Module:links: Difference between revisions

Jump to navigation Jump to search

Module:links (view source)

Revision as of 10:38, 6 August 2024

28,518 bytes added ,  6 August
no edit summary
No edit summary
No edit summary
Line 2: Line 2:


--[=[
--[=[
[[Unsupported titles]] and pages with high
[[Unsupported titles]], pages with high memory usage,
memory usage are listed at [[Module:links/data]].
extraction modules and part-of-speech names are listed
at [[Module:links/data]].


Other modules used:
Other modules used:
Line 10: Line 11:
[[Module:languages]] and its submodules
[[Module:languages]] and its submodules
[[Module:gender and number]]
[[Module:gender and number]]
[[Module:utilities]]
[[Module:debug/track]]
[[Module:string]]
]=]
]=]
local m_str_utils = require("Module:string utilities")
local pron_qualifier_module = "Module:pron qualifier"


-- These are prefixed with u to avoid confusion with the default string methods
local anchorEncode = mw.uri.anchorEncode
-- of the same name.
local concat = table.concat
local usub = mw.ustring.sub
local decode_entities = m_str_utils.decode_entities
local decode_uri = m_str_utils.decode_uri
local find = string.find
local encode_entities = require("Module:string/encode entities") -- Can't yet replace, as the [[Module:string utilities]] version no longer has automatic double-encoding prevention, which requires changes here to account for.
local insert = table.insert
local match = string.match
local new_title = mw.title.new
local remove = table.remove
local shallowcopy = require("Module:table").shallowcopy
local split = m_str_utils.split
local sub = string.sub
local toNFC = mw.ustring.toNFC
local trim -- defined below
local ulower = m_str_utils.lower
local umatch = m_str_utils.match
local unstrip = mw.text.unstrip
local u = m_str_utils.char
local TEMP_UNDERSCORE = u(0xFFF0)


local table_insert = table.insert
local function track(page, code)
local table_concat = table.concat
local tracking_page = "links/" .. page
if code then
require("Module:debug/track"){tracking_page, tracking_page .. "/" .. code}
else
require("Module:debug/track")(tracking_page)
end
return true
end
 
do
local _trim = m_str_utils.trim
-- Unconditionally trimmed charset.
local always_trim =
"\194\128-\194\159" .. -- U+0080-009F (C1 control characters)
"\194\173" .. -- U+00AD (soft hyphen)
"\226\128\170-\226\128\174" .. -- U+202A-202E (directionality formatting characters)
"\226\129\166-\226\129\169" -- U+2066-2069 (directionality formatting characters)
-- Standard trimmed charset.
local standard_trim = "%s" .. -- (default whitespace charset)
"\226\128\139-\226\128\141" .. -- U+200B-200D (zero-width spaces)
always_trim
-- If there are non-whitespace characters, trim all characters in `standard_trim`.
-- Otherwise, only trim the characters in `always_trim`.
trim = function(text)
if text == "" then
return text
end
local trimmed = _trim(text, standard_trim)
if trimmed ~= "" then
return trimmed
end
return _trim(text, always_trim)
end
end
 
local function escape(text, str)
local rep
repeat
text, rep = text:gsub("\\\\(\\*" .. str .. ")", "\5%1")
until rep == 0
return (text:gsub("\\" .. str, "\6"))
end
 
local function unescape(text, str)
return (text
:gsub("\5", "\\")
:gsub("\6", str))
end


local ignore_cap = {
-- Remove bold, italics, soft hyphens, strip markers and HTML tags.
["ko"] = true,
local function remove_formatting(str)
}
str = str
:gsub("('*)'''(.-'*)'''", "%1%2")
:gsub("('*)''(.-'*)''", "%1%2")
:gsub("­", "")
return (unstrip(str)
:gsub("<[^<>]+>", ""))
end


local phonetic_extraction = {
--[==[Takes an input and splits on a double slash (taking account of escaping backslashes).]==]
["th"] = "Module:th",
function export.split_on_slashes(text)
["km"] = "Module:km",
text = split(escape(text, "//"), "//", true) or {}
}
for i, v in ipairs(text) do
text[i] = unescape(v, "//")
if v == "" then
text[i] = false
end
end
return text
end


local pos_tags = {
-- Does the work of export.get_fragment, but can be called directly to avoid unnecessary checks for embedded links.
["a"] = "adjective",
local function get_fragment(text)
["adv"] = "adverb",
text = escape(text, "#")
["int"] = "interjection",
-- Replace numeric character references with the corresponding character (&#29; → '),
["n"] = "noun",
-- as they contain #, which causes the numeric character reference to be
["pron"] = "pronoun",
-- misparsed (wa'a → wa&#29;a → pagename wa&, fragment 29;a).
["v"] = "verb",
text = decode_entities(text)
["vi"] = "intransitive verb",
local target, fragment = text:match("^(..-)#(.+)$")
["vt"] = "transitive verb",
target = target or text
["vti"] = "transitive and intransitive verb",
target = unescape(target, "#")
}
fragment = fragment and unescape(fragment, "#")
return target, fragment
end


local unsupported_titles
--[==[Takes a link target and outputs the actual target and the fragment (if any).]==]
function export.getLinkPage(target, lang)
function export.get_fragment(text)
unsupported_titles = unsupported_titles or mw.loadData("Module:links/data").unsupported_titles
-- If there are no embedded links, process input.
if unsupported_titles[target] then
local open = find(text, "[[", 1, true)
return "Unsupported titles/" .. unsupported_titles[target]
if not open then
return get_fragment(text)
end
local close = find(text, "]]", open + 2, true)
if not close then
return get_fragment(text)
-- If there is one, but it's redundant (i.e. encloses everything with no pipe), remove and process.
elseif open == 1 and close == #text - 1 and not find(text, "|", 3, true) then
return get_fragment(sub(text, 3, -3))
end
end
-- Otherwise, return the input.
return text
end


-- If the link contains unexpanded template parameters, then don't create a link.
local pos_tags
if target:find("{{{") then
function export.get_link_page(target, lang, sc, plain)
if not target then
return nil
return nil
end
target = remove_formatting(target)
-- Check if the target is an interwiki link.
if target:match(":") and target ~= ":" then
-- If this is an a link to another namespace or an interwiki link, ensure there's an initial colon and then return what we have (so that it works as a conventional link, and doesn't do anything weird like add the term to a category.)
local prefix = target:gsub("^:*(.-):.*", ulower)
if (
mw.loadData("Module:data/namespaces")[prefix] or
mw.loadData("Module:data/interwikis")[prefix]
) then
return ":" .. target:gsub("^:+", ""), nil, {}
end
-- Convert any escaped colons
target = target:gsub("\\:", ":")
end
-- Check if the term is reconstructed and remove any asterisk. Otherwise, handle the escapes.
local reconstructed, escaped
if not plain then
target, reconstructed = target:gsub("^%*(.)", "%1")
end
target, escaped = target:gsub("^(\\-)\\%*", "%1*")
if not require("Module:utilities").check_object("script", true, sc) or sc:getCode() == "None" then
sc = lang:findBestScript(target)
end
end


if target:sub(1, 1) == ":" or target:sub(1, 2) == "w:" or target:sub(1, 10) == "wikipedia:" then
-- Remove carets if they are used to capitalize parts of transliterations (unless they have been escaped).
return target
if (not sc:hasCapitalization()) and sc:isTransliterated() and target:match("%^") then
target = escape(target, "^")
:gsub("%^", "")
target = unescape(target, "^")
end
end


-- Remove diacritics from the page name
-- Get the entry name for the language.
target = lang:makeEntryName(target)
target = lang:makeEntryName(target, sc)
 
