("[^%d]") then -- if RFC has anything but digits
if id:match("[^%d]") then -- if RFC has anything but digits
err_cat = ' ' .. set_message ('err_bad_rfc'); -- set an error message
set_message ('err_bad_rfc'); -- set an error message
options.coins_list_t['RFC'] = nil; -- when error, unset so not included in COinS
else -- RFC is only digits
else -- RFC is only digits
local id_num = tonumber (id); -- convert id to a number for range testing
local id_num = tonumber (id); -- convert id to a number for range testing
if 1 > id_num or handler.id_limit < id_num then -- if RFC is outside test limit boundaries
if 1 > id_num or handler.id_limit < id_num then -- if RFC is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_rfc'); -- set an error message
set_message ('err_bad_rfc'); -- set an error message
options.coins_list_t['RFC'] = nil; -- when error, unset so not included in COinS
end
end
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access});
end
end


Line 1,228: Line 1,355:
local access = options.access;
local access = options.access;
local handler = options.handler;
local handler = options.handler;
local err_cat = ''; -- presume that S2CID is valid
local id_num;
local id_num;
local text;
local text;
Line 1,237: Line 1,363:
id_num = tonumber (id_num); -- convert id_num to a number for range testing
id_num = tonumber (id_num); -- convert id_num to a number for range testing
if handler.id_limit < id_num then -- if S2CID is outside test limit boundaries
if handler.id_limit < id_num then -- if S2CID is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_s2cid'); -- set an error message
set_message ('err_bad_s2cid'); -- set an error message
options.coins_list_t['S2CID'] = nil; -- when error, unset so not included in COinS
end
end
else -- when id format incorrect
else -- when id format incorrect
err_cat = ' ' .. set_message ('err_bad_s2cid'); -- set an error message
set_message ('err_bad_s2cid'); -- set an error message
options.coins_list_t['S2CID'] = nil; -- when error, unset so not included in COinS
end
end


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


return text;
return text;
Line 1,254: Line 1,381:


9-digit form of ISBN-10; uses same check-digit validation when SBN is prefixed with an additional '0' to make 10 digits
9-digit form of ISBN-10; uses same check-digit validation when SBN is prefixed with an additional '0' to make 10 digits
sbn value not made part of COinS metadata because we don't have a url or isn't a COinS-defined identifier (rft.xxx)
or an identifier registered at info-uri.info (info:)


]]
]]
Line 1,261: Line 1,391:
local ignore_invalid = options.accept;
local ignore_invalid = options.accept;
local handler = options.handler;
local handler = options.handler;
local function return_result (check, err_type) -- local function to handle the various returns
local function return_result (check, err_type) -- local function to handle the various returns
local SBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
local SBN = internal_link_id ({link = handler.link, label = handler.label, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator});
prefix = handler.prefix, id = id, separator = handler.separator});
if not ignore_invalid then -- if not ignoring SBN errors
if not ignore_invalid then -- if not ignoring SBN errors
if not check then
if not check then
return SBN .. set_message ('err_bad_sbn', {err_type}, false, ' '); -- display an error message
options.coins_list_t['SBN'] = nil; -- when error, unset so not included in COinS; not really necessary here because sbn not made part of COinS
set_message ('err_bad_sbn', {err_type}); -- display an error message
return SBN;
end
end
else
else
Line 1,278: Line 1,410:
end
end


