Wiktionary gdwiktionary https://gd.wiktionary.org/wiki/Pr%C3%AComh-Dhuilleag MediaWiki 1.47.0-wmf.18 case-sensitive Meadhan Sònraichte Deasbaireachd Cleachdaiche Deasbaireachd a' chleachdaiche Wiktionary An deasbaireachd aig Wiktionary Faidhle Deasbaireachd an fhaidhle MediaWiki Deasbaireachd MediaWiki Teamplaid Deasbaireachd na teamplaid Cobhair Deasbaireachd na cobharach Roinn-seòrsa Deasbaireachd na roinn-seòrsa TimedText TimedText talk Mòideal Deasbaireachd mòideil Event Event talk 10 18085 89189 2025-11-01T15:05:51Z en>Surjection 0 Protected "[[Template:en-head]]": Highly visible template/module ([Edit=Allow only autoconfirmed users] (indefinite) [Move=Allow only autoconfirmed users] (indefinite)) 89189 wikitext text/x-wiki {{#invoke:en-headword|show}}<!-- --><noinclude>{{documentation}}</noinclude> b61rollxgc9fbrkcs5t760xnqz1hw58 89190 89189 2026-09-02T18:50:44Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Template:en-head]] 89189 wikitext text/x-wiki {{#invoke:en-headword|show}}<!-- --><noinclude>{{documentation}}</noinclude> b61rollxgc9fbrkcs5t760xnqz1hw58 828 18086 89191 2026-01-11T23:45:01Z en>Benwing2 0 check_missing should be done through checkredlinks=1 in [[Module:headword]] 89191 Scribunto text/plain local export = {} local pos_functions = {} --[==[ Author from 2020 on: mostly Benwing2, with significant contributions from Theknightwho. Based on a prior version by Rua (by now mostly rewritten), with contributions from Erutuon and others (see history for full attribution). ]==] local force_cat = false -- for testing; if true, categories appear in non-mainspace pages local require = require local require_when_needed = require("Module:require when needed") local en_utilities_module = "Module:en-utilities" local headword_utilities_module = "Module:headword utilities" local headword_module = "Module:headword" local inflection_utilities_module = "Module:inflection utilities" local parse_utilities_module = "Module:parse utilities" local JSON_module = "Module:JSON" local labels_module = "Module:labels" local links_module = "Module:links" local parameters_module = "Module:parameters" local string_utilities_module = "Module:string utilities" local table_module = "Module:table" local utilities_module = "Module:utilities" local yesno_module = "Module:yesno" local iut = require_when_needed(inflection_utilities_module) local put = require_when_needed(parse_utilities_module) local m_headword_utilities = require_when_needed(headword_utilities_module) local add_links_to_multiword_term = require_when_needed(headword_utilities_module, "add_links_to_multiword_term") local add_suffix = require_when_needed(en_utilities_module, "add_suffix") local apply_link_modifiers = require_when_needed(headword_utilities_module, "apply_link_modifiers") local concat = table.concat local deepEquals = require_when_needed(table_module, "deepEquals") local dump = mw.dumpObject local format_categories = require_when_needed(utilities_module, "format_categories") local full_headword = require_when_needed(headword_module, "full_headword") local get_label_info = require_when_needed(labels_module, "get_label_info") local get_link_page = require_when_needed(links_module, "get_link_page") local glossary_link = require_when_needed(headword_utilities_module, "glossary_link") local insert = table.insert local insertIfNot = require_when_needed(table_module, "insertIfNot") local ipairs = ipairs local is_regular_plural = require_when_needed(en_utilities_module, "is_regular_plural") local list_to_set = require_when_needed(table_module, "listToSet") local pairs = pairs local process_params = require_when_needed(parameters_module, "process") local remove = table.remove local remove_links = require_when_needed(links_module, "remove_links") local replacement_escape = require_when_needed(string_utilities_module, "replacement_escape") local shallowCopy = require_when_needed(table_module, "shallowCopy") local singularize = require_when_needed(en_utilities_module, "singularize") local split = require_when_needed(string_utilities_module, "split") local toJSON = require_when_needed(JSON_module, "toJSON") local toNFD = mw.ustring.toNFD local type = type local ulen = require_when_needed(string_utilities_module, "len") local ulower = require_when_needed(string_utilities_module, "lower") local umatch = require_when_needed(string_utilities_module, "match") local u = require_when_needed(string_utilities_module, "char") local ugsub = require_when_needed(string_utilities_module, "gsub") local lang = require("Module:languages").getByCode("en") local langname = lang:getCanonicalName() local list_param = {list = true, disallow_holes = true} local list_allow_holes = {list = true, allow_holes = true} local boolean_param = {type = "boolean"} local function ine(val) if val == "" then return nil else return val end end local function track(page) require("Module:debug/track")("en-headword/" .. page) return true end ------------------------------------------- UTILITY FUNCTIONS ------------------------------------------ -- Parse and return an inflection not requiring additional processing. The raw arguments come from `args[field]`, which -- is parsed for inline modifiers. local function parse_inflection(args, field, is_head) local argfield = field if type(argfield) == "table" then argfield = argfield[1] end return m_headword_utilities.parse_term_list_with_modifiers { paramname = field, forms = args[argfield], splitchar = ",", is_head = is_head, } end -- Insert the parsed inflections in `terms` (as parsed by `parse_inflection`) into `data.inflections`, with label -- `label` and optional accelerator spec `accel`. local function insert_inflection(data, terms, label, accel, no_label) for _, termobj in ipairs(terms) do m_headword_utilities.remove_termobj_field_modifiers(termobj) end m_headword_utilities.insert_inflection { headdata = data, terms = terms, label = label, no_label = no_label, accel = accel and {form = accel} or nil, } end -- Insert a fixed label `label` into the inflections for `data`. If `originating_term` is supplied, copy the qualifiers, -- labels and references from it into the fixed label. local function insert_fixed_inflection(data, label, originating_term) m_headword_utilities.insert_fixed_inflection { headdata = data, originating_term = originating_term, label = label, } end -- Parse and insert an inflection not requiring additional processing into `data.inflections`. The raw arguments come -- from `args[field]`, which is parsed for inline modifiers. `label` is the label that the inflections are given; -- `accel` is the accelerator form, or nil. local function parse_and_insert_inflection(data, args, field, label, accel) m_headword_utilities.parse_and_insert_inflection { headdata = data, forms = args[field], paramname = field, splitchar = ",", label = label, accel = accel and {form = accel} or nil, } end -- These functions are used directly in the <> format as well as in the utility functions #2 below. local function compute_double_last_cons_stem(term) local last_cons = term:match("([bcdfghjklmnpqrstvwxyzBCDFGHJKLMNPQRSTVWXYZ])$") if not last_cons then error("Verb stem '" .. term .. "' must end in a consonant to use ++") end return term .. last_cons end local function compute_plusplus_s_form(term, default_s_form) if term:find("[szx]$") then -- regas -> regasses, derez -> derezzes return compute_double_last_cons_stem(term) .. "es" else return default_s_form end end -- The main entry point. -- This is the only function that can be invoked from a template. function export.show(frame) local iparams = { [1] = true, } local iargs = require("Module:parameters").process(frame.args, iparams) local parargs = frame:getParent().args local poscat = iargs[1] local pos_in_1 = not poscat if pos_in_1 then poscat = ine(parargs[1]) or mw.title.getCurrentTitle().fullText == "Template:en-head" and "interjection" or error("Part of speech must be specified in 1=") poscat = require(headword_module).canonicalize_pos(poscat) end local indexing_poscat = pos_in_1 and "head" or poscat local params = { ["head"] = list_param, ["id"] = true, ["json"] = boolean_param, ["sort"] = true, ["splithyph"] = boolean_param, ["nosplithyph"] = boolean_param, ["hyphspace"] = boolean_param, ["nolink"] = boolean_param, ["nolinkhead"] = {type = "boolean_param", alias_of = "nolink"}, ["suffix"] = boolean_param, ["nosuffix"] = boolean_param, ["nomultiwordcat"] = boolean_param, ["abbr"] = list_param, ["the"] = true, ["def"] = {alias_of = "the"}, ["pagename"] = true, -- for testing } if pos_in_1 then params[1] = {required = true} -- required but ignored as already processed above end local pos_data = pos_functions[indexing_poscat] local pos_func if pos_data then local pos_params = pos_data.params if pos_params then for key, val in pairs(pos_params) do params[key] = val end end pos_func = pos_data.func end local args = process_params(parargs, params) -- Account for unsupported titles, e.g. 'C|N>K' instead of 'Unsupported titles/C through N to K'. local pagename = args.pagename or mw.loadData("Module:headword/data").pagename local user_specified_heads = parse_inflection(args, "head", "is_head") local heads = user_specified_heads local autohead if args.nolink or not pagename:find("[ '%-]") then autohead = pagename else local en_no_split_apostrophe_words = list_to_set { "one's", "someone's", "he's", "she's", "it's", } local en_include_hyphen_prefixes = list_to_set { -- We don't include things that are also words even though they are often (perhaps mostly) prefixes, e.g. -- "be", "counter", "cross", "extra", "half", "mid", "over", "pan", "under". "acro", "acousto", "Afro", "agro", "anarcho", "angio", "Anglo", "ante", "anti", "arch", "auto", "bi", "bio", "cis", "co", "cryo", "crypto", "de", "demi", "eco", "electro", "Euro", "ex", "Greco", "hemi", "hydro", "hyper", "hypo", "infra", "Indo", "inter", "intra", "Judeo", "macro", "meta", "micro", "mini", "multi", "neo", "neuro", "non", "para", "peri", "post", "pre", "pro", "proto", "pseudo", "re", "semi", "sub", "super", "trans", "un", "vice", } local function is_english(term) local title = mw.title.new(term) if title and title.exists then local content = title:getContent() if content and content:find("==English==\n") then return true end end return false end local function en_split_hyphen_when_space(word) if not word:find("-", nil, true) then return nil end if args.hyphspace then return "[[" .. word:gsub("%-+", " ") .. "|" .. word .. "]]" end if args.nosplithyph then return "[[" .. word .. "]]" end if not args.splithyph then local space_word = word:gsub("%-+", " ") if is_english(space_word) then return "[[" .. space_word .. "|" .. word .. "]]" end if is_english(word) then return "[[" .. word .. "]]" end end return nil end local function en_split_apostrophe(word) local base = word:match("^(.*)'s$") if base then return "[[" .. base .. "]][[-'s|'s]]" end -- Only treat final apostrophe as possessive if preceded by something that looks like a plural ending in /z/. -- In particular we don't want to do it for words like [[truckin']]. base = word:match("^(.*[sxz])'$") if base then if base:find("s$") then local sg = singularize(base) if is_english(sg) then return "[[" .. sg .. "|" .. base .. "]][[-'|']]" end end return "[[" .. base .. "]][[-'|']]" end return "[[" .. word .. "]]" end autohead = add_links_to_multiword_term(pagename, { split_hyphen_when_space = en_split_hyphen_when_space, split_apostrophe = en_split_apostrophe, no_split_apostrophe_words = en_no_split_apostrophe_words, include_hyphen_prefixes = en_include_hyphen_prefixes, }) end if not heads[1] then heads = {{term = autohead}} else for _, headobj in ipairs(heads) do local head = headobj.term if head:find("^~") then head = apply_link_modifiers(autohead, head:sub(2), lang) headobj.term = head elseif head:find("^[!?]$") then -- If explicit head= just consists of ! or ?, add it to the end of the default head. headobj.term = autohead .. head end if head == autohead then track("redundant-head") end end end -- handle the=/def= if args.the == "~" then local newheads = {} for _, headobj in ipairs(heads) do local barehead = shallowCopy(headobj) insert(newheads, barehead) headobj.term = "the " .. headobj.term insert(newheads, headobj) end heads = newheads elseif args.the then local the = require(yesno_module)(args.the) if the then for _, headobj in ipairs(heads) do headobj.term = "the " .. headobj.term end end end local data = { lang = lang, pos_category = poscat, categories = {}, heads = heads, user_specified_heads = user_specified_heads, -- We use our own splitting algorithm so the redundant head cat will be inaccurate. no_redundant_head_cat = true, inflections = {}, nomultiwordcat = args.nomultiwordcat, sort_key = args.sort, pagename = pagename, id = args.id, force_cat_output = force_cat, } local function inscat(cat) insert(data.categories, langname .. " " .. cat) end local is_suffix = false if args.suffix or not args.nosuffix and pagename:find("^%-") and not pagename:find("^%-%-") and poscat ~= "suffix forms" then is_suffix = true data.pos_category = "suffixes" local singular_poscat = singularize(poscat) inscat(singular_poscat .. "-forming suffixes") insert(data.inflections, {label = singular_poscat .. "-forming suffix"}) end if pos_func then pos_func(args, data, is_suffix) end local extra_categories = {} if pagename:find("[Qq]") then -- Check for q not followed by u. We want to exclude things like [[13q deletion syndrome]] and [[BFOQ]] that -- don't have a lowercase letter on either side, as well as things like [[& seq.]] and [[acq.]] that are -- abbreviations for words containing a following u. -- -- Approximate range of combining diacritics; we want to remove them so the checks below for -- a lowercase letter next to the q aren't tripped up by diacritics on the letter. local u300 = u(0x0300) local u36F = u(0x036F) local pagename_no_diacritics = ugsub(toNFD(pagename), "[" .. u300 .. "-" .. u36F .. "]", "") if pagename_no_diacritics:find("[Qq][a-tv-z]") or pagename_no_diacritics:find("[a-z]q[^u.]") or pagename_no_diacritics:find("[a-z]q$") then inscat("words containing Q not followed by U") end end -- toNFD performs decomposition, so letters that decompose to an ASCII -- vowel and a diacritic, such as é, are counted as vowels and do not do not -- need to be included in the pattern. if not umatch(ulower(toNFD(pagename)), "[aeiouyæœøəªºαεηιουω]") then inscat("words spelled without vowels") end if pagename:find("yre$") then inscat('words ending in "-yre"') end if not pagename:find(" ") and ulen(pagename) >= 25 then insert(extra_categories, "Long " .. langname .. " words") end if pagename:find("^[^aeiou ]*a[^aeiou ]*e[^aeiou ]*i[^aeiou ]*o[^aeiou ]*u[^aeiou ]*$") then inscat("words that use all vowels in alphabetical order") end parse_and_insert_inflection(data, args, "abbr", "abbreviation") if args.json then return toJSON(data) end return full_headword(data) .. (extra_categories[1] and format_categories(extra_categories, lang, args.sort) or "") end local function make_default_comparative(word) if word == "good" or word == "well" then return {"better"} elseif word == "bad" or word == "badly" then return {"worse"} elseif word == "far" then return {"further", "farther"} else return {add_suffix(word, "r")} end end local function make_default_superlative(word) if word == "good" or word == "well" then return {"best"} elseif word == "bad" or word == "badly" then return {"worst"} elseif word == "far" then return {"furthest", "farthest"} else return {add_suffix(word, "st.superlative")} end end -- This function does the common work between adjectives and adverbs. local function process_comparative_args(data, args, plpos) local pagename = data.pagename local comps = parse_inflection(args, 1) local sups = parse_inflection(args, "sup") local outcomps, outsups if args.componly then if comps[1] then error("Can't specify comparatives of comparative-only " .. plpos) end insert(data.inflections, {label = glossary_link("comparative") .. " form only"}) insert(data.categories, langname .. " comparative-only " .. plpos) -- Set to empty list so we don't get any comparatives output, but process superlatives if specified. outcomps = {} if not sups[1] then -- Set to empty list so we don't get any superlatives output unless explicitly given. outsups = {} end elseif args.suponly then if comps[1] or sups[1] then error("Can't specify comparatives or superlatives of or superlative-only " .. plpos) end insert(data.inflections, {label = glossary_link("superlative") .. " form only"}) insert(data.categories, langname .. " superlative-only " .. plpos) return end -- If the first parameter is ?, then don't show anything, just return. if comps[1] and comps[1].term == "?" then if comps[2] then error("Can't specify additional comparatives along with '?'") end if sups[1] then error("Can't specify superlatives along with '?' for the comparative") end return end if comps[1] and comps[1].term == "-" then local hyphencomp = remove(comps, 1) -- Remove the "-" but retain for qualifiers, labels, references -- Not (generally) comparable; may occasionally have a comparative if comps[1] then insert_fixed_inflection(data, "not generally <<comparable>>", hyphencomp) elseif not sups[1] then insert_fixed_inflection(data, "not <<comparable>>", hyphencomp) insert(data.categories, langname .. " uncomparable " .. plpos) return else -- No comparative, but a superlative. insert_inflection() will correctly generate 'no comparative' if we -- pass in "-" as the value. outcomps = {hyphencomp} end elseif not comps[1] then comps = {{term = "more"}} end if not outcomps then -- not if we set `outcomps` to "-" above or processed a comparative-only term outcomps = {} -- Go over each parameter given and create a comparative and superlative form. for _, compobj in ipairs(comps) do local comp = compobj.term if comp == "-" then error("Comparative of '-' only allowed as first comparative") end if comp == "+" then comp = "+more" elseif comp == "more" and pagename ~= "many" and pagename ~= "much" then comp = "+more" elseif comp == "further" and pagename ~= "far" then comp = "+further" elseif comp == "better" and pagename ~= "good" and pagename ~= "well" then comp = "+better" elseif comp:find("~") then comp = comp:gsub("~", replacement_escape(pagename)) end compobj.origterm = comp if comp == "+more" then comp = "more [[" .. pagename .. "]]" elseif comp == "+further" then comp = {"further [[" .. pagename .. "]]", "farther [[" .. pagename .. "]]"} elseif comp == "+better" then comp = "better [[" .. pagename .. "]]" elseif comp == "er" then -- Add -er. comp = add_suffix(pagename, "r") elseif comp == "ier" then if pagename:sub(-1) ~= "y" then error("Can't specify 'ier' comparative unless the term ends with 'y': " .. pagename) end comp = pagename:gsub("e?y$", "ier") elseif comp:find("^%+") then local special = m_headword_utilities.get_special_indicator(comp, "noerror") if special then comp = m_headword_utilities.handle_multiword(pagename, special, make_default_comparative) end end if type(comp) == "table" and not comp[2] then comp = comp[1] end if type(comp) == "table" then for i = 1, #comp - 1 do local outobj = shallowCopy(compobj) outobj.term = comp[i] insert(outcomps, outobj) end compobj.term = comp[#comp] insert(outcomps, compobj) else compobj.term = comp insert(outcomps, compobj) end end end if sups[1] and sups[1].term == "-" then if sups[2] then error("Can't specify '-' as superlative followed by further values") end -- No superlative. insert_inflection() will correctly generate 'no superlative' if we pass in "-" as the value. outsups = sups else if not sups[1] then sups = {{term = "+"}} end end -- `outsups` will be set if we set `outsups` to "-" above or processed a comparative-only term without superlatives. if not outsups then outsups = {} local function process_sup(sup, special, supobj, compobj) if special then sup = m_headword_utilities.handle_multiword(pagename, special, make_default_superlative) elseif sup == "-" or sup == "+" then error(("Internal error: Superlative value of '%s' should have been handled earlier"):format(sup)) elseif sup == "+most" then sup = "most [[" .. pagename .. "]]" elseif sup == "+furthest" then sup = {"furthest [[" .. pagename .. "]]", "farthest [[" .. pagename .. "]]"} elseif sup == "+best" then sup = "best [[" .. pagename .. "]]" elseif sup == "est" then -- Add -est. sup = add_suffix(pagename, "st.superlative") elseif sup == "iest" then if pagename:sub(-1) ~= "y" then error("Can't specify 'iest' superlative unless the term ends with 'y': " .. pagename) end sup = pagename:gsub("e?y$", "iest") end if type(sup) == "table" and not sup[2] then sup = sup[1] end if compobj then supobj = shallowCopy(supobj) supobj = m_headword_utilities.combine_termobj_qualifiers_labels(supobj, compobj) end if type(sup) == "table" then for i = 1, #sup - 1 do local outobj = shallowCopy(supobj) outobj.term = sup[i] insert(outsups, outobj) end supobj.term = sup[#sup] insert(outsups, supobj) else supobj.term = sup insert(outsups, supobj) end end for _, supobj in ipairs(sups) do local sup = supobj.term if sup == "-" then error("Superlative of '-' only allowed as first superlative") end if sup == "+" then if not comps[1] then error("Superlative of '+' can't be specified when there are no comparatives") end for _, compobj in ipairs(comps) do local comp = compobj.origterm local special if comp == "+more" then sup = "+most" elseif comp == "+further" then sup = "+furthest" elseif comp == "+better" then sup = "+best" elseif comp == "er" then sup = "est" elseif comp == "ier" then sup = "iest" else if comp:find("^%+") then special = m_headword_utilities.get_special_indicator(comp, "noerror") end if not special then -- If the full comparative was given, then derive the superlative by replacing -er with -- -est. if comp:sub(-2) == "er" then sup = comp:sub(1, -3) .. "est" else error(("The superlative cannot be derived automatically from comparative '%s' because it doesn't end in -er"):format(comp)) end end end process_sup(sup, special, supobj, compobj) end else local special = m_headword_utilities.get_special_indicator(sup, "noerror") -- Do some work here rather than in process_sup() so we don't end up double-processing a term with a '~' -- in it or a term that happens to be 'most' or similar after substitution of ~ in the comparative. if not special then if sup == "most" and pagename ~= "many" and pagename ~= "much" then sup = "+most" elseif sup == "furthest" and pagename ~= "far" then sup = "+furthest" elseif sup == "best" and pagename ~= "good" and pagename ~= "well" then sup = "+best" elseif sup:find("~") then sup = sup:gsub("~", replacement_escape(pagename)) end end process_sup(sup, special, supobj) end end end insert_inflection(data, outcomps, "<<comparative>>", "comparative") insert_inflection(data, outsups, "<<superlative>>", "superlative") end pos_functions["adjectives"] = { params = { [1] = list_param, ["comp_qual"] = {list = "comp\1_qual", allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the comparative value", }, ["sup"] = list_param, ["sup_qual"] = {list = "sup\1_qual", allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the superlative value", }, ["componly"] = boolean_param, ["suponly"] = boolean_param, }, func = function(args, data) -- Process the comparatives and superlatives. process_comparative_args(data, args, "adjectives") end, } pos_functions["adverbs"] = { params = { [1] = list_param, ["comp_qual"] = {list = "comp\1_qual", allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the comparative value", }, ["sup"] = list_param, ["sup_qual"] = {list = "sup\1_qual", allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the superlative value", }, ["componly"] = boolean_param, ["suponly"] = boolean_param, }, func = function(args, data) -- Process the comparatives and superlatives. process_comparative_args(data, args, "adverbs") end, } local function escape(str) return (str:gsub("\\([:#])", "\\\\%1") :gsub("[:#]", "\\%0")) end local function canonicalize_plural(pl, pagename, pos) if pl == "+" then return escape(add_suffix(pagename, "s.plural", pos)) elseif pl == "++" then return escape(compute_plusplus_s_form(pagename, add_suffix(pagename, "s.plural", pos))) elseif pl == "*" then return escape(pagename) elseif pl == "ies" then if pagename:sub(-1) == "y" then return escape(pagename:gsub("e?y$", pl)) end error("Can't specify 'ies' plural unless the term ends with 'y'.") elseif pl == "s" or pl == "es" or pl == "'s" then return escape(pagename .. pl) end end local function do_nouns(args, data, pos) local pagename = data.pagename pos = pos or "noun" local plurals = parse_inflection(args, 1) local function insert_plurale_tantum_inflections(is_plural_only, originating_label) if args.sg[1] then insert_fixed_inflection(data, "normally plural", originating_label) parse_and_insert_inflection(data, args, "sg", "singular") elseif is_plural_only then insert_fixed_inflection(data, "plural only", originating_label) end if args.attr[1] then parse_and_insert_inflection(data, args, "attr", "attributive") end end local function first_pl_term() return plurals[1] and plurals[1].term or nil end if first_pl_term() == "p" then -- plurale tantum if plurals[2] then error("With plurale tantum noun, can't specify more than one plural") end data.genders = {"p"} -- this should auto-insert the correct 'pluralia tantum' category insert_plurale_tantum_inflections("plural only", plurals[1]) return end local function inscat(cat) insert(data.categories, langname .. " " .. cat) end local need_default_plural = pos == "noun" if first_pl_term() == "sp" then -- construed as singular or plural sp = remove(plurals, 1) -- Remove the "sp" but retain it for its qualifiers, labels, references inscat("nouns construed as singular or plural") data.genders = {"s", "p"} -- this should auto-insert the correct 'pluralia tantum' category insert_plurale_tantum_inflections(nil, sp) need_default_plural = false elseif first_pl_term() == "-" then -- Uncountable noun; may occasionally have a plural local hyphpl = remove(plurals, 1) -- Remove the "-" but retain for qualifiers, labels, references inscat("uncountable nouns") -- If plural forms were given explicitly, then show "usually" if plurals[1] then insert_fixed_inflection(data, "usually <<uncountable>>", hyphpl) else insert_fixed_inflection(data, "<<uncountable>>", hyphpl) end need_default_plural = false elseif first_pl_term() == "#" then -- Usually countable (e.g., "grilled cheese") local hashpl = remove(plurals, 1) -- Remove the "#" but retain for qualifiers, labels, references insert_fixed_inflection(data, "usually <<countable>>", hashpl) inscat("uncountable nouns") inscat("countable nouns") -- If no plural was given, add a default one now if not plurals[1] then plurals[1] = {term = escape(add_suffix(pagename, "s.plural", pos))} end elseif first_pl_term() == "~" then -- Mixed countable/uncountable noun, always has a plural local tildepl = remove(plurals, 1) -- Remove the "~" but retain for qualifiers, labels, references insert_fixed_inflection(data, "<<countable>> and <<uncountable>>", tildepl) inscat("uncountable nouns") inscat("countable nouns") -- If no plural was given, add a default one now if not plurals[1] then plurals[1] = {term = escape(add_suffix(pagename, "s.plural", pos))} end end -- Plural is unknown if first_pl_term() == "?" then local questionpl = remove(plurals, 1) -- Remove the "?" but retain for qualifiers, labels, references -- Not desired; see [[Wiktionary:Tea_room/2021/August#"Plural unknown or uncertain"]] -- insert_fixed_inflection(data, "plural unknown or uncertain", questionpl) inscat("nouns with unknown or uncertain plurals") if plurals[1] then error("Can't specify explicit plurals along with '?' for unknown/uncertain plural") end return end -- Plural is not attested if first_pl_term() == "!" then local exclampl = remove(plurals, 1) -- Remove the "!" but retain for qualifiers, labels, references insert_fixed_inflection(data, "plural not attested", exclampl) inscat("nouns with unattested plurals") if plurals[1] then error("Can't specify explicit plurals along with '!' for unattested plural") end return end -- If no plural was given, maybe add a default one, otherwise (when "-" was given or proper noun) return. if not plurals[1] then if not need_default_plural then inscat("uncountable nouns") return end plurals[1] = {term = escape(add_suffix(pagename, "s.plural", pos))} end -- There are plural forms to show, so show them. inscat("countable nouns") local irregular, indeclinable for i, pl in ipairs(plurals) do local canon_pl = canonicalize_plural(pl.term, pagename, pos) if canon_pl then pl.term = canon_pl end local pl_term = get_link_page(pl.term, lang) if not (pagename:find(" ") or is_regular_plural(pl_term, pagename)) then irregular = true if pl_term == pagename then indeclinable = true end end end if irregular then inscat("nouns with irregular plurals") end if indeclinable then inscat("indeclinable nouns") end insert_inflection(data, plurals, "plural", "p") end -- Return the parameters to be used for nouns and proper nouns. Currently the same. local noun_params = { [1] = list_param, ["pl\1qual"] = {list = true, allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the plural", }, -- The following four only used for pluralia tantum (1=p) ["sg"] = list_param, ["attr"] = list_param, } pos_functions["nouns"] = { params = noun_params, func = do_nouns, } pos_functions["proper nouns"] = { params = noun_params, func = function(args, data) return do_nouns(args, data, "proper noun") end, } local function base_default_verb_forms(verb) return escape(add_suffix(verb, "s.verb")), escape(add_suffix(verb, "ing")), escape(add_suffix(verb, "d")) end local function default_verb_forms(verb) local full_s_form, full_ing_form, full_ed_form = base_default_verb_forms(verb) if verb:find(" ") then local first, rest = verb:match("^(.-)( .*)$") local first_s_form, first_ing_form, first_ed_form = base_default_verb_forms(first) return full_s_form, full_ing_form, full_ed_form, first_s_form .. rest, first_ing_form .. rest, first_ed_form .. rest, first, rest else return full_s_form, full_ing_form, full_ed_form, nil, nil, nil, nil, nil end end local function compute_double_last_cons_stem_of_split_verb(verb, ending) local first, rest = verb:match("^(.-)( .*)$") if not first then error("Verb '" .. verb .. "' must have a space in it to use **") end local last_cons = first:match("([bcdfghjklmnpqrstvwxyzBCDFGHJKLMNPQRSTVWXYZ])$") if not last_cons then error("First word '" .. first .. "' must end in a consonant to use **") end return first .. last_cons .. ending .. rest end local function check_non_nil_star_form(form, pagename) if form == nil then error("Verb '" .. pagename .. "' must have a space in it to use *, **, *l, *! or *'") end return form end local function sub_tilde(form, pagename) if not form then return nil end if form:find("~") then form = form:gsub("~", replacement_escape(pagename)) end return form end local deprecated_qual_replaced_by_inline_modifier = { list = true, allow_holes = true, replaced_by = false, instead = "use an inline modifier <q:...> or <l:...> on the value" } pos_functions["verbs"] = { params = { [1] = {list = "pres_3sg", disallow_holes = true}, ["pres_3sg\1_qual"] = deprecated_qual_replaced_by_inline_modifier, [2] = {list = "pres_ptc", disallow_holes = true}, ["pres_ptc\1_qual"] = deprecated_qual_replaced_by_inline_modifier, [3] = {list = "past", disallow_holes = true}, ["past\1_qual"] = deprecated_qual_replaced_by_inline_modifier, [4] = {list = "past_ptc", allow_holes = true}, ["past_ptc\1_qual"] = deprecated_qual_replaced_by_inline_modifier, ["noautolinkverb"] = boolean_param, ["angle_bracket"] = boolean_param, }, func = function(args, data) -- Get parameters local par1s local par2s = parse_inflection(args, {2, "pres_ptc"}) local par3s = parse_inflection(args, {3, "past"}) local par4s = parse_inflection(args, {4, "past_ptc"}) local pres_3sgs, pres_ptcs, pasts, past_ptcs local pagename = data.pagename ------------------------------------------- UTILITY FUNCTIONS #2 ------------------------------------------ -- These functions are used in both in the separate-parameter format and in the override params such as past_ptc2=. local full_default_s, full_default_ing, full_default_ed, split_default_s, split_default_ing, split_default_ed local lemma local function set_lemma_and_default_forms(the_lemma) lemma = the_lemma full_default_s, full_default_ing, full_default_ed, split_default_s, split_default_ing, split_default_ed, lemma_first, lemma_rest = default_verb_forms(the_lemma) end local function canonicalize_s_form(form) if form == "+" then error("Internal error: Should not see '+' here") elseif form == "^" then return full_default_s elseif form == "*" then return check_non_nil_star_form(split_default_s, lemma) elseif form == "++" then return compute_plusplus_s_form(lemma, full_default_s) elseif form == "**" then if lemma:find("^[^ ]*[szx] ") then return compute_double_last_cons_stem_of_split_verb(lemma, "es") else return check_non_nil_star_form(split_default_s, lemma) end elseif form == "+!" then return lemma .. "s" elseif form == "*!" then return check_non_nil_star_form(lemma_first) .. "s" .. lemma_rest elseif form == "+'" then return lemma .. "'s" elseif form == "*'" then return check_non_nil_star_form(lemma_first) .. "'s" .. lemma_rest elseif form == "+l" then if lemma:find("[szx]$") then return {{term = full_default_s, l = {"US"}}, {term = compute_plusplus_s_form(lemma, full_default_s), l = {"UK"}}} else return compute_plusplus_s_form(lemma, full_default_s) end elseif form == "*l" then if lemma:find("^[^ ]*[szx] ") then return {{term = check_non_nil_star_form(split_default_s, lemma), l = {"US"}}, {term = compute_double_last_cons_stem_of_split_verb(lemma, "es"), l = {"UK"}}} else return check_non_nil_star_form(split_default_s, lemma) end else return sub_tilde(form, lemma) end end local function canonicalize_ing_form(form) if form == "+" then error("Internal error: Should not see '+' here") elseif form == "^" then return full_default_ing elseif form == "*" then return check_non_nil_star_form(split_default_ing, lemma) elseif form == "++" then return compute_double_last_cons_stem(lemma) .. "ing" elseif form == "**" then return compute_double_last_cons_stem_of_split_verb(lemma, "ing") elseif form == "+!" then return lemma .. "ing" elseif form == "*!" then return check_non_nil_star_form(lemma_first) .. "ing" .. lemma_rest elseif form == "+'" then return lemma .. "'ing" elseif form == "*'" then return check_non_nil_star_form(lemma_first) .. "'ing" .. lemma_rest elseif form == "+l" then return {{term = full_default_ing, l = {"US"}}, {term = compute_double_last_cons_stem(lemma) .. "ing", l = {"UK"}}} elseif form == "*l" then return {{term = check_non_nil_star_form(split_default_ing, lemma), l = {"US"}}, {term = compute_double_last_cons_stem_of_split_verb(lemma, "ing"), l = {"UK"}}} else return sub_tilde(form, lemma) end end local function canonicalize_ed_form(form) if form == "+" then error("Internal error: Should not see '+' here") elseif form == "^" then return full_default_ed elseif form == "*" then return check_non_nil_star_form(split_default_ed, lemma) elseif form == "++" then return compute_double_last_cons_stem(lemma) .. "ed" elseif form == "+!" then return lemma .. "ed" elseif form == "*!" then return check_non_nil_star_form(lemma_first) .. "ed" .. lemma_rest elseif form == "+'" then return {{term = lemma .. "'d"}, {term = lemma .. "'ed"}} elseif form == "*'" then return {{term = check_non_nil_star_form(lemma_first) .. "'d" .. lemma_rest}, {term = check_non_nil_star_form(lemma_first) .. "'ed" .. lemma_rest}} elseif form == "**" then return compute_double_last_cons_stem_of_split_verb(lemma, "ed") elseif form == "+l" then return {{term = full_default_ed, l = {"US"}}, {term = compute_double_last_cons_stem(lemma) .. "ed", l = {"UK"}}} elseif form == "*l" then return {{term = check_non_nil_star_form(split_default_ed, lemma), l = {"US"}}, {term = compute_double_last_cons_stem_of_split_verb(lemma, "ed"), l = {"UK"}}} else return sub_tilde(form, lemma) end end -- FIXME: options should be "+", "*", "++", "**", "+n", "*n", "++n" and "**n", but not "n" local function canonicalize_en_form(form) if form == "n" then track("n4") return add_suffix(lemma, "n") end return canonicalize_ed_form(form) end --------------------------------- MAIN PARSING/CONJUGATING CODE -------------------------------- local is_angle_bracket = args.angle_bracket if is_angle_bracket then if par2s[1] or par3s[1] or par4s[1] then error("Can't specify explicit values for 2=, 3= or 4= along with the angle-bracket format") end elseif is_angle_bracket == nil and not par2s[1] and not par3s[1] and not par4s[1] and not args[1][2] and args[1][1] and args[1][1]:find("<") then if put.term_contains_top_level_html(args[1][1]) then -- Often, term_contains_top_level_html() returns true on the angle-bracket format, which would -- make the pcall() below succeed but leave the angle brackets as-is. Check for this and only do the -- pcall() if term_contains_top_level_html() returns false. is_angle_bracket = true else -- If it's ambiguous whether it's an angle-bracket format or separate params with an inline modifier, -- try to parse as the latter. If an error occurs, treat as the former. local ok ok, par1s = pcall(parse_inflection, args, {1, "pres_3sg"}) if not ok then par1s = nil is_angle_bracket = true end end end if is_angle_bracket then -------------------------- ANGLE-BRACKET FORMAT -------------------------- -- (0) Expand multiword term with angle brackets just on the first word. local arg11 = args[1][1] if arg11:find("^<.