Linguifex

Module:links: Difference between revisions

← Older edit
No edit summary
Tag: Reverted
Undo revision 523165 by Sware (talk)
Tag: Undo
 
(94 intermediate revisions by 2 users not shown)
Line 5: Line 5:
extraction modules and part-of-speech names are listed
extraction modules and part-of-speech names are listed
at [[Module:links/data]].
at [[Module:links/data]].
 
Other modules used:
Other modules used:
[[Module:script utilities]]
[[Module:script utilities]]
Line 13: Line 13:
]=]
]=]


-- These are prefixed with u to avoid confusion with the default string methods
local anchors_module = "Module:anchors"
-- of the same name.
local form_of_module = "Module:form of"
local gender_and_number_module = "Module:getn"
local languages_module = "Module:languages"
local load_module = "Module:load"
local memoize_module = "Module:memoize"
local pages_module = "Module:pages"
local pron_qualifier_module = "Module:pron qualifier"
local scripts_module = "Module:scripts"
local script_utilities_module = "Module:script utilities"
local string_encode_entities_module = "Module:string/encode entities"
local string_utilities_module = "Module:string utilities"
local table_module = "Module:table"
local utilities_module = "Module:utilities"
 
local concat = table.concat
local concat = table.concat
local encode = mw.text.encode
local find = string.find
local find = mw.ustring.find
local get_current_title = mw.title.getCurrentTitle
local get_entities = require("Module:utilities").get_entities
local gsub = mw.ustring.gsub
local insert = table.insert
local insert = table.insert
local lower = mw.ustring.lower
local ipairs = ipairs
local split = mw.text.split
local match = string.match
local new_title = mw.title.new
local pairs = pairs
local remove = table.remove
local sub = string.sub
local toNFC = mw.ustring.toNFC
local toNFC = mw.ustring.toNFC
local trim = mw.text.trim
local tostring = tostring
local type = type
local unstrip = mw.text.unstrip
local unstrip = mw.text.unstrip
local NAMESPACE = get_current_title().nsText
local function anchor_encode(...)
anchor_encode = require(memoize_module)(mw.uri.anchorEncode, true)
return anchor_encode(...)
end
local function decode_entities(...)
decode_entities = require(string_utilities_module).decode_entities
return decode_entities(...)
end
local function decode_uri(...)
decode_uri = require(string_utilities_module).decode_uri
return decode_uri(...)
end
-- 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 function encode_entities(...)
encode_entities = require(string_encode_entities_module)
return encode_entities(...)
end
local function extend(...)
extend = require(table_module).extend
return extend(...)
end
local function find_best_script_without_lang(...)
find_best_script_without_lang = require(scripts_module).findBestScriptWithoutLang
return find_best_script_without_lang(...)
end
local function format_categories(...)
format_categories = require(utilities_module).format_categories
return format_categories(...)
end
local function format_genders(...)
format_genders = require(gender_and_number_module).format_genders
return format_genders(...)
end
local function format_qualifiers(...)
format_qualifiers = require(pron_qualifier_module).format_qualifiers
return format_qualifiers(...)
end
local function get_current_L2(...)
get_current_L2 = require(pages_module).get_current_L2
return get_current_L2(...)
end
local function get_lang(...)
get_lang = require(languages_module).getByCode
return get_lang(...)
end
local function get_script(...)
get_script = require(scripts_module).getByCode
return get_script(...)
end
local function language_anchor(...)
language_anchor = require(anchors_module).language_anchor
return language_anchor(...)
end
local function load_data(...)
load_data = require(load_module).load_data
return load_data(...)
end
local function request_script(...)
request_script = require(script_utilities_module).request_script
return request_script(...)
end
local function shallow_copy(...)
shallow_copy = require(table_module).shallowCopy
return shallow_copy(...)
end
local function split(...)
split = require(string_utilities_module).split
return split(...)
end
local function tag_text(...)
tag_text = require(script_utilities_module).tag_text
return tag_text(...)
end
local function tag_translit(...)
tag_translit = require(script_utilities_module).tag_translit
return tag_translit(...)
end
local function trim(...)
trim = require(string_utilities_module).trim
return trim(...)
end
local function u(...)
u = require(string_utilities_module).char
return u(...)
end
local function ulower(...)
ulower = require(string_utilities_module).lower
return ulower(...)
end
local function umatch(...)
umatch = require(string_utilities_module).match
return umatch(...)
end
local m_headword_data
local function get_headword_data()
m_headword_data = load_data("Module:headword/data")
return m_headword_data
end
local function selective_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`.
selective_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
return selective_trim(...)
end


local function escape(text, str)
local function escape(text, str)
Line 41: Line 210:
end
end


-- Trim only if there are non-whitespace characters.
-- Remove bold, italics, soft hyphens, strip markers and HTML tags.
local function cond_trim(text)
local function remove_formatting(str)
-- Include all conventional whitespace + zero-width space.
str = str
if find(text, "[^%s​]") then
:gsub("('*)'''(.-'*)'''", "%1%2")
text = trim(text, "%s​")
:gsub("('*)''(.-'*)''", "%1%2")
:gsub("­", "")
return (unstrip(str)
:gsub("<[^<>]+>", ""))
end
 
--[==[Takes an input and splits on a double slash (taking account of escaping backslashes).]==]
function export.split_on_slashes(text)
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
 
--[==[Takes a wikilink and outputs the link target and display text. By default, the link target will be returned as a title object, but if `allow_bad_target` is set it will be returned as a string, and no check will be performed as to whether it is a valid link target.]==]
function export.get_wikilink_parts(text, allow_bad_target)
-- TODO: replace `allow_bad_target` with `allow_unsupported`, with support for links to unsupported titles, including escape sequences.
if (                        -- Filters out anything but "[[...]]" with no intermediate "[[" or "]]".
not match(text, "^()%[%[") or -- Faster than sub(text, 1, 2) ~= "[[".
find(text, "[[", 3, true) or
find(text, "]]", 3, true) ~= #text - 1
) then
return nil, nil
end
local pipe, title, display = find(text, "|", 3, true)
if pipe then
title, display = sub(text, 3, pipe - 1), sub(text, pipe + 1, -3)
else
title = sub(text, 3, -3)
display = title
end
if allow_bad_target then
return title, display
end
title = new_title(title)
-- No title object means the target is invalid.
if title == nil then
return nil, nil
-- If the link target starts with "#" then mw.title.new returns a broken
-- title object, so grab the current title and give it the correct fragment.
elseif title.prefixedText == "" then
local fragment = title.fragment
if fragment == "" then -- [[#]] isn't valid
return nil, nil
end
title = get_current_title()
title.fragment = fragment
end
return title, display
end
 
-- Does the work of export.get_fragment, but can be called directly to avoid unnecessary checks for embedded links.
local function get_fragment(text)
text = escape(text, "#")
-- Replace numeric character references with the corresponding character (&#39; → '),
-- as they contain #, which causes the numeric character reference to be
-- misparsed (wa'a → wa&#39;a → pagename wa&, fragment 39;a).
text = decode_entities(text)
local target, fragment = text:match("^(.-)#(.+)$")
target = target or text
target = unescape(target, "#")
fragment = fragment and unescape(fragment, "#")
return target, fragment
end
 
