Content deleted Content added
No edit summary |
m Removed protection from "Module:Citation/CS1/Utilities/sandbox" |
||
(44 intermediate revisions by 4 users not shown) | |||
Line 1:
--[[
History of changes since last sync:
]]
Line 8 ⟶ 6:
local z = {
error_ids_t = {}; -- list of error identifiers; used to prevent duplication of certain errors; local to this module
error_msgs_t = {}; -- sequence table of error messages
prop_keys_t = {}; -- for adding classes to the citation's <cite> tag
};
Line 28 ⟶ 27:
]]
local function is_set (
return not (var == nil or var == '');
end
Line 39 ⟶ 38:
]]
local function in_array (
if needle == nil then
return false;
end
for n, v in ipairs
if v == needle then
return n;
Line 49 ⟶ 48:
end
return false;
end
--[[--------------------------< H A S _ A C C E P T _ A S _ W R I T T E N >------------------------------------
When <str> is wholly wrapped in accept-as-written markup, return <str> without markup and true; return <str> and false else
with allow_empty = false, <str> must have at least one character inside the markup
with allow_empty = true, <str> the markup frame can be empty like (()) to distinguish an empty template parameter from the specific condition "has no applicable value" in citation-context.
After further evaluation the two cases might be merged at a later stage, but should be kept separated for now.
]]
local function has_accept_as_written (str, allow_empty)
if not is_set (str) then
return str, false;
end
local count;
if true == allow_empty then
str, count = str:gsub ('^%(%((.*)%)%)$', '%1'); -- allows (()) to be an empty set
else
str, count = str:gsub ('^%(%((.+)%)%)$', '%1');
end
return str, 0 ~= count;
end
Line 54 ⟶ 80:
--[[--------------------------< S U B S T I T U T E >----------------------------------------------------------
Populates numbered arguments in a message string using an argument table. <args> may be a single string or a
sequence table of multiple strings.
]]
local function substitute (
return args and mw.message.newRawMessage (
end
Line 65 ⟶ 92:
--[[--------------------------< E R R O R _ C O M M E N T >----------------------------------------------------
Wraps error messages with CSS markup according to the state of hidden. <content> may be a single string or a
sequence table of multiple strings.
]]
local function error_comment (
return substitute
end
--[[--------------------------< H Y P H E N _ T O _ D A S H >--------------------------------------------------
Converts a hyphen, endash, emdash to endash under certain conditions. The hyphen/en/em must separate
like items; unlike items are returned unmodified. These forms are modified:
letter - letter (A-B)
digit - digit (4-5)
digit separator digit - digit separator digit (4.1-4.5 or 4-1-4-5)
letterdigit - letterdigit (A1-A5) (an optional separator between letter and
digit is supported – a.1-a.5 or a-1-a-5)
digitletter - digitletter (5a-5d) (an optional separator between letter and
digit is supported – 5.a-5.d or 5-a-5-d)
any other forms are returned unmodified.
str may be a comma- or semicolon-separated list of page ranges with/without single pages
]]
local function hyphen_to_dash (str)
if not is_set (str) then
return str;
end
str = str:gsub ("(%(%(.-%)%))", function(m) return m:gsub(",", ","):gsub(";", ";") end) -- replace commas and semicolons in accept-as-written markup with similar unicode characters so they'll be ignored during the split
str = str:gsub ('&[nm]dash;', {['–'] = '–', ['—'] = '—'}); -- replace — and – entities with their characters; semicolon mucks up the text.split
str = str:gsub ('-', '-'); -- replace HTML numeric entity with hyphen character
str = str:gsub (' ', ' '); -- replace entity with generic keyboard space character
local out = {};
local list = mw.text.split (str, '%s*[,;]%s*'); -- split str at comma or semicolon separators if there are any
local accept; -- boolean
for _, item in ipairs (list) do -- for each item in the list
item, accept = has_accept_as_written (item); -- remove accept-this-as-written markup when it wraps all of item
if not accept and mw.ustring.match (item, '^%w*[%.%-]?%w+%s*[—–-]%s*%w*[%.%-]?%w+$') then -- if a hyphenated range or has endash or emdash separators
if mw.ustring.match (item, '^%a+[%.%-]?%d+%s*[—–-]%s*%a+[%.%-]?%d+$') or -- letterdigit hyphen letterdigit (optional separator between letter and digit)
mw.ustring.match (item, '^%d+[%.%-]?%a+%s*[—–-]%s*%d+[%.%-]?%a+$') or -- digitletter hyphen digitletter (optional separator between digit and letter)
mw.ustring.match (item, '^%d+[%.%-]%d+%s*[—–-]%s*%d+[%.%-]%d+$') then -- digit separator digit hyphen digit separator digit
item = mw.ustring.gsub (item, '(%w*[%.%-]?%w+)%s*[—–-]%s*(%w*[%.%-]?%w+)', '<span class="nowrap">%1 –</span> <span class="nowrap">%2</span>'); -- replace hyphen/dash, with spaced endash