*>$") and pagename:find(" ") then local first, rest = pagename:match("^(.-)( .*)$") arg11 = first .. arg11 .. rest end -- (1) Parse the indicator specs inside of angle brackets. local function parse_indicator_spec(angle_bracket_spec) local inside = angle_bracket_spec:match("^<(.*)>$") assert(inside) local segments = put.parse_balanced_segment_run(inside, "[", "]") local comma_separated_groups = put.split_alternating_runs(segments, ",") if #comma_separated_groups > 4 then error("Too many comma-separated parts in indicator spec, expected at most 4: " .. angle_bracket_spec) end local function fetch_footnotes(separated_group) local footnotes for j = 2, #separated_group - 1, 2 do if separated_group[j + 1] ~= "" then error("Extraneous text after bracketed footnotes: '" .. concat(separated_group) .. "'") end if not footnotes then footnotes = {} end insert(footnotes, separated_group[j]) end return footnotes end local function fetch_specs(comma_separated_group) if not comma_separated_group then return {{term = "+"}} end local specs = {} local colon_separated_groups = put.split_alternating_runs(comma_separated_group, ":") for _, colon_separated_group in ipairs(colon_separated_groups) do local form = colon_separated_group[1] if form == "*" or form == "**" or form == "*l" or form == "*!" or form == "*'" then error("*, **, *l, *! and *' not allowed inside of indicator specs: " .. angle_bracket_spec) end if form == "" then form = "+" end local termobj = { term = form } local footnotes = fetch_footnotes(colon_separated_group) if footnotes then for _, footnote in ipairs(footnotes) do m_headword_utilities.add_footnote_to_termobj(termobj, footnote) end end insert(specs, termobj) end return specs end local s_specs = fetch_specs(comma_separated_groups[1]) local ing_specs = fetch_specs(comma_separated_groups[2]) local ed_specs = fetch_specs(comma_separated_groups[3]) local en_specs = fetch_specs(comma_separated_groups[4]) return { forms = {}, s_specs = s_specs, ing_specs = ing_specs, ed_specs = ed_specs, en_specs = en_specs, } end local parse_props = { parse_indicator_spec = parse_indicator_spec, } local alternant_multiword_spec = iut.parse_inflected_text(arg11, parse_props) -- (2) Check for user-specified brackets; remove any links from the lemma, but remember the original -- form so we can use it below in the 'lemma_linked' form. -- Check to see if there are brackets in the pre-text or post-text. If so, use the linked lemma (with the -- verb autolinked unless noautolinkverb is given). Otherwise, use the default headword algorithm. local function check_bracket(val) if val:find("%[%[") then alternant_multiword_spec.saw_bracket = true end end for _, alternant_or_word_spec in ipairs(alternant_multiword_spec.alternant_or_word_specs) do check_bracket(alternant_or_word_spec.before_text) if alternant_or_word_spec.alternants then for _, multiword_spec in ipairs(alternant_or_word_spec.alternants) do for _, word_spec in ipairs(multiword_spec.word_specs) do check_bracket(word_spec.before_text) end check_bracket(multiword_spec.post_text) end end end check_bracket(alternant_multiword_spec.post_text) iut.map_word_specs(alternant_multiword_spec, function(base) if base.lemma == "" then base.lemma = pagename end base.orig_lemma = base.lemma base.lemma = remove_links(base.lemma) if args.noautolinkverb or base.orig_lemma:find("%[%[") then base.linked_lemma = base.orig_lemma else base.linked_lemma = "[[" .. base.orig_lemma .. "]]" end end) -- (3) Conjugate the verbs according to the indicator specs parsed above. local all_verb_slots = { lemma = "infinitive", lemma_linked = "infinitive", s_form = "3|s|pres", ing_form = "pres|ptcp", ed_form = "past", en_form = "past|ptcp", } local function conjugate_verb(base) local function process_specs(slot, specs, canon_func, default_values, default_already_formobj) local function insert_termobj_into_slot(termobj) local formobj = m_headword_utilities.convert_termobj_to_formobj(termobj) -- If the form is -, don't insert any forms, which will result in there being no overall forms -- (in fact it will be nil). We check for that down below and substitute a single "-" as the -- form, which in turn gets turned into special labels like "no present participle". if formobj.form == "-" then if formobj.footnotes then error("Unable to preserve footnotes specified on missing form '-': FIXME: " .. dump(formobj.footnotes)) end else iut.insert_form(base.forms, slot, formobj) end end local function canonicalize_and_insert(arg) local canon_arg = canon_func(arg) if type(canon_arg) == "string" then arg.term = canon_arg insert_termobj_into_slot(arg) else for _, canon in ipairs(canon_arg) do m_headword_utilities.combine_termobj_qualifiers_labels(canon, arg) insert_termobj_into_slot(canon) end end end for _, arg in ipairs(specs) do if arg.term == "+" then if default_values then -- will be nil if past tense specified as - and no past ptc given for _, val in ipairs(default_values) do val = shallowCopy(val) if default_already_formobj then local argformobj = m_headword_utilities.convert_termobj_to_formobj(arg) val.footnotes = iut.combine_footnotes(val.footnotes, argformobj.footnotes) iut.insert_form(base.forms, slot, val) else m_headword_utilities.combine_termobj_qualifiers_labels(val, arg) canonicalize_and_insert(val) end end end else canonicalize_and_insert(arg) end end end set_lemma_and_default_forms(base.lemma) local all_part_default_specs = {} local function process_and_canonicalize_s_form(arg) local form = arg.term if form == "+" then error("Internal error: '+' should have been converted to '^' by now") end if form == "*" or form == "**" or form == "*l" or form == "*!" or form == "*'" then error(("Internal error: '%s' should have already thrown an error"):format(form)) end if form == "^" or form == "++" or form == "+l" or form == "+!" or form == "+'" then insert(all_part_default_specs, shallowCopy(arg)) end return canonicalize_s_form(form) end process_specs("s_form", base.s_specs, process_and_canonicalize_s_form, {{term = "^"}}) if not all_part_default_specs[1] then all_part_default_specs[1] = {term = "^"} end process_specs("ing_form", base.ing_specs, function(arg) return canonicalize_ing_form(arg.term) end, all_part_default_specs) process_specs("ed_form", base.ed_specs, function(arg) return canonicalize_ed_form(arg.term) end, all_part_default_specs) process_specs("en_form", base.en_specs, function(arg) return canonicalize_en_form(arg.term) end, base.forms.ed_form, "default already formobj") iut.insert_form(base.forms, "lemma", {form = base.lemma}) -- Add linked version of lemma for use in head=. We write this in a general fashion in case -- there are multiple lemma forms (which isn't possible currently at this level, although it's -- possible overall using the ((...,...)) notation). iut.insert_forms(base.forms, "lemma_linked", iut.map_forms(base.forms.lemma, function(form) if form == base.lemma and base.linked_lemma:find("%[%[") then return base.linked_lemma else return form end end)) end local inflect_props = { slot_table = all_verb_slots, inflect_word_spec = conjugate_verb, } iut.inflect_multiword_or_alternant_multiword_spec(alternant_multiword_spec, inflect_props) -- (4) Fetch the forms and put the conjugated lemmas in data.heads if not explicitly given. local function fetch_termobjs(slot) local forms = alternant_multiword_spec.forms[slot] -- See above. This should only occur if the user explicitly used - for a spec. if not forms or not forms[1] then return {{term = "-"}} end local termobjs = {} for _, formobj in ipairs(forms) do insert(termobjs, m_headword_utilities.convert_formobj_to_termobj(formobj)) end return termobjs end pres_3sgs = fetch_termobjs("s_form") pres_ptcs = fetch_termobjs("ing_form") pasts = fetch_termobjs("ed_form") past_ptcs = fetch_termobjs("en_form") -- Use the "linked" form of the lemma as the head if no head= explicitly given and the user specified -- brackets in one of the lemmas. Otherwise we use the default headword-linking algorithm. if not data.user_specified_heads[1] and alternant_multiword_spec.saw_bracket then data.heads = {} for _, lemma_obj in ipairs(alternant_multiword_spec.forms.lemma_linked) do insert(data.heads, m_headword_utilities.convert_formobj_to_termobj(lemma_obj)) end end else -------------------------- SEPARATE-PARAM FORMAT -------------------------- set_lemma_and_default_forms(pagename) par1s = par1s or parse_inflection(args, {1, "pres_3sg"}) pres_3sgs = {} pres_ptcs = {} pasts = {} past_ptcs = {} if not par1s[1] then par1s = {{term = "+"}} end if not par2s[1] then par2s = {{term = "+"}} end if not par3s[1] then par3s = {{term = "+"}} end if not par4s[1] then par4s = {{term = "+"}} end local function process_argument(args, dest, canon_func, default_values, default_already_canonicalized) local function canonicalize_and_insert(arg) local canon_arg = canon_func(arg) if type(canon_arg) == "string" then arg.term = canon_arg m_headword_utilities.insert_termobj_combining_duplicates(dest, arg) else for _, canon in ipairs(canon_arg) do m_headword_utilities.combine_termobj_qualifiers_labels(canon, arg) m_headword_utilities.insert_termobj_combining_duplicates(dest, canon) end end end for _, arg in ipairs(args) do if arg.term == "+" then for _, val in ipairs(default_values) do val = shallowCopy(val) m_headword_utilities.combine_termobj_qualifiers_labels(val, arg) if default_already_canonicalized then m_headword_utilities.insert_termobj_combining_duplicates(dest, val) else canonicalize_and_insert(val) end end else canonicalize_and_insert(arg) end end end local all_part_default_specs = {} local function process_and_canonicalize_s_form(arg) local form = arg.term if form == "+" then error("Internal error: '+' should have been converted to '^' by now") end if form == "^" or form == "++" or form == "+l" or form == "+!" or form == "+'" or form == "*" or form == "**" or form == "*l" or form == "*!" or form == "*'" then insert(all_part_default_specs, shallowCopy(arg)) end return canonicalize_s_form(form) end process_argument(par1s, pres_3sgs, process_and_canonicalize_s_form, {{term = "^"}}) if not all_part_default_specs[1] then all_part_default_specs[1] = {term = "^"} end process_argument(par2s, pres_ptcs, function(arg) return canonicalize_ing_form(arg.term) end, all_part_default_specs) process_argument(par3s, pasts, function(arg) return canonicalize_ed_form(arg.term) end, all_part_default_specs) process_argument(par4s, past_ptcs, function(arg) return canonicalize_en_form(arg.term) end, pasts, "default already canonicalized") end ------------------------------------------- INSERT INFLECTIONS ------------------------------------------ insert_inflection(data, pres_3sgs, "third-person singular simple present", "s-verb-form") insert_inflection(data, pres_ptcs, "present participle", "ing-form") if deepEquals(pasts, past_ptcs) then insert_inflection(data, pasts, "simple past and past participle", "ed-form", "no simple past or past participle") else insert_inflection(data, pasts, "simple past", "spast") insert_inflection(data, past_ptcs, "past participle", "past|part") end if pagename:find(" ") then -- Check for placeholder "it" local words = split(pagename, " ") for _, word in ipairs(words) do if word == "it" or word == "its" or word == "it's" then insert(data.categories, langname .. ' terms with placeholder "it"') break end end -- Check for phrasal verbs local phrasal_adverbs = list_to_set{ -- NOTE: This should only contain common phrasal adverbs, not random words like [[low]], -- [[adrift]], etc. "aback", "about", "above", "across", "after", "against", "ahead", "along", "apart", "around", "as", "aside", "at", "away", "back", "before", "behind", "below", "between", "beyond", "by", "down", "for", "forth", "from", "in", "into", "of", "off", "on", "onto", "out", "over", "past", "round", "through", "to", "together", "towards", "under", "up", "upon", "with", "without", } local allowed_non_adverb_words = list_to_set{ "it", "one", "oneself", "someone", } local base = pagename local seen_adverbs = {} -- Only consider a verb to be phrasal if it consists of a single base verb followed exclusively by either -- adverbs from `phrasal_adverbs` or placeholder words from `allowed_non_adverb_words`, where at -- least one following word is from `phrasal_adverbs` (hence [[can it]] is not a phrasal verb). while true do local prev, word = base:match("^(.+) (.-)$") if not prev then break end if phrasal_adverbs[word] then insert(seen_adverbs, word) elseif allowed_non_adverb_words[word] then -- do nothing else break end base = prev end if not base:find(" ") and seen_adverbs[1] then insert(data.categories, langname .. " phrasal verbs") for i = #seen_adverbs, 1, -1 do insert(data.categories, langname .. ' phrasal verbs formed with "' .. seen_adverbs[i] .. '"') end end end end, } ----------------------------------------------------------------------------------------- -- Suffix forms -- ----------------------------------------------------------------------------------------- pos_functions["suffix forms"] = { params = { [1] = {required = true, list = true, disallow_holes = true}, }, func = function(args, data, is_suffix) local suffix_type = {} for _, typ in ipairs(args[1]) do insert(suffix_type, typ .. "-forming suffix") end insert(data.inflections, {label = "non-lemma form of " .. m_table.serialCommaJoin(suffix_type, {conj = "or"})}) end, } return export ceeoihcznsqch6eb3fm46g3ha2kkyh2 89192 89191 2026-09-02T18:50:57Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Module:en-headword]] 89191 Scribunto text/plain local export = {} local pos_functions = {} --[==[ Author from 2020 on: mostly Benwing2, with significant contributions from Theknightwho. Based on a prior version by Rua (by now mostly rewritten), with contributions from Erutuon and others (see history for full attribution). ]==] local force_cat = false -- for testing; if true, categories appear in non-mainspace pages local require = require local require_when_needed = require("Module:require when needed") local en_utilities_module = "Module:en-utilities" local headword_utilities_module = "Module:headword utilities" local headword_module = "Module:headword" local inflection_utilities_module = "Module:inflection utilities" local parse_utilities_module = "Module:parse utilities" local JSON_module = "Module:JSON" local labels_module = "Module:labels" local links_module = "Module:links" local parameters_module = "Module:parameters" local string_utilities_module = "Module:string utilities" local table_module = "Module:table" local utilities_module = "Module:utilities" local yesno_module = "Module:yesno" local iut = require_when_needed(inflection_utilities_module) local put = require_when_needed(parse_utilities_module) local m_headword_utilities = require_when_needed(headword_utilities_module) local add_links_to_multiword_term = require_when_needed(headword_utilities_module, "add_links_to_multiword_term") local add_suffix = require_when_needed(en_utilities_module, "add_suffix") local apply_link_modifiers = require_when_needed(headword_utilities_module, "apply_link_modifiers") local concat = table.concat local deepEquals = require_when_needed(table_module, "deepEquals") local dump = mw.dumpObject local format_categories = require_when_needed(utilities_module, "format_categories") local full_headword = require_when_needed(headword_module, "full_headword") local get_label_info = require_when_needed(labels_module, "get_label_info") local get_link_page = require_when_needed(links_module, "get_link_page") local glossary_link = require_when_needed(headword_utilities_module, "glossary_link") local insert = table.insert local insertIfNot = require_when_needed(table_module, "insertIfNot") local ipairs = ipairs local is_regular_plural = require_when_needed(en_utilities_module, "is_regular_plural") local list_to_set = require_when_needed(table_module, "listToSet") local pairs = pairs local process_params = require_when_needed(parameters_module, "process") local remove = table.remove local remove_links = require_when_needed(links_module, "remove_links") local replacement_escape = require_when_needed(string_utilities_module, "replacement_escape") local shallowCopy = require_when_needed(table_module, "shallowCopy") local singularize = require_when_needed(en_utilities_module, "singularize") local split = require_when_needed(string_utilities_module, "split") local toJSON = require_when_needed(JSON_module, "toJSON") local toNFD = mw.ustring.toNFD local type = type local ulen = require_when_needed(string_utilities_module, "len") local ulower = require_when_needed(string_utilities_module, "lower") local umatch = require_when_needed(string_utilities_module, "match") local u = require_when_needed(string_utilities_module, "char") local ugsub = require_when_needed(string_utilities_module, "gsub") local lang = require("Module:languages").getByCode("en") local langname = lang:getCanonicalName() local list_param = {list = true, disallow_holes = true} local list_allow_holes = {list = true, allow_holes = true} local boolean_param = {type = "boolean"} local function ine(val) if val == "" then return nil else return val end end local function track(page) require("Module:debug/track")("en-headword/" .. page) return true end ------------------------------------------- UTILITY FUNCTIONS ------------------------------------------ -- Parse and return an inflection not requiring additional processing. The raw arguments come from `args[field]`, which -- is parsed for inline modifiers. local function parse_inflection(args, field, is_head) local argfield = field if type(argfield) == "table" then