--[==[Takes a link target and outputs the actual target and the fragment (if any).]==]
function export.get_fragment(text)
-- If there are no embedded links, process input.
local open = find(text, "[[", nil, true)
if not open then
return get_fragment(text)
end
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
-- Otherwise, return the input.
return text
return text
end
end


local ignore_cap
--[==[
local pos_tags
Given a link target as passed to `full_link()`, get the actual page that the target refers to. This removes
function export.getLinkPage(target, lang, sc, plain)
bold, italics, strip markets and HTML; calls `makeEntryName()` for the language in question; converts targets
beginning with `*` to the Reconstruction namespace; and converts appendix-constructed languages to the Appendix
namespace. Returns up to three values:
# the actual page to link to, or {nil} to not link to anything;
# how the target should be displayed as, if the user didn't explicitly specify any display text; generally the
  same as the original target, but minus any anti-asterisk !!;
# the value `true` if the target had a backslash-escaped * in it (FIXME: explain this more clearly).
]==]
function export.get_link_page_with_auto_display(target, lang, sc, plain)
local orig_target = target
 
if not target then
if not target then
return nil
return nil
end
end
 
-- Remove bold, italics, soft hyphens, strip markers and HTML tags.
target = remove_formatting(target)
target = target
 
:gsub("('*)'''(.-'*)'''", "%1%2")
if target:sub(1, 1) == ":" then
:gsub("('*)''(.-'*)''", "%1%2")
local link_target = target:sub(2)
:gsub("­", "")
link_target = (lang:hasType("conlang") and "Contionary:" or "wikt:") .. link_target
target = unstrip(target)
return link_target, orig_target
:gsub("<[^<>]+>", "")
end
 
-- Check if the target is an interwiki link.
local prefix = target:match("^(.-):")
if target:match(":") and target ~= ":" then
-- Convert any escaped colons
local m_utildata = mw.loadData("Module:utilities/data")
target = target:gsub("\\:", ":")
-- 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.)
if prefix then
local prefix = target:gsub("^:*(.-):.*", lower)
-- If this is an a link to another namespace or an interwiki link, ensure there's an initial colon and then
if m_utildata.namespaces[prefix] or m_utildata.interwikis[prefix] then
-- return what we have (so that it works as a conventional link, and doesn't do anything weird like add the term
return ":" .. target:gsub("^:+", ""), nil, {}
-- to a category.)
prefix = ulower(trim(prefix))
if prefix ~= "" and (
load_data("Module:data/namespaces")[prefix] or
load_data("Module:data/interwikis")[prefix]
) then
return target, orig_target
end
end
-- Convert any escaped colons
target = target:gsub("\\:", ":")
end
end
 
-- Check if the term is reconstructed and remove any asterisk. Otherwise, handle the escapes.
-- Check if the term is reconstructed and remove any asterisk. Also check for anti-asterisk (!!).
local reconstructed, escaped
-- Otherwise, handle the escapes.
local reconstructed, escaped, anti_asterisk
if not plain then
if not plain then
target, reconstructed = target:gsub("^%*(.)", "%1")
target, reconstructed = target:gsub("^%*(.)", "%1")
if reconstructed == 0 then
target, anti_asterisk = target:gsub("^!!(.)", "%1")
if anti_asterisk == 1 then
-- Remove !! from original. FIXME! We do it this way because the call to remove_formatting() above
-- may cause non-initial !! to be interpreted as anti-asterisks. We should surely move the
-- remove_formatting() call later.
orig_target = orig_target:gsub("^!!", "")
end
end
end
target, escaped = target:gsub("^(\\-)\\%%*", "%1*")
if reconstructed == 0 and lang:hasType("reconstructed") then
orig_target = "*" .. target
reconstructed = 1
end
end
target, escaped = target:gsub("^(\\-)\\%*", "%1*")
if not require("Module:utilities").check_object("script", true, sc) or sc:getCode() == "None" then
if not (sc and sc:getCode() ~= "None") then
sc = lang:findBestScript(target)
sc = lang:findBestScript(target)
end
end
 
-- Remove carets if they are used to capitalize parts of transliterations (unless they have been escaped).
-- Remove carets if they are used to capitalize parts of transliterations (unless they have been escaped).
if (not sc:hasCapitalization()) and sc:isTransliterated() and target:match("%^") then
if (not sc:hasCapitalization()) and sc:isTransliterated() and target:match("%^") then
Line 94: Line 374:
target = unescape(target, "^")
target = unescape(target, "^")
end
end
 
-- Get the entry name for the language.
-- Get the entry name for the language.
target = lang:makeEntryName(target, sc)
target = lang:makeEntryName(target, sc, reconstructed == 1 or lang:hasType("appendix-constructed"))
 
-- If the link contains unexpanded template parameters, then don't create a link.
-- If the link contains unexpanded template parameters, then don't create a link.
if target:find("{{{") then
if target:match("{{{.-}}}") then
-- FIXME: Should we return the original target as the default display value (second return value)?
return nil
return nil
end
end
 
if target:sub(1, 1) == "/" then
-- Link to appendix for reconstructed terms and terms in appendix-only languages. Plain links interpret *
return ":" .. target
-- literally, however.
if not lang:hasType("conlang") then
elseif target:find("^Reconstruction:") then
if lang:hasType("appendix-constructed") then
return target
target = "wikt:Appendix:" .. lang:getFullName() .. "/" .. target
elseif reconstructed == 1 then -- asterisk found
-- Link to appendix for reconstructed terms and terms in appendix-only languages. Plain links interpret * literally, however.
if lang:getFullCode() == "und" then
elseif reconstructed == 1 then
-- Return the original target as default display value. If we don't do this, we wrongly get
if lang:getNonEtymologicalCode() == "und" then
-- [Term?] displayed instead.
return nil
return nil, orig_target
else
end
target = "Reconstruction:" .. lang:getNonEtymologicalName() .. "/" .. target
target = "wikt:Reconstruction:" .. lang:getFullName() .. "/" .. target
elseif anti_asterisk ~= 1 and (lang:hasType("reconstructed") or lang:getFamilyCode() == "qfa-sub") then
--error("The specified language " .. lang:getCanonicalName()
--.. " is unattested, while the given term does not begin with '*' to indicate that it is reconstructed.")
orig_target = "*" .. target
end
end
-- Reconstructed languages and substrates require an initial *.
else
elseif lang:hasType("reconstructed") or lang:getFamilyCode() == "qfa-sub" then
if reconstructed == 1 then
local check, m_utildata = target:match("^:*([^:]*):"), mw.loadData("Module:utilities/data")
target = "*" .. target
check = check and lower(check)
if m_utildata.interwikis[check] or m_utildata.namespaces[check] then
return target
else
error("The specified language " .. lang:getCanonicalName()
.. " is unattested, while the given word is not marked with '*' to indicate that it is reconstructed.")
end
end
elseif lang:hasType("appendix-constructed") then
target = "Appendix:" .. lang:getNonEtymologicalName() .. "/" .. target
end
end
target = (lang:hasType("conlang") and "Contionary:" or "wikt:") .. target


return target, escaped > 0
return target, orig_target, escaped > 0
end
 
function export.get_link_page(target, lang, sc, plain)
local target, auto_display, escaped = export.get_link_page_with_auto_display(target, lang, sc, plain)
return target, escaped
end
end


-- Make a link from a given link's parts
-- Make a link from a given link's parts
local function makeLink(link, lang, sc, id, allow_self_link, isolated, plain)
local function make_link(link, lang, sc, id, isolated, cats, no_alt_ast, plain)
-- Convert percent encoding to plaintext.
-- Convert percent encoding to plaintext.
link.target = mw.uri.decode(link.target, "PATH")
link.target = link.target and decode_uri(link.target, "PATH")
link.fragment = link.fragment and mw.uri.decode(link.fragment, "PATH")
link.fragment = link.fragment and decode_uri(link.fragment, "PATH")
 