-- If the link contains unexpanded template parameters, then don't create a link.
if target:find("{{{") then
return nil
end


if target:sub(1, 1) == "/" then
if target:sub(1, 1) == "/" then
return ":" .. target
return ":" .. target
end


return target
elseif target:find("^Reconstruction:") then
end
return target


-- Make a language-specific link from given link's parts
-- Link to appendix for reconstructed terms and terms in appendix-only languages. Plain links interpret * literally, however.
local function makeLangLink(link, lang, id, allow_self_link)
elseif reconstructed == 1 then
-- Temporary tracking code
if lang:getFullCode() == "und" then
local langCode = lang:getCode()
return nil
if langCode == "se" or langCode == "sia" or langCode:find("^sm[ajns]$")
else
or langCode:find("^sj[dektu]$") then
target = "Reconstruction:" .. lang:getFullName() .. "/" .. target
if link.display and link.display:find("'") then
end
elseif link.target and link.target:find("'") then
-- Reconstructed languages and substrates require an initial *.
elseif lang:hasType("reconstructed") or lang:getFamilyCode() == "qfa-sub" then
local check = target:match("^:*([^:]*):")
check = check and ulower(check)
if (
mw.loadData("Module:data/namespaces")[check] or
mw.loadData("Module:data/interwikis")[check]
) then
return target
else
error("The specified language " .. lang:getCanonicalName()
.. " is unattested, while the given term does not begin with '*' to indicate that it is reconstructed.")
end
end
elseif lang:hasType("appendix-constructed") then
target = "Appendix:" .. lang:getFullName() .. "/" .. target
end
end


-- Find fragments (when link didn't come from parseLink).
return target, escaped > 0
end
 
-- Make a link from a given link's parts
local function make_link(link, lang, sc, id, isolated, plain, cats, no_alt_ast)
-- Convert percent encoding to plaintext.
link.target = decode_uri(link.target, "PATH")
link.fragment = link.fragment and decode_uri(link.fragment, "PATH")
-- Find fragments (if one isn't already set).
-- Prevents {{l|en|word#Etymology 2|word}} from linking to [[word#Etymology 2#English]].
-- Prevents {{l|en|word#Etymology 2|word}} from linking to [[word#Etymology 2#English]].
if link.fragment == nil then
-- # can be escaped as \#.
-- Replace numeric character references with the corresponding character (&#29; → '),
if link.target and link.fragment == nil then
-- as they contain #, which causes the numeric character reference to be
link.target, link.fragment = get_fragment(link.target)
-- misparsed (wa'a → wa&#29;a → pagename wa&, fragment 29;a).
link.target = link.target:gsub("&#(%d+);",
function(number) return mw.ustring.char(tonumber(number)) end)
local first, second = link.target:match("^([^#]+)#(.+)$")
if first then
link.target, link.fragment = first, second
end
end
end


-- If there is no display form, then create a default one
-- Create a default display form.
if not link.display then
local auto_display = link.target
link.display = link.target
-- Process the target
local escaped
link.target, escaped = export.get_link_page(link.target, lang, sc, plain)


-- Strip the prefix from the displayed form
-- If the display is the target and the reconstruction * has been escaped, remove the escaping backslash.
-- TODO: other interwiki links?
if escaped then
if link.display:sub(1, 1) == ":" and not mw.loadData("Module:links/data").unsupported_titles[link.display] then
auto_display = auto_display:gsub("\\([^\\]*%*)", "%1", 1)
link.display = link.display:sub(2) -- remove colon from beginning
end
else
local prefix = link.display:match("^([^:]+):")
-- Process the display form.
local prefixes = {
if link.display then
w = true,
local orig_display = link.display
wikipedia = true,
link.display = lang:makeDisplayText(link.display, sc, true)
}
if cats then
 
auto_display = lang:makeDisplayText(auto_display, sc)
if prefixes[prefix] then
-- If the alt text is the same as what would have been automatically generated, then the alt parameter is redundant (e.g. {{l|en|foo|foo}}, {{l|en|w:foo|foo}}, but not {{l|en|w:foo|w:foo}}).
link.display = link.display:sub(#prefix + 2) -- remove prefix plus colon
-- If they're different, but the alt text could have been entered as the term parameter without it affecting the target page, then the target parameter is redundant (e.g. {{l|ru|фу|фу́}}).
-- If `no_alt_ast` is true, use pcall to catch the error which will be thrown if this is a reconstructed lang and the alt text doesn't have *.
if link.display == auto_display then
insert(cats, lang:getFullName() .. " links with redundant alt parameters")
else
local ok, check
if no_alt_ast then
ok, check = pcall(export.get_link_page, orig_display, lang, sc, plain)
else
ok = true
check = export.get_link_page(orig_display, lang, sc, plain)
end
if ok and link.target == check then
insert(cats, lang:getFullName() .. " links with redundant target parameters")
end
end
end
end
end
else
link.display = lang:makeDisplayText(auto_display, sc)
end
end
-- Process the target
link.target = link.nocont and link.target:gsub("^%*", "Reconstruction:" .. lang:getCanonicalName() .. "/") or link.target
link.target = export.getLinkPage(link.target, lang)
if not link.target then
if not link.target then
return link.display
return link.display
end
end
 
-- If the target is the same as the current page and there is no sense id
-- If the target is the same as the current page, there is no sense id
-- and linking to the same page hasn't been turned on, then return a "self-link"
-- and either the language code is "und" or the current L2 is the current
-- like the software does.
-- language then return a "self-link" like the software does.
if not (allow_self_link or id) and link.target:gsub("^:", "") == mw.title.getCurrentTitle().prefixedText then
if link.target == mw.title.getCurrentTitle().prefixedText then
return "<strong class=\"selflink\">" .. link.display .. "</strong>"
local fragment, current_L2 = link.fragment, require("Module:utilities").get_current_L2()
if (
fragment and fragment == current_L2 or
not (id or fragment) and (lang:getFullCode() == "und" or lang:getFullName() == current_L2)
) then
return tostring(mw.html.create("strong")
:addClass("selflink")
:wikitext(link.display))
end
end
end


--[[
-- Add fragment. Do not add a section link to "Undetermined", as such sections do not exist and are invalid.
Add fragment
-- TabbedLanguages handles links without a section by linking to the "last visited" section, but adding
Do not add a section link to "Undetermined", as such sections do not exist and are invalid.
-- "Undetermined" would break that feature. For localized prefixes that make syntax error, please use the
TabbedLanguages handles links without a section by linking to the "last visited" section,
-- format: ["xyz"] = true.
but adding "Undetermined" would break that feature.
local prefix = link.target:match("^:*([^:]+):")
For localized prefixes that make syntax error, please use the format: ["xyz"] = true,
prefix = prefix and ulower(prefix)
]]
local prefix = link.target:match("^:?([^:]+):")
local prefixes = {
w = true,
wikipedia = true,
Category = true,
}


if not prefixes[prefix] then
if prefix ~= "category" and not (prefix and mw.loadData("Module:data/interwikis")[prefix]) then
if (link.fragment or link.target:find("#$")) and not plain then
track("fragment", lang:getFullCode())
if cats then
insert(cats, lang:getFullName() .. " links with manual fragments")
end
end


if not link.fragment and lang:getCode() ~= "und" then
if (not link.fragment) and lang:getFullCode() ~= "und" then
if id then
if id then
link.fragment = require("Module:utilities").make_id(lang, id)
link.fragment = require("Module:senseid").anchor(lang, id)
elseif not mw.ustring.find(link.target, "^Appendix:")
elseif not (link.target:find("^Appendix:") or link.target:find("^Reconstruction:") or plain) then
and not mw.ustring.find(link.target, "^Reconstruction:") then
link.fragment = lang:getFullName()
link.fragment = lang:getCanonicalName()
end
end
elseif plain and id then
link.fragment = id
end
end
end
-- Put inward-facing square brackets around a link to isolated spacing character(s).
if isolated and #link.display > 0 and not umatch(decode_entities(link.display), "%S") then
link.display = "&#x5D;" .. link.display .. "&#x5B;"
end


