Linguifex

Module:links: Difference between revisions

← Older edit
No edit summary
No edit summary
 
(40 intermediate revisions by the same user not shown)
Line 12: Line 12:
[[Module:gender and number]]
[[Module:gender and number]]
]=]
]=]
local m_str_utils = require("Module:string utilities")
 
local anchors_module = "Module:anchors"
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 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 anchor_encode = require("Module:memoize")(mw.uri.anchorEncode, true)
local concat = table.concat
local concat = table.concat
local decode_entities = m_str_utils.decode_entities
local decode_uri = m_str_utils.decode_uri
local find = string.find
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 get_current_title = mw.title.getCurrentTitle
local insert = table.insert
local insert = table.insert
local ipairs = ipairs
local ipairs = ipairs
local load_data = mw.loadData
local match = string.match
local match = string.match
local new_title = mw.title.new
local new_title = mw.title.new
local pairs = pairs
local pairs = pairs
local remove = table.remove
local remove = table.remove
local shallow_copy = require("Module:table").shallowCopy
local split = m_str_utils.split
local sub = string.sub
local sub = string.sub
local toNFC = mw.ustring.toNFC
local toNFC = mw.ustring.toNFC
local tostring = tostring
local tostring = tostring
local trim -- defined below
local type = type
local type = type
local ulower = m_str_utils.lower
local umatch = m_str_utils.match
local unstrip = mw.text.unstrip
local unstrip = mw.text.unstrip
local u = m_str_utils.char


local TEMP_UNDERSCORE = u(0xFFF0)
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


do
local function find_best_script_without_lang(...)
local _trim = m_str_utils.trim
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.
-- Unconditionally trimmed charset.
local always_trim =
local always_trim =
"\194\128-\194\159" .. -- U+0080-009F (C1 control characters)
"\194\128-\194\159" ..   -- U+0080-009F (C1 control characters)
"\194\173" .. -- U+00AD (soft hyphen)
"\194\173" ..           -- U+00AD (soft hyphen)
"\226\128\170-\226\128\174" .. -- U+202A-202E (directionality formatting characters)
"\226\128\170-\226\128\174" .. -- U+202A-202E (directionality formatting characters)
"\226\129\166-\226\129\169" -- U+2066-2069 (directionality formatting characters)
"\226\129\166-\226\129\169" -- U+2066-2069 (directionality formatting characters)
 
-- Standard trimmed charset.
-- Standard trimmed charset.
local standard_trim = "%s" .. -- (default whitespace charset)
local standard_trim = "%s" .. -- (default whitespace charset)
"\226\128\139-\226\128\141" .. -- U+200B-200D (zero-width spaces)
"\226\128\139-\226\128\141" .. -- U+200B-200D (zero-width spaces)
always_trim
always_trim
 
-- If there are non-whitespace characters, trim all characters in `standard_trim`.
-- If there are non-whitespace characters, trim all characters in `standard_trim`.
-- Otherwise, only trim the characters in `always_trim`.
-- Otherwise, only trim the characters in `always_trim`.
trim = function(text)
selective_trim = function(text)
if text == "" then
if text == "" then
return text
return text
end
end
local trimmed = _trim(text, standard_trim)
local trimmed = trim(text, standard_trim)
if trimmed ~= "" then
if trimmed ~= "" then
return trimmed
return trimmed
end
end
return _trim(text, always_trim)
return trim(text, always_trim)
end
end
return selective_trim(...)
end
end


Line 105: Line 230:
end
end
return text
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
end