elseif mw.ustring.match (item, '^%d+%s*[—–-]%s*%d+$') or -- digit hyphen digit
mw.ustring.match (item, '^%a+%s*[—–-]%s*%a+$') then -- letter hyphen letter
item = mw.ustring.gsub (item, '(%w+)%s*[—–-]%s*(%w+)', '<span class="nowrap">%1–</span>%2'); -- replace hyphen/emdash with endash, remove extraneous space characters
else
-- item = mw.ustring.gsub (item, '%s*[—–-]%s*', '–'); -- disabled; here when 'unlike' items so return <item> as is
end
end
table.insert (out, item); -- add the (possibly modified) item to the output table
end
local temp_str = ''; -- concatenate the output table into a comma separated string
temp_str, accept = has_accept_as_written (table.concat (out, ', ')); -- remove accept-this-as-written markup when it wraps all of concatenated out
if accept then
temp_str = has_accept_as_written (str); -- when global markup removed, return original str; do it this way to suppress boolean second return value
return temp_str:gsub(",", ","):gsub(";", ";");
else
return temp_str:gsub(",", ","):gsub(";", ";"); -- else, return assembled temp_str
end
end
Line 77 ⟶ 168:
Makes a wikilink; when both link and display text is provided, returns a wikilink in the form [[L|D]]; if only
link is provided (or link and display are the same), returns a wikilink in the form [[L]]; if neither are
provided or link is omitted, returns an empty string. ]=]
local function make_wikilink (link, display)
if not is_set (link) then return '' end
return table.concat ({'[[', link, '|', display, ']]'});
else
return table.concat ({'[[', link, ']]'});
end
end
Line 97 ⟶ 186:
--[[--------------------------< S E T _ M E S S A G E >----------------------------------------------------------
Sets an error
call, inserts the resulting message in z.error_msgs_t{} sequence table, and returns the error message.
<error_id> – key value for appropriate error handler in ~/Configuration error_conditions{} table
<arguments> – may be a single string or a sequence table of multiple strings to be subsititued into error_conditions[error_id].message
<raw> – boolean
true – causes this function to return the error message not wrapped in visible-error, hidden-error span tag;
returns error_conditions[error_id].hidden as a second return value
does not add message to z.error_msgs_t sequence table
false, nil – adds message wrapped in visible-error, hidden-error span tag to z.error_msgs_t
returns the error message wrapped in visible-error, hidden-error span tag; there is no second return value
<prefix> – string to be prepended to <message> -- TODO: remove support for these unused(?) arguments?
<suffix> – string to be appended to <message>
TODO: change z.error_cats_t and z.maint_cats_t to have the form cat_name = true? this to avoid dups without having to have an extra table
]]
local added_maint_cats = {} -- list of maintenance categories that have been added to z.maint_cats_t; TODO: figure out how to delete this table
local function set_message (error_id, arguments, raw, prefix, suffix)
local error_state = cfg.error_conditions[error_id];
prefix = prefix or
suffix = suffix or
if error_state == nil then
error
elseif is_set (error_state.category) then
if error_state.message then -- when error_state.message defined, this is an error message
table.insert (
else
if not added_maint_cats[error_id] then
added_maint_cats[error_id] = true; -- note that we've added this category
table.insert (z.
end
return; -- because no message, nothing more to do
Line 127 ⟶ 227:
end
local message = substitute (
message = table.concat (
Line 144 ⟶ 244:
});
z.
if z.
in_array (error_id, {'err_bare_url_missing_title', 'err_trans_missing_title'}) then -- and this error is one of these
return '', false; -- don't bother because one flavor of missing title is sufficient
end
message = table.concat ({
if
return message, error_state.hidden; -- return message not wrapped in visible-error, hidden-error span tag
end
table.insert (z.error_msgs_t, message); -- add it to the messages sequence table
return message; -- and done; return value generally not used but is used as a flag in various functions of ~/Identifiers
end
Line 187 ⟶ 289:
end
if is_set (args[alias]) then
if value ~= nil and selected ~= alias then -- if we have already selected one of the aliases
local skip;
for _, v in ipairs (error_list) do -- spin through the error list to see if we've added this alias
if v == alias then
skip = true;
Line 197 ⟶ 299:
end
if not skip then -- has not been added so
table.insert (
end
else
Line 210 ⟶ 312:
--[[--------------------------< A D D _ M A I N T _ C A T >------------------------------------------------------
Adds a category to z.
To prevent duplication, the added_maint_cats table lists the categories by key that have been added to z.