-- This allows linking to pages like [[sms:a]] without it being treated weirdly.
link.target = link.target:gsub("^(:?)(.*)", function(m1, m2)
link.target = link.target:gsub(":", "&#x3a;")
return m1 .. encode_entities(m2, "#%&+/:<=>@[\\]_{|}")
end
end)
link.fragment = link.fragment and encode_entities(remove_formatting(link.fragment), "#%&+/:<=>@[\\]_{|}")


return "[[" .. (link.nocont and "wikt:" or "Contionary:") .. link.target .. "#" .. lang:getCanonicalName() .. "|" .. link.display .. "]]"
return "[[" .. (link.nocont and "wikt:" or "Contionary:") .. link.target .. (link.fragment and "#" .. link.fragment or "") .. "|" .. link.display .. "]]"
end
end




-- Split a link into its parts
-- Split a link into its parts
local function parseLink(linktext)
local function parse_link(linktext)
local link = { target = linktext }
local link = {target = linktext}
local first, second = link.target:match("^([^|]+)|(.+)$")
 
local target = link.target
link.target, link.display = target:match("^(..-)|(.+)$")
if not link.target then
link.target = target
link.display = target
end


if first then
-- There's no point in processing these, as they aren't real links.
link.target = first
local target_lower = link.target:lower()
link.display = second
for _, false_positive in ipairs({"category", "cat", "file", "image"}) do
else
if target_lower:match("^" .. false_positive .. ":") then return nil end
link.display = link.target
end
end


first, second = link.target:match("^(.+)#(.+)$")
link.display = decode_entities(link.display)
link.target, link.fragment = get_fragment(link.target)


if first then
-- So that make_link does not look for a fragment again.
link.target = first
if not link.fragment then
link.fragment = second
else
-- So that makeLangLink does not look for a fragment again
link.fragment = false
link.fragment = false
end
end
Line 187: Line 367:
end
end


-- Find embedded links and ensure they link to the correct section.
local function process_embedded_links(text, data, plain)
-- Process the non-linked text.
text = data.lang:makeDisplayText(text, data.sc[1], true)


-- Creates a basic wikilink to the given term. If the text already contains
-- If the text begins with * and another character, then act as if each link begins with *. However, don't do this if the * is contained within a link at the start. E.g. `|*[[foo]]` would set all_reconstructed to true, while `|[[*foo]]` would not.
-- links, these are replaced with links to the correct section.
local all_reconstructed = false
function export.language_link(data, allow_self_link)
if not plain then
if type(data) ~= "table" then
-- anchorEncode removes links etc.
error("The first argument to the function language_link must be a table. See Module:links/documentation for more information.")
if anchorEncode(text):sub(1, 1) == "*" then
all_reconstructed = true
end
-- Otherwise, handle any escapes.
text = text:gsub("^(\\-)\\%*", "%1*")
end
end


local text = data.term
if data.alt then
 
track("alt-ignored")
if ignore_cap[data.lang:getCode()] and text then
mw.log("(from Module:links)", "text with embedded wikilinks:", text,
text = text:gsub("%^", "")
"ignored alt:", data.alt, "lang:", data.lang:getFullCode())
if data.cats then
insert(data.cats, data.lang:getFullName() .. " links with ignored alt parameters")
end
end
if data.id then
track("id-ignored")
mw.log("(from Module:links)", "text with embedded wikilinks:", text,
"ignored id:", data.id, "lang:", data.lang:getFullCode())
if data.cats then
insert(data.cats, data.lang:getFullName() .. " links with ignored id parameters")
end
end
end


-- If the text begins with * and another character,
local function process_link(space1, linktext, space2)
-- then act as if each link begins with *
local capture = "[[" .. linktext .. "]]"
local allReconstructed = false


if text:find("^*.") then
local link = parse_link(linktext)
allReconstructed = true
end


-- Do we have embedded wikilinks?
--Return unprocessed false positives untouched (e.g. categories).
if text:find("[[", nil, true) then
if not link then return capture end


if data.alt then
if all_reconstructed and not link.target:find("^%*") then
mw.log("(from Module:links)", "text with embedded wikilinks:", text,
link.target = "*" .. link.target
"ignored alt:", data.alt, "lang:", data.lang:getCode())
end
end


if data.id then
linktext = make_link(link, data.lang, data.sc, data.id, false, plain)
mw.log("(from Module:links)", "text with embedded wikilinks:", text,
:gsub("^%[%[", "\3")
"ignored id:", data.id, "lang:", data.lang:getCode())
:gsub("%]%]$", "\4")
end


-- Begins and ends with a wikilink tag
return space1 .. linktext .. space2
if text:find("^%[%[(.+)%]%]$") then
end
-- There are no [ ] in between.
-- This makes the wikilink tag redundant.
if text:find("^%[%[[^%[%]]+%]%]$") then
else
local temp = text:gsub("^%[%[(.+)%]%]$", "%1")
temp = temp:gsub("%]%], %[%[", "|")
end
end


text = text:gsub("%[%[([^%]]+)%]%]",
-- Use chars 1 and 2 as temporary substitutions, so that we can use charsets. These are converted to chars 3 and 4 by process_link, which means we can convert any remaining chars 1 and 2 back to square brackets (i.e. those not part of a link).
function(linktext)
text = text
local link = parseLink(linktext)
:gsub("%[%[", "\1")
:gsub("%]%]", "\2")
-- If the script uses ^ to capitalize transliterations, make sure that any carets preceding links are on the inside, so that they get processed with the following text.
if text:match("%^") and not data.sc:hasCapitalization() and data.sc:isTransliterated() then
text = escape(text, "^")
:gsub("%^\1", "\1%^")
text = unescape(text, "^")
end
text = text:gsub("\1(%s*)([^\1\2]-)(%s*)\2", process_link)


if allReconstructed then
-- Remove the extra * at the beginning of a language link if it's immediately followed by a link whose display begins with * too.
link.target = "*" .. link.target
if all_reconstructed then
end
text = text:gsub("^%*\3([^|\1-\4]+)|%*", "\3%1|*")
end


return makeLangLink(link, data.lang, data.id, allow_self_link)
return (text
end)
:gsub("[\1\3]", "[[")
:gsub("[\2\4]", "]]"))
end