Line 114: Line 276:
-- misparsed (wa'a → wa'a → pagename wa&, fragment 39;a).
-- misparsed (wa'a → wa'a → pagename wa&, fragment 39;a).
text = decode_entities(text)
text = decode_entities(text)
local target, fragment = text:match("^(..-)#(.+)$")
local target, fragment = text:match("^(.-)#(.+)$")
target = target or text
target = target or text
target = unescape(target, "#")
target = unescape(target, "#")
Line 124: Line 286:
function export.get_fragment(text)
function export.get_fragment(text)
-- If there are no embedded links, process input.
-- If there are no embedded links, process input.
local open = find(text, "[[", 1, true)
local open = find(text, "[[", nil, true)
if not open then
if not open then
return get_fragment(text)
return get_fragment(text)
Line 131: Line 293:
if not close then
if not close then
return get_fragment(text)
return get_fragment(text)
-- If there is one, but it's redundant (i.e. encloses everything with no pipe), remove and process.
-- 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
elseif open == 1 and close == #text - 1 and not find(text, "|", 3, true) then
return get_fragment(sub(text, 3, -3))
return get_fragment(sub(text, 3, -3))
Line 139: Line 301:
end
end


local pos_tags
--[==[
function export.get_link_page(target, lang, sc, plain)
Given a link target as passed to `full_link()`, get the actual page that the target refers to. This removes
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
 
target = remove_formatting(target)
target = remove_formatting(target)


-- Check if the target is an interwiki link.
if target:sub(1, 1) == ":" then
if target:match(":") and target ~= ":" then
-- FIXME, the auto_display (second return value) should probably remove the colon
-- 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.)
return target:sub(2), orig_target
local prefix = target:gsub("^:*(.-):.*", ulower)
end
if (
 
load_data("Module:data/namespaces")[prefix] or
local prefix = target:match("^(.-):")
load_data("Module:data/interwikis")[prefix]
-- Convert any escaped colons
) then
target = target:gsub("\\:", ":")
return ":" .. target:gsub("^:+", ""), nil, {}
if prefix 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.)
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
end
target, escaped = target:gsub("^(\\-)\\%*", "%1*")
target, escaped = target:gsub("^(\\-)\\%*", "%1*")
 
if not require("Module:utilities").check_object("script", true, sc) or sc:getCode() == "None" then
if reconstructed == 0 and lang:hasType("reconstructed") and not lang:hasType("conlang") then
orig_target = "*" .. target
reconstructed = 1
end
if not (sc and sc:getCode() ~= "None") then
sc = lang:findBestScript(target)
sc = lang:findBestScript(target)
end
end
Line 180: Line 375:


-- 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:getFullCode() == "und" then
-- [Term?] displayed instead.
return nil
return nil, orig_target
elseif lang:hasType("conlang") then
target = "*" .. target
elseif not lang:hasType("conlang") then
target = "Reconstruction:" .. lang:getFullName() .. "/" .. target
end
-- 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 (
load_data("Module:data/namespaces")[check] or
load_data("Module:data/interwikis")[check]
) then
return target
else
if lang:hasType("conlang") then
target = "*" .. target
elseif not lang:hasType("conlang") then
target = "Reconstruction:" .. lang:getFullName() .. "/" .. target
end
end
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
elseif lang:hasType("appendix-constructed") then
target = "Appendix:" .. lang:getFullName() .. "/" .. target
end
end
target = (lang:hasType("conlang") and "Contionary:" or "wikt:") .. target
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 make_link(link, lang, sc, id, isolated, plain, cats, no_alt_ast)
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 = decode_uri(link.target, "PATH")
link.target = link.target and decode_uri(link.target, "PATH")
link.fragment = link.fragment and decode_uri(link.fragment, "PATH")
link.fragment = link.fragment and decode_uri(link.fragment, "PATH")
 
-- Find fragments (if one isn't already set).
-- 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]].
Line 240: Line 425:
link.target, link.fragment = get_fragment(link.target)
link.target, link.fragment = get_fragment(link.target)
end
end
-- Process the target
local auto_display, escaped
link.target, auto_display, escaped = export.get_link_page_with_auto_display(link.target, lang, sc, plain)


-- Create a default display form.
-- Create a default display form.
local auto_display = link.target
-- If the target is "" then it's a link like [[#English]], which refers to the current page.
if auto_display == "" then
-- Process the target
auto_display = (m_headword_data or get_headword_data()).pagename
local escaped
end
link.target, escaped = export.get_link_page(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.
Line 252: Line 440:
auto_display = auto_display:gsub("\\([^\\]*%*)", "%1", 1)
auto_display = auto_display:gsub("\\([^\\]*%*)", "%1", 1)
end
end
 