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


if  9 ~= ident:len() then
if  9 ~= ident:len() then
Line 1,305: Line 1,437:
local id = options.id;
local id = options.id;
local handler = options.handler;
local handler = options.handler;
local err_cat = ''; -- presume that SSRN is valid
local id_num;
local id_num;
local text;
local text;
Line 1,314: Line 1,445:
id_num = tonumber (id_num); -- convert id_num to a number for range testing
id_num = tonumber (id_num); -- convert id_num to a number for range testing
if 100 > id_num or handler.id_limit < id_num then -- if SSRN is outside test limit boundaries
if 100 > id_num or handler.id_limit < id_num then -- if SSRN is outside test limit boundaries
err_cat = ' ' .. set_message ('err_bad_ssrn'); -- set an error message
set_message ('err_bad_ssrn'); -- set an error message
options.coins_list_t['SSRN'] = nil; -- when error, unset so not included in COinS
end
end
else -- when id format incorrect
else -- when id format incorrect
err_cat = ' ' .. set_message ('err_bad_ssrn'); -- set an error message
set_message ('err_bad_ssrn'); -- set an error message
options.coins_list_t['SSRN'] = nil; -- when error, unset so not included in COinS
end
end
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
text = external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = handler.access}) .. err_cat;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode, access = options.access});


return text;
return text;
Line 1,342: Line 1,475:
   
   
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_message ('err_bad_usenet_id') -- add an error message if the message id is invalid
set_message ('err_bad_usenet_id') -- add an error message if the message id is invalid
options.coins_list_t['USENETID'] = nil; -- when error, unset so not included in COinS
end  
end  
return text
return text;
end
end


Line 1,362: Line 1,496:
local id = options.id;
local id = options.id;
local handler = options.handler;
local handler = options.handler;
local err_cat = '';
 
if id:match('^%d%d%d%d%d%d%d%d$') then -- is this identifier using temporary format?
if id:match('^%d%d%d%d%d%d%d%d$') then -- is this identifier using temporary format?
set_message ('maint_zbl'); -- yes, add maint cat
set_message ('maint_zbl'); -- yes, add maint cat
elseif not id:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then -- not temporary, is it normal format?
elseif not id:match('^%d?%d?%d?%d%.%d%d%d%d%d$') then -- not temporary, is it normal format?
err_cat = ' ' .. set_message ('err_bad_zbl'); -- no, set an error message
set_message ('err_bad_zbl'); -- no, set an error message
options.coins_list_t['ZBL'] = nil; -- when error, unset so not included in COinS
end
end
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
return external_link_id ({link = handler.link, label = handler.label, q = handler.q, redirect = handler.redirect,
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode}) .. err_cat;
prefix = handler.prefix, id = id, separator = handler.separator, encode = handler.encode});
end
end


Line 1,414: Line 1,548:
if is_set (access_level) then
if is_set (access_level) then
if not in_array (access_level, cfg.keywords_lists['id-access']) then -- exact match required
if not in_array (access_level, cfg.keywords_lists['id-access']) then -- exact match required
table.insert (z.message_tail, { set_message ('invalid_param_val', {access_param, access_level}, true) } );
set_message ('err_invalid_param_val', {access_param, access_level});
access_level = nil; -- invalid so unset
access_level = nil; -- invalid so unset
end
end
if not is_set (id_list[k]) then -- identifier access-level must have a matching identifier
if not is_set (id_list[k]) then -- identifier access-level must have a matching identifier
table.insert (z.message_tail, { set_message ('err_param_access_requires_param', {k:lower()}, true) } ); -- parameter name is uppercase in cfg.id_handlers (k); lowercase for error message
set_message ('err_param_access_requires_param', {k:lower()}); -- parameter name is uppercase in cfg.id_handlers (k); lowercase for error message
end
end
id_accesses_list[k] = cfg.keywords_xlate[access_level]; -- get translated keyword
id_accesses_list[k] = cfg.keywords_xlate[access_level]; -- get translated keyword
Line 1,432: Line 1,566:
render the identifiers into a sorted sequence table
render the identifiers into a sorted sequence table


<id_list_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value
<ID_list_coins_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value
<options_t> is a table of is a table of various k/v option pairs provided in the call to new_build_id_list();
<options_t> is a table of various k/v option pairs provided in the call to new_build_id_list();
modified by this function and passed to all identifier rendering functions
modified by this function and passed to all identifier rendering functions
<access_levels_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value (if valid)
<access_levels_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value (if valid)


returns a sequence table of sorted (by hkey) rendered identifier strings
returns a sequence table of sorted (by hkey - 'handler' key) rendered identifier strings