-- Remove the extra * at the beginning if it's immediately followed
local function handle_redundant_wikilink(data)
-- by a link whose display begins with * too
local text = data.term
if allReconstructed then
if ( -- Filters out anything but "[[...]]" with no intermediate "[[" or "]]".
text = text:gsub("^%*%[%[([^|%]]+)|%*", "[[%1|*")
not match(text, "^()%[%[") or -- Faster than sub(text, 1, 2) ~= "[[".
find(text, "[[", 3, true) or
find(text, "]]", 3, true) ~= #text - 1
) then
return
end
text = sub(text, 3, -3)
-- A pipe at the start of an embedded link is treated as part of the target (e.g. [[|foo]]: "|foo").
-- FIXME: This should be handled via a proper escape sequence.
local pipe = find(text, "|", 2, true)
local term = pipe and sub(text, 1, pipe - 1) or text
local title = new_title(term)
if title then
local namespace = title.namespace
-- Categories and files are false-positives.
if namespace == 6 or namespace == 14 then
return
end
end
end
data.term, data.alt = term, pipe and pipe ~= #text and sub(text, pipe + 1) or nil
if data.cats then
insert(data.cats, data.lang:getFullName() .. " links with redundant wikilinks")
end
end
--[==[Creates a basic link to the given term. It links to the language section (such as <code>==English==</code>), but it does not add language and script wrappers, so any code that uses this function should call the <code class="n">[[Module:script utilities#tag_text|tag_text]]</code> from [[Module:script utilities]] to add such wrappers itself at some point.
The first argument, <code class="n">data</code>, may contain the following items, a subset of the items used in the <code class="n">data</code> argument of <code class="n">full_link</code>. If any other items are included, they are ignored.
{ {
term = entry_to_link_to,
alt = link_text_or_displayed_text,
lang = language_object,
id = sense_id,
} }
; <code class="n">term</code>
: Text to turn into a link. This is generally the name of a page. The text can contain wikilinks already embedded in it. These are processed individually just like a single link would be. The <code class="n">alt</code> argument is ignored in this case.
; <code class="n">alt</code> (''optional'')
: The alternative display for the link, if different from the linked page. If this is {{code|lua|nil}}, the <code class="n">text</code> argument is used instead (much like regular wikilinks). If <code class="n">text</code> contains wikilinks in it, this argument is ignored and has no effect. (Links in which the alt is ignored are tracked with the tracking template {{whatlinkshere|tracking=links/alt-ignored}}.)
; <code class="n">lang</code>
: The [[Module:languages#Language objects|language object]] for the term being linked. If this argument is defined, the function will determine the language's canonical name (see [[Template:language data documentation]]), and point the link or links in the <code class="n">term</code> to the language's section of an entry, or to a language-specific senseid if the <code class="n">id</code> argument is defined.
; <code class="n">id</code> (''optional'')
: Sense id string. If this argument is defined, the link will point to a language-specific sense id ({{ll|en|identifier|id=HTML}}) created by the template {{temp|senseid}}. A sense id consists of the language's canonical name, a hyphen (<code>-</code>), and the string that was supplied as the <code class="n">id</code> argument. This is useful when a term has more than one sense in a language. If the <code class="n">term</code> argument contains wikilinks, this argument is ignored. (Links in which the sense id is ignored are tracked with the tracking template {{whatlinkshere|tracking=links/id-ignored}}.)
The second argument is as follows:
; <code class="n">allow_self_link</code>
: If {{code|lua|true}}, the function will also generate links to the current page. The default ({{code|lua|false}}) will not generate a link but generate a bolded "self link" instead.
The following special options are processed for each link (both simple text and with embedded wikilinks):
* The target page name will be processed to generate the correct entry name. This is done by the [[Module:languages#makeEntryName|makeEntryName]] function in [[Module:languages]], using the <code class="n">entry_name</code> replacements in the language's data file (see [[Template:language data documentation]] for more information). This function is generally used to automatically strip dictionary-only diacritics that are not part of the normal written form of a language.
* If the text starts with <code class="n">*</code>, then the term is considered a reconstructed term, and a link to the Reconstruction: namespace will be created. If the text contains embedded wikilinks, then <code class="n">*</code> is automatically applied to each one individually, while preserving the displayed form of each link as it was given. This allows linking to phrases containing multiple reconstructed terms, while only showing the * once at the beginning.
* If the text starts with <code class="n">:</code>, then the link is treated as "raw" and the above steps are skipped. This can be used in rare cases where the page name begins with <code class="n">*</code> or if diacritics should not be stripped. For example:
** {{temp|l|en|*nix}} links to the nonexistent page [[Reconstruction:English/nix]] (<code class="n">*</code> is interpreted as a reconstruction), but {{temp|l|en|:*nix}} links to [[*nix]].
** {{temp|l|sl|Franche-Comté}} links to the nonexistent page [[Franche-Comte]] (<code>é</code> is converted to <code>e</code> by <code class="n">makeEntryName</code>), but {{temp|l|sl|:Franche-Comté}} links to [[Franche-Comté]].]==]
function export.language_link(data)
if type(data) ~= "table" then
error("The first argument to the function language_link must be a table. See Module:links/documentation for more information.")
-- Do we have a redundant wikilink? If so, remove it.
elseif data.term then
handle_redundant_wikilink(data)
-- Nothing to process, return nil.
elseif not data.alt then
return nil
end
local text = data.term
-- If we don't have a script, get one.
if not data.sc then
data.sc = data.lang:findBestScript(data.alt or text)
end
-- Do we have embedded wikilinks? If so, they need to be processed individually.
if text then
local open = find(text, "[[", 1, true)
if open and find(text, "]]", open + 2, true) then
return process_embedded_links(text, data)
end
end
-- If not, make a link using the parameters.
text = text and trim(text)
data.alt = data.alt and trim(data.alt)
return make_link({target = text, display = data.alt, fragment = data.fragment}, data.lang, data.sc, data.id, true, nil, data.cats, data.no_alt_ast)
end
function export.plain_link(data)
if type(data) ~= "table" then
error("The first argument to the function language_link must be a table. See Module:links/documentation for more information.")
-- Do we have a redundant wikilink? If so, remove it.
elseif data.term then
handle_redundant_wikilink(data)
-- Only have alt (or nothing), just return it.
else
else
-- There is no embedded wikilink, make a link using the parameters.
return data.alt
text = makeLangLink({ target = text, display = data.alt, nocont = data.nocont}, data.lang, data.id, allow_self_link)
end
-- Make sure the language is "und".
local lang = data.lang
if not lang or lang:getCode() ~= "und" then
data.lang = require("Module:languages").getByCode("und")
end
local text = data.term
-- If we don't have a script, get one.
if not data.sc then
data.sc = require("Module:scripts").findBestScriptWithoutLang(data.alt or text)
end
-- Do we have embedded wikilinks? If so, they need to be processed individually.
local open = find(text, "[[", 1, true)
if open and find(text, "]]", open + 2, true) then
return process_embedded_links(text, data)
end
-- If not, make a link using the parameters.
text = trim(text)
data.alt = data.alt and trim(data.alt)
return make_link({target = text, display = data.alt, fragment = data.fragment}, data.lang, data.sc, data.id, true, true)
end
 
--[==[Replace any links with links to the correct section, but don't link the whole text if no embedded links are found. Returns the display text form.]==]
function export.embedded_language_links(data)
if type(data) ~= "table" then
error("The first argument to the function language_link must be a table. See Module:links/documentation for more information.")
end
end


return text
local text = data.term
-- If we don't have a script, get one.
if not data.sc then
data.sc = data.lang:findBestScript(text)
end
-- Do we have embedded wikilinks? If so, they need to be processed individually.
local open = find(text, "[[", 1, true)
if open and find(text, "]]", open + 2, true) then
return process_embedded_links(text, data)
end
-- If not, return the display text.
text = trim(text)
-- FIXME: Double-escape any percent-signs, because we don't want to treat non-linked text as having percent-encoded characters. This is a hack: percent-decoding should come out of [[Module:languages]] and only dealt with in this module, as it's specific to links.
text = text:gsub("%%", "%%25")
return (data.lang:makeDisplayText(text, data.sc, true))
end
end


function export.mark(text, itemType, face, lang)
function export.mark(text, item_type, face, lang)
local tag = { "", "" }
local tag = { "", "" }


if itemType == "gloss" then
if item_type == "gloss" then
tag = { '<span class="mention-gloss-double-quote">“</span><span class="mention-gloss">',
tag = { '<span class="mention-gloss-double-quote">“</span><span class="mention-gloss">',
'</span><span class="mention-gloss-double-quote">”</span>' }
'</span><span class="mention-gloss-double-quote">”</span>' }
elseif itemType == "tr" then
elseif item_type == "tr" then
if face == "term" then
if face == "term" then
tag = { '<span lang="' .. lang:getCode() .. '" class="tr mention-tr Latn">',
tag = { '<span lang="' .. lang:getFullCode() .. '" class="tr mention-tr Latn">',
'</span>' }
'</span>' }
else
else
tag = { '<span lang="' .. lang:getCode() .. '" class="tr Latn">', '</span>' }
tag = { '<span lang="' .. lang:getFullCode() .. '" class="tr Latn">', '</span>' }
end
end
elseif itemType == "ts" then
elseif item_type == "ts" then
tag = { '<span class="ts mention-ts Latn">/', '/</span>' }
-- \226\129\160 = word joiner (zero-width non-breaking space) U+2060
elseif itemType == "pos" then
tag = { '<span class="ts mention-ts Latn">/\226\129\160', '\226\129\160/</span>' }
elseif item_type == "pos" then
tag = { '<span class="ann-pos">', '</span>' }
tag = { '<span class="ann-pos">', '</span>' }
elseif itemType == "annotations" then
elseif item_type == "annotations" then
tag = { '<span class="mention-gloss-paren annotation-paren">(</span>',
tag = { '<span class="mention-gloss-paren annotation-paren">(</span>',
'<span class="mention-gloss-paren annotation-paren">)</span>' }
'<span class="mention-gloss-paren annotation-paren">)</span>' }
Line 286: Line 619:
end
end