-- Process the display form.
-- Process the display form.
if link.display then
if link.display then
Line 280: Line 468:
link.display = lang:makeDisplayText(auto_display, sc)
link.display = lang:makeDisplayText(auto_display, sc)
end
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, there is no sense id
-- If the target is the same as the current page, there is no sense id
-- and either the language code is "und" or the current L2 is the current
-- and either the language code is "und" or the current L2 is the current
-- language then return a "self-link" like the software does.
-- language then return a "self-link" like the software does.
if link.target == mw.title.getCurrentTitle().prefixedText then
if link.target == get_current_title().prefixedText then
local fragment, current_L2 = link.fragment, require("Module:pages").get_current_L2()
local fragment, current_L2 = link.fragment, get_current_L2()
if (
if (
fragment and fragment == current_L2 or
fragment and fragment == current_L2 or
not (id or fragment) and (lang:getFullCode() == "und" or lang:getFullName() == current_L2)
not (id or fragment) and (lang:getFullCode() == "und" or lang:getFullName() == current_L2)
) then
) then
return tostring(mw.html.create("strong")
return tostring(mw.html.create("strong")
:addClass("selflink")
:addClass("selflink")
Line 316: Line 504:
if not link.fragment then
if not link.fragment then
if id then
if id then
link.fragment = lang:getFullCode() == "und" and anchor_encode(id) or require("Module:anchors").language_anchor(lang, id)
link.fragment = lang:getFullCode() == "und" and anchor_encode(id) or language_anchor(lang, id)
elseif lang:getFullCode() ~= "und" and not (link.target:find("^Appendix:") or link.target:find("^Reconstruction:")) then
elseif lang:getFullCode() ~= "und" and not (link.target:match("^Appendix:") or link.target:match("^Reconstruction:")) then
link.fragment = anchor_encode(lang:getFullName())
link.fragment = anchor_encode(lang:getFullName())
end
end
end
end
end
end
 
-- Put inward-facing square brackets around a link to isolated spacing character(s).
-- 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
if isolated and #link.display > 0 and not umatch(decode_entities(link.display), "%S") then
Line 331: Line 519:
return m1 .. encode_entities(m2, "#%&+/:<=>@[\\]_{|}")
return m1 .. encode_entities(m2, "#%&+/:<=>@[\\]_{|}")
end)
end)
 
link.fragment = link.fragment and encode_entities(remove_formatting(link.fragment), "#%&+/:<=>@[\\]_{|}")
link.fragment = link.fragment and encode_entities(remove_formatting(link.fragment), "#%&+/:<=>@[\\]_{|}")
local asterisk = lang:hasType("reconstructed") and not lang:hasType("conlang") and not link.target:match("^%*") and "*" or ""
return "[[" ..
link.target:gsub("^[^:]", ":%0") .. (link.fragment and "#" .. link.fragment or "") .. "|" .. link.display .. "]]"
return "[[" .. asterisk .. link.target .. (link.fragment and "#" .. link.fragment or "") .. "|" .. link.display .. "]]"
end
end


Line 341: Line 528:
-- Split a link into its parts
-- Split a link into its parts
local function parse_link(linktext)
local function parse_link(linktext)
local link = {target = linktext}
local link = { target = linktext }


local target = link.target
local target = link.target
Line 352: Line 539:
-- 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 _, false_positive in ipairs({"category", "cat", "file", "image"}) do
for _, false_positive in ipairs({ "category", "cat", "file", "image" }) do
if target_lower:match("^" .. false_positive .. ":") then return nil end
if target_lower:match("^" .. false_positive .. ":") then
return nil
end
end
end


Line 365: Line 554:


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, 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 all_reconstructed 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.
Line 383: Line 585:
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:getFullCode())
if data.cats then
insert(data.cats, data.lang:getFullName() .. " links with ignored alt parameters")
end
end
if data.id then
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


local function process_link(space1, linktext, space2)
local function process_link(space1, linktext, space2)
local capture = "[[" .. linktext .. "]]"
local capture = "[[" .. linktext .. "]]"
local link = parse_link(linktext)
local link = parse_link(linktext)