-- Find fragments (when link didn't come from parseLink).
-- 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]].
-- # can be escaped as \#.
-- # can be escaped as \#.
if link.target then
if link.target and link.fragment == nil then
link.target = escape(link.target, "#")
link.target, link.fragment = get_fragment(link.target)
if link.fragment == nil then
-- Replace numeric character references with the corresponding character (&#29; → '),
-- as they contain #, which causes the numeric character reference to be
-- misparsed (wa'a → wa&#29;a → pagename wa&, fragment 29;a).
link.target = get_entities(link.target)
local first, second = link.target:match("^([^#]+)#(.+)$")
if first then
link.target, link.fragment = first, second
end
end
link.target = unescape(link.target, "#")
link.fragment = link.fragment and unescape(link.fragment, "#")
end
end
 
-- If there is no display form, then create a default one.
-- Process the target
if not link.display then
local auto_display, escaped
link.display = link.target
link.target, auto_display, escaped = export.get_link_page_with_auto_display(link.target, lang, sc, plain)
 
-- Create a default display form.
-- If the target is "" then it's a link like [[#English]], which refers to the current page.
if auto_display == "" then
auto_display = (m_headword_data or get_headword_data()).pagename
end
end
 
local display_is_target = link.display == link.target
-- Process the target
local escaped
link.target, escaped = export.getLinkPage(link.target, lang, sc, plain)
-- If the display is the target and the reconstruction * has been escaped, remove the escaping backslash.
-- If the display is the target and the reconstruction * has been escaped, remove the escaping backslash.
if display_is_target and escaped then
if escaped then
link.display = link.display:gsub("\\([^\\]*%*)", "%1", 1)
auto_display = auto_display:gsub("\\([^\\]*%*)", "%1", 1)
end
end
 
-- Process the display form.
-- Process the display form.
link.display = lang:makeDisplayText(link.display, sc, not display_is_target)
if link.display then
local orig_display = link.display
link.display = lang:makeDisplayText(link.display, sc, true)
if cats then
auto_display = lang:makeDisplayText(auto_display, sc)
-- 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}}).
-- 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
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
end
end
else
link.display = lang:makeDisplayText(auto_display, sc)
end
 
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 == mw.title.getCurrentTitle().prefixedText then
if link.target == get_current_title().prefixedText then
return tostring(mw.html.create("strong")
local fragment, current_L2 = link.fragment, get_current_L2()
:addClass("selflink")
if (
:wikitext(link.display))
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. TabbedLanguages handles links without a section by linking to the "last visited" section, but adding "Undetermined" would break that feature. For localized prefixes that make syntax error, please use the format: ["xyz"] = true.
-- Add fragment. Do not add a section link to "Undetermined", as such sections do not exist and are invalid.
local prefix, lower_prefix = link.target:match("^:*([^:]+):")
-- TabbedLanguages handles links without a section by linking to the "last visited" section, but adding
-- "Undetermined" would break that feature. For localized prefixes that make syntax error, please use the
local m_utildata
-- format: ["xyz"] = true.
if prefix then
local prefix = link.target:match("^:*([^:]+):")
lower_prefix = lower(prefix)
prefix = prefix and ulower(prefix)
m_utildata = mw.loadData("Module:utilities/data")
 
end
if prefix ~= "category" and not (prefix and load_data("Module:data/interwikis")[prefix]) then
if (link.fragment or link.target:sub(-1) == "#") and not plain then
if not (m_utildata and m_utildata.interwikis[lower_prefix]) then
if cats then
insert(cats, lang:getFullName() .. " links with manual fragments")
if (not link.fragment) and lang:getNonEtymologicalCode() ~= "und" then
end
end
 
if not link.fragment then
if id then
if id then
link.fragment = require("Module:senseid").anchor(lang, id)
link.fragment = lang:getFullCode() == "und" and anchor_encode(id) or language_anchor(lang, id)
elseif not (link.target:find("^Appendix:") or link.target:find("^Reconstruction:") or plain) then
elseif lang:getFullCode() ~= "und" and not (link.target:match("^Appendix:") or link.target:match("^Reconstruction:")) then
link.fragment = lang:getNonEtymologicalName()
link.fragment = anchor_encode(lang:getFullName())
end
end
elseif plain and id then
link.fragment = id
end
end
end
end
 
if isolated then
-- Put inward-facing square brackets around a link to isolated spacing character(s).
link.display = mw.loadData("Module:links/data").display_change[link.display] or link.display
if isolated and #link.display > 0 and not umatch(decode_entities(link.display), "%S") then
link.display = "&#x5D;" .. link.display .. "&#x5B;"
end
end
 
link.target = link.target:gsub("^(:?)(.*)", function(m1, m2)
link.target = link.target:gsub("^(:?)(.*)", function(m1, m2)
return m1 .. encode(m2, "#%%&+/:<=>@[\\%]_{|}")
return m1 .. encode_entities(m2, "#%&+/:<=>@[\\]_{|}")
end)
end)
link.fragment = link.fragment and encode(link.fragment, "#%%&+/:<=>@[\\%]_{|}")
 
link.fragment = link.fragment and encode_entities(remove_formatting(link.fragment), "#%&+/:<=>@[\\]_{|}")
return "[[" .. link.target .. (link.fragment and "#" .. link.fragment or "") .. "|" .. link.display .. "]]"
return "[[" ..
link.target:gsub("^[^:]", ":%0") .. (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
-- Prevent characters whose HTML entities are unsupported titles from being incorrectly recognised as the entity if they are in a link being re-parsed (e.g. "&" becomes "&amp;" when returned, but "&amp;" is also an unsupported title. If "&" is given as a link which is then re-parsed, we don't want it to be perceived as "&amp;".)
link.target, link.display = target:match("^(..-)|(.+)$")
if link.target:match("&[^;]+;") then
if not link.target then
local unsupported_titles = mw.loadData("Module:links/data").unsupported_titles
link.target = target
if unsupported_titles[second] and unsupported_titles[second] ~= first then
link.display = target
link.target = get_entities(link.target)
first, second = link.target:match("^([^|]+)|(.+)$")
end
end
end


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


return link
return link
end
local function check_params_ignored_when_embedded(alt, lang, id, cats)
if alt then
if cats then
insert(cats, lang:getFullName() .. " links with ignored alt parameters")
end
end
if id then
if cats then
insert(cats, lang:getFullName() .. " links with ignored id parameters")
end
end
end
end


-- Find embedded links and ensure they link to the correct section.
-- Find embedded links and ensure they link to the correct section.
local function process_embedded_links(text, data, allow_self_link, plain)
local function process_embedded_links(text, alt, lang, sc, id, cats, no_alt_ast, plain)
-- Process the non-linked text.
-- Process the non-linked text.
text = data.lang:makeDisplayText(text, data.sc[1], true)
text = lang:makeDisplayText(text, sc, true)
 
-- 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 allReconstructed to true, while `|[[*foo]]` would not.
-- 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.
local allReconstructed = false
local all_reconstructed = false
if not plain then
if not plain then
if require("Module:utilities").get_plaintext(text:gsub("%[%[.-%]%]", ".")):match("^*.") then
-- anchor_encode removes links etc.
allReconstructed = true
if anchor_encode(text):sub(1, 1) == "*" then
all_reconstructed = true
end
end
-- Otherwise, handle any escapes.
-- Otherwise, handle any escapes.
text = text:gsub("^(\\-)\\%*", "%1*")
text = text:gsub("^(\\-)\\%*", "%1*")
end
end
 
if data.alt then
check_params_ignored_when_embedded(alt, lang, id, cats)
mw.log("(from Module:links)", "text with embedded wikilinks:", text,
 