argfield = argfield[1] end return m_headword_utilities.parse_term_list_with_modifiers { paramname = field, forms = args[argfield], splitchar = ",", is_head = is_head, } end -- Insert the parsed inflections in `terms` (as parsed by `parse_inflection`) into `data.inflections`, with label -- `label` and optional accelerator spec `accel`. local function insert_inflection(data, terms, label, accel, no_label) for _, termobj in ipairs(terms) do m_headword_utilities.remove_termobj_field_modifiers(termobj) end m_headword_utilities.insert_inflection { headdata = data, terms = terms, label = label, no_label = no_label, accel = accel and {form = accel} or nil, } end -- Insert a fixed label `label` into the inflections for `data`. If `originating_term` is supplied, copy the qualifiers, -- labels and references from it into the fixed label. local function insert_fixed_inflection(data, label, originating_term) m_headword_utilities.insert_fixed_inflection { headdata = data, originating_term = originating_term, label = label, } end -- Parse and insert an inflection not requiring additional processing into `data.inflections`. The raw arguments come -- from `args[field]`, which is parsed for inline modifiers. `label` is the label that the inflections are given; -- `accel` is the accelerator form, or nil. local function parse_and_insert_inflection(data, args, field, label, accel) m_headword_utilities.parse_and_insert_inflection { headdata = data, forms = args[field], paramname = field, splitchar = ",", label = label, accel = accel and {form = accel} or nil, } end -- These functions are used directly in the <> format as well as in the utility functions #2 below. local function compute_double_last_cons_stem(term) local last_cons = term:match("([bcdfghjklmnpqrstvwxyzBCDFGHJKLMNPQRSTVWXYZ])$") if not last_cons then error("Verb stem '" .. term .. "' must end in a consonant to use ++") end return term .. last_cons end local function compute_plusplus_s_form(term, default_s_form) if term:find("[szx]$") then -- regas -> regasses, derez -> derezzes return compute_double_last_cons_stem(term) .. "es" else return default_s_form end end -- The main entry point. -- This is the only function that can be invoked from a template. function export.show(frame) local iparams = { [1] = true, } local iargs = require("Module:parameters").process(frame.args, iparams) local parargs = frame:getParent().args local poscat = iargs[1] local pos_in_1 = not poscat if pos_in_1 then poscat = ine(parargs[1]) or mw.title.getCurrentTitle().fullText == "Template:en-head" and "interjection" or error("Part of speech must be specified in 1=") poscat = require(headword_module).canonicalize_pos(poscat) end local indexing_poscat = pos_in_1 and "head" or poscat local params = { ["head"] = list_param, ["id"] = true, ["json"] = boolean_param, ["sort"] = true, ["splithyph"] = boolean_param, ["nosplithyph"] = boolean_param, ["hyphspace"] = boolean_param, ["nolink"] = boolean_param, ["nolinkhead"] = {type = "boolean_param", alias_of = "nolink"}, ["suffix"] = boolean_param, ["nosuffix"] = boolean_param, ["nomultiwordcat"] = boolean_param, ["abbr"] = list_param, ["the"] = true, ["def"] = {alias_of = "the"}, ["pagename"] = true, -- for testing } if pos_in_1 then params[1] = {required = true} -- required but ignored as already processed above end local pos_data = pos_functions[indexing_poscat] local pos_func if pos_data then local pos_params = pos_data.params if pos_params then for key, val in pairs(pos_params) do params[key] = val end end pos_func = pos_data.func end local args = process_params(parargs, params) -- Account for unsupported titles, e.g. 'C|N>K' instead of 'Unsupported titles/C through N to K'. local pagename = args.pagename or mw.loadData("Module:headword/data").pagename local user_specified_heads = parse_inflection(args, "head", "is_head") local heads = user_specified_heads local autohead if args.nolink or not pagename:find("[ '%-]") then autohead = pagename else local en_no_split_apostrophe_words = list_to_set { "one's", "someone's", "he's", "she's", "it's", } local en_include_hyphen_prefixes = list_to_set { -- We don't include things that are also words even though they are often (perhaps mostly) prefixes, e.g. -- "be", "counter", "cross", "extra", "half", "mid", "over", "pan", "under". "acro", "acousto", "Afro", "agro", "anarcho", "angio", "Anglo", "ante", "anti", "arch", "auto", "bi", "bio", "cis", "co", "cryo", "crypto", "de", "demi", "eco", "electro", "Euro", "ex", "Greco", "hemi", "hydro", "hyper", "hypo", "infra", "Indo", "inter", "intra", "Judeo", "macro", "meta", "micro", "mini", "multi", "neo", "neuro", "non", "para", "peri", "post", "pre", "pro", "proto", "pseudo", "re", "semi", "sub", "super", "trans", "un", "vice", } local function is_english(term) local title = mw.title.new(term) if title and title.exists then local content = title:getContent() if content and content:find("==English==\n") then return true end end return false end local function en_split_hyphen_when_space(word) if not word:find("-", nil, true) then return nil end if args.hyphspace then return "[[" .. word:gsub("%-+", " ") .. "|" .. word .. "]]" end if args.nosplithyph then return "[[" .. word .. "]]" end if not args.splithyph then local space_word = word:gsub("%-+", " ") if is_english(space_word) then return "[[" .. space_word .. "|" .. word .. "]]" end if is_english(word) then return "[[" .. word .. "]]" end end return nil end local function en_split_apostrophe(word) local base = word:match("^(.*)'s$") if base then return "[[" .. base .. "]][[-'s|'s]]" end -- Only treat final apostrophe as possessive if preceded by something that looks like a plural ending in /z/. -- In particular we don't want to do it for words like [[truckin']]. base = word:match("^(.*[sxz])'$") if base then if base:find("s$") then local sg = singularize(base) if is_english(sg) then return "[[" .. sg .. "|" .. base .. "]][[-'|']]" end end return "[[" .. base .. "]][[-'|']]" end return "[[" .. word .. "]]" end autohead = add_links_to_multiword_term(pagename, { split_hyphen_when_space = en_split_hyphen_when_space, split_apostrophe = en_split_apostrophe, no_split_apostrophe_words = en_no_split_apostrophe_words, include_hyphen_prefixes = en_include_hyphen_prefixes, }) end if not heads[1] then heads = {{term = autohead}} else for _, headobj in ipairs(heads) do local head = headobj.term if head:find("^~") then head = apply_link_modifiers(autohead, head:sub(2), lang) headobj.term = head elseif head:find("^[!?]$") then -- If explicit head= just consists of ! or ?, add it to the end of the default head. headobj.term = autohead .. head end if head == autohead then track("redundant-head") end end end -- handle the=/def= if args.the == "~" then local newheads = {} for _, headobj in ipairs(heads) do local barehead = shallowCopy(headobj) insert(newheads, barehead) headobj.term = "the " .. headobj.term insert(newheads, headobj) end heads = newheads elseif args.the then local the = require(yesno_module)(args.the) if the then for _, headobj in ipairs(heads) do headobj.term = "the " .. headobj.term end end end local data = { lang = lang, pos_category = poscat, categories = {}, heads = heads, user_specified_heads = user_specified_heads, -- We use our own splitting algorithm so the redundant head cat will be inaccurate. no_redundant_head_cat = true, inflections = {}, nomultiwordcat = args.nomultiwordcat, sort_key = args.sort, pagename = pagename, id = args.id, force_cat_output = force_cat, } local function inscat(cat) insert(data.categories, langname .. " " .. cat) end local is_suffix = false if args.suffix or not args.nosuffix and pagename:find("^%-") and not pagename:find("^%-%-") and poscat ~= "suffix forms" then is_suffix = true data.pos_category = "suffixes" local singular_poscat = singularize(poscat) inscat(singular_poscat .. "-forming suffixes") insert(data.inflections, {label = singular_poscat .. "-forming suffix"}) end if pos_func then pos_func(args, data, is_suffix) end local extra_categories = {} if pagename:find("[Qq]") then -- Check for q not followed by u. We want to exclude things like [[13q deletion syndrome]] and [[BFOQ]] that -- don't have a lowercase letter on either side, as well as things like [[& seq.]] and [[acq.]] that are -- abbreviations for words containing a following u. -- -- Approximate range of combining diacritics; we want to remove them so the checks below for -- a lowercase letter next to the q aren't tripped up by diacritics on the letter. local u300 = u(0x0300) local u36F = u(0x036F) local pagename_no_diacritics = ugsub(toNFD(pagename), "[" .. u300 .. "-" .. u36F .. "]", "") if pagename_no_diacritics:find("[Qq][a-tv-z]") or pagename_no_diacritics:find("[a-z]q[^u.]") or pagename_no_diacritics:find("[a-z]q$") then inscat("words containing Q not followed by U") end end -- toNFD performs decomposition, so letters that decompose to an ASCII -- vowel and a diacritic, such as é, are counted as vowels and do not do not -- need to be included in the pattern. if not umatch(ulower(toNFD(pagename)), "[aeiouyæœøəªºαεηιουω]") then inscat("words spelled without vowels") end if pagename:find("yre$") then inscat('words ending in "-yre"') end if not pagename:find(" ") and ulen(pagename) >= 25 then insert(extra_categories, "Long " .. langname .. " words") end if pagename:find("^[^aeiou ]*a[^aeiou ]*e[^aeiou ]*i[^aeiou ]*o[^aeiou ]*u[^aeiou ]*$") then inscat("words that use all vowels in alphabetical order") end parse_and_insert_inflection(data, args, "abbr", "abbreviation") if args.json then return toJSON(data) end return full_headword(data) .. (extra_categories[1] and format_categories(extra_categories, lang, args.sort) or "") end local function make_default_comparative(word) if word == "good" or word == "well" then return {"better"} elseif word == "bad" or word == "badly" then return {"worse"} elseif word == "far" then return {"further", "farther"} else return {add_suffix(word, "r")} end end local function make_default_superlative(word) if word == "good" or word == "well" then return {"best"} elseif word == "bad" or word == "badly" then return {"worst"} elseif word == "far" then return {"furthest", "farthest"} else return {add_suffix(word, "st.superlative")} end end -- This function does the common work between adjectives and adverbs. local function process_comparative_args(data, args, plpos) local pagename = data.pagename local comps = parse_inflection(args, 1) local sups = parse_inflection(args, "sup") local outcomps, outsups if args.componly then if comps[1] then error("Can't specify comparatives of comparative-only " .. plpos) end insert(data.inflections, {label = glossary_link("comparative") .. " form only"}) insert(data.categories, langname .. " comparative-only " .. plpos) -- Set to empty list so we don't get any comparatives output, but process superlatives if specified. outcomps = {} if not sups[1] then -- Set to empty list so we don't get any superlatives output unless explicitly given. outsups = {} end elseif args.suponly then if comps[1] or sups[1] then error("Can't specify comparatives or superlatives of or superlative-only " .. plpos) end insert(data.inflections, {label = glossary_link("superlative") .. " form only"}) insert(data.categories, langname .. " superlative-only " .. plpos) return end -- If the first parameter is ?, then don't show anything, just return. if comps[1] and comps[1].term == "?" then if comps[2] then error("Can't specify additional comparatives along with '?'") end if sups[1] then error("Can't specify superlatives along with '?' for the comparative") end return end if comps[1] and comps[1].term == "-" then local hyphencomp = remove(comps, 1) -- Remove the "-" but retain for qualifiers, labels, references -- Not (generally) comparable; may occasionally have a comparative if comps[1] then insert_fixed_inflection(data, "not generally <<comparable>>", hyphencomp) elseif not sups[1] then insert_fixed_inflection(data, "not <<comparable>>", hyphencomp) insert(data.categories, langname .. " uncomparable " .. plpos) return else -- No comparative, but a superlative. insert_inflection() will correctly generate 'no comparative' if we -- pass in "-" as the value. outcomps = {hyphencomp} end elseif not comps[1] then comps = {{term = "more"}} end if not outcomps then -- not if we set `outcomps` to "-" above or processed a comparative-only term outcomps = {} -- Go over each parameter given and create a comparative and superlative form. for _, compobj in ipairs(comps) do local comp = compobj.term if comp == "-" then error("Comparative of '-' only allowed as first comparative") end if comp == "+" then comp = "+more" elseif comp == "more" and pagename ~= "many" and pagename ~= "much" then comp = "+more" elseif comp == "further" and pagename ~= "far" then comp = "+further" elseif comp == "better" and pagename ~= "good" and pagename ~= "well" then comp = "+better" elseif comp:find("~") then comp = comp:gsub("~", replacement_escape(pagename)) end compobj.origterm = comp if comp == "+more" then comp = "more [[" .. pagename .. "]]" elseif comp == "+further" then comp = {"further [[" .. pagename .. "]]", "farther [[" .. pagename .. "]]"} elseif comp == "+better" then comp = "better [[" .. pagename .. "]]" elseif comp == "er" then -- Add -er. comp = add_suffix(pagename, "r") elseif comp == "ier" then if pagename:sub(-1) ~= "y" then error("Can't specify 'ier' comparative unless the term ends with 'y': " .. pagename) end comp = pagename:gsub("e?y$", "ier") elseif comp:find("^%+") then local special = m_headword_utilities.get_special_indicator(comp, "noerror") if special then comp = m_headword_utilities.handle_multiword(pagename, special, make_default_comparative) end end if type(comp) == "table" and not comp[2] then comp = comp[1] end if type(comp) == "table" then for i = 1, #comp - 1 do local outobj = shallowCopy(compobj) outobj.term = comp[i] insert(outcomps, outobj) end compobj.term = comp[#comp] insert(outcomps, compobj) else compobj.term = comp insert(outcomps, compobj) end end end if sups[1] and sups[1].term == "-" then if sups[2] then error("Can't specify '-' as superlative followed by further values") end -- No superlative. insert_inflection() will correctly generate 'no superlative' if we pass in "-" as the value. outsups = sups else if not sups[1] then sups = {{term = "+"}} end end -- `outsups` will be set if we set `outsups` to "-" above or processed a comparative-only term without superlatives. if not outsups then outsups = {} local function process_sup(sup, special, supobj, compobj) if special then sup = m_headword_utilities.handle_multiword(pagename, special, make_default_superlative) elseif sup == "-" or sup == "+" then error(("Internal error: Superlative value of '%s' should have been handled earlier"):format(sup)) elseif sup == "+most" then sup = "most [[" .. pagename .. "]]" elseif sup == "+furthest" then sup = {"furthest [[" .. pagename .. "]]", "farthest [[" .. pagename .. "]]"} elseif sup == "+best" then sup = "best [[" .. pagename .. "]]" elseif sup == "est" then -- Add -est. sup = add_suffix(pagename, "st.superlative") elseif sup == "iest" then if pagename:sub(-1) ~= "y" then error("Can't specify 'iest' superlative unless the term ends with 'y': " .. pagename) end sup = pagename:gsub("e?y$", "iest") end if type(sup) == "table" and not sup[2] then sup = sup[1] end if compobj then supobj = shallowCopy(supobj) supobj = m_headword_utilities.combine_termobj_qualifiers_labels(supobj, compobj) end if type(sup) == "table" then for i = 1, #sup - 1 do local outobj = shallowCopy(supobj) outobj.term = sup[i] insert(outsups, outobj) end supobj.term = sup[#sup] insert(outsups, supobj) else supobj.term = sup insert(outsups, supobj) end end for _, supobj in ipairs(sups) do local sup = supobj.term if sup == "-" then error("Superlative of '-' only allowed as first superlative") end if sup == "+" then if not comps[1] then error("Superlative of '+' can't be specified when there are no comparatives") end for _, compobj in ipairs(comps) do local comp = compobj.origterm local special if comp == "+more" then sup = "+most" elseif comp == "+further" then sup = "+furthest" elseif comp == "+better" then sup = "+best" elseif comp == "er" then sup = "est" elseif comp == "ier" then sup = "iest" else if comp:find("^%+") then special = m_headword_utilities.get_special_indicator(comp, "noerror") end if not special then -- If the full comparative was given, then derive the superlative by replacing -er with -- -est. if comp:sub(-2) == "er" then sup = comp:sub(1, -3) .. "est" else error(("The superlative cannot be derived automatically from comparative '%s' because it doesn't end in -er"):format(comp)) end end end process_sup(sup, special, supobj, compobj) end else local special = m_headword_utilities.get_special_indicator(sup, "noerror") -- Do some work here rather than in process_sup() so we don't end up double-processing a term with a '~' -- in it or a term that happens to be 'most' or similar after substitution of ~ in the comparative. if not special then if sup == "most" and pagename ~= "many" and pagename ~= "much" then sup = "+most" elseif sup == "furthest" and pagename ~= "far" then sup = "+furthest" elseif sup == "best" and pagename ~= "good" and pagename ~= "well" then sup = "+best" elseif sup:find("~") then sup = sup:gsub("~", replacement_escape(pagename)) end end process_sup(sup, special, supobj) end end end insert_inflection(data, outcomps, "<<comparative>>", "comparative") insert_inflection(data, outsups, "<<superlative>>", "superlative") end pos_functions["adjectives"] = { params = { [1] = list_param, ["comp_qual"] = {list = "comp\1_qual", allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the comparative value", }, ["sup"] = list_param, ["sup_qual"] = {list = "sup\1_qual", allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the superlative value", }, ["componly"] = boolean_param, ["suponly"] = boolean_param, }, func = function(args, data) -- Process the comparatives and superlatives. process_comparative_args(data, args, "adjectives") end, } pos_functions["adverbs"] = { params = { [1] = list_param, ["comp_qual"] = {list = "comp\1_qual", allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the comparative value", }, ["sup"] = list_param, ["sup_qual"] = {list = "sup\1_qual", allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the superlative