--Return unprocessed false positives untouched (e.g. categories).
-- Return unprocessed false positives untouched (e.g. categories).
if not link then return capture end
if not link then
return capture
end


if all_reconstructed and not link.target:find("^%*") then
if all_reconstructed then
link.target = "*" .. link.target
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
end


linktext = make_link(link, data.lang, data.sc, data.id, false, plain)
linktext = make_link(link, lang, sc, id, false, nil, no_alt_ast, plain)
:gsub("^%[%[", "\3")
:gsub("^%[%[", "\3")
:gsub("%]%]$", "\4")
:gsub("%]%]$", "\4")
Line 423: Line 622:
: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%^")
Line 437: Line 640:
return (text
return (text
:gsub("[\1\3]", "[[")
:gsub("[\1\3]", "[[")
:gsub("[\2\4]", "]]"))
:gsub("[\2\4]", "]]")
)
end
end


local function handle_redundant_wikilink(data)
local function simple_link(term, fragment, alt, lang, sc, id, cats, no_alt_ast, srwc)
local text = data.term
local plain
if ( -- Filters out anything but "[[...]]" with no intermediate "[[" or "]]".
if lang == nil then
not match(text, "^()%[%[") or -- Faster than sub(text, 1, 2) ~= "[[".
lang, plain = get_lang("und"), true
find(text, "[[", 3, true) or
end
find(text, "]]", 3, true) ~= #text - 1
 
) then
-- Get the link target and display text. If the term is the empty string, treat the input as a link to the current page.
return
if term == "" then
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
text = sub(text, 3, -3)
if alt then
-- A pipe at the start of an embedded link is treated as part of the target (e.g. [[|foo]]: "|foo").
alt = selective_trim(alt)
-- FIXME: This should be handled via a proper escape sequence.
if alt == "" then
local pipe = find(text, "|", 2, true)
alt = nil
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
end
data.term, data.alt = term, pipe and pipe ~= #text and sub(text, pipe + 1) or nil
-- If there's nothing to process, return nil.
if data.cats then
if not (term or alt) then
if data.suppress_redundant_wikilink_cat and data.suppress_redundant_wikilink_cat(data.term, data.alt) then
return nil
return
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
end
insert(data.cats, data.lang:getFullName() .. " links with redundant wikilinks")
term = selective_trim(term)
end
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 498: Line 739:
function export.language_link(data)
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(
-- Do we have a redundant wikilink? If so, remove it.
"The first argument to the function language_link must be a table. See Module:links/documentation for more information.")
elseif data.term then
handle_redundant_wikilink(data)
-- Nothing to process, return nil.
elseif not data.alt then
return nil
end
end
 
local text = data.term
-- Categorize links to "und".
local lang, cats = data.lang, data.cats
-- If we don't have a script, get one.
if cats and lang:getCode() == "und" then
if not data.sc then
insert(cats, "Undetermined language links")
data.sc = data.lang:findBestScript(data.alt or text)
end
end
 
-- Do we have embedded wikilinks? If so, they need to be processed individually.
return simple_link(
if text then
data.term,
local open = find(text, "[[", 1, true)
data.fragment,
if open and find(text, "]]", open + 2, true) then
data.alt,
return process_embedded_links(text, data)
lang,
end
data.sc,
end
data.id,
cats,
-- If not, make a link using the parameters.
data.no_alt_ast,
text = text and trim(text)
data.suppress_redundant_wikilink_cat
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
end


function export.plain_link(data)
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(
-- Do we have a redundant wikilink? If so, remove it.
"The first argument to the function plain_link must be a table. See Module:links/documentation for more information.")
elseif data.term then
handle_redundant_wikilink(data)
-- Only have alt (or nothing), just return it.
else
return data.alt
end
end
 
-- Make sure the language is "und".
return simple_link(
local lang = data.lang
data.term,
if not lang or lang:getCode() ~= "und" then
data.fragment,
data.lang = require("Module:languages").getByCode("und")
data.alt,
end
nil,
data.sc,
local text = data.term
data.id,
data.cats,
-- If we don't have a script, get one.
data.no_alt_ast,
if not data.sc then
data.suppress_redundant_wikilink_cat
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
end