"ignored alt:", data.alt, "lang:", data.lang:getNonEtymologicalCode())
local function process_link(space1, linktext, space2)
end
if data.id then
mw.log("(from Module:links)", "text with embedded wikilinks:", text,
"ignored id:", data.id, "lang:", data.lang:getNonEtymologicalCode())
end
local function processLink(space1, linktext, space2)
local capture = "[[" .. linktext .. "]]"
local capture = "[[" .. linktext .. "]]"
local link = parse_link(linktext)
linktext = get_entities(linktext)
 
-- Return unprocessed false positives untouched (e.g. categories).
local link = parseLink(linktext)
if not link then
return capture
--Return unprocessed false positives untouched (e.g. categories).
if not link then return capture end
if allReconstructed and not link.target:find("^%*") then
link.target = "*" .. link.target
end
end
 
linktext = makeLink(link, data.lang, data.sc, data.id, allow_self_link, false, plain)
if all_reconstructed then
if link.target:find("^!!") then
-- Check for anti-asterisk !! at the beginning of a target, indicating that a reconstructed term
-- wants a part of the term to link to a non-reconstructed term, e.g. Old English
-- {{ang-noun|m|head=*[[!!Crist|Cristes]] [[!!mæsseǣfen]]}}.
link.target = link.target:sub(3)
-- Also remove !! from the display, which may have been copied from the target (as in mæsseǣfen in
-- the example above).
link.display = link.display:gsub("^!!", "")
elseif not link.target:match("^%*") then
link.target = "*" .. link.target
end
end
 
linktext = make_link(link, lang, sc, id, false, nil, no_alt_ast, plain)
:gsub("^%[%[", "\3")
:gsub("^%[%[", "\3")
:gsub("%]%]$", "\4")
:gsub("%]%]$", "\4")
 
return space1 .. linktext .. space2
return space1 .. linktext .. space2
end
end
 
-- Use chars 1 and 2 as temporary substitutions, so that we can use charsets. These are converted to chars 3 and 4 by processLink, which means we can convert any remaining chars 1 and 2 back to square brackets (i.e. those not part of a link).
-- 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).
text = text
text = text
:gsub("%[%[", "\1")
:gsub("%[%[", "\1")
:gsub("%]%]", "\2")
: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 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
if (
text:find("^", nil, true) and
not sc:hasCapitalization() and
sc:isTransliterated()
) then
text = escape(text, "^")
text = escape(text, "^")
:gsub("%^\1", "\1%^")
:gsub("%^\1", "\1%^")
text = unescape(text, "^")
text = unescape(text, "^")
end
end
text = text:gsub("\1(%s*)([^\1\2]-)(%s*)\2", processLink)
text = text:gsub("\1(%s*)([^\1\2]-)(%s*)\2", process_link)
 
-- Remove the extra * at the beginning of a language link if it's immediately followed by a link whose display begins with * too.
-- Remove the extra * at the beginning of a language link if it's immediately followed by a link whose display begins with * too.
if allReconstructed then
if all_reconstructed then
text = text:gsub("^%*\3([^|\1-\4]+)|%*", "\3%1|*")
text = text:gsub("^%*\3([^|\1-\4]+)|%*", "\3%1|*")
end
end
 
return (text
return (text
:gsub("[\1\3]", "[[")
:gsub("[\1\3]", "[[")
:gsub("[\2\4]", "]]"))
:gsub("[\2\4]", "]]")
)
end
end


local function handle_redundant_wikilink(text, alt)
local function simple_link(term, fragment, alt, lang, sc, id, cats, no_alt_ast, srwc)
local temp = text:match("^%[%[(.-)%]%]$")
local plain
if not temp then
if lang == nil then
return text, alt
lang, plain = get_lang("und"), true
end
end
local temp_lower = temp:lower()
 
for _, falsePositive in ipairs({"category", "cat", "file", "image"}) do
-- Get the link target and display text. If the term is the empty string, treat the input as a link to the current page.
if temp_lower:match("^" .. falsePositive .. ":") then
if term == "" then
return text, alt
term = get_current_title().prefixedText
elseif term then
local new_term, new_alt = export.get_wikilink_parts(term, true)
if new_term then
check_params_ignored_when_embedded(alt, lang, id, cats)
-- [[|foo]] links are treated as plaintext "[[|foo]]".
-- FIXME: Pipes should be handled via a proper escape sequence, as they can occur in unsupported titles.
if new_term == "" then
term, alt = nil, term
else
local title = new_title(new_term)
if title then
local ns = title.namespace
-- File: and Category: links should be returned as-is.
if ns == 6 or ns == 14 then
return term
end
end
term, alt = new_term, new_alt
if cats then
if not (srwc and srwc(term, alt)) then
insert(cats, lang:getFullName() .. " links with redundant wikilinks")
end
end
end
end
end
end
end
-- Note: it's possible for "[[" or "]]" to be uninvolved in links, so we need to check for both individually (e.g. "[[aaa]] bb]]" would not have a redundant wikilink).
if alt then
if temp and not (temp:find("%[%[") or temp:find("%]%]")) then
alt = selective_trim(alt)
text, alt = temp:match("^([^|]+)|?(.-)$")
if alt == "" then
if alt == "" then
alt = nil
alt = nil
end
end
end
end
return text, alt
-- If there's nothing to process, return nil.
if not (term or alt) then
return nil
end
 
-- If there is no script, get one.
if not sc then
sc = lang:findBestScript(alt or term)
end
 
-- Embedded wikilinks need to be processed individually.
if term then
local open = find(term, "[[", nil, true)
if open and find(term, "]]", open + 2, true) then
return process_embedded_links(term, alt, lang, sc, id, cats, no_alt_ast, plain)
end
term = selective_trim(term)
end
 
-- If not, make a link using the parameters.
return make_link({
target = term,
display = alt,
fragment = fragment
}, lang, sc, id, true, cats, no_alt_ast, plain)
end
end


Line 372: Line 730:
: 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}}.)
: 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:
The second argument is as follows:
; <code class="n">allowSelfLink</code>
; <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.
: 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 following special options are processed for each link (both simple text and with embedded wikilinks):
Line 380: Line 738:
** {{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|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]].]==]
** {{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, allow_self_link)
function export.language_link(data)
if type(data) ~= "table" then
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.")
error(
-- Nothing to process, return nil.
"The first argument to the function language_link must be a table. See Module:links/documentation for more information.")
elseif not (data.term or data.alt) then
return nil
end
end
 
local text = data.term
-- Categorize links to "und".
local lang, cats = data.lang, data.cats
data.sc = data.sc or data.lang:findBestScript(text)
if cats and lang:getCode() == "und" then
insert(cats, "Undetermined language links")
ignore_cap = ignore_cap or mw.loadData("Module:links/data").ignore_cap
if (ignore_cap[data.lang:getCode()] or ignore_cap[data.lang:getNonEtymologicalCode()]) and text then
text = text:gsub("%^", "")
end
end
 