value", }, ["componly"] = boolean_param, ["suponly"] = boolean_param, }, func = function(args, data) -- Process the comparatives and superlatives. process_comparative_args(data, args, "adverbs") end, } local function escape(str) return (str:gsub("\\([:#])", "\\\\%1") :gsub("[:#]", "\\%0")) end local function canonicalize_plural(pl, pagename, pos) if pl == "+" then return escape(add_suffix(pagename, "s.plural", pos)) elseif pl == "++" then return escape(compute_plusplus_s_form(pagename, add_suffix(pagename, "s.plural", pos))) elseif pl == "*" then return escape(pagename) elseif pl == "ies" then if pagename:sub(-1) == "y" then return escape(pagename:gsub("e?y$", pl)) end error("Can't specify 'ies' plural unless the term ends with 'y'.") elseif pl == "s" or pl == "es" or pl == "'s" then return escape(pagename .. pl) end end local function do_nouns(args, data, pos) local pagename = data.pagename pos = pos or "noun" local plurals = parse_inflection(args, 1) local function insert_plurale_tantum_inflections(is_plural_only, originating_label) if args.sg[1] then insert_fixed_inflection(data, "normally plural", originating_label) parse_and_insert_inflection(data, args, "sg", "singular") elseif is_plural_only then insert_fixed_inflection(data, "plural only", originating_label) end if args.attr[1] then parse_and_insert_inflection(data, args, "attr", "attributive") end end local function first_pl_term() return plurals[1] and plurals[1].term or nil end if first_pl_term() == "p" then -- plurale tantum if plurals[2] then error("With plurale tantum noun, can't specify more than one plural") end data.genders = {"p"} -- this should auto-insert the correct 'pluralia tantum' category insert_plurale_tantum_inflections("plural only", plurals[1]) return end local function inscat(cat) insert(data.categories, langname .. " " .. cat) end local need_default_plural = pos == "noun" if first_pl_term() == "sp" then -- construed as singular or plural sp = remove(plurals, 1) -- Remove the "sp" but retain it for its qualifiers, labels, references inscat("nouns construed as singular or plural") data.genders = {"s", "p"} -- this should auto-insert the correct 'pluralia tantum' category insert_plurale_tantum_inflections(nil, sp) need_default_plural = false elseif first_pl_term() == "-" then -- Uncountable noun; may occasionally have a plural local hyphpl = remove(plurals, 1) -- Remove the "-" but retain for qualifiers, labels, references inscat("uncountable nouns") -- If plural forms were given explicitly, then show "usually" if plurals[1] then insert_fixed_inflection(data, "usually <<uncountable>>", hyphpl) else insert_fixed_inflection(data, "<<uncountable>>", hyphpl) end need_default_plural = false elseif first_pl_term() == "#" then -- Usually countable (e.g., "grilled cheese") local hashpl = remove(plurals, 1) -- Remove the "#" but retain for qualifiers, labels, references insert_fixed_inflection(data, "usually <<countable>>", hashpl) inscat("uncountable nouns") inscat("countable nouns") -- If no plural was given, add a default one now if not plurals[1] then plurals[1] = {term = escape(add_suffix(pagename, "s.plural", pos))} end elseif first_pl_term() == "~" then -- Mixed countable/uncountable noun, always has a plural local tildepl = remove(plurals, 1) -- Remove the "~" but retain for qualifiers, labels, references insert_fixed_inflection(data, "<<countable>> and <<uncountable>>", tildepl) inscat("uncountable nouns") inscat("countable nouns") -- If no plural was given, add a default one now if not plurals[1] then plurals[1] = {term = escape(add_suffix(pagename, "s.plural", pos))} end end -- Plural is unknown if first_pl_term() == "?" then local questionpl = remove(plurals, 1) -- Remove the "?" but retain for qualifiers, labels, references -- Not desired; see [[Wiktionary:Tea_room/2021/August#"Plural unknown or uncertain"]] -- insert_fixed_inflection(data, "plural unknown or uncertain", questionpl) inscat("nouns with unknown or uncertain plurals") if plurals[1] then error("Can't specify explicit plurals along with '?' for unknown/uncertain plural") end return end -- Plural is not attested if first_pl_term() == "!" then local exclampl = remove(plurals, 1) -- Remove the "!" but retain for qualifiers, labels, references insert_fixed_inflection(data, "plural not attested", exclampl) inscat("nouns with unattested plurals") if plurals[1] then error("Can't specify explicit plurals along with '!' for unattested plural") end return end -- If no plural was given, maybe add a default one, otherwise (when "-" was given or proper noun) return. if not plurals[1] then if not need_default_plural then inscat("uncountable nouns") return end plurals[1] = {term = escape(add_suffix(pagename, "s.plural", pos))} end -- There are plural forms to show, so show them. inscat("countable nouns") local irregular, indeclinable for i, pl in ipairs(plurals) do local canon_pl = canonicalize_plural(pl.term, pagename, pos) if canon_pl then pl.term = canon_pl end local pl_term = get_link_page(pl.term, lang) if not (pagename:find(" ") or is_regular_plural(pl_term, pagename)) then irregular = true if pl_term == pagename then indeclinable = true end end end if irregular then inscat("nouns with irregular plurals") end if indeclinable then inscat("indeclinable nouns") end insert_inflection(data, plurals, "plural", "p") end -- Return the parameters to be used for nouns and proper nouns. Currently the same. local noun_params = { [1] = list_param, ["pl\1qual"] = {list = true, allow_holes = true, replaced_by = false, instead = "use <l:...> or <q:...> inline modifier on the plural", }, -- The following four only used for pluralia tantum (1=p) ["sg"] = list_param, ["attr"] = list_param, } pos_functions["nouns"] = { params = noun_params, func = do_nouns, } pos_functions["proper nouns"] = { params = noun_params, func = function(args, data) return do_nouns(args, data, "proper noun") end, } local function base_default_verb_forms(verb) return escape(add_suffix(verb, "s.verb")), escape(add_suffix(verb, "ing")), escape(add_suffix(verb, "d")) end local function default_verb_forms(verb) local full_s_form, full_ing_form, full_ed_form = base_default_verb_forms(verb) if verb:find(" ") then local first, rest = verb:match("^(.-)( .*)$") local first_s_form, first_ing_form, first_ed_form = base_default_verb_forms(first) return full_s_form, full_ing_form, full_ed_form, first_s_form .. rest, first_ing_form .. rest, first_ed_form .. rest, first, rest else return full_s_form, full_ing_form, full_ed_form, nil, nil, nil, nil, nil end end local function compute_double_last_cons_stem_of_split_verb(verb, ending) local first, rest = verb:match("^(.-)( .*)$") if not first then error("Verb '" .. verb .. "' must have a space in it to use **") end local last_cons = first:match("([bcdfghjklmnpqrstvwxyzBCDFGHJKLMNPQRSTVWXYZ])$") if not last_cons then error("First word '" .. first .. "' must end in a consonant to use **") end return first .. last_cons .. ending .. rest end local function check_non_nil_star_form(form, pagename) if form == nil then error("Verb '" .. pagename .. "' must have a space in it to use *, **, *l, *! or *'") end return form end local function sub_tilde(form, pagename) if not form then return nil end if form:find("~") then form = form:gsub("~", replacement_escape(pagename)) end return form end local deprecated_qual_replaced_by_inline_modifier = { list = true, allow_holes = true, replaced_by = false, instead = "use an inline modifier <q:...> or <l:...> on the value" } pos_functions["verbs"] = { params = { [1] = {list = "pres_3sg", disallow_holes = true}, ["pres_3sg\1_qual"] = deprecated_qual_replaced_by_inline_modifier, [2] = {list = "pres_ptc", disallow_holes = true}, ["pres_ptc\1_qual"] = deprecated_qual_replaced_by_inline_modifier, [3] = {list = "past", disallow_holes = true}, ["past\1_qual"] = deprecated_qual_replaced_by_inline_modifier, [4] = {list = "past_ptc", allow_holes = true}, ["past_ptc\1_qual"] = deprecated_qual_replaced_by_inline_modifier, ["noautolinkverb"] = boolean_param, ["angle_bracket"] = boolean_param, }, func = function(args, data) -- Get parameters local par1s local par2s = parse_inflection(args, {2, "pres_ptc"}) local par3s = parse_inflection(args, {3, "past"}) local par4s = parse_inflection(args, {4, "past_ptc"}) local pres_3sgs, pres_ptcs, pasts, past_ptcs local pagename = data.pagename ------------------------------------------- UTILITY FUNCTIONS #2 ------------------------------------------ -- These functions are used in both in the separate-parameter format and in the override params such as past_ptc2=. local full_default_s, full_default_ing, full_default_ed, split_default_s, split_default_ing, split_default_ed local lemma local function set_lemma_and_default_forms(the_lemma) lemma = the_lemma full_default_s, full_default_ing, full_default_ed, split_default_s, split_default_ing, split_default_ed, lemma_first, lemma_rest = default_verb_forms(the_lemma) end local function canonicalize_s_form(form) if form == "+" then error("Internal error: Should not see '+' here") elseif form == "^" then return full_default_s elseif form == "*" then return check_non_nil_star_form(split_default_s, lemma) elseif form == "++" then return compute_plusplus_s_form(lemma, full_default_s) elseif form == "**" then if lemma:find("^[^ ]*[szx] ") then return compute_double_last_cons_stem_of_split_verb(lemma, "es") else return check_non_nil_star_form(split_default_s, lemma) end elseif form == "+!" then return lemma .. "s" elseif form == "*!" then return check_non_nil_star_form(lemma_first) .. "s" .. lemma_rest elseif form == "+'" then return lemma .. "'s" elseif form == "*'" then return check_non_nil_star_form(lemma_first) .. "'s" .. lemma_rest elseif form == "+l" then if lemma:find("[szx]$") then return {{term = full_default_s, l = {"US"}}, {term = compute_plusplus_s_form(lemma, full_default_s), l = {"UK"}}} else return compute_plusplus_s_form(lemma, full_default_s) end elseif form == "*l" then if lemma:find("^[^ ]*[szx] ") then return {{term = check_non_nil_star_form(split_default_s, lemma), l = {"US"}}, {term = compute_double_last_cons_stem_of_split_verb(lemma, "es"), l = {"UK"}}} else return check_non_nil_star_form(split_default_s, lemma) end else return sub_tilde(form, lemma) end end local function canonicalize_ing_form(form) if form == "+" then error("Internal error: Should not see '+' here") elseif form == "^" then return full_default_ing elseif form == "*" then return check_non_nil_star_form(split_default_ing, lemma) elseif form == "++" then return compute_double_last_cons_stem(lemma) .. "ing" elseif form == "**" then return compute_double_last_cons_stem_of_split_verb(lemma, "ing") elseif form == "+!" then return lemma .. "ing" elseif form == "*!" then return check_non_nil_star_form(lemma_first) .. "ing" .. lemma_rest elseif form == "+'" then return lemma .. "'ing" elseif form == "*'" then return check_non_nil_star_form(lemma_first) .. "'ing" .. lemma_rest elseif form == "+l" then return {{term = full_default_ing, l = {"US"}}, {term = compute_double_last_cons_stem(lemma) .. "ing", l = {"UK"}}} elseif form == "*l" then return {{term = check_non_nil_star_form(split_default_ing, lemma), l = {"US"}}, {term = compute_double_last_cons_stem_of_split_verb(lemma, "ing"), l = {"UK"}}} else return sub_tilde(form, lemma) end end local function canonicalize_ed_form(form) if form == "+" then error("Internal error: Should not see '+' here") elseif form == "^" then return full_default_ed elseif form == "*" then return check_non_nil_star_form(split_default_ed, lemma) elseif form == "++" then return compute_double_last_cons_stem(lemma) .. "ed" elseif form == "+!" then return lemma .. "ed" elseif form == "*!" then return check_non_nil_star_form(lemma_first) .. "ed" .. lemma_rest elseif form == "+'" then return {{term = lemma .. "'d"}, {term = lemma .. "'ed"}} elseif form == "*'" then return {{term = check_non_nil_star_form(lemma_first) .. "'d" .. lemma_rest}, {term = check_non_nil_star_form(lemma_first) .. "'ed" .. lemma_rest}} elseif form == "**" then return compute_double_last_cons_stem_of_split_verb(lemma, "ed") elseif form == "+l" then return {{term = full_default_ed, l = {"US"}}, {term = compute_double_last_cons_stem(lemma) .. "ed", l = {"UK"}}} elseif form == "*l" then return {{term = check_non_nil_star_form(split_default_ed, lemma), l = {"US"}}, {term = compute_double_last_cons_stem_of_split_verb(lemma, "ed"), l = {"UK"}}} else return sub_tilde(form, lemma) end end -- FIXME: options should be "+", "*", "++", "**", "+n", "*n", "++n" and "**n", but not "n" local function canonicalize_en_form(form) if form == "n" then track("n4") return add_suffix(lemma, "n") end return canonicalize_ed_form(form) end --------------------------------- MAIN PARSING/CONJUGATING CODE -------------------------------- local is_angle_bracket = args.angle_bracket if is_angle_bracket then if par2s[1] or par3s[1] or par4s[1] then error("Can't specify explicit values for 2=, 3= or 4= along with the angle-bracket format") end elseif is_angle_bracket == nil and not par2s[1] and not par3s[1] and not par4s[1] and not args[1][2] and args[1][1] and args[1][1]:find("<") then if put.term_contains_top_level_html(args[1][1]) then -- Often, term_contains_top_level_html() returns true on the angle-bracket format, which would -- make the pcall() below succeed but leave the angle brackets as-is. Check for this and only do the -- pcall() if term_contains_top_level_html() returns false. is_angle_bracket = true else -- If it's ambiguous whether it's an angle-bracket format or separate params with an inline modifier, -- try to parse as the latter. If an error occurs, treat as the former. local ok ok, par1s = pcall(parse_inflection, args, {1, "pres_3sg"}) if not ok then par1s = nil is_angle_bracket = true end end end if is_angle_bracket then -------------------------- ANGLE-BRACKET FORMAT -------------------------- -- (0) Expand multiword term with angle brackets just on the first word. local arg11 = args[1][1] if arg11:find("^<.*>$") and pagename:find(" ") then local first, rest = pagename:match("^(.-)( .*)$") arg11 = first .. arg11 .. rest end -- (1) Parse the indicator specs inside of angle brackets. local function parse_indicator_spec(angle_bracket_spec) local inside = angle_bracket_spec:match("^<(.*)>$") assert(inside) local segments = put.parse_balanced_segment_run(inside, "[", "]") local comma_separated_groups = put.split_alternating_runs(segments, ",") if #comma_separated_groups > 4 then error("Too many comma-separated parts in indicator spec, expected at most 4: " .. angle_bracket_spec) end local function fetch_footnotes(separated_group) local footnotes for j = 2, #separated_group - 1, 2 do if separated_group[j + 1] ~= "" then error("Extraneous text after bracketed footnotes: '" .. concat(separated_group) .. "'") end if not footnotes then footnotes = {} end insert(footnotes, separated_group[j]) end return footnotes end local function fetch_specs(comma_separated_group) if not comma_separated_group then return {{term = "+"}} end local specs = {} local colon_separated_groups = put.split_alternating_runs(comma_separated_group, ":") for _, colon_separated_group in ipairs(colon_separated_groups) do local form = colon_separated_group[1] if form == "*" or form == "**" or form == "*l" or form == "*!" or form == "*'" then error("*, **, *l, *! and *' not allowed inside of indicator specs: " .. angle_bracket_spec) end if form == "" then form = "+" end local termobj = { term = form } local footnotes = fetch_footnotes(colon_separated_group) if footnotes then for _, footnote in ipairs(footnotes) do m_headword_utilities.add_footnote_to_termobj(termobj, footnote) end end insert(specs, termobj) end return specs end local s_specs = fetch_specs(comma_separated_groups[1]) local ing_specs = fetch_specs(comma_separated_groups[2]) local ed_specs = fetch_specs(comma_separated_groups[3]) local en_specs = fetch_specs(comma_separated_groups[4]) return { forms = {}, s_specs = s_specs, ing_specs = ing_specs, ed_specs = ed_specs, en_specs = en_specs, } end local parse_props = { parse_indicator_spec = parse_indicator_spec, } local alternant_multiword_spec = iut.parse_inflected_text(arg11, parse_props) -- (2) Check for user-specified brackets; remove any links from the lemma, but remember the original -- form so we can use it below in the 'lemma_linked' form. -- Check to see if there are brackets in the pre-text or post-text. If so, use the linked lemma (with the -- verb autolinked unless noautolinkverb is given). Otherwise, use the default headword algorithm. local function check_bracket(val) if val:find("%[%[") then alternant_multiword_spec.saw_bracket = true end end for _, alternant_or_word_spec in ipairs(alternant_multiword_spec.alternant_or_word_specs) do check_bracket(alternant_or_word_spec.before_text) if alternant_or_word_spec.alternants then for _, multiword_spec in ipairs(alternant_or_word_spec.alternants) do for _, word_spec in ipairs(multiword_spec.word_specs) do check_bracket(word_spec.before_text) end check_bracket(multiword_spec.post_text) end end end check_bracket(alternant_multiword_spec.post_text) iut.map_word_specs(alternant_multiword_spec, function(base) if base.lemma == "" then base.lemma = pagename end base.orig_lemma = base.lemma base.lemma = remove_links(base.lemma) if args.noautolinkverb or base.orig_lemma:find("%[%[") then base.linked_lemma = base.orig_lemma else base.linked_lemma = "[[" .. base.orig_lemma .. "]]" end end) -- (3) Conjugate the verbs according to the indicator specs parsed above. local all_verb_slots = { lemma = "infinitive", lemma_linked = "infinitive", s_form = "3|s|pres", ing_form = "pres|ptcp", ed_form = "past", en_form = "past|ptcp", } local function conjugate_verb(base) local function process_specs(slot, specs, canon_func, default_values, default_already_formobj) local function insert_termobj_into_slot(termobj) local formobj = m_headword_utilities.convert_termobj_to_formobj(termobj) -- If the form is -, don't insert any forms, which will result in there being no overall forms -- (in fact it will be nil). We check for that down below and substitute a single "-" as the -- form, which in turn gets turned into special labels like "no present participle". if formobj.form == "-" then if formobj.footnotes then error("Unable to preserve footnotes specified on missing form '-': FIXME: " .. dump(formobj.footnotes)) end else iut.insert_form(base.forms, slot, formobj) end