]]
]]


local function build_id_list (id_list_t, options_t, access_levels_t)
local function build_id_list (ID_list_coins_t, options_t, access_levels_t)
local new_list_t = {};
local ID_list_t = {};
local accept;
local accept;
local func_map = { --function map points to functions associated with hkey identifier
local func_map = { --function map points to functions associated with hkey identifier
Line 1,459: Line 1,593:
['JSTOR'] = jstor,
['JSTOR'] = jstor,
['LCCN'] = lccn,
['LCCN'] = lccn,
['MEDRXIV'] = medrxiv,
['MR'] = mr,
['MR'] = mr,
['OCLC'] = oclc,
['OCLC'] = oclc,
Line 1,465: Line 1,600:
['PMC'] = pmc,
['PMC'] = pmc,
['PMID'] = pmid,
['PMID'] = pmid,
['RFC']  = rfc,
['RFC']  = rfc,
['S2CID'] = s2cid,
['S2CID'] = s2cid,
['SBN'] = sbn,
['SBN'] = sbn,
['SSRN'] = ssrn,
['SSRN'] = ssrn,
['USENETID'] = usenet_id,
['USENETID'] = usenet_id,
['ZBL'] = zbl,
['ZBL'] = zbl,
}
}
 
 
for hkey, v in pairs (id_list_t) do
for hkey, v in pairs (ID_list_coins_t) do
v, accept = has_accept_as_written (v); -- remove accept-as-written markup if present; accept is boolean true when markup removed; false else
v, accept = has_accept_as_written (v); -- remove accept-as-written markup if present; accept is boolean true when markup removed; false else
-- every function gets the options table with value v and accept boolean
-- every function gets the options table with value v and accept boolean
options_t.hkey = hkey; -- ~/Configuration handler key
options_t.hkey = hkey; -- ~/Configuration handler key
options_t.id = v; -- add that identifier value to the options table
options_t.id = v; -- add that identifier value to the options table
options_t.accept = accept; -- add the accept boolean flag
options_t.accept = accept; -- add the accept boolean flag
options_t.access = access_levels_t[hkey]; -- add the access level for those that have an |<identifier-access= parameter
options_t.access = access_levels_t[hkey]; -- add the access level for those that have an |<identifier-access= parameter
options_t.handler = cfg.id_handlers[hkey];
options_t.handler = cfg.id_handlers[hkey];
if func_map[hkey] then
options_t.coins_list_t = ID_list_coins_t; -- pointer to ID_list_coins_t; for |asin= and |ol=; also to keep erroneous values out of the citation's metadata
table.insert (new_list_t, {hkey, func_map[hkey] (options_t)}); -- call the function and add the results to the output sequence table
options_t.coins_list_t[hkey] = v; -- id value without accept-as-written markup for metadata
if options_t.handler.access and not in_array (options_t.handler.access, cfg.keywords_lists['id-access']) then
error (cfg.messages['unknown_ID_access'] .. options_t.handler.access); -- here when handler access key set to a value not listed in list of allowed id access keywords
end
 
if func_map[hkey] then
local id_text = func_map[hkey] (options_t); -- call the function to get identifier text and any error message
table.insert (ID_list_t, {hkey, id_text}); -- add identifier text to the output sequence table
else
error (cfg.messages['unknown_ID_key'] .. hkey); -- here when func_map doesn't have a function for hkey
end
end
 
local function comp (a, b) -- used by following table.sort()
return a[1]:lower() < b[1]:lower(); -- sort by hkey
end
 
table.sort (ID_list_t, comp); -- sequence table of tables sort
for k, v in ipairs (ID_list_t) do -- convert sequence table of tables to simple sequence table of strings
ID_list_t[k] = v[2]; -- v[2] is the identifier rendering from the call to the various functions in func_map{}
end
return ID_list_t;
end
 