-- Do we have a redundant wikilink? If so, remove it.
return simple_link(
if text then
data.term,
text, data.alt = handle_redundant_wikilink(text, data.alt)
data.fragment,
end
data.alt,
lang,
-- Do we have embedded wikilinks?
data.sc,
if text and text:find("%[%[.-%]%]") then
data.id,
text = process_embedded_links(text, data, allow_self_link)
cats,
-- If not, make a link using the parameters.
data.no_alt_ast,
else
data.suppress_redundant_wikilink_cat
text = text and cond_trim(text)
)
data.alt = data.alt and cond_trim(data.alt)
text = makeLink({ target = text, display = data.alt }, data.lang, data.sc, data.id, allow_self_link, true)
end
return text
end
end


function export.plain_link(data, allow_self_link)
function export.plain_link(data)
if type(data) ~= "table" then
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.")
error(
-- Nothing to process, return nil.
"The first argument to the function plain_link must be a table. See Module:links/documentation for more information.")
elseif not (data.term or data.alt) then
return nil
-- Only have alt, just return it.
elseif not data.term then
return data.alt
end
end
 
local text = data.term
return simple_link(
if (not data.lang) or data.lang:getNonEtymologicalCode() ~= "und" then
data.term,
data.lang = require("Module:languages").getByCode("und")
data.fragment,
end
data.alt,
data.sc = data.sc or require("Module:scripts").findBestScriptWithoutLang(text)
nil,
data.sc,
-- Do we have a redundant wikilink? If so, remove it.
data.id,
if text then
data.cats,
text, data.alt = handle_redundant_wikilink(text, data.alt)
data.no_alt_ast,
end
data.suppress_redundant_wikilink_cat
)
-- Do we have embedded wikilinks?
if text:find("%[%[.-%]%]") then
text = process_embedded_links(text, data, allow_self_link, true)
-- If not, make a link using the parameters.
else
text = cond_trim(text)
data.alt = data.alt and cond_trim(data.alt)
text = makeLink({ target = text, display = data.alt }, data.lang, data.sc, data.id, allow_self_link, true, true)
end
return text
end
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.]==]
--[==[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, allow_self_link)
function export.embedded_language_links(data)
if type(data) ~= "table" then
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.")
error(
"The first argument to the function embedded_language_links must be a table. See Module:links/documentation for more information.")
end
 
local term, lang, sc = data.term, data.lang, data.sc
 
-- If we don't have a script, get one.
if not sc then
sc = lang:findBestScript(term)
end
end
 
local text = data.term
-- Do we have embedded wikilinks? If so, they need to be processed individually.
data.sc = data.sc or data.lang:findBestScript(text)
local open = find(term, "[[", nil, true)
if open and find(term, "]]", open + 2, true) then
-- Do we have embedded wikilinks?
return process_embedded_links(term, data.alt, lang, sc, data.id, data.cats, data.no_alt_ast)
if text:find("%[%[.-%]%]") then
text = process_embedded_links(text, data, allow_self_link)
else
-- If there are no embedded wikilinks, return the display text.
text = cond_trim(text)
text = (data.lang:makeDisplayText(text, data.sc, true))
end
end
 
return text
-- If not, return the display text.
term = selective_trim(term)
-- 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.
term = term:gsub("%%", "%%25")
return lang:makeDisplayText(term, 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:getNonEtymologicalCode() .. '" class="tr mention-tr Latn">',
tag = { '<span lang="' .. lang:getFullCode() .. '" class="tr mention-tr Latn">',
'</span>' }
'</span>' }
else
else
tag = { '<span lang="' .. lang:getNonEtymologicalCode() .. '" class="tr Latn">', '</span>' }
tag = { '<span lang="' .. lang:getFullCode() .. '" class="tr Latn">', '</span>' }
end
end
elseif itemType == "ts" then
elseif item_type == "ts" then
-- \226\129\160 = word joiner (zero-width non-breaking space) U+2060
-- \226\129\160 = word joiner (zero-width non-breaking space) U+2060
tag = { '<span class="ts mention-ts Latn">/\226\129\160', '\226\129\160/</span>' }
tag = { '<span class="ts mention-ts Latn">/\226\129\160', '\226\129\160/</span>' }
elseif itemType == "pos" then
elseif item_type == "pos" then
tag = { '<span class="ann-pos">', '</span>' }
tag = { '<span class="ann-pos">', '</span>' }
elseif itemType == "annotations" then
elseif item_type == "non-gloss" then
tag = { '<span class="ann-non-gloss">', '</span>' }
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>' }
elseif item_type == "infl" then
tag = { '<span class="ann-infl">', '</span>' }
end
end


Line 501: Line 843:
end
end


--[==[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.  
local pos_tags
 
--[==[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:
* The first argument is a table possessing some or all of the following keys:
*:; <code class="n">genders</code>
*:; <code class="n">genders</code>
Line 510: Line 854:
*:: Gloss that translates the term in the link, or gives some other descriptive information.
*:: Gloss that translates the term in the link, or gives some other descriptive information.
*:; <code class="n">pos</code>
*:; <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.
*:: Part of speech of the linked term. If the given argument matches one of the aliases in `pos_aliases` in [[Module:headword/data]], or consists of a part of speech or alias followed by `f` (for a non-lemma form), expand it appropriately. Otherwise, just show the given text as it is.
*:; <code class="n">ng</code>
*:: Arbitrary non-gloss descriptive text for the link. This should be used in preference to putting descriptive text in `gloss` or `pos`.
*:; <code class="n">lit</code>
*:; <code class="n">lit</code>
*:: Literal meaning of the term, if the usual meaning is figurative or idiomatic.
*:: Literal meaning of the term, if the usual meaning is figurative or idiomatic.
*:; <code class="n">infl</code>
*:: Table containing a list of grammar tags in the style of [[Module:form of]] `tagged_inflections`.
*: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.
*: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.]==]
* The second argument is a string. Valid values are listed in [[Module:script utilities/data]] "data.translit" table.]==]
Line 529: Line 877:


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


Line 545: Line 899:


if data.tr[1] and data.ts[1] then
if data.tr[1] and data.ts[1] then
insert(annotations,
insert(annotations, tag_translit(data.tr[1], data.lang, kind) .. " " .. export.mark(data.ts[1], "ts"))
require("Module:script utilities").tag_translit(data.tr[1], data.lang, kind)
.. " " .. export.mark(data.ts[1], "ts"))
elseif data.ts[1] then
elseif data.ts[1] then
insert(annotations, export.mark(data.ts[1], "ts"))
insert(annotations, export.mark(data.ts[1], "ts"))
else
else
insert(annotations,
insert(annotations, tag_translit(data.tr[1], data.lang, kind))
require("Module:script utilities").tag_translit(data.tr[1], data.lang, kind))
end
end
end
end
Line 564: Line 915:
if data.pos then
if data.pos then
-- debug category for pos= containing transcriptions
-- debug category for pos= containing transcriptions
if data.pos:find("/[^><]*/") then
if data.pos:match("/[^><]-/") then
data.pos = data.pos .. "[[Category:links likely containing transcriptions in pos]]"
data.pos = data.pos .. "[[Category:links likely containing transcriptions in pos]]"
end
end


pos_tags = pos_tags or mw.loadData("Module:links/data").pos_tags
-- Canonicalize part of speech aliases as well as non-lemma aliases like 'nf' or 'nounf' for "noun form".
insert(annotations, export.mark(pos_tags[data.pos] or data.pos, "pos"))
pos_tags = pos_tags or (m_headword_data or get_headword_data()).pos_aliases
local pos = pos_tags[data.pos]
if not pos and data.pos:find("f$") then
local pos_form = data.pos:sub(1, -2)
-- We only expand something ending in 'f' if the result is a recognized non-lemma POS.
pos_form = (pos_tags[pos_form] or pos_form) .. " form"
if (m_headword_data or get_headword_data()).nonlemmas[pos_form .. "s"] then
pos = pos_form
end
end
insert(annotations, export.mark(pos or data.pos, "pos"))
end
 