end local function canonicalize_and_insert(arg) local canon_arg = canon_func(arg) if type(canon_arg) == "string" then arg.term = canon_arg insert_termobj_into_slot(arg) else for _, canon in ipairs(canon_arg) do m_headword_utilities.combine_termobj_qualifiers_labels(canon, arg) insert_termobj_into_slot(canon) end end end for _, arg in ipairs(specs) do if arg.term == "+" then if default_values then -- will be nil if past tense specified as - and no past ptc given for _, val in ipairs(default_values) do val = shallowCopy(val) if default_already_formobj then local argformobj = m_headword_utilities.convert_termobj_to_formobj(arg) val.footnotes = iut.combine_footnotes(val.footnotes, argformobj.footnotes) iut.insert_form(base.forms, slot, val) else m_headword_utilities.combine_termobj_qualifiers_labels(val, arg) canonicalize_and_insert(val) end end end else canonicalize_and_insert(arg) end end end set_lemma_and_default_forms(base.lemma) local all_part_default_specs = {} local function process_and_canonicalize_s_form(arg) local form = arg.term if form == "+" then error("Internal error: '+' should have been converted to '^' by now") end if form == "*" or form == "**" or form == "*l" or form == "*!" or form == "*'" then error(("Internal error: '%s' should have already thrown an error"):format(form)) end if form == "^" or form == "++" or form == "+l" or form == "+!" or form == "+'" then insert(all_part_default_specs, shallowCopy(arg)) end return canonicalize_s_form(form) end process_specs("s_form", base.s_specs, process_and_canonicalize_s_form, {{term = "^"}}) if not all_part_default_specs[1] then all_part_default_specs[1] = {term = "^"} end process_specs("ing_form", base.ing_specs, function(arg) return canonicalize_ing_form(arg.term) end, all_part_default_specs) process_specs("ed_form", base.ed_specs, function(arg) return canonicalize_ed_form(arg.term) end, all_part_default_specs) process_specs("en_form", base.en_specs, function(arg) return canonicalize_en_form(arg.term) end, base.forms.ed_form, "default already formobj") iut.insert_form(base.forms, "lemma", {form = base.lemma}) -- Add linked version of lemma for use in head=. We write this in a general fashion in case -- there are multiple lemma forms (which isn't possible currently at this level, although it's -- possible overall using the ((...,...)) notation). iut.insert_forms(base.forms, "lemma_linked", iut.map_forms(base.forms.lemma, function(form) if form == base.lemma and base.linked_lemma:find("%[%[") then return base.linked_lemma else return form end end)) end local inflect_props = { slot_table = all_verb_slots, inflect_word_spec = conjugate_verb, } iut.inflect_multiword_or_alternant_multiword_spec(alternant_multiword_spec, inflect_props) -- (4) Fetch the forms and put the conjugated lemmas in data.heads if not explicitly given. local function fetch_termobjs(slot) local forms = alternant_multiword_spec.forms[slot] -- See above. This should only occur if the user explicitly used - for a spec. if not forms or not forms[1] then return {{term = "-"}} end local termobjs = {} for _, formobj in ipairs(forms) do insert(termobjs, m_headword_utilities.convert_formobj_to_termobj(formobj)) end return termobjs end pres_3sgs = fetch_termobjs("s_form") pres_ptcs = fetch_termobjs("ing_form") pasts = fetch_termobjs("ed_form") past_ptcs = fetch_termobjs("en_form") -- Use the "linked" form of the lemma as the head if no head= explicitly given and the user specified -- brackets in one of the lemmas. Otherwise we use the default headword-linking algorithm. if not data.user_specified_heads[1] and alternant_multiword_spec.saw_bracket then data.heads = {} for _, lemma_obj in ipairs(alternant_multiword_spec.forms.lemma_linked) do insert(data.heads, m_headword_utilities.convert_formobj_to_termobj(lemma_obj)) end end else -------------------------- SEPARATE-PARAM FORMAT -------------------------- set_lemma_and_default_forms(pagename) par1s = par1s or parse_inflection(args, {1, "pres_3sg"}) pres_3sgs = {} pres_ptcs = {} pasts = {} past_ptcs = {} if not par1s[1] then par1s = {{term = "+"}} end if not par2s[1] then par2s = {{term = "+"}} end if not par3s[1] then par3s = {{term = "+"}} end if not par4s[1] then par4s = {{term = "+"}} end local function process_argument(args, dest, canon_func, default_values, default_already_canonicalized) local function canonicalize_and_insert(arg) local canon_arg = canon_func(arg) if type(canon_arg) == "string" then arg.term = canon_arg m_headword_utilities.insert_termobj_combining_duplicates(dest, arg) else for _, canon in ipairs(canon_arg) do m_headword_utilities.combine_termobj_qualifiers_labels(canon, arg) m_headword_utilities.insert_termobj_combining_duplicates(dest, canon) end end end for _, arg in ipairs(args) do if arg.term == "+" then for _, val in ipairs(default_values) do val = shallowCopy(val) m_headword_utilities.combine_termobj_qualifiers_labels(val, arg) if default_already_canonicalized then m_headword_utilities.insert_termobj_combining_duplicates(dest, val) else canonicalize_and_insert(val) end end else canonicalize_and_insert(arg) end end end local all_part_default_specs = {} local function process_and_canonicalize_s_form(arg) local form = arg.term if form == "+" then error("Internal error: '+' should have been converted to '^' by now") end if form == "^" or form == "++" or form == "+l" or form == "+!" or form == "+'" or form == "*" or form == "**" or form == "*l" or form == "*!" or form == "*'" then insert(all_part_default_specs, shallowCopy(arg)) end return canonicalize_s_form(form) end process_argument(par1s, pres_3sgs, process_and_canonicalize_s_form, {{term = "^"}}) if not all_part_default_specs[1] then all_part_default_specs[1] = {term = "^"} end process_argument(par2s, pres_ptcs, function(arg) return canonicalize_ing_form(arg.term) end, all_part_default_specs) process_argument(par3s, pasts, function(arg) return canonicalize_ed_form(arg.term) end, all_part_default_specs) process_argument(par4s, past_ptcs, function(arg) return canonicalize_en_form(arg.term) end, pasts, "default already canonicalized") end ------------------------------------------- INSERT INFLECTIONS ------------------------------------------ insert_inflection(data, pres_3sgs, "third-person singular simple present", "s-verb-form") insert_inflection(data, pres_ptcs, "present participle", "ing-form") if deepEquals(pasts, past_ptcs) then insert_inflection(data, pasts, "simple past and past participle", "ed-form", "no simple past or past participle") else insert_inflection(data, pasts, "simple past", "spast") insert_inflection(data, past_ptcs, "past participle", "past|part") end if pagename:find(" ") then -- Check for placeholder "it" local words = split(pagename, " ") for _, word in ipairs(words) do if word == "it" or word == "its" or word == "it's" then insert(data.categories, langname .. ' terms with placeholder "it"') break end end -- Check for phrasal verbs local phrasal_adverbs = list_to_set{ -- NOTE: This should only contain common phrasal adverbs, not random words like [[low]], -- [[adrift]], etc. "aback", "about", "above", "across", "after", "against", "ahead", "along", "apart", "around", "as", "aside", "at", "away", "back", "before", "behind", "below", "between", "beyond", "by", "down", "for", "forth", "from", "in", "into", "of", "off", "on", "onto", "out", "over", "past", "round", "through", "to", "together", "towards", "under", "up", "upon", "with", "without", } local allowed_non_adverb_words = list_to_set{ "it", "one", "oneself", "someone", } local base = pagename local seen_adverbs = {} -- Only consider a verb to be phrasal if it consists of a single base verb followed exclusively by either -- adverbs from `phrasal_adverbs` or placeholder words from `allowed_non_adverb_words`, where at -- least one following word is from `phrasal_adverbs` (hence [[can it]] is not a phrasal verb). while true do local prev, word = base:match("^(.+) (.-)$") if not prev then break end if phrasal_adverbs[word] then insert(seen_adverbs, word) elseif allowed_non_adverb_words[word] then -- do nothing else break end base = prev end if not base:find(" ") and seen_adverbs[1] then insert(data.categories, langname .. " phrasal verbs") for i = #seen_adverbs, 1, -1 do insert(data.categories, langname .. ' phrasal verbs formed with "' .. seen_adverbs[i] .. '"') end end end end, } ----------------------------------------------------------------------------------------- -- Suffix forms -- ----------------------------------------------------------------------------------------- pos_functions["suffix forms"] = { params = { [1] = {required = true, list = true, disallow_holes = true}, }, func = function(args, data, is_suffix) local suffix_type = {} for _, typ in ipairs(args[1]) do insert(suffix_type, typ .. "-forming suffix") end insert(data.inflections, {label = "non-lemma form of " .. m_table.serialCommaJoin(suffix_type, {conj = "or"})}) end, } return export ceeoihcznsqch6eb3fm46g3ha2kkyh2 10 18087 89193 2016-10-25T17:52:03Z en>TheDaveRoss 0 Changed protection level for "[[Template:en-adv]]": Adjusting the protection level from 'sysop' to 'template editor'. ([Edit=Allow only template editors and administrators] (indefinite) [Move=Allow only template editors and administrators] (indefinite)) 89193 wikitext text/x-wiki {{#invoke:en-headword|show|adverbs}}<!-- --><noinclude>{{documentation}}</noinclude> k03tupdv25re1unn1137r1dcfanl9gk 89194 89193 2026-09-02T18:52:16Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Template:en-adv]] 89193 wikitext text/x-wiki {{#invoke:en-headword|show|adverbs}}<!-- --><noinclude>{{documentation}}</noinclude> k03tupdv25re1unn1137r1dcfanl9gk 10 18088 89195 2024-05-31T04:55:05Z en>Benwing2 0 use [[Module:audio]] with greatly expanded functionality; default caption to "Audio" and put a colon after the caption if not already 89195 wikitext text/x-wiki {{ {{#if:{{{lang|}}}|check deprecated lang param usage|no deprecated lang param usage}}|lang={{{lang|}}}|1=<!-- -->{{#invoke:audio|show}}<!-- -->}}<noinclude>{{documentation}}</noinclude> m5e7v618pe7zo812h4lo5dzfuuhjzjh 89196 89195 2026-09-02T18:52:55Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Template:audio]] 89195 wikitext text/x-wiki {{ {{#if:{{{lang|}}}|check deprecated lang param usage|no deprecated lang param usage}}|lang={{{lang|}}}|1=<!-- -->{{#invoke:audio|show}}<!-- -->}}<noinclude>{{documentation}}</noinclude> m5e7v618pe7zo812h4lo5dzfuuhjzjh 89209 89196 2026-09-02T18:54:55Z Altronic 4137 89209 wikitext text/x-wiki {{#invoke:audio|show}}<noinclude>{{documentation}}</noinclude> 4i9yj48nuiay70elt2q8tcv9amzj5ah 10 18089 89197 2023-01-12T03:35:55Z en>Theknightwho 0 89197 wikitext text/x-wiki {{ {{#if:{{{lang|}}}|check deprecated lang param usage|no deprecated lang param usage}}|lang={{{lang|}}}|<!-- -->{{#invoke:rhymes|show}}<!-- -->}}<!-- --><noinclude>{{documentation}}</noinclude> 3ygdtqzxe484ssk7zq07a5phzlirxjt 89198 89197 2026-09-02T18:53:00Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Template:rhymes]] 89197 wikitext text/x-wiki {{ {{#if:{{{lang|}}}|check deprecated lang param usage|no deprecated lang param usage}}|lang={{{lang|}}}|<!-- -->{{#invoke:rhymes|show}}<!-- -->}}<!-- --><noinclude>{{documentation}}</noinclude> 3ygdtqzxe484ssk7zq07a5phzlirxjt 828 18090 89199 2026-08-11T21:06:33Z en>Benwing2 0 support `parse_lang_prefix` in parse_term_with_modifiers() and parse_term_list_with_modifiers() 89199 Scribunto text/plain local export = {} local fun_is_callable_module = "Module:fun/isCallable" local languages_module = "Module:languages" local links_module = "Module:links" local parse_utilities_module = "Module:parse utilities" local string_pattern_escape_module = "Module:string/patternEscape" local string_replacement_escape_module = "Module:string/replacementEscape" local string_utilities_module = "Module:string utilities" local table_module = "Module:table" local dump = mw.dumpObject local unpack = unpack or table.unpack -- Lua 5.2 compatibility local insert = table.insert local concat = table.concat local remove = table.remove local sort = table.sort local function deepEquals(...) deepEquals = require(table_module).deepEquals return deepEquals(...) end local function escape_wikicode(...) escape_wikicode = require(parse_utilities_module).escape_wikicode return escape_wikicode(...) end local function extend(...) extend = require(table_module).extend return extend(...) end local function get_lang(...) get_lang = require(languages_module).getByCode return get_lang(...) end local function insert_if_not(...) insert_if_not = require(table_module).insertIfNot return insert_if_not(...) end local function is_callable(...) is_callable = require(fun_is_callable_module) return is_callable(...) end local function parse_inline_modifiers(...) parse_inline_modifiers = require(parse_utilities_module).parse_inline_modifiers return parse_inline_modifiers(...) end local function pattern_escape(...) pattern_escape = require(string_pattern_escape_module) return pattern_escape(...) end local function replacement_escape(...) replacement_escape = require(string_replacement_escape_module) return replacement_escape(...) 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 term_contains_top_level_html(...) term_contains_top_level_html = require(parse_utilities_module).term_contains_top_level_html return term_contains_top_level_html(...) end local function ugsub(...) ugsub = require(string_utilities_module).gsub return ugsub(...) end local function umatch(...) umatch = require(string_utilities_module).match return umatch(...) end local param_mods = { id = {}, -- disabled when `is_head = true` q = {type = "qualifier"}, qq = {type = "qualifier"}, l = {type = "labels"}, ll = {type = "labels"}, -- [[Module:headword]] expects part references in `.refs`. ref = {item_dest = "refs", type = "references", store = "insert-flattened"}, } local optional_param_mods = { g = {item_dest = "genders", type = "genders"}, alt = {}, lang = {type = "language"}, sc = {type = "script"}, t = {item_dest = "gloss"}, gloss = {}, pos = {}, lit = {}, tr = {}, ts = {}, face = {}, nolinkinfl = {type = "boolean"}, } local optional_headword_param_mods = { sc = {type = "script"}, tr = {}, ts = {}, } --[==[ Parse a single inflection or headword form or list of such forms. In either case, inline modifiers may be attached. `data` is an object with the following fields: * `val`: The raw value to parse. Required. * `paramname`: The name of the parameter from which the value was taken; used in error messages. Required. * `is_head`: We are parsing a headword parameter (a value which goes into the `heads` field of `data`). This changes the allowed modifiers, disabling the `id` modifier and only allowing a subset of optional modifiers. * `frob`: An optional function of one value to apply to the form after inline modifiers have been removed (i.e. to apply to the `.term` field of the returned object). * `include_mods`: List of extra inline modifiers to include, besides the default ones (see below). Each list item is either a string specifying a recognized extra inline modifier (see `optional_param_mods` in the code), or a two-item list of modifier name and modifier spec, where the spec should follow the syntax for modifier specs in `parse_inline_modifiers` in [[Module:parse utilities]]. * `exclude_mods`: List of default inline modifiers to not include. * `splitchar`: If specified, the value in `val` can be a list of forms to parse, separated by the value of `splitchar` (which is a Lua pattern, as in `parse_inline_modifiers` in [[Module:parse utilities]]). Most commonly, `splitchar` is a single comma and the values are comma-separated (in this case, splitting will not happen if a space follows the comma). * `parse_lang_prefix`: If specified, allow a language prefix to precede a form, and if found, store into the `.lang` field of the returned object. * `preserve_splitchar`, `delimiter_key`, `escape_fun`, `unescape_fun`, `pre_normalize_modifiers`: As in `parse_inline_modifiers` in [[Module:parse utilities]]. Returns an object suitable for storing as one element of one of the lists in `headdata.inflections`, where `headdata` is the structure passed to [[Module:headword]]. If `splitchar` is specified, howeve, the return value is a list of such objects. The following default inline modifiers are currently recognized: * `q`: Left qualifier. * `qq`: Right qualifier. * `l`: Comma-separated list of left labels. No space should follow the comma. * `ll`: Comma-separated list of right labels. No space should follow the comma. * `ref`: Reference or references. See {{tl|IPA}} for the syntax. * `id`: Sense ID, in case there are multiple senses. See {{tl|l}}. The following are the recognized additional inline modifiers: * `g`: Comma-separated list of genders. * `alt`: Display text. * `lang`: Language code of language of the form, if different from the language of the headword. * `sc`: Script code of script of the form. Almost never needed. * `t`: Gloss for the form. * `gloss`: Gloss for the form (alias for `t`). * `pos`: Part of speech of the form. * `lit`: Literal meaning of the form. * `tr`: Manual transliteration of the form. * `ts`: Transcription of the form, for languages where the transliteration differs markedly from the pronunciation. * `face`: Face to display the form in, e.g. {"hypothetical"} for a hypothetical form (unlinkable and displayed in italics). * `nolinkinfl`: Make the form unlinkable. ]==] function export.parse_term_with_modifiers(data) local paramname, val, frob = data.paramname, data.val, data.frob local function generate_obj(term, parse_err) if frob then term = frob(term, parse_err) end if data.parse_lang_prefix and term:find(":") then return require(parse_utilities_module).generate_obj_maybe_parsing_lang_prefix { term = term, paramname = paramname, parse_lang_prefix = true, parse_err = parse_err, } else return {term = term} end end -- Check for inline modifier, e.g. מרים<tr:Miryem>. But exclude top-level HTML entry with <span ...