-- Format the annotations (things following the linked term)
--[==[Formats the annotations that are displayed with a link created by {{code|lua|full_link}}. Annotations are the extra bits of information that are displayed following the linked term, and include things such as gender, transliteration, gloss and so on.
* The first argument is a table possessing some or all of the following keys:
*:; <code class="n">genders</code>
*:: Table containing a list of gender specifications in the style of [[Module:gender and number]].
*:; <code class="n">tr</code>
*:: Transliteration.
*:; <code class="n">gloss</code>
*:: Gloss that translates the term in the link, or gives some other descriptive information.
*:; <code class="n">pos</code>
*:: Part of speech of the linked term. If the given argument matches one of the templates in [[:Category:Part of speech tags]], then call that to show a part-of-speech tag. Otherwise, just show the given text as it is.
*:; <code class="n">lit</code>
*:: Literal meaning of the term, if the usual meaning is figurative or idiomatic.
*:Any of the above values can be omitted from the <code class="n">info</code> argument. If a completely empty table is given (with no annotations at all), then an empty string is returned.
* The second argument is a string. Valid values are listed in [[Module:script utilities/data]] "data.translit" table.]==]
function export.format_link_annotations(data, face)
function export.format_link_annotations(data, face)
local output = {}
local output = {}
Line 292: Line 638:
-- Interwiki link
-- Interwiki link
if data.interwiki then
if data.interwiki then
table_insert(output, data.interwiki)
insert(output, data.interwiki)
end
end


Line 301: Line 647:


if data.genders and #data.genders > 0 then
if data.genders and #data.genders > 0 then
local m_gen = require("Module:getn")
local m_gen = require("Module:gender and number")
table_insert(output, "&nbsp;" .. m_gen.format_list(data.genders, data.lang))
insert(output, "&nbsp;" .. m_gen.format_list(data.genders, data.lang))
end
end


Line 308: Line 654:


-- Transliteration and transcription
-- Transliteration and transcription
if data.tr or data.ts then
if data.tr and data.tr[1] or data.ts and data.ts[1] then
local kind
local kind
if face == "term" then
if face == "term" then
Line 316: Line 662:
end
end


if data.tr and data.ts then
if data.tr[1] and data.ts[1] then
table_insert(annotations,
insert(annotations,
require("Module:script utilities").tag_translit(data.tr, data.lang, kind)
require("Module:script utilities").tag_translit(data.tr[1], data.lang, kind)
.. " " .. export.mark(data.ts, "ts"))
.. " " .. export.mark(data.ts[1], "ts"))
elseif data.ts then
elseif data.ts[1] then
table_insert(annotations, export.mark(data.ts, "ts"))
insert(annotations, export.mark(data.ts[1], "ts"))
else
else
table_insert(annotations,
insert(annotations,
require("Module:script utilities").tag_translit(data.tr, data.lang, kind))
require("Module:script utilities").tag_translit(data.tr[1], data.lang, kind))
end
end
end
end
Line 330: Line 676:
-- Gloss/translation
-- Gloss/translation
if data.gloss then
if data.gloss then
table_insert(annotations, export.mark(data.gloss, "gloss"))
insert(annotations, export.mark(data.gloss, "gloss"))
end
end


Line 340: Line 686:
end
end


table_insert(annotations, export.mark(pos_tags[data.pos] or data.pos, "pos"))
pos_tags = pos_tags or mw.loadData("Module:links/data").pos_tags
insert(annotations, export.mark(pos_tags[data.pos] or data.pos, "pos"))
end
end


-- Literal/sum-of-parts meaning
-- Literal/sum-of-parts meaning
if data.lit then
if data.lit then
table_insert(annotations, "literally " .. export.mark(data.lit, "gloss"))
insert(annotations, "literally " .. export.mark(data.lit, "gloss"))
end
end


if #annotations > 0 then
if #annotations > 0 then
table_insert(output, " " .. export.mark(table_concat(annotations, ", "), "annotations"))
insert(output, " " .. export.mark(concat(annotations, ", "), "annotations"))
end
 
return concat(output)
end
 
-- Add any left or right regular or accent qualifiers, labels or references to a formatted term. `data` is the object
-- specifying the term, which should optionally contain:
-- * a language object in `lang`; required if any accent qualifiers or labels are given;
-- * left regular qualifiers in `q` (an array of strings or a single string); an empty array or blank string will be
--  ignored;
-- * right regular qualifiers in `qq` (an array of strings or a single string); an empty array or blank string will be
--  ignored;
-- * left accent qualifiers in `a` (an array of strings); an empty array will be ignored;
-- * right accent qualifiers in `aa` (an array of strings); an empty array will be ignored;
-- * left labels in `l` (an array of strings); an empty array will be ignored;
-- * right labels in `ll` (an array of strings); an empty array will be ignored;
-- * references in `refs`, an array either of strings (formatted reference text) or objects containing fields `text`
--  (formatted reference text) and optionally `name` and/or `group`.
-- `formatted` is the formatted version of the term itself.
local function add_qualifiers_and_refs_to_term(data, formatted)
local q = data.q
if type(q) == "string" then
q = {q}
end
local qq = data.qq
if type(qq) == "string" then
qq = {qq}
end
if q and q[1] or qq and qq[1] or data.a and data.a[1] or data.aa and data.aa[1] or data.l and data.l[1] or
data.ll and data.ll[1] or data.refs and data.refs[1] then
formatted = require(pron_qualifier_module).format_qualifiers {
lang = data.lang,
text = formatted,
q = q,
qq = qq,
a = data.a,
aa = data.aa,
l = data.l,
ll = data.ll,
refs = data.refs,
}
end
end