-- Inflection data
if data.infl then
local m_form_of = require(form_of_module)
-- Split tag sets manually, since tagged_inflections creates a numbered list, and we do not want that.
local infl_outputs = {}
local tag_sets = m_form_of.split_tag_set(data.infl)
for _, tag_set in ipairs(tag_sets) do
table.insert(infl_outputs,
m_form_of.tagged_inflections({ tags = tag_set, lang = data.lang, nocat = true, nolink = true, nowrap = true }))
end
insert(annotations, export.mark(table.concat(infl_outputs, "; "), "infl"))
end
 
-- Non-gloss text
if data.ng then
insert(annotations, export.mark(data.ng, "non-gloss"))
end
end


Line 575: Line 954:
if data.lit then
if data.lit then
insert(annotations, "literally " .. export.mark(data.lit, "gloss"))
insert(annotations, "literally " .. export.mark(data.lit, "gloss"))
end
-- Provide a hook to insert additional annotations such as nested inflections.
if data.postprocess_annotations then
data.postprocess_annotations {
data = data,
annotations = annotations
}
end
end


Line 584: Line 971:
end
end


--[==[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}}.
-- Encode certain characters to avoid various delimiter-related issues at various stages. We need to encode < and >
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}}:
-- 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 accel_char_map
local function get_accel_char_map()
accel_char_map = {
["%"] = ".",
[" "] = "_",
["_"] = u(0xFFF0),
["<"] = "&lt;",
[">"] = "&gt;",
}
return accel_char_map
end
 
local function encode_accel_param_chars(param)
return (param:gsub("[% <>_]", accel_char_map or get_accel_char_map()))
end
 
local function encode_accel_param(prefix, param)
if not param then
return ""
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 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
 
local function insert_if_not_blank(list, item)
if item == "" then
return
end
insert(list, item)
end
 
local function get_class(lang, tr, accel, nowrap)
if not accel and not nowrap then
return ""
end
local classes = {}
if accel then
insert(classes, "form-of lang-" .. lang:getFullCode())
local form = accel.form
if form then
insert(classes, encode_accel_param_chars(form) .. "-form-of")
end
insert_if_not_blank(classes, encode_accel_param("gender-", accel.gender))
insert_if_not_blank(classes, encode_accel_param("pos-", accel.pos))
insert_if_not_blank(classes, encode_accel_param("transliteration-", accel.translit or (tr ~= "-" and tr or nil)))
insert_if_not_blank(classes, encode_accel_param("target-", accel.target))
insert_if_not_blank(classes, encode_accel_param("origin-", accel.lemma))
insert_if_not_blank(classes, encode_accel_param("origin_transliteration-", accel.lemma_translit))
if accel.no_store then
insert(classes, "form-of-nostore")
end
end
if nowrap then
insert(classes, nowrap)
end
return concat(classes, " ")
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 = 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
 
return formatted
end
 
 
--[==[
Creates a full link, with annotations (see `[[#format_link_annotations|format_link_annotations]]`), in the style of {{tl|l}} or {{tl|m}}.
The first argument, `data`, must be a table. It contains the various elements that can be supplied as parameters to {{tl|l}} or {{tl|m}}:
{ {
{ {
term = entry_to_link_to,
term = entry_to_link_to,
Line 591: Line 1,101:
lang = language_object,
lang = language_object,
sc = script_object,
sc = script_object,
track_sc = boolean,
no_nonstandard_sc_cat = boolean,
fragment = link_fragment,
id = sense_id,
id = sense_id,
genders = { "gender1", "gender2", ... },
genders = { "gender1", "gender2", ... },
tr = transliteration,
tr = transliteration,
respect_link_tr = boolean,
ts = transcription,
ts = transcription,
gloss = gloss,
gloss = gloss,
pos = part_of_speech_tag,
pos = part_of_speech_tag,
ng = non-gloss text,
lit = literal_translation,
lit = literal_translation,
infl = { "form_of_grammar_tag1", "form_of_grammar_tag2", ... },
no_alt_ast = boolean,
accel = {accelerated_creation_tags},
accel = {accelerated_creation_tags},
interwiki = interwiki,
interwiki = interwiki,
pretext = "text_at_beginning" or nil,
posttext = "text_at_end" or nil,
q = { "left_qualifier1", "left_qualifier2", ...} or "left_qualifier",
qq = { "right_qualifier1", "right_qualifier2", ...} or "right_qualifier",
l = { "left_label1", "left_label2", ...},
ll = { "right_label1", "right_label2", ...},
a = { "left_accent_qualifier1", "left_accent_qualifier2", ...},
aa = { "right_accent_qualifier1", "right_accent_qualifier2", ...},
refs = { "formatted_ref1", "formatted_ref2", ...} or { {text = "text", name = "name", group = "group"}, ... },
show_qualifiers = boolean,
} }
} }
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.
Any one of the items in the `data` table may be {nil}, but an error will be shown if neither `term` nor `alt` nor `tr`
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}}.
is present. Thus, calling {full_link{ term = term, lang = lang, sc = sc }}, where `term` is the page to link to (which
may have diacritics that will be stripped and/or embedded bracketed links) and `lang` is a
[[Module:languages#Language objects|language object]] from [[Module:languages]], will give a plain link similar to the
one produced by the template {{tl|l}}, and calling {full_link( { term = term, lang = lang, sc = sc }, "term" )} will
give a link similar to the one produced by the template {{tl|m}}.
 
The function will:
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.
* Try to determine the script, based on the characters found in the `term` or `alt` argument, if the script was not
* 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.
  given. If a script is given and `track_sc` is {true}, it will check whether the input script is the same as the one
* 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.
  which would have been automatically generated and add the category [[:Category:LANG terms with redundant script codes]]
* Generate a transliteration, based on the alt or term arguments, if the script is not Latin and no transliteration was provided.
  if yes, or [[:Category:LANG terms with non-redundant manual script codes]] if no. This should be used when the input
* Add the annotations (transliteration, gender, gloss etc.) after the link.]==]
  script object is directly determined by a template's `sc` parameter.
function export.full_link(data, face, allow_self_link, no_check_redundant_translit)
* Call `[[#language_link|language_link]]` on the `term` or `alt` forms, to remove diacritics in the page name, process
-- Prevent data from being destructively modified.
  any embedded wikilinks and create links to Reconstruction or Appendix pages when necessary.
local data = require("Module:table").shallowcopy(data)
* Call `[[Module:script utilities#tag_text]]` to add the appropriate language and script tags to the term and
  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, no transliteration was
  provided in `tr` and the combination of the term's language and script support automatic transliteration. The
  transliteration itself will be linked if both `.respect_link_tr` is specified and the language of the term has the
  `link_tr` property set for the script of the term; but not otherwise.
* Add the annotations (transliteration, gender, gloss, etc.) after the link.
* If `no_alt_ast` 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 `pretext` or `posttext` is specified, this is text to (respectively) prepend or append to the output, directly
  before processing qualifiers, labels and references. This can be used to add arbitrary extra text inside of the
  qualifiers, labels and references.
* If `show_qualifiers` is specified or the `show_qualifiers` argument is given, then left and right qualifiers, accent
  qualifiers, labels and references will be displayed, otherwise they will be ignored. (This is because a fair amount of
  code stores qualifiers, labels and/or references in these fields and displays them itself, rather than expecting
  {full_link()} to display them.)]==]
function export.full_link(data, face, allow_self_link, show_qualifiers)
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. "
.. "See Module:links/documentation for more information.")
.. "See Module:links/documentation for more information.")
end
end
 
local terms = {true}
-- Prevent data from being destructively modified.
local data = shallow_copy(data)
 
-- FIXME: this shouldn't be added to `data`, as that means the input table needs to be cloned.
data.cats = {}
 