>, -- <sup> or similar in it. if (val:find("<", nil, true) or data.splitchar) and not term_contains_top_level_html(val) and -- don't parse inline modifiers if is_head and the value begins with a ~ (link modifier syntax) (not data.is_head or not val:find("^~")) then local param_mods = param_mods if data.is_head then param_mods = shallow_copy(param_mods) param_mods.id = nil end if data.include_mods or data.exclude_mods then if not data.is_head then -- already copied when data.is_head param_mods = shallow_copy(param_mods) end if data.include_mods then local optional_mods = data.is_head and optional_headword_param_mods or optional_param_mods for _, mod in ipairs(data.include_mods) do if type(mod) == "table" then if #mod ~= 2 then error(("Internal error: Modifier spec %s in `include_mods` should be of length 2"):format( dump(mod))) end local modkey, modvalue = unpack(mod) param_mods[modkey] = modvalue elseif not optional_mods[mod] then error(("Internal error: Unrecognized modifier spec %s in `include_mods`"):format( dump(mod))) else param_mods[mod] = optional_mods[mod] end end end if data.exclude_mods then for _, mod in ipairs(data.exclude_mods) do if not param_mods[mod] then error(("Internal error: Modifier spec %s in `exclude_mods` not found among existing modifiers" ):format(dump(mod))) else param_mods[mod] = nil end end end end return parse_inline_modifiers(val, { paramname = paramname, param_mods = param_mods, generate_obj = generate_obj, splitchar = data.splitchar, preserve_splitchar = data.preserve_splitchar, delimiter_key = data.delimiter_key, escape_fun = data.escape_fun, unescape_fun = data.unescape_fun, pre_normalize_modifiers = data.pre_normalize_modifiers, }) else local retval = generate_obj(val) if data.splitchar then retval = {retval} end return retval end end --[==[ Parse a list of inflection forms that may have inline modifiers attached. `data` is an object with the following fields: * `forms`: The list of raw values to parse. Required. * `paramname`: The name of the first parameter from which the value was taken; used in error messages. If this is a two-element list, the first element is the first parameter and the second element is the prefix of the remaining parameters. Parameter names that are numbers are handled correctly, as are those with \1 in it marking where the parameter index goes. Required. * `qualifiers`: If specified, a possibly gappy list of left qualifiers to add to the parsed terms (for compatibility purposes). * `splitchar`: As in `parse_term_with_modifiers()`. The resulting per-term lists will be flattened. * `frob`, `include_mods`, `exclude_mods`, `is_head`, `preserve_splitchar`, `parse_lang_prefix`, `delimiter_key`, `escape_fun`, `unescape_fun`, `pre_normalize_modifiers`: As in `parse_term_with_modifiers()`. Returns a list of objects, suitable for storing as one of the lists in `headdata.inflections` (once a label is added), where `headdata` is the structure passed to [[Module:headword]]. ]==] function export.parse_term_list_with_modifiers(data) local paramname, forms = data.paramname, data.forms local qualifiers = data.qualifiers local first, restpref if type(paramname) == "table" then first = paramname[1] restpref = paramname[2] else first = paramname restpref = paramname end local terms = {} data = shallow_copy(data) for i, val in ipairs(forms) do data.paramname = i == 1 and first or type(restpref) == "number" and restpref + i - 1 or restpref:find("\1", nil, true) and restpref:gsub("\1", tostring(i)) or restpref .. i data.val = val local parsed = export.parse_term_with_modifiers(data) if qualifiers and qualifiers[i] then if data.splitchar then for _, term in ipairs(parsed) do term.q = {qualifiers[i]} end else parsed.q = {qualifiers[i]} end end if data.splitchar then extend(terms, parsed) else terms[i] = parsed end end return terms end --[==[ Construct a link to [[Appendix:Glossary]] for `entry`. If `text` is specified, it is the display text; otherwise, `entry` is used. ]==] function export.glossary_link(entry, text) text = text or entry return "[[Appendix:Glossary#" .. entry .. "|" .. text .. "]]" end function export.replace_glossary_links_in_label(label) if label:find("<<", nil, true) then label = label:gsub("<<(.-)|(.-)>>", export.glossary_link):gsub("<<(.-)>>", export.glossary_link) end return label end function export.insert_fixed_inflection(data) local headdata, origterm, label = data.headdata, data.originating_term, data.label local inflobj = data.inflobj or headdata inflobj.inflections = inflobj.inflections or {} if not origterm then insert(inflobj.inflections, { label = export.replace_glossary_links_in_label(label) }) else if origterm.id then error(("It doesn't make sense to pass in an ID '%s' for label '%s' in conjunction with a term value '%s'" ):format(origterm.id, label, origterm.term)) end -- Preserve qualifiers, labels, references origterm.term = nil origterm.label = export.replace_glossary_links_in_label(label) insert(inflobj.inflections, origterm) end end --[==[ Insert previously-parsed terms into an `inflections` field. The `inflections` field will be initialized if needed. `data` is an object with the following fields: * `headdata`: The headword structure passed to [[Module:headword]]. Required. * `inflobj`: The object whose `inflections` field the terms are inserted into. Defaults to `headdata`. Only needs to be set for nested inflections, which are specified for an inflection object rather than the headword data structure as a whole. * `terms`: The list of parsed terms. If {nil} or omitted, nothing happens unless `request` is set. * `label`: The label that the inflections are given; any parts of the label surrounded in <<...>> are linked to the glossary. (If the contents of <<...> contain a | in them, they are a two-part link.) Required. * `no_label`: If the term is {"-"} and there are no other terms, insert a fixed label with this value. Defaults to {"no "} plus the label. * `usually_no_label`: If the term is {"-"} and there are other terms, insert a fixed label with this value. Defaults to {"usually no "} plus the label. * `accel`: If specified, a full accelerator object to add to the inflections. * `request`: If specified and no terms are given, insert a label with a request for inflections to be given. * `enable_auto_translit`: If specified and terms are given, display automatic transliteration of the terms. The return value indicates whether the inflection exists and how many terms are in it. It is an object with the following fields: * `exists`: {"yes"} if one or more terms were specified; {"no"} if the value was given as {"-"}; {"usually no"} if the first value was given as {"-"} but additional terms were supplied; otherwise {nil}, indicating that the status is unspecified. * `numterms`: Number of terms in the inflection. Will be 0 unless `exists` has the value {"yes"} or {"usually no"}. * `request`: True if no terms were specified but a term request was inserted into the inflection (because `data.request` was specified). Otherwise {nil}. ]==] function export.insert_inflection(data) local headdata, terms, label = data.headdata, data.terms, data.label local inflobj = data.inflobj or headdata local retval = {} if terms and terms[1] then if terms[1].term == "-" then if terms[2] then export.insert_fixed_inflection { headdata = headdata, inflobj = inflobj, originating_term = terms[1], label = data.usually_no_label or "usually no " .. label, } remove(terms, 1) retval.numterms = #terms retval.exists = "usually no" else export.insert_fixed_inflection { headdata = headdata, inflobj = inflobj, originating_term = terms[1], label = data.no_label or "no " .. label, } retval.numterms = 0 retval.exists = "no" return retval end else retval.numterms = #terms retval.exists = "yes" end if data.check_missing then error("check_missing support removed; use checkredlinks=true in [[Module:headword]]") end terms.label = export.replace_glossary_links_in_label(label) if data.accel then terms.accel = data.accel end terms.enable_auto_translit = data.enable_auto_translit inflobj.inflections = inflobj.inflections or {} insert(inflobj.inflections, terms) elseif data.request then inflobj.inflections = inflobj.inflections or {} insert(inflobj.inflections, { label = export.replace_glossary_links_in_label(label), request = true, }) retval.numterms = 0 -- retval.exists == nil retval.request = true else retval.numterms = 0 -- retval.exists == nil end return retval end --[==[ Parse raw arguments from `forms` for inline modifiers, and insert the resulting terms (which should not require significant additional processing) into `headdata.inflections`. `data` is an object with the following fields: * `forms`: The list of raw values to parse. If {nil} or omitted, nothing happens. * `headdata`: The headword structure passed to [[Module:headword]]. Required. * `paramname`: As in `parse_term_list_with_modifiers()`. Required. * `label`: As in `insert_inflection()`. Required. * `qualifiers`, `frob`, `include_mods`, `exclude_mods`, `is_head`, `splitchar`, `preserve_splitchar`, `delimiter_key`, `escape_fun`, `unescape_fun`, `pre_normalize_modifiers`: As in `parse_term_list_with_modifiers()`. * `accel`: As in `insert_inflection()`. Return value is as in `insert_inflection()`. ]==] function export.parse_and_insert_inflection(data) local forms = data.forms if forms and forms[1] then data = shallow_copy(data) data.forms = forms data.terms = export.parse_term_list_with_modifiers(data) return export.insert_inflection(data) end return { numterms = 0 } end --[==[ Combine two sets of qualifiers or labels. If either is {nil}, just return the other, and if both are {nil}, return {nil}. ]==] function export.combine_qualifiers_or_labels(quals1, quals2) if not quals1 and not quals2 then return nil end if not quals1 then return quals2 end if not quals2 then return quals1 end local combined = shallow_copy(quals1) for _, note in ipairs(quals2) do insert_if_not(combined, note) end return combined end --[==[ Combine the qualifiers, labels, references and ID's of two term objects. `destobj` is the "destination term object" into which the combined properties are written, and `srcobj` is the "source object" into which the properties are merged. `destobj` is side-effected (but the lists inside of `destobj` are not); if this is undesirable, make sure to shallow-copy `destobj` first. If both objects have values for a given qualifier, label or reference, the values of `destobj` come first. If both objects have a value for `id`, the values must match or an error is thrown; otherwise, the resulting value of `id` comes from whichever one is defined. '''NOTE:''' This may not be the correct behavior when deduplicating a list of term objects. See `insert_termobj_combining_duplicates` for a different approach. ]==] function export.combine_termobj_qualifiers_labels(destobj, srcobj) destobj.q = export.combine_qualifiers_or_labels(destobj.q, srcobj.q) destobj.qq = export.combine_qualifiers_or_labels(destobj.qq, srcobj.qq) destobj.l = export.combine_qualifiers_or_labels(destobj.l, srcobj.l) destobj.ll = export.combine_qualifiers_or_labels(destobj.ll, srcobj.ll) destobj.refs = export.combine_qualifiers_or_labels(destobj.refs, srcobj.refs) if destobj.id and srcobj.id and destobj.id ~= srcobj.id then -- FIXME: We probably want to pass in an error function error(("Can't specify two different ID's %s and %s when combining objects"):format(srcobj.id, destobj.id)) end destobj.id = destobj.id or srcobj.id return destobj end function export.termobj_has_qualifiers_or_labels(obj) return obj.q and obj.q[1] or obj.qq and obj.qq[1] or obj.l and obj.l[1] or obj.ll and obj.ll[1] or obj.refs and obj.refs[1] end local function one_ancillary_property_equal(prop1, prop2) local prop1_is_nil = not prop1 or not prop1[1] local prop2_is_nil = not prop2 or not prop2[1] if prop1_is_nil and prop2_is_nil then return true end if prop1_is_nil or prop2_is_nil then return false end return deepEquals(prop1, prop2) end function export.termobj_ancillary_properties_equal(obj1, obj2) return one_ancillary_property_equal(obj1.q, obj2.q) and one_ancillary_property_equal(obj1.qq, obj2.qq) and one_ancillary_property_equal(obj1.l, obj2.l) and one_ancillary_property_equal(obj1.ll, obj2.ll) and one_ancillary_property_equal(obj1.refs, obj2.refs) and obj1.id == obj2.id end function export.convert_termobj_to_formobj(termobj) local formobj = { form = termobj.term, translit = termobj.tr, } local footnotes local function mods_to_footnote(mod_prefix, mod_vals) if mod_vals and mod_vals[1] then footnotes = footnotes or {} for _, val in ipairs(mod_vals) do insert(footnotes, "[" .. mod_prefix .. ":" .. val .. "]") end end end mods_to_footnote("q", termobj.q) mods_to_footnote("qq", termobj.qq) mods_to_footnote("l", termobj.l) mods_to_footnote("ll", termobj.ll) mods_to_footnote("ref", termobj.refs) mods_to_footnote("id", termobj.id and {termobj.id} or nil) formobj.footnotes = footnotes return formobj end local recognized_multi_mods = { q = "q", qq = "qq", l = "l", ll = "ll", ref = "refs", } local recognized_single_mods = { id = "id", } function export.add_footnote_to_termobj(termobj, footnote) local stripped_footnote = footnote:match("^%[(.*)%]$") if not stripped_footnote then error("Internal error: Footnote should be surrounded by brackets at this stage: " .. footnote) end local prefix, rest = stripped_footnote:match("^([a-z]+):(.+)$") local field, is_multi if prefix then if recognized_multi_mods[prefix] then field = recognized_multi_mods[prefix] is_multi = true elseif recognized_single_mods[prefix] then field = recognized_single_mods[prefix] is_multi = false end end if not field then rest = stripped_footnote field = "l" is_multi = true end if is_multi then if not termobj[field] then termobj[field] = {} end insert(termobj[field], rest) else if termobj[field] and termobj[field] ~= rest then error(("Can't set two values for '%s': '%s' and '%s'"):format(field, termobj[field], rest)) end termobj[field] = rest end end function export.convert_formobj_to_termobj(formobj) local termobj = { term = formobj.form, tr = formobj.translit, } if formobj.footnotes then for _, footnote in ipairs(formobj.footnotes) do export.add_footnote_to_termobj(termobj, footnote) end end return termobj end local function extract_termobj_field_modifiers(fieldval) return fieldval:match("^([*+]?)(.*)$") end function export.remove_termobj_field_modifiers(termobj) local function remove_field_modifiers(field) if termobj[field] and termobj[field][1] then local any_field_modifiers = false for _, val in ipairs(termobj[field]) do local field_mods, _ = extract_termobj_field_modifiers(val) if field_mods ~= "" then any_field_modifiers = true break end end local new_field = {} if any_field_modifiers then for _, val in ipairs(termobj[field]) do local _, field_without_mods = extract_termobj_field_modifiers(val) insert_if_not(new_field, field_without_mods) end termobj[field] = new_field end end end remove_field_modifiers("q") remove_field_modifiers("qq") remove_field_modifiers("l") remove_field_modifiers("ll") remove_field_modifiers("refs") end function export.insert_termobj_combining_duplicates(destobjs, termobj) for _, destobj in ipairs(destobjs) do if destobj.term == termobj.term and destobj.tr == termobj.tr then -- Form already present; maybe combine footnotes. local function combine_field_values(field) if termobj[field] and termobj[field][1] then -- Check to see if there are existing values with *; if so, remove them. if destobj[field] and destobj[field][1] then local any_values_with_asterisk = false for _, val in ipairs(destobj[field]) do local field_mods, _ = extract_termobj_field_modifiers(val) if field_mods:find("%*") then any_values_with_asterisk = true break end end if any_values_with_asterisk then local filtered_values = {} for _, val in ipairs(destobj[field]) do local field_mods, _ = extract_termobj_field_modifiers(val) if not field_mods:find("%*") then insert(filtered_values, val) end end if filtered_values[1] then destobj[field] = filtered_values else destobj[field] = nil end end end local any_footnotes_with_plus = false for _, val in ipairs(termobj[field]) do local field_mods, _ = extract_termobj_field_modifiers(val) if field_mods:find("%+") then any_footnotes_with_plus = true break end end if any_footnotes_with_plus then if not destobj[field] then destobj[field] = {} else destobj[field] = shallow_copy(destobj[field]) end for _, val in ipairs(termobj[field]) do local already_seen = false local field_mods, field_without_mods = extract_termobj_field_modifiers(val) if field_mods:find("%+") then for _, existing_val in ipairs(destobj[field]) do local _, existing_field_without_mods = extract_termobj_field_modifiers(existing_val) if existing_field_without_mods == field_without_mods then already_seen = true break end end if not already_seen then insert(destobj[field], val) end end end end end end combine_field_values("q") combine_field_values("qq") combine_field_values("l") combine_field_values("ll") combine_field_values("refs") if destobj.id and termobj.id and destobj.id ~= termobj.id then -- FIXME: We probably want to pass in an error function error(("Can't specify two different ID's %s and %s when combining objects"):format(termobj.id, destobj.id)) end destobj.id = destobj.id or termobj.id return end end insert(destobjs, termobj) end export.allowed_special_indicators = { ["first"] = true, ["first-second"] = true, ["first-last"] = true, ["second"] = true, ["last"] = true, ["each"] = true, ["+"] = true, -- requests the default behavior with preposition handling } --[==[ Check for special indicators (values such as {"+first"} or {"+first-last"} that are used in a `pl`, `f`, etc. argument and indicate how to inflect a multiword term). If `form` is such an indicator, the return value is `form` minus the initial `+` sign; otherwise, if form begins with a `+` sign, an error is thrown; otherwise the return value is nil. ]==] function export.get_special_indicator(form, noerror) if form:find("^%+") then form = form:gsub("^%+", "") if not export.allowed_special_indicators[form] then if noerror then return nil end local indicators = {} for indic, _ in pairs(export.allowed_special_indicators) do insert(indicators, "+" .. indic) end sort(indicators) error("Special inflection indicator beginning with '+' can only be " .. mw.text.listToText(indicators) .. ": +" .. form) end return form end return nil end local function add_endings(bases, endings) local retval = {} if type(bases) ~= "table" then bases = {bases} end if type(endings) ~= "table" then endings = {endings} end for _, base in ipairs(bases) do for _, ending in ipairs(endings) do insert(retval, base .. ending) end end return retval end --[==[ Inflect a possibly multiword or hyphenated term `form` using the function `inflect`, which is a function of one argument that is called on a single word to inflect and should return either the inflected word or a list of inflected words. `special` indicates how to inflect the multiword term and should be e.g. {"first"} to inflect only the first word, {"first-last"} to inflect the first and last words, {"each"} to inflect each word, etc. See `allowed_special_indicators` above for the possibilities. If `special` is `+`, or is omitted and the term is multiword (i.e. containing a space character), and `prepositions` is supplied, the function checks for multiword or hyphenated terms containing the prepositions in `prepositions`, e.g. Italian [[senso di marcia]] or [[medaglia d'oro]] or Portuguese [[tartaruga-do-mar]]. If such a term is found, only the first word is inflected. Otherwise, the default is {"first-last"}. `prepositions` is a list of Lua patterns matching prepositions. The patterns will automatically have the separator character (space or hyphen) added to the left side but not the right side, so they should contain a space character (which will automatically be converted to the appropriate separator) on the right side unless the preposition is joined on the right side with an apostrophe. Examples of preposition patterns for Italian are {"di "}, {"sull'"} and {"d?all[oae] "} (which matches {"dallo "}, {"dalle "}, {"alla "}, etc.). The return value is always either a list of inflected multiword or hyphenated terms, or nil if `special` is omitted and `form` is not multiword. (If `special` is specified and `form` is not multiword or hyphenated, an error results.) ]==] function export.handle_multiword(form, special, inflect, prepositions, sep) sep = sep or form:find(" ") and " " or "%-" local raw_sep = sep == " " and " " or "-" -- Used to add regex version of separator in the replacement portion of ugsub() or :gsub() local sep_replacement = sep == " " and " " or "%%-" -- Given a Lua pattern, replace space with the appropriate separator. local function hack_re(re) if sep == " " then return re end return (re:gsub(" ", sep_replacement)) end if special == "first" then local first, rest = form:match(hack_re("^(.