Line 567: Line 784:
function export.embedded_language_links(data)
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
end


local text = data.term
local term, lang, sc = data.term, data.lang, data.sc
 
-- If we don't have a script, get one.
-- If we don't have a script, get one.
if not data.sc then
if not sc then
data.sc = data.lang:findBestScript(text)
sc = lang:findBestScript(term)
end
end
 
-- Do we have embedded wikilinks? If so, they need to be processed individually.
-- Do we have embedded wikilinks? If so, they need to be processed individually.
local open = find(text, "[[", 1, true)
local open = find(term, "[[", nil, true)
if open and find(text, "]]", open + 2, true) then
if open and find(term, "]]", open + 2, true) then
return process_embedded_links(text, data)
return process_embedded_links(term, data.alt, lang, sc, data.id, data.cats, data.no_alt_ast)
end
end
 
-- If not, return the display text.
-- If not, return the display text.
text = trim(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.
-- 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")
term = term:gsub("%%", "%%25")
return (data.lang:makeDisplayText(text, data.sc, true))
return lang:makeDisplayText(term, sc, true)
end
end


Line 596: Line 814:
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>' }
if type(text) == "string" and text:find("^''[^'].*''$") then
-- Temporary tracking for mention glosses that are entirely italicized or bolded, which is probably
-- wrong. (Note that this will also find bolded mention glosses since they use triple apostrophes.)
end
elseif item_type == "tr" then
elseif item_type == "tr" then
if face == "term" then
if face == "term" then
Line 612: Line 826:
elseif item_type == "pos" then
elseif item_type == "pos" then
tag = { '<span class="ann-pos">', '</span>' }
tag = { '<span class="ann-pos">', '</span>' }
elseif item_type == "non-gloss" then
tag = { '<span class="ann-non-gloss">', '</span>' }
elseif item_type == "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>' }
elseif item_type == "infl" then
tag = { '<span class="ann-infl">', '</span>' }
end
end


Line 624: Line 842:
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 633: Line 853:
*:: 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 652: Line 876:


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 668: Line 898:


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 687: Line 914:
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 load_data("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 698: Line 953:
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 705: Line 968:


return concat(output)
return concat(output)
end
-- 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 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
end


Line 724: Line 1,067:
local q = data.q
local q = data.q
if type(q) == "string" then
if type(q) == "string" then
q = {q}
q = { q }
end
end
local qq = data.qq
local qq = data.qq
if type(qq) == "string" then
if type(qq) == "string" then
qq = {qq}
qq = { qq }
end
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
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
data.ll and data.ll[1] or data.refs and data.refs[1] then
formatted = require(pron_qualifier_module).format_qualifiers {
formatted = format_qualifiers {
lang = data.lang,
lang = data.lang,
text = formatted,
text = formatted,
Line 749: Line 1,092:




--[==[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}}:
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 758: Line 1,102:
track_sc = boolean,
track_sc = boolean,
no_nonstandard_sc_cat = boolean,
no_nonstandard_sc_cat = boolean,
fragment = link_fragment
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,
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",
q = { "left_qualifier1", "left_qualifier2", ...} or "left_qualifier",
qq = { "right_qualifier1", "right_qualifier2", ...} or "right_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"}, ... },
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. 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.
* 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.
* 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).
* Call `[[#language_link|language_link]]` on the `term` or `alt` forms, to remove diacritics in the page name, process
* 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.]==]
  any embedded wikilinks and create links to Reconstruction or Appendix pages when necessary.
* 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)
function export.full_link(data, face, allow_self_link, show_qualifiers)
if type(data) ~= "table" then
error("The first argument to the function full_link must be a table. "
.. "See Module:links/documentation for more information.")
end
-- Prevent data from being destructively modified.
-- Prevent data from being destructively modified.
local data = shallow_copy(data)
local data = shallow_copy(data)


if type(data) ~= "table" then
-- FIXME: this shouldn't be added to `data`, as that means the input table needs to be cloned.
error("The first argument to the function full_link must be a table. "
data.cats = {}
.. "See Module:links/documentation for more information.")
 