 
--[[--------------------------< O P T I O N S _ C H E C K >----------------------------------------------------
 
check that certain option parameters have their associated identifier parameters with values
 
<ID_list_coins_t> is a table of k/v pairs where k is same as key in cfg.id_handlers and v is the assigned value
<ID_support_t> is a sequence table of tables created in citation0() where each subtable has four elements:
[1] is the support parameter's assigned value; empty string if not set
[2] is a text string same as key in cfg.id_handlers
[3] is cfg.error_conditions key used to create error message
[4] is original ID support parameter name used to create error message
returns nothing; on error emits an appropriate error message
 
]]


-- TODO: also retrieve identifier validity status, OL A/M/W/X type and ASIN TLD info from the corresponding function call for improved metadata generation in COinS() in ~/COinS
local function options_check (ID_list_coins_t, ID_support_t)
 
for _, v in ipairs (ID_support_t) do
else
if is_set (v[1]) and not ID_list_coins_t[v[2]] then -- when support parameter has a value but matching identifier parameter is missing or empty
error (cfg.messages['unknown_ID_key'] .. ' ' .. hkey); -- here when func_map doesn't have a function for hkey
set_message (v[3], (v[4])); -- emit the appropriate error message
end
end
end
end
local function comp (a, b) -- used by following table.sort()
return a[1]:lower() < b[1]:lower(); -- sort by hkey
end
table.sort (new_list_t, comp); -- sequence table of tables sort
for k, v in ipairs (new_list_t) do -- convert sequence table of tables to simple sequence table of strings
new_list_t[k] = v[2];
end
return new_list_t;
end
end


Line 1,512: Line 1,676:
]]
]]


local function identifier_lists_get (args, options_t)
local function identifier_lists_get (args_t, options_t, ID_support_t)
local ID_list_coins_t = extract_ids (args); -- get a table of identifiers and their values for use locally and for use in COinS
local ID_list_coins_t = extract_ids (args_t); -- get a table of identifiers and their values for use locally and for use in COinS
local ID_access_levels_t = extract_id_access_levels (args, ID_list_coins_t); -- get a table of identifier access levels
options_check (ID_list_coins_t, ID_support_t); -- ID support parameters must have matching identifier parameters
local ID_access_levels_t = extract_id_access_levels (args_t, ID_list_coins_t); -- get a table of identifier access levels
local ID_list_t = build_id_list (ID_list_coins_t, options_t, ID_access_levels_t); -- get a sequence table of rendered identifier strings
local ID_list_t = build_id_list (ID_list_coins_t, options_t, ID_access_levels_t); -- get a sequence table of rendered identifier strings
-- TODO: add code to retrieve identifier validity status, OL A/M/W/X type and ASIN TLD info from ID_list_t and add this to ID_list_coins_t for improved metadata generation in COinS() in ~/COinS


return ID_list_t, ID_list_coins_t; -- return the tables
return ID_list_t, ID_list_coins_t; -- return the tables
Line 1,550: Line 1,713:
auto_link_urls = auto_link_urls, -- table of identifier URLs to be used when auto-linking |title=
auto_link_urls = auto_link_urls, -- table of identifier URLs to be used when auto-linking |title=
identifier_lists_get = identifier_lists_get, -- experiment to replace individual calls to build_id_list, extract_ids, extract_id_access_levels
identifier_lists_get = identifier_lists_get, -- experiment to replace individual calls to build_id_list(), extract_ids, extract_id_access_levels
is_embargoed = is_embargoed;
is_embargoed = is_embargoed;
set_selected_modules = set_selected_modules;
set_selected_modules = set_selected_modules;
}
}
Retrieved from "https://en.bharatpedia.org/wiki/Special:MobileDiff/431507"