return table_concat(output)
return formatted
end
end


-- A version of {{l}} or {{m}} that can be called from other modules too
 
function export.full_link(data, face, allow_self_link, no_check_redundant_translit)
--[==[Creates a full link, with annotations (see <code class="n">[[#format_link_annotations|format_link_annotations]]</code>), in the style of {{temp|l}} or {{temp|m}}.
The first argument, <code class="n">data</code>, must be a table. It contains the various elements that can be supplied as parameters to {{temp|l}} or {{temp|m}}:
{ {
term = entry_to_link_to,
alt = link_text_or_displayed_text,
lang = language_object,
sc = script_object,
track_sc = boolean,
fragment = link_fragment
id = sense_id,
genders = { "gender1", "gender2", ... },
tr = transliteration,
ts = transcription,
gloss = gloss,
pos = part_of_speech_tag,
lit = literal_translation,
no_alt_ast = boolean,
accel = {accelerated_creation_tags},
interwiki = interwiki,
q = { "left_qualifier1", "left_qualifier2", ...} or "left_qualifier",
qq = { "right_qualifier1", "right_qualifier2", ...} or "right_qualifier",
refs = { "formatted_ref1", "formatted_ref2", ...} or { {text = "text", name = "name", group = "group"}, ... },
} }
Any one of the items in the <code class="n">data</code> table may be {{code|lua|nil}}, but an error will be shown if neither <code class="n">term</code> nor <code class="n">alt</code> nor <code class="n">tr</code> is present.
Thus, calling {{code|lua|2=full_link{ term = term, lang = lang, sc = sc } }}, where <code class="n">term</code> is an entry name, <code class="n">lang</code>  is a [[Module:languages#Language objects|language object]] from [[Module:languages]], and <code class="n">sc</code> is a [[Module:scripts#Script objects|script object]] from [[Module:scripts]], will give a plain link similar to the one produced by the template {{temp|l}}, and calling {{code|lua|2=full_link( { term = term, lang = lang, sc = sc }, "term" )}} will give a link similar to the one produced by the template {{temp|m}}.
The function will:
* Try to determine the script, based on the characters found in the term or alt argument, if the script was not given. If a script is given and <code class="n">track_sc</code> is {{code|lua|true}}, it will check whether the input script is the same as the one which would have been automatically generated and add the category [[:Category:Terms with redundant script codes]] if yes, or [[:Category:Terms with non-redundant manual script codes]] if no. This should be used when the input script object is directly determined by a template's <code class="n">sc=</code> parameter.
* Call <code class="n">[[#language_link|language_link]]</code> on the term or alt forms, to remove diacritics in the page name, process any embedded wikilinks and create links to Reconstruction or Appendix pages when necessary.
* Call <code class="n">[[Module:script utilities#tag_text]]</code> to add the appropriate language and script tags to the term, and to italicize terms written in the Latin script if necessary. Accelerated creation tags, as used by [[WT:ACCEL]], are included.
* Generate a transliteration, based on the alt or term arguments, if the script is not Latin and no transliteration was provided.
* Add the annotations (transliteration, gender, gloss etc.) after the link.
* If <code class="n">no_alt_ast</code> is specified, then the alt text does not need to contain an asterisk if the language is reconstructed. This should only be used by modules which really need to allow links to reconstructions that don't display asterisks (e.g. number boxes).
* If <code class="n">show_qualifiers</code> is specified, left and right qualifiers and references will be displayed. (This is for compatibility reasons, since a fair amount of code stores qualifiers and/or references in these fields and displays them itself, expecting {{code|lua|full_link()}} to ignore them.]==]
function export.full_link(data, face, allow_self_link, show_qualifiers)
-- Prevent data from being destructively modified.
local data = shallowcopy(data)
 
if type(data) ~= "table" then
if type(data) ~= "table" then
error("The first argument to the function full_link must be a table. "
error("The first argument to the function full_link must be a table. "
Line 362: Line 786:
end
end


local terms = {true}
-- Generate multiple forms if applicable.
for _, param in ipairs{"term", "alt"} do
if type(data[param]) == "string" and data[param]:find("//") then
data[param] = export.split_on_slashes(data[param])
elseif type(data[param]) == "string" and not (type(data.term) == "string" and data.term:find("//")) then
data[param] = data.lang:generateForms(data[param])
else
data[param] = {}
end
end
for _, param in ipairs{"sc", "tr", "ts"} do
data[param] = {data[param]}
end
for _, param in ipairs{"term", "alt", "sc", "tr", "ts"} do
for i in pairs(data[param]) do
terms[i] = true
end
end
-- Create the link
-- Create the link
local output = {}
local output = {}
local categories = {}
data.cats = {}
local link = ""
local link = ""
local annotations
local annotations


--local m_utilities = require("Module:utilities")
for i in ipairs(terms) do
-- Is there any text to show?
if (data.term[i] or data.alt[i]) then
-- Try to detect the script if it was not provided
local display_term = data.alt[i] or data.term[i]
local best = data.lang:findBestScript(display_term)
if (
best:getCode() == "None" and
require("Module:scripts").findBestScriptWithoutLang(display_term):getCode() ~= "None"
) then
insert(data.cats, data.lang:getFullName() .. " terms in nonstandard scripts")
end
if not data.sc[i] then
data.sc[i] = best
-- Track uses of sc parameter.
elseif data.track_sc then
if data.sc[i]:getCode() == best:getCode() then
insert(data.cats, data.lang:getFullName() .. " terms with redundant script codes")
else
insert(data.cats, data.lang:getFullName() .. " terms with non-redundant manual script codes")
end
end
 
-- If using a discouraged character sequence, add to maintenance category
if data.sc[i]:hasNormalizationFixes() == true then
if (data.term[i] and data.sc[i]:fixDiscouragedSequences(toNFC(data.term[i])) ~= toNFC(data.term[i])) or (data.alt[i] and data.sc[i]:fixDiscouragedSequences(toNFC(data.alt[i])) ~= toNFC(data.alt[i])) then
insert(data.cats, "Pages using discouraged character sequences")
end
end


-- Is there any text to show?
local class = ""
if (data.term or data.alt) and data.term ~= "-" then
-- Try to detect the script if it was not provided
if not data.sc then
data.sc = require("Module:scripts").findBestScript(data.alt or data.term, data.lang)
else
-- Track uses of sc parameter
local best = require("Module:scripts").findBestScript(data.alt or data.term, data.lang)
end


local class = ""
-- Encode certain characters to avoid various delimiter-related issues at various stages. We need to encode < and >
-- because they end up forming part of CSS class names inside of <span ...> and will interfere with finding the end
-- of the HTML tag. I first tried converting them to URL encoding, i.e. %3C and %3E; they then appear in the URL as
-- %253C and %253E, which get mapped back to %3C and %3E when passed to [[Module:accel]]. But mapping them to &lt;
-- and &gt; somehow works magically without any further work; they appear in the URL as < and >, and get passed to
-- [[Module:accel]] as < and >. I have no idea who along the chain of calls is doing the encoding and decoding. If
-- someone knows, please modify this comment appropriately!
local encode_accel_char_map = {
["%"] = ".",
[" "] = "_",
["_"] = TEMP_UNDERSCORE,
["<"] = "&lt;",
[">"] = "&gt;",
}
local function encode_accel_param_chars(param)
local retval = param:gsub("[% <>_]", encode_accel_char_map) -- discard second return value
return retval
end


local function encode_accel_param(prefix, param)
local function encode_accel_param(prefix, param)
-- This is decoded again by [[WT:ACCEL]].
if not param then
return param and prefix .. param:gsub("%%", "."):gsub(" ", "_") or ""
return ""
end
end
if type(param) == "table" then
local filled_params = {}
-- There may be gaps in the sequence, especially for translit params.
local maxindex = 0
for k, v in pairs(param) do
if type(k) == "number" and k > maxindex then
maxindex = k
end
end
for i=1,maxindex do
filled_params[i] = param[i] or ""
end
-- [[Module:accel]] splits these up again.
param = concat(filled_params, "*~!")
end
-- This is decoded again by [[WT:ACCEL]].
return prefix .. encode_accel_param_chars(param)
end


if data.accel then
if data.accel then
local form = data.accel.form and data.accel.form .. "-form-of" or ""
local form = data.accel.form and encode_accel_param_chars(data.accel.form) .. "-form-of" or ""
local gender = encode_accel_param("gender-", data.accel.gender)
local gender = encode_accel_param("gender-", data.accel.gender)
local pos = encode_accel_param("pos-", data.accel.pos)
local pos = encode_accel_param("pos-", data.accel.pos)
local translit = encode_accel_param("transliteration-", data.accel.translit)
local translit = encode_accel_param("transliteration-",
local lemma = encode_accel_param("origin-", data.accel.lemma)
data.accel.translit or (data.tr[i] ~= "-" and data.tr[i] or nil))
local lemma_translit = encode_accel_param("origin_transliteration-", data.accel.lemma_translit)
local target = encode_accel_param("target-", data.accel.target)
local no_store = data.accel.no_store and "form-of-nostore" or ""
local lemma = encode_accel_param("origin-", data.accel.lemma)
local lemma_translit = encode_accel_param("origin_transliteration-", data.accel.lemma_translit)
local no_store = data.accel.no_store and "form-of-nostore" or ""


local accel =
local accel =
form .. " " ..
form .. " " ..
gender .. " " ..
gender .. " " ..
pos .. " " ..
pos .. " " ..
translit .. " " ..
translit .. " " ..
lemma .. " " ..
target .. " " ..
lemma_translit .. " " ..
lemma .. " " ..
no_store .. " "
lemma_translit .. " " ..
no_store .. " "


class = "form-of lang-" .. data.lang:getCode() .. " " .. accel
class = "form-of lang-" .. data.lang:getFullCode() .. " " .. accel
end
end