-- Categorize links to "und".
local lang, cats = data.lang, data.cats
if cats and lang:getCode() == "und" then
insert(cats, "Undetermined language links")
end
end


local terms = {true}
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] = export.split_on_slashes(data[param])
data[param] = export.split_on_slashes(data[param])
elseif type(data[param]) == "string" and not (type(data.term) == "string" and data.term:find("//")) then
elseif type(data[param]) == "string" and not (type(data.term) == "string" and data.term:find("//", nil, true)) then
data[param] = data.lang:generateForms(data[param])
if not data.no_generate_forms then
data[param] = lang:generateForms(data[param])
else
data[param] = { data[param] }
end
else
else
data[param] = {}
data[param] = {}
Line 805: Line 1,195:
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 = {}
data.cats = {}
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])


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 display_term = data.alt[i] or data.term[i]
local display_term = data.alt[i] or data.term[i]
local best = data.lang:findBestScript(display_term)
local best = lang:findBestScript(display_term)
-- no_nonstandard_sc_cat is intended for use in [[Module:interproject]]
-- no_nonstandard_sc_cat is intended for use in [[Module:interproject]]
if (
if (
not data.no_nonstandard_sc_cat and
not data.no_nonstandard_sc_cat and
best:getCode() == "None" and
best:getCode() == "None" and
require("Module:scripts").findBestScriptWithoutLang(display_term):getCode() ~= "None"
find_best_script_without_lang(display_term):getCode() ~= "None"
) then
) then
insert(data.cats, data.lang:getFullName() .. " terms in nonstandard scripts")
insert(cats, lang:getFullName() .. " terms in nonstandard scripts")
end
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.
-- Track uses of sc parameter.
elseif data.track_sc then
elseif data.track_sc then
if data.sc[i]:getCode() == best:getCode() then
if data.sc[i]:getCode() == best:getCode() then
insert(data.cats, data.lang:getFullName() .. " terms with redundant script codes")
insert(cats, lang:getFullName() .. " terms with redundant script codes")
else
else
insert(data.cats, data.lang:getFullName() .. " terms with non-redundant manual script codes")
insert(cats, lang:getFullName() .. " terms with non-redundant manual script codes")
end
end
end
end
Line 849: Line 1,240:
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(data.cats, "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
["_"] = TEMP_UNDERSCORE,
-- simple_link can return nil, so check if a link has been generated.
["<"] = "&lt;",
if link then
[">"] = "&gt;",
-- Add "nowrap" class to prefixes in order to prevent wrapping after the hyphen
}
local nowrap
local function encode_accel_param_chars(param)
local display_term = data.alt[i] or data.term[i]
local retval = param:gsub("[% <>_]", encode_accel_char_map) -- discard second return value
if display_term and (display_term:find("^%-") or display_term:find("^־")) then -- Hebrew maqqef -- FIXME, use hyphens from [[Module:affix]]
return retval
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 = 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:getFullCode() .. " " .. 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],
fragment = data.fragment,
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,
cats = data.cats,
no_alt_ast = data.no_alt_ast,
suppress_redundant_wikilink_cat = data.suppress_redundant_wikilink_cat,
}
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
--[[ 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).
-- 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
-- If there are multiple terms, break the loop instead.
-- If there are multiple terms, break the loop instead.
if i > 1 then
if i > 1 then
remove(output)
remove(output)
break
break
elseif mw.title.getCurrentTitle().nsText ~= "Template" then
elseif NAMESPACE ~= "Template" then
insert(data.cats, data.lang:getFullName() .. " term requests")
insert(cats, lang:getFullName() .. " term requests")
end
end
link = "<small>[Term?]</small>"
link = "<small>[Term?]</small>"
Line 964: Line 1,286:
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
else
local phonetic_extraction = load_data("Module:links/data").phonetic_extraction
-- 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).
phonetic_extraction = phonetic_extraction[data.lang:getCode()] or phonetic_extraction[data.lang:getFullCode()]
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
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]))
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


elseif (data.term[1] or data.alt[1]) and data.sc[1]:isTransliterated() then
if not nevercalltr then
-- Track whenever there is manual translit. The categories below like 'terms with redundant transliterations'
-- Try to generate a transliteration.
-- aren't sufficient because they only work with reference to automatic translit and won't operate at all in
local text = data.alt[1] or data.term[1]
-- languages without any automatic translit, like Persian and Hebrew.
if not link_tr then
text = export.remove_links(text, true)
end