]]
local function add_maint_cat (key, arguments)
if not added_maint_cats [key] then
added_maint_cats [key] = true; -- note that we've added this category
table.insert
end
end
--[[--------------------------< A D D _ P R O P _ C A T >--------------------------------------------------------
Adds a category to z.prop_cats_t using names from the configuration file with additional text if any.
foreign_lang_source and foreign_lang_source_2 keys have a language code appended to them so that multiple languages
may be categorized but multiples of the same language are not categorized.
added_prop_cats is a table declared in page scope variables above
]]
local added_prop_cats = {}; -- list of property categories that have been added to z.prop_cats_t
local function add_prop_cat (key, arguments, key_modifier)
local key_modified = key .. ((key_modifier and key_modifier) or ''); -- modify <key> with <key_modifier> if present and not nil
if not added_prop_cats [key_modified] then
added_prop_cats [key_modified] = true; -- note that we've added this category
table.insert (z.prop_cats_t, substitute (cfg.prop_cats [key], arguments)); -- make name then add to table
table.insert (z.prop_keys_t, 'cs1-prop-' .. key); -- convert key to class for use in the citation's <cite> tag
end
end
Line 234 ⟶ 359:
]]
local function safe_for_italics (
if not is_set (str) then return str end
if str:sub (1, 1) == "'" then str = "<span></span>" .. str; end
return str:gsub ('\n', ' '); -- Remove newlines as they break italics.
end
Line 256 ⟶ 378:
local function wrap_style (key, str)
if not is_set (
return "";
elseif in_array
str = safe_for_italics (
end
return substitute (
end
--[[--------------------------<
make a separated list of items using provided separators.
Line 272 ⟶ 394:
<sep_list_pair> - typically '<space>and<space>'
<sep_list_end> - typically '<comma><space>and<space>' or '<comma><space>&<space>'
defaults to cfg.presentation['sep_list'], cfg.presentation['sep_list_pair'], and cfg.presentation['sep_list_end']
if <sep_list_end> is specified, <sep_list> and <sep_list_pair> must also be supplied
]]
local function
local list = '';
if not sep_list then -- set the defaults
sep_list = cfg.presentation['sep_list'];
sep_list_pair = cfg.presentation['sep_list_pair'];
sep_list_end = cfg.presentation['sep_list_end'];
end
if 2 >= count then
list = table.concat (list_seq, sep_list_pair); -- insert separator between two items; returns list_seq[1] then only one item
elseif 2 < count then
list = table.concat (list_seq, sep_list, 1, count - 1); -- concatenate all but last item with plain list separator
list = table.concat ({list, list_seq[count]}, sep_list_end); -- concatenate last item onto end of <list> with final separator
end
Line 286 ⟶ 418:
return list;
end
--[[--------------------------< S E L E C T _ O N E >----------------------------------------------------------
Line 307 ⟶ 440:
if index ~= nil then index = tostring(index); end
for _, alias in ipairs (
if alias:match ('#') then -- if this alias can be enumerated
if '1' == index then -- when index is 1 test for enumerated and non-enumerated aliases
value, selected = is_alias_used (args, alias, index, false, value, selected, error_list); -- first test for non-enumerated alias
end
value, selected = is_alias_used (args, alias, index, true, value, selected, error_list);
else
value, selected = is_alias_used (args, alias, index, false, value, selected, error_list);
end
end
if #error_list > 0 and 'none' ~= error_condition then -- for cases where this code is used outside of extract_names()
for i, v in ipairs (error_list) do
error_list[i] = wrap_style ('parameter', v);
end
table.insert (error_list, wrap_style ('parameter', selected));
set_message (error_condition, {make_sep_list (#error_list, error_list)});
end
Line 358 ⟶ 476:
local function remove_wiki_link (str)
return (str:gsub
return l:gsub
end));
end
Line 395 ⟶ 513:
D = mw.text.trim (D, '%s|'); -- trim white space and pipe characters
return wl_type, D, L or '';
end
Line 423 ⟶ 540:
local flag;
while true do
if argument:find (
argument, flag = argument:gsub ("%'%'%'%'%'", "");
elseif argument:find (
argument, flag=argument:gsub ("%'%'%'%'", "");
elseif argument:find (
argument, flag=argument:gsub ("%'%'%'", "");
elseif argument:find (
argument, flag = argument:gsub ("%'%'", "");
else
break;
Line 457 ⟶ 574:
return {
add_maint_cat = add_maint_cat, -- exported functions
add_prop_cat = add_prop_cat,
error_comment = error_comment,
has_accept_as_written = has_accept_as_written,
hyphen_to_dash = hyphen_to_dash,
in_array = in_array,
is_set = is_set,
is_wikilink = is_wikilink,
make_sep_list = make_sep_list,
make_wikilink = make_wikilink,
remove_wiki_link = remove_wiki_link,
|