-- Only make a link if the term has been given, otherwise just show the alt text without a link
-- Only make a link if the term has been given, otherwise just show the alt text without a link
link = require("Module:script utilities").tag_text(
local term_data = {
data.term and export.language_link(data, allow_self_link)
term = data.term[i],
or data.alt, data.lang, data.sc, face, class)
alt = data.alt[i],
else
lang = data.lang,
--[[ No term to show.
sc = data.sc[i],
Is there at least a transliteration we can work from? ]]
fragment = data.fragment,
link = require("Module:script utilities").request_script(data.lang, data.sc)
id = data.id,
genders = data.genders,
if not data.sc then
tr = data.tr[i],
data.sc = require("Module:scripts").findBestScript(data.alt or data.term, data.lang)
ts = data.ts[i],
gloss = data.gloss,
pos = data.pos,
lit = data.lit,
accel = data.accel,
interwiki = data.interwiki,
cats = data.cats,
no_alt_ast = data.no_alt_ast
}
link = require("Module:script utilities").tag_text(
data.term[i] and export.language_link(term_data)
or data.alt[i], data.lang, data.sc[i], face, class)
else
else
-- Track uses of sc parameter
--[[ No term to show.
local best = require("Module:scripts").findBestScript(data.alt or data.term, data.lang)
Is there at least a transliteration we can work from? ]]
end
link = require("Module:script utilities").request_script(data.lang, data.sc[i])
-- No link to show, and no transliteration either. Show a term request (unless it's a substrate, as they rarely take terms).
if link == "" or link == "-" or link == "?" or not data.tr or data.tr == "-" then
if (link == "" or (not data.tr[i]) or data.tr[i] == "-") and data.lang:getFamilyCode() ~= "qfa-sub" then
-- No link to show, and no transliteration either. Show a term request.
-- If there are multiple terms, break the loop instead.
local category = ""
if i > 1 then
 
remove(output)
if mw.title.getCurrentTitle().nsText ~= "Template" then
break
table_insert(categories, "[[Category:" .. data.lang:getCanonicalName() .. " term requests]]")
elseif mw.title.getCurrentTitle().nsText ~= "Template" then
insert(data.cats, data.lang:getFullName() .. " term requests")
end
link = "<small>[Term?]</small>"
end
end
link = "<small>[Term?]</small>"
end
end
insert(output, link)
if i < #terms then insert(output, "<span class=\"Zsym mention\" style=\"font-size:100%;\">／</span>") end
end
end


table_insert(output, link)
-- TODO: Currently only handles the first transliteration, pending consensus on how to handle multiple translits for multiple forms, as this is not always desirable (e.g. traditional/simplified Chinese).
if data.tr[1] == "" or data.tr[1] == "-" then
data.tr[1] = nil


if data.tr == "" or data.tr == "-" then
else
data.tr = nil
local phonetic_extraction = mw.loadData("Module:links/data").phonetic_extraction
phonetic_extraction = phonetic_extraction[data.lang:getCode()] or phonetic_extraction[data.lang:getFullCode()]


elseif phonetic_extraction[data.lang:getCode()] then
if phonetic_extraction then
local m_phonetic = require(phonetic_extraction[data.lang:getCode()])
data.tr[1] = data.tr[1] or require(phonetic_extraction).getTranslit(export.remove_links(data.alt[1] or data.term[1]))
data.tr = data.tr or m_phonetic.getTranslit(export.remove_links(data.term))


elseif (data.term or data.alt) and not data.sc:getCode():find("Lati?n") then
elseif (data.term[1] or data.alt[1]) and data.sc[1]:isTransliterated() then
-- Track whenever there is manual translit. The categories below like 'terms with redundant transliterations'
-- aren't sufficient because they only work with reference to automatic translit and won't operate at all in
-- languages without any automatic translit, like Persian and Hebrew.
if data.tr[1] then
track("manual-tr", data.lang:getFullCode())
end
 
-- Try to generate a transliteration, unless transliteration has been supplied and data.no_check_redundant_translit is
-- given. (Checking for redundant transliteration can use up significant amounts of memory so we don't want to do it
-- if memory is tight. `no_check_redundant_translit` is currently set when called ultimately from
-- {{multitrans|...|no-check-redundant-translit=1}}.)
if not (data.tr[1] and data.no_check_redundant_translit) then
local text = data.alt[1] or data.term[1]
if not data.lang:link_tr(data.sc[1]) then
text = export.remove_links(text, true)
end


-- Try to generate a transliteration, unless transliteration has been supplied and either
local automated_tr, tr_categories
-- no_check_redundant_translit is given or we are in a high-memory entry. (Checking for redundant
automated_tr, data.tr_fail, tr_categories = data.lang:transliterate(text, data.sc[1])
-- transliteration can use up significant amounts of memory so we don't want to do it if memory
-- is tight. `no_check_redundant_translit` is currently set when called ultimately from
-- {{multitrans|...|no-check-redundant-translit=1}}.)
if not data.tr then
local automated_tr = data.lang:transliterate(export.remove_links(data.alt or data.term), data.sc)


if automated_tr then
if automated_tr or data.tr_fail then
local manual_tr = data.tr
local manual_tr = data.tr[1]


if manual_tr then
if manual_tr then
if manual_tr == automated_tr then
if (export.remove_links(manual_tr) == export.remove_links(automated_tr)) and (not data.tr_fail) then
table_insert(categories,
insert(data.cats, data.lang:getFullName() .. " terms with redundant transliterations")
"[[Category:Terms with redundant transliterations]]"
elseif not data.tr_fail then
.. "[[Category:Terms with redundant transliterations/" .. data.lang:getCode() .. "]]")
-- Prevents Arabic root categories from flooding the tracking categories.
else
if mw.title.getCurrentTitle().nsText ~= "Category" then
-- Prevents Arabic root categories from flooding the tracking categories.
insert(data.cats, data.lang:getFullName() .. " terms with non-redundant manual transliterations")
if mw.title.getCurrentTitle().nsText ~= "Category" then
end
table_insert(categories,
end
"[[Category:Terms with manual transliterations different from the automated ones]]"
end
.. "[[Category:Terms with manual transliterations different from the automated ones/" .. data.lang:getCode() .. "]]")
if (not manual_tr) or data.lang:overrideManualTranslit(data.sc[1]) then
data.tr[1] = automated_tr
for _, category in ipairs(tr_categories) do
insert(data.cats, category)
end
end
end
end
end
if (not manual_tr) or data.lang:overrideManualTranslit() then
data.tr = automated_tr
end
end
end
end
Line 481: Line 1,013:


-- Link to the transliteration entry for languages that require this
-- Link to the transliteration entry for languages that require this
if data.tr and data.lang:link_tr() then
if data.tr[1] and data.lang:link_tr(data.sc[1]) and not (data.tr[1]:match("%[%[(.-)%]%]") or data.tr_fail) then
data.tr = require("Module:script utilities").tag_text(
data.tr[1] = export.language_link{
export.language_link({ lang = data.lang, term = data.tr, nocont = data.nocont}, allow_self_link),
lang = data.lang,
data.lang, data.sc, face, class)
term = data.tr[1],
--data.tr = export.language_link { lang = data.lang, term = data.tr, nocont = data.nocont}
sc = require("Module:scripts").getByCode("Latn")
elseif data.tr then
}
data.tr = "''" .. data.tr .. "''"
elseif data.tr[1] and not (data.lang:link_tr(data.sc[1]) or data.tr_fail) then
-- Remove the pseudo-HTML tags added by remove_links.
data.tr[1] = data.tr[1]:gsub("</?link>", "")
end
end
if data.tr[1] and not umatch(data.tr[1], "[^%s%p]") then data.tr[1] = nil end
insert(output, export.format_link_annotations(data, face))


table_insert(output, export.format_link_annotations(data, face))
local categories = #data.cats > 0 and require("Module:utilities").format_categories(data.cats, data.lang, "-", nil, nil, data.sc) or ""


return table_concat(output) .. table_concat(categories)
output = concat(output)
if show_qualifiers then
output = add_qualifiers_and_refs_to_term(data, output)
end
return output .. categories
end
end


--[[ Strips links: deletes category links,
--[==[Replaces all wikilinks with their displayed text, and removes any categories. This function can be invoked either from a template or from another module.
the targets of piped links,
-- Strips links: deletes category links, the targets of piped links, and any double square brackets involved in links (other than file links, which are untouched). If `tag` is set, then any links removed will be given pseudo-HTML tags, which allow the substitution functions in [[Module:languages]] to properly subdivide the text in order to reduce the chance of substitution failures in modules which scrape pages like [[Module:zh-translit]].
and all double square brackets. ]]
-- FIXME: This is quite hacky. We probably want this to be integrated into [[Module:languages]], but we can't do that until we know that nothing is pushing pipe linked transliterations through it for languages which don't have link_tr set.
function export.remove_links(text)
* <code><nowiki>[[page|displayed text]]</nowiki></code> &rarr; <code><nowiki>displayed text</nowiki></code>
* <code><nowiki>[[page and displayed text]]</nowiki></code> &rarr; <code><nowiki>page and displayed text</nowiki></code>
* <code><nowiki>[[Category:English lemmas|WORD]]</nowiki></code> &rarr; ''(nothing)'']==]
function export.remove_links(text, tag)
if type(text) == "table" then
if type(text) == "table" then
text = text.args[1]
text = text.args[1]
Line 507: Line 1,051:
end
end


text = mw.ustring.gsub(text, "%[%[Category:[^|%]]-|?[^|%]]-%]%]", "")
text = text
text = text:gsub("%[%[[^|%]]-|", "")
:gsub("%[%[", "\1")
text = text:gsub("%[%[", "")
:gsub("%]%]", "\2")
text = text:gsub("%]%]", "")


return text
-- Parse internal links for the display text.
end
text = text:gsub("(\1)([^\1\2]-)(\2)",
function(c1, c2, c3)
-- Don't remove files.
for _, false_positive in ipairs({"file", "image"}) do
if c2:lower():match("^" .. false_positive .. ":") then return c1 .. c2 .. c3 end
end
-- Remove categories completely.
for _, false_positive in ipairs({"category", "cat"}) do
if c2:lower():match("^" .. false_positive .. ":") then return "" end
end
-- In piped links, remove all text before the pipe, unless it's the final character (i.e. the pipe trick), in which case just remove the pipe.
c2 = c2:match("^[^|]*|(.+)") or c2:match("([^|]+)|$") or c2
if tag then
return "<link>" .. c2 .. "</link>"
else
return c2
end
end)


function export.english_links(text)
text = text
local lang = require("Module:languages").getByCode("en")
:gsub("\1", "[[")
:gsub("\2", "]]")


-- Parentheses around function call to remove second return value, the
return text
-- number of replacements.
return (text:gsub("%[%[([^%]]+)%]%]",
function(linktext)
local link = parseLink(linktext)
return makeLangLink(link, lang, nil, true, false)
end))
end
 
function export.light_link(data)
local language_names = mw.loadData("Module:languages/code to canonical name")
local script_codes = mw.loadData("Module:scripts/codes")
 
if data.langCode then
data.langName = language_names[data.langCode] or error('The language code "' .. data.langCode .. '" is not recognized.')
else
error('Language code is required.')
end
 
if not data.term then
error('Term to link to is required.')
end
 
if data.scCode then
if not script_codes[data.scCode] then
error('The script code "' .. data.sc .. '" is not recognized.')
end
else
error("The function light_link requires a script code.")
end
 
local fragment
if data.id then
fragment = data.langName .. "-" .. mw.uri.encode(data.id, "WIKI")
else
fragment = data.langName
end
 
return table_concat {
'<span class="', data.scCode, '" lang="', data.langCode,
'">[[', data.term, "#", fragment, "|", (data.alt or data.term), "]]</span>"
}
end
end


--[=[
--[=[
For example, Norwegian_Bokm.C3.A5l → Norwegian_Bokmål. 0xC3 and 0xA5 are the
This decodes old section encodings.
hexadecimal-base representation of the two bytes used to encode the character
For example, Norwegian_Bokm.C3.A5l → Norwegian_Bokmål.
å in the UTF-8 encoding:
It isn't picky about whether the section encodings represent the UTF-8 encoding
11000011 10100101
of a real Unicode character, so it will mangle section names that contain
a period followed by two uppercase hex characters. At least such section names
are probably pretty rare.


Note that the bytes used to represent a character are actually different from
Wiktionary adds an additional id="" attribute for sections
the Unicode codepoint. For å, the codepoint is 0xE5. The bits (digits) that
using a legacy encoding, if it is different from the modern minimally modified attribute.
actually spell the codepoint are found in the brackets: 110[00011] 10[100101].
It is like percent encoding (URI or URL encoding) except with "." instead of "%".
For further explanation, see [[w:UTF-8#Description]].
See [[mw:Manual:$wgFragmentMode]] and the code that does the encoding at
https://gerrit.wikimedia.org/r/plugins/gitiles/mediawiki/core/+/7bf779524ab1fd8e1d74f79ea4840564d48eea4d/includes/parser/Sanitizer.php#893
]=]
]=]
-- The character class %x should not be used, as it includes the characters a-f,
-- which do not occur in these anchor encodings.
local capitalHex = "[0-9A-F]"
local function decodeAnchor(anchor)
return (anchor:gsub("%.(" .. capitalHex .. capitalHex .. ")",
function(hexByte)
return string.char(tonumber(hexByte, 16))
end))
end


function export.section_link(link)
function export.section_link(link)
Line 589: Line 1,101:
error("The first argument to section_link was a " .. type(link) .. ", but it should be a string.")
error("The first argument to section_link was a " .. type(link) .. ", but it should be a string.")
end
end
 
link = link:gsub("_", " ")
link = link:gsub("_", " ")
 
local target, section = link:match("(.-)#(.*)")
local numberSigns = require("Module:string").count(link, "#")
 
if not target then
if numberSigns > 1 then
error("The section link should only contain one number sign (#).")
end
 
link = mw.uri.decode(link, "WIKI")
local page, section = link:match("^([^#]*)#(.+)$")
if page == "" then
page = nil
end
 
if section then
section = decodeAnchor(section)
 
-- URI-encode (percent-encode) section to allow square brackets and
-- other dodgy characters in section name.
-- If not percent-encoded, they prevent the parser from creating a link.
-- Decode percent-encoding in the displayed text
if page then
return "[[" .. page .. "#" .. mw.uri.encode(section, "WIKI")
.. "|" .. page .. " §&nbsp;" .. section .. "]]"
else
return "[[#" .. mw.uri.encode(section, "WIKI")
.. "|§&nbsp;" .. section .. "]]"
end
else
error("The function “section_link” could not find a number sign marking a section name.")
error("The function “section_link” could not find a number sign marking a section name.")
end
end
return export.plain_link{
term = target,
fragment = section,
alt = link:gsub("#", " §&nbsp;", 1)
}
end
end


return export
return export
Retrieved from "https://linguifex.com/wiki/Special:MobileDiff/374694"

Navigation menu