-- Categorize links to "und".
local lang, cats = data.lang, data.cats
if cats and lang:getCode() == "und" then
insert(cats, "Undetermined language links")
end
 
local terms = { true }
 
-- Generate multiple forms if applicable.
-- Generate multiple forms if applicable.
for _, param in ipairs{"term", "alt"} do
for _, param in ipairs { "term", "alt" } do
if type(data[param]) == "string" and data[param]:find("//") then
if type(data[param]) == "string" and data[param]:find("//", nil, true) then
data[param] = escape(data[param], "//")
data[param] = export.split_on_slashes(data[param])
data[param] = split(data[param], "//") or {}
elseif type(data[param]) == "string" and not (type(data.term) == "string" and data.term:find("//", nil, true)) then
for i, subparam in ipairs(data[param]) do
if not data.no_generate_forms then
data[param][i] = unescape(subparam, "//")
data[param] = lang:generateForms(data[param])
if subparam == "" then
else
data[param][i] = nil
data[param] = { data[param] }
end
end
end
elseif type(data[param]) == "string" and not (type(data.term) == "string" and data.term:find("//")) then
data[param] = data.lang:generateForms(data[param])
else
else
data[param] = {}
data[param] = {}
end
end
end
end
 
for _, param in ipairs{"sc", "tr", "ts"} do
for _, param in ipairs { "sc", "tr", "ts" } do
data[param] = {data[param]}
data[param] = { data[param] }
end
end
 
for _, param in ipairs{"term", "alt", "sc", "tr", "ts"} do
for _, param in ipairs { "term", "alt", "sc", "tr", "ts" } do
for i in pairs(data[param]) do
for i in pairs(data[param]) do
terms[i] = true
terms[i] = true
end
end
end
end
 
-- Create the link
-- Create the link
local output = {}
local output = {}
local categories = {}
local id, no_alt_ast, srwc, accel, nevercalltr = data.id, data.no_alt_ast, data.suppress_redundant_wikilink_cat,
local link = ""
data.accel, data.never_call_transliteration_module
local annotations
local link_tr = data.respect_link_tr and lang:link_tr(data.sc[1])
 
local phonetic_extraction = mw.loadData("Module:links/data").phonetic_extraction
phonetic_extraction = phonetic_extraction[data.lang:getCode()] or phonetic_extraction[data.lang:getNonEtymologicalCode()]
for i in ipairs(terms) do
for i in ipairs(terms) do
local link
-- Is there any text to show?
-- Is there any text to show?
if (data.term[i] or data.alt[i]) then
if (data.term[i] or data.alt[i]) then
-- Try to detect the script if it was not provided
-- Try to detect the script if it was not provided
local best = data.lang:findBestScript(data.alt[i] or data.term[i])
local display_term = data.alt[i] or data.term[i]
local best = lang:findBestScript(display_term)
-- no_nonstandard_sc_cat is intended for use in [[Module:interproject]]
if (
not data.no_nonstandard_sc_cat and
best:getCode() == "None" and
find_best_script_without_lang(display_term):getCode() ~= "None"
) then
insert(cats, lang:getFullName() .. " terms in nonstandard scripts")
end
if not data.sc[i] then
if not data.sc[i] then
data.sc[i] = best
data.sc[i] = best
-- Track uses of sc parameter.
elseif data.track_sc then
if data.sc[i]:getCode() == best:getCode() then
insert(cats, lang:getFullName() .. " terms with redundant script codes")
else
insert(cats, lang:getFullName() .. " terms with non-redundant manual script codes")
end
end
end
 
-- If using a discouraged character sequence, add to maintenance category
-- If using a discouraged character sequence, add to maintenance category
if data.sc[i]:hasNormalizationFixes() == true then
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
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(categories, "Pages using discouraged character sequences")
insert(cats, "Pages using discouraged character sequences")
end
end
end
end


local class = ""
link = simple_link(
 
data.term[i],
-- Encode certain characters to avoid various delimiter-related issues at various stages. We need to encode < and >
data.fragment,
-- because they end up forming part of CSS class names inside of <span ...> and will interfere with finding the end
data.alt[i],
-- of the HTML tag. I first tried converting them to URL encoding, i.e. %3C and %3E; they then appear in the URL as
lang,
-- %253C and %253E, which get mapped back to %3C and %3E when passed to [[Module:accel]]. But mapping them to &lt;
data.sc[i],
-- and &gt; somehow works magically without any further work; they appear in the URL as < and >, and get passed to
id,
-- [[Module:accel]] as < and >. I have no idea who along the chain of calls is doing the encoding and decoding. If
cats,
-- someone knows, please modify this comment appropriately!
no_alt_ast,
local encode_accel_char_map = {
srwc
["%"] = ".",
)
[" "] = "_",
end
["<"] = "&lt;",
-- simple_link can return nil, so check if a link has been generated.
[">"] = "&gt;",
if link then
}
-- Add "nowrap" class to prefixes in order to prevent wrapping after the hyphen
local function encode_accel_param_chars(param)
local nowrap
local retval = param:gsub("[% <>]", encode_accel_char_map) -- discard second return value
local display_term = data.alt[i] or data.term[i]
return retval
if display_term and (display_term:find("^%-") or display_term:find("^־")) then -- Hebrew maqqef -- FIXME, use hyphens from [[Module:affix]]
nowrap = "nowrap"
end
end


local function encode_accel_param(prefix, param)
link = tag_text(link, lang, data.sc[i], face, get_class(lang, data.tr[i], accel, nowrap))
if not param then
return ""
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 = table.concat(filled_params, "*~!")
end
-- This is decoded again by [[WT:ACCEL]].
return prefix .. encode_accel_param_chars(param)
end
 
if data.accel then
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 pos = encode_accel_param("pos-", data.accel.pos)
local translit = encode_accel_param("transliteration-",
data.accel.translit or (data.tr[i] ~= "-" and data.tr[i] or nil))
local target = encode_accel_param("target-", data.accel.target)
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 =
form .. " " ..
gender .. " " ..
pos .. " " ..
translit .. " " ..
target .. " " ..
lemma .. " " ..
lemma_translit .. " " ..
no_store .. " "
 
class = "form-of lang-" .. data.lang:getNonEtymologicalCode() .. " " .. accel
end
-- Only make a link if the term has been given, otherwise just show the alt text without a link
local term_data = {term = data.term[i], alt = data.alt[i], lang = data.lang, sc = data.sc[i], id = data.id, genders = data.genders, tr = data.tr[i], ts = data.ts[i], gloss = data.gloss, pos = data.pos, lit = data.lit, accel = data.accel, interwiki = data.interwiki}
link = require("Module:script utilities").tag_text(
data.term[i] and export.language_link(term_data, allow_self_link)
or data.alt[i], data.lang, data.sc[i], face, class)
else
else
--[[ No term to show.
--[[ No term to show.
Is there at least a transliteration we can work from? ]]
Is there at least a transliteration we can work from? ]]
link = require("Module:script utilities").request_script(data.lang, data.sc[i])
link = request_script(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 (not data.tr[i]) or data.tr[i] == "-") and data.lang:getFamilyCode() ~= "qfa-sub" then
if (link == "" or (not data.tr[i]) or data.tr[i] == "-") and lang:getFamilyCode() ~= "qfa-sub" then
-- No link to show, and no transliteration either. Show a term request (unless it's a substrate, as they rarely take terms).
-- If there are multiple terms, break the loop instead.
if mw.title.getCurrentTitle().nsText ~= "Template" then
if i > 1 then
insert(categories, data.lang:getNonEtymologicalName() .. " term requests")
remove(output)
break
elseif NAMESPACE ~= "Template" then
insert(cats, lang:getFullName() .. " term requests")
end
end
link = "<small>[Term?]</small>"
link = "<small>[Term?]</small>"
Line 759: Line 1,284:
end
end
insert(output, link)
insert(output, link)
if i < #terms then insert(output, "<span class=\"Zsym mention\" style=\"font-size:100%;\">／</span>") end
if i < #terms then insert(output, "<span class=\"Zsym mention\" style=\"font-size:100%;\">&nbsp;/ </span>") end
end
end
 