-- Try to generate a transliteration.
local automated_tr = lang:transliterate(text, data.sc[1])
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


local automated_tr, tr_categories
if automated_tr then
automated_tr, data.tr_fail, tr_categories = data.lang:transliterate(text, data.sc[1])
local manual_tr = data.tr[1]


if automated_tr or data.tr_fail then
if manual_tr then
local manual_tr = data.tr[1]
if export.remove_links(manual_tr) == export.remove_links(automated_tr) then
insert(cats, lang:getFullName() .. " terms with redundant transliterations")
else
-- Prevents Arabic root categories from flooding the tracking categories.
if NAMESPACE ~= "Category" then
insert(cats,
lang:getFullName() .. " terms with non-redundant manual transliterations")
end
end
end


if manual_tr then
if not manual_tr or lang:overrideManualTranslit(data.sc[1]) then
if (export.remove_links(manual_tr) == export.remove_links(automated_tr)) and (not data.tr_fail) then
data.tr[1] = automated_tr
insert(data.cats, data.lang:getFullName() .. " terms with redundant transliterations")
elseif not data.tr_fail then
-- Prevents Arabic root categories from flooding the tracking categories.
if mw.title.getCurrentTitle().nsText ~= "Category" then
insert(data.cats, data.lang:getFullName() .. " terms with non-redundant manual transliterations")
end
end
end
end
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
Line 1,014: Line 1,342:


-- 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(data.sc[1]) 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{
data.tr[1] = simple_link(
lang = data.lang,
data.tr[1],
term = data.tr[1],
nil,
suppress_redundant_wikilink_cat = data.suppress_redundant_wikilink_cat,
nil,
sc = require("Module:scripts").getByCode("Latn")
lang,
}
get_script("Latn"),
elseif data.tr[1] and not (data.lang:link_tr(data.sc[1]) or data.tr_fail) then
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>", "")
Line 1,029: Line 1,362:
insert(output, export.format_link_annotations(data, face))
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 ""
if data.pretext then
insert(output, 1, data.pretext)
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)
output = concat(output)
if show_qualifiers then
if show_qualifiers or data.show_qualifiers then
output = add_qualifiers_and_refs_to_term(data, output)
output = add_qualifiers_and_refs_to_term(data, output)
end
end
Line 1,061: Line 1,401:
function(c1, c2, c3)
function(c1, c2, c3)
-- Don't remove files.
-- Don't remove files.
for _, false_positive in ipairs({"file", "image"}) do
for _, false_positive in ipairs({ "file", "image" }) do
if c2:lower():match("^" .. false_positive .. ":") 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 _, false_positive in ipairs({"category", "cat"}) do
for _, false_positive in ipairs({ "category", "cat" }) do
if c2:lower():match("^" .. false_positive .. ":") then return "" end
if c2:lower():match("^" .. false_positive .. ":") then return "" end
end
end
Line 1,083: Line 1,423:
return text
return text
end
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
]=]


function export.section_link(link)
function export.section_link(link)
Line 1,103: Line 1,428:
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("_", " ")
local target, section = get_fragment((link:gsub("_", " ")))
local target, section = link:match("(.-)#(.*)")
 
if not section then
if not target then
error("No \"#\" delineating a section name")
error("The function “section_link” could not find a number sign marking a section name.")
end
end
 
return export.plain_link{
return simple_link(
term = target,
target,
fragment = section,
section,
alt = link:gsub("#", " §&nbsp;", 1)
target .. " §&nbsp;" .. section
}
)
end
end


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