-- 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).
-- When suppress_tr is true, do not show or generate any transliteration
if data.tr[1] == "" or data.tr[1] == "-" then
if data.suppress_tr then
data.tr[1] = nil
data.tr[1] = nil
else
-- 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
else
local phonetic_extraction = load_data("Module:links/data").phonetic_extraction
phonetic_extraction = phonetic_extraction[lang:getCode()] or phonetic_extraction[lang:getFullCode()]
if phonetic_extraction then
data.tr[1] = data.tr[1] or
require(phonetic_extraction).getTranslit(export.remove_links(data.alt[1] or data.term[1]))
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
local full_code = lang:getFullCode()
end
if not nevercalltr then
-- Try to generate a transliteration.
local text = data.alt[1] or data.term[1]
if not link_tr then
text = export.remove_links(text, true)
end


elseif phonetic_extraction then
local automated_tr = lang:transliterate(text, data.sc[1])
local m_phonetic = require(phonetic_extraction)
data.tr[1] = data.tr[1] or m_phonetic.getTranslit(export.remove_links(data.alt[1] or data.term[1]))


elseif (data.term[1] or data.alt[1]) and data.sc[1]:isTransliterated() then
if automated_tr then
-- Try to generate a transliteration, unless transliteration has been supplied and no_check_redundant_translit is
local manual_tr = data.tr[1]
-- 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 no_check_redundant_translit) then
local text = data.alt[1] or data.term[1]
if not data.lang:link_tr() then
text = export.remove_links(text, true)
end
local automated_tr, tr_categories
automated_tr, data.tr_fail, tr_categories = data.lang:transliterate(text, data.sc[1])
if automated_tr or data.tr_fail then
local manual_tr = data.tr[1]


if manual_tr then
if manual_tr then
if (export.remove_links(manual_tr) == export.remove_links(automated_tr)) and (not data.tr_fail) then
if export.remove_links(manual_tr) == export.remove_links(automated_tr) then
insert(categories, "Terms with redundant transliterations")
insert(cats, lang:getFullName() .. " terms with redundant transliterations")
insert(categories, "Terms with redundant transliterations/" .. data.lang:getNonEtymologicalCode())
else
elseif not data.tr_fail then
-- Prevents Arabic root categories from flooding the tracking categories.
-- Prevents Arabic root categories from flooding the tracking categories.
if NAMESPACE ~= "Category" then
if mw.title.getCurrentTitle().nsText ~= "Category" then
insert(cats,
insert(categories, "Terms with manual transliterations different from the automated ones")
lang:getFullName() .. " terms with non-redundant manual transliterations")
insert(categories, "Terms with manual transliterations different from the automated ones/" .. data.lang:getNonEtymologicalCode())
end
end
end
end
end
end


if (not manual_tr) or data.lang:overrideManualTranslit() then
if not manual_tr or lang:overrideManualTranslit(data.sc[1]) then
data.tr[1] = automated_tr
data.tr[1] = automated_tr
for _, category in ipairs(tr_categories) do
end
insert(categories, category)
end
end
end
end
Line 809: Line 1,341:
end
end
end
end
 
-- Link to the transliteration entry for languages that require this
-- Link to the transliteration entry for languages that require this
if data.tr[1] and data.lang:link_tr() and not (data.tr[1]:match("%[%[(.-)%]%]") or data.tr_fail) then
if data.tr[1] and link_tr and not data.tr[1]:match("%[%[(.-)%]%]") then
data.tr[1] = export.language_link{lang = data.lang, term = data.tr[1]}
data.tr[1] = simple_link(
elseif data.tr[1] and not (data.lang:link_tr() or data.tr_fail) then
data.tr[1],
nil,
nil,
lang,
get_script("Latn"),
nil,
cats,
no_alt_ast,
srwc
)
elseif data.tr[1] and not link_tr then
-- Remove the pseudo-HTML tags added by remove_links.
-- Remove the pseudo-HTML tags added by remove_links.
data.tr[1] = data.tr[1]:gsub("</?link>", "")
data.tr[1] = data.tr[1]:gsub("</?link>", "")
end
end
if data.tr[1] and gsub(data.tr[1], "[%s%p]", ""):len() == 0 then data.tr[1] = nil 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))
insert(output, export.format_link_annotations(data, face))
 
categories = #categories > 0 and require("Module:utilities").format_categories(categories, data.lang, "-", nil, nil, data.sc) or ""
if data.pretext then
insert(output, 1, data.pretext)
return concat(output) .. categories
end
if data.posttext then
insert(output, data.posttext)
end
 
local categories = cats[1] and format_categories(cats, lang, "-", nil, nil, data.sc) or ""
 
output = concat(output)
if show_qualifiers or data.show_qualifiers then
output = add_qualifiers_and_refs_to_term(data, output)
end
return output .. categories
end
end


Line 840: Line 1,393:
return ""
return ""
end
end
 
text = text
text = text
:gsub("%[%[", "\1")
:gsub("%[%[", "\1")
Line 849: Line 1,402:
function(c1, c2, c3)
function(c1, c2, c3)
-- Don't remove files.
-- Don't remove files.
for _, falsePositive in ipairs({"file", "image"}) do
for _, false_positive in ipairs({ "file", "image" }) do
if c2:lower():match("^" .. falsePositive .. ":") then return c1 .. c2 .. c3 end
if c2:lower():match("^" .. false_positive .. ":") then return c1 .. c2 .. c3 end
end
end
-- Remove categories completely.
-- Remove categories completely.
for _, falsePositive in ipairs({"category", "cat"}) do
for _, false_positive in ipairs({ "category", "cat" }) do
if c2:lower():match("^" .. falsePositive .. ":") then return "" end
if c2:lower():match("^" .. false_positive .. ":") then return "" end
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.
-- 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.
Line 864: Line 1,417:
end
end
end)
end)
 
text = text
text = text
:gsub("\1", "[[")
:gsub("\1", "[[")
Line 870: Line 1,423:


return text
return text
end
--[=[
This decodes old section encodings.
For example, Norwegian_Bokm.C3.A5l → Norwegian_Bokmål.
It isn't picky about whether the section encodings represent the UTF-8 encoding
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.
Wiktionary adds an additional id="" attribute for sections
using a legacy encoding, if it is different from the modern minimally modified attribute.
It is like percent encoding (URI or URL encoding) except with "." instead of "%".
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
end


Line 903: Line 1,430:
end
end


link = link:gsub("_", " ")
local target, section = get_fragment((link:gsub("_", " ")))
 
local numberSigns = select(2, link:gsub("#", ""))


if numberSigns > 1 then
if not section then
error("The section link should only contain one number sign (#).")
error("No \"#\" delineating a section name")
end
end


link = mw.uri.decode(link, "WIKI")
return simple_link(
local page, section = link:match("^([^#]*)#(.+)$")
target,
if page == "" then
section,
page = nil
target .. " §&nbsp;" .. section
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.")
end
end
end


return export
return export
Retrieved from "https://linguifex.com/wiki/Module:links"