-)( .*)$")) if not first then error("Special indicator 'first' can only be used with a multiword term: " .. form) end return add_endings(inflect(first), rest) elseif special == "second" then local first, second, rest = form:match(hack_re("^([^ ]+ )([^ ]+)( .*)$")) if not first then error("Special indicator 'second' can only be used with a term with three or more words: " .. form) end return add_endings(add_endings({first}, inflect(second)), rest) elseif special == "first-second" then local first, space, second, rest = form:match(hack_re("^([^ ]+)( )([^ ]+)( .*)$")) if not first then error("Special indicator 'first-second' can only be used with a term with three or more words: " .. form) end return add_endings(add_endings(add_endings(inflect(first), space), inflect(second)), rest) elseif special == "each" then local terms = split(form, sep) if #terms < 2 then error("Special indicator 'each' can only be used with a multiword term: " .. form) end for i, term in ipairs(terms) do terms[i] = inflect(term) if i > 1 then terms[i] = add_endings(raw_sep, terms[i]) end end local result = "" for _, term in ipairs(terms) do result = add_endings(result, term) end return result elseif special == "first-last" then local first, middle, last = form:match(hack_re("^(.-)( .* )(.-)$")) if not first then first, middle, last = form:match(hack_re("^(.-)( )(.*)$")) end if not first then error("Special indicator 'first-last' can only be used with a multiword term: " .. form) end return add_endings(add_endings(inflect(first), middle), inflect(last)) elseif special == "last" then local rest, last = form:match(hack_re("^(.* )(.-)$")) if not rest then error("Special indicator 'last' can only be used with a multiword term: " .. form) end return add_endings(rest, inflect(last)) elseif special and special ~= "+" then error("Unrecognized special=" .. special) end -- Only do default behavior if special indicator '+' explicitly given or separator is space; otherwise we will -- break existing behavior with hyphenated words. if (special == "+" or sep == " ") and form:find(sep) then if prepositions then -- check for prepositions in the middle of the word; do it this way so we can handle -- more than one word before the preposition (and usually inflect each word) for _, prep in ipairs(prepositions) do local first, space_prep_rest = umatch(form, hack_re("^(.-)( " .. prep .. ".*)$")) if first then return add_endings(inflect(first), space_prep_rest) end end end -- multiword or hyphenated expressions default to first-last; we need to pass in the separator to avoid -- problems with multiword terms containing hyphens in the individual words return export.handle_multiword(form, "first-last", inflect, prepositions, sep) end return nil end local function link_hyphen_split_component(word, data) if data.link_hyphen_split_component then return data.link_hyphen_split_component(word) else return "[[" .. word .. "]]" end end -- Default function to split a word on apostrophes. Don't split apostrophes at the beginning or end of a word (e.g. -- [['ndrangheta]] or [[po']]). Handle multiple apostrophes correctly, e.g. [[l'altr'ieri]] -> [[l']][altr']][[ieri]]. function export.default_split_apostrophe(word, data) local apostrophe_parts = split(word, "'", true, true) local linked_apostrophe_parts = {} local apostrophes_at_beginning = "" local i = 1 -- Apostrophes at beginning get attached to the first word after (which will always exist but may -- be blank if the word consists only of apostrophes). while i < #apostrophe_parts do -- <, not <=, in case the word consists only of apostrophes local apostrophe_part = apostrophe_parts[i] i = i + 1 if apostrophe_part == "" then apostrophes_at_beginning = apostrophes_at_beginning .. "'" else break end end apostrophe_parts[i] = apostrophes_at_beginning .. apostrophe_parts[i] -- Now, do the remaining parts. A blank part indicates more than one apostrophe in a row; we join -- all of them to the preceding word. while i <= #apostrophe_parts do local apostrophe_part = apostrophe_parts[i] if apostrophe_part == "" then linked_apostrophe_parts[#linked_apostrophe_parts] = linked_apostrophe_parts[#linked_apostrophe_parts] .. "'" elseif i == #apostrophe_parts then insert(linked_apostrophe_parts, apostrophe_part) else insert(linked_apostrophe_parts, apostrophe_part .. "'") end i = i + 1 end for j, tolink in ipairs(linked_apostrophe_parts) do linked_apostrophe_parts[j] = link_hyphen_split_component(tolink, data) end return concat(linked_apostrophe_parts) end --[=[ Auto-add links to a word that should not have spaces but may have hyphens and/or apostrophes. We split off final punctuation, then split on hyphens if `data.split_hyphen` is given, and also split on apostrophes if `data.split_apostrophe` is given. We only split on hyphens if they are in the middle of the word, not at the beginning or end (hyphens at the beginning or end indicate suffixes or prefixes, respectively). `include_hyphen_prefixes`, if given, is a set of prefixes (not including the final hyphen) where we should include the final hyphen in the prefix. Hence, e.g. if "anti" is in the set, a Portuguese word like [[anti-herói]] "anti-hero" will be split [[anti-]][[herói]] (whereas a word like [[código-fonte]] "source code" will be split as [[código]]-[[fonte]]). If `data.split_apostrophe` is specified, we split on apostrophes unless `data.no_split_apostrophe_words` is given and the word is in the specified set, such as French [[c'est]] and [[quelqu'un]]. If `data.split_apostrophe` is true, the default algorithm applies, which splits on all apostrophes except those at the beginning and end of a word (as in Italian [['ndrangheta]] or [[po']]), and includes the apostrophe in the link to its left (so we auto-split French [[l'eau]] as [[l']][[eau]] and [[l'altr'ieri]] as [[l']][altr']][[ieri]]). If `data.split_apostrophe` is specified but not `true`, it should be a function of one argument that does custom apostrophe-splitting. The argument is the word to split, and the return value should be the split and linked word. ]=] local function add_single_word_links(space_word, data, term_has_spaces) local space_word_no_punct, punct local punct_pattern = data.punctuation if punct_pattern and is_callable(punct_pattern) then space_word_no_punct, punct = punct_pattern(space_word) else if punct_pattern == nil then punct_pattern = "[,;:?!]" end space_word_no_punct, punct = umatch(space_word, "^(.*)(" .. punct_pattern .. ")$") end space_word_no_punct = space_word_no_punct or space_word punct = punct or "" local words if space_word_no_punct:sub(1, 1) == "-" or space_word_no_punct:sub(-1) == "-" then -- don't split prefixes and suffixes words = {space_word_no_punct} else local splitter if term_has_spaces then splitter = data.split_hyphen_when_space else splitter = data.split_hyphen_when_no_space end if is_callable(splitter) then words = splitter(space_word_no_punct) if type(words) == "string" then return words .. punct end end end if not words then local split_hyphen if term_has_spaces then split_hyphen = data.split_hyphen_when_space else split_hyphen = data.split_hyphen_when_no_space if split_hyphen == nil then -- default to true; use `false` to avoid this split_hyphen = true end end if split_hyphen then words = split(space_word_no_punct, "-", true, true) else words = {space_word_no_punct} end end local linked_words = {} for j, word in ipairs(words) do if j < #words and data.include_hyphen_prefixes and data.include_hyphen_prefixes[word] then word = "[[" .. word .. "-]]" elseif j > 1 and data.include_hyphen_suffixes and data.include_hyphen_suffixes[word] then word = "[[-" .. word .. "]]" else -- Don't split on apostrophes if the word is in `no_split_apostrophe_words`. if (not data.no_split_apostrophe_words or not data.no_split_apostrophe_words[word]) and data.split_apostrophe and word:find("'", nil, true) then if data.split_apostrophe == true then word = export.default_split_apostrophe(word, data) else -- custom apostrophe splitter/linker word = data.split_apostrophe(word) end elseif word ~= "" then -- avoid -[[]]- (e.g. f--k) word = link_hyphen_split_component(word, data) end if j < #words then word = word .. "-" end end insert(linked_words, word) end return concat(linked_words) .. punct end --[=[ Auto-add links to a multiword term. `data` contains fields customizing how to do this. By default we proceed as follows: (1) If the term already has embedded links in it, they are left unchanged. (2) Otherwise, if there are spaces present, we split on spaces and link each word separately. (3) If a given space-separated component ends in punctuation (defaulting to [,;:?!]), it is separated off, the remainder of the algorithm run, and the punctuation pasted back on. (4) If there are hyphens in a given space-separated component, we may link each hyphenated term separately depending on the settings in `data`. Normally the hyphens are not included in the linked terms, but this can be overridden for specific prefixes and/or suffixes. By default, if there are spaces in the multiword term, we do not link hyphenated components (because of cases like "boire du petit-lait" where "petit-lait" should be linked as a whole), but do so otherwise (e.g. for "avant-avant-hier"); this can overridden for cases like "croyez-le ou non". Cases where only some of the hyphens should be split can always be handled by explicitly specifying the head (e.g. "Nord-Pas-de-Calais" given as head=[[Nord]]-[[Pas-de-Calais]]). (5) If there are apostrophes in a given component, we may link each apostrophe-separated term separately depending on the settings in `data`, including the apostrophe in the link to its left (so we split "de l'eau" as "[[de]] [[l']][[eau]]"). The settings in `data` are as follows: `split_hyphen_when_no_space`: Whether to split on hyphens when the term has no spaces. Defaults to true if set to `nil`. This can be a function of one argument, to implement a custom splitting algorithm for hyphen-separated terms. If this returns [FIXME: FINISH ME ...] If `data.split_apostrophe` is specified, we split on apostrophes unless `data.no_split_apostrophe_words` is given and the word is in the specified set, such as French [[c'est]] and [[quelqu'un]]. If `data.split_apostrophe` is true, the default algorithm applies, which splits on all apostrophes except those at the beginning and end of a word (as in Italian [['ndrangheta]] or [[po']]), and includes the apostrophe in the link to its left (so we auto-split French [[l'eau]] as [[l']][[eau]] and [[l'altr'ieri]] as [[l']][altr']][[ieri]]). If `data.split_apostrophe` is specified but not `true`, it should be a function of one argument that does custom apostrophe-splitting. The argument is the word to split, and the return value should be the split and linked word. We don't always split on hyphens because of cases like "boire du petit-lait" where "petit-lait" should be linked as a whole, but provide the option to do it for cases like "croyez-le ou non". If there's no space, however, then it makes sense to split on hyphens by `no_split_apostrophe_words` and `include_hyphen_prefixes` allow for special-case handling of particular words and are as described in the comment above add_single_word_links(). ]=] function export.add_links_to_multiword_term(term, data) if term:match("[%[%]]") then return term end local words = split(term, " ", true, true) local term_has_spaces = #words > 1 local linked_words = {} for _, word in ipairs(words) do insert(linked_words, add_single_word_links(word, data, term_has_spaces)) end local retval = concat(linked_words, " ") -- If we ended up with a single link consisting of the entire term, -- remove the link. return retval:match("^%[%[([^%[%]]*)%]%]$") or retval end local function canonicalize_begin_end_spec(spec) local from, to = spec:match("^(.-):(.*)$") if not from then from = spec to = "" end return from, to end --[==[ Given a `linked_term` that is the output of add_links_to_multiword_term(), apply modifications as given in `modifier_spec` to change the link destination of subterms (normally single-word non-lemma forms; sometimes collections of adjacent words). This is usually used to link non-lemma forms to their corresponding lemma, but can also be used to replace a span of adjacent separately-linked words to a single multiword lemma. The format of `modifier_spec` is one or more semicolon-separated subterm specs, where each such spec is of the form SUBTERM:DEST, where SUBTERM is one or more words in the `linked_term` but without brackets in them, and DEST is the corresponding link destination to link the subterm to. Any occurrence of ~ in DEST is replaced with SUBTERM. Alternatively, a single modifier spec can be of the form BEGIN[FROM:TO], which is equivalent to writing BEGINFROM:BEGINTO (see example below). For example, given the source phrase [[il bue che dice cornuto all'asino]] "the pot calling the kettle black" (literally "the ox that calls the donkey horned/cuckolded"), the result of calling add_links_to_multiword_term() is [[il]] [[bue]] [[che]] [[dice]] [[cornuto]] [[all']][[asino]]. With a modifier_spec of 'dice:dire', the result is [[il]] [[bue]] [[che]] [[dire|dice]] [[cornuto]] [[all']][[asino]]. Here, based on the modifier spec, the non-lemma form [[dice]] is replaced with the two-part link [[dire|dice]]. Another example: given the source phrase [[chi semina vento raccoglie tempesta]] "sow the wind, reap the whirlwind" (literally (he) who sows wind gathers [the] tempest"). The result of calling add_links_to_multiword_term() is [[chi]] [[semina]] [[vento]] [[raccoglie]] [[tempesta]], and with a modifier_spec of 'semina:~re; raccoglie:~re', the result is [[chi]] [[seminare|semina]] [[vento]] [[raccogliere|raccoglie]] [[tempesta]]. Here we use the ~ notation to stand for the non-lemma form in the destination link. A more complex example is [[se non hai altri moccoli puoi andare a letto al buio]], which becomes [[se]] [[non]] [[hai]] [[altri]] [[moccoli]] [[puoi]] [[andare]] [[a]] [[letto]] [[al]] [[buio]] after calling add_links_to_multiword_term(). With the following modifier_spec: 'hai:avere; altr[i:o]; moccol[i:o]; puoi: potere; andare a letto:~; al buio:~', the result of applying the spec is [[se]] [[non]] [[avere|hai]] [[altro|altri]] [[moccolo|moccoli]] [[potere|puoi]] [[andare a letto]] [[al buio]]. Here, we rely on the alternative notation mentioned above for e.g. 'altr[i:o]', which is equivalent to 'altri:altro', and link multiword subterms using e.g. 'andare a letto:~'. (The code knows how to handle multiword subexpressions properly, and if the link text and destination are the same, only a single-part link is formed.) ]==] function export.apply_link_modifiers(linked_term, modifier_spec, lang) local split_modspecs = split(modifier_spec, "%s*;%s*") for j, modspec in ipairs(split_modspecs) do local id if modspec:find("<") then local rest rest, id = modspec:match("^(.*)<id:(.-)>$") if rest then modspec = rest end end local subterm, dest, otherlang local begin_spec, rest, end_spec = modspec:match("^%[(.-)%]([^:]*)%[(.-)%]$") if begin_spec then local begin_from, begin_to = canonicalize_begin_end_spec(begin_spec) local end_from, end_to = canonicalize_begin_end_spec(end_spec) subterm = begin_from .. rest .. end_from dest = begin_to .. rest .. end_to end if not subterm then rest, end_spec = modspec:match("^([^:]*)%[(.-)%]$") if rest then local end_from, end_to = canonicalize_begin_end_spec(end_spec) subterm = rest .. end_from dest = rest .. end_to end end if not subterm then begin_spec, rest = modspec:match("^%[(.-)%]([^:]*)$") if begin_spec then local begin_from, begin_to = canonicalize_begin_end_spec(begin_spec) subterm = begin_from .. rest dest = begin_to .. rest end end if not subterm then subterm, dest = modspec:match("^(.-)%s*:%s*(.*)$") if subterm and subterm ~= "^" and subterm ~= "$" then local langdest -- Parse off an initial language code (e.g. 'en:Higgs', 'la:minūtia' or 'grc:σκατός'). Also handle -- Wikipedia prefixes ('w:Abatemarco' or 'w:it:Colle Val d'Elsa'). otherlang, langdest = dest:match("^([A-Za-z0-9._-]+):([^ ].*)$") if otherlang == "w" then local foreign_wikipedia, foreign_term = langdest:match("^([A-Za-z0-9._-]+):([^ ].*)$") if foreign_wikipedia then otherlang = otherlang .. ":" .. foreign_wikipedia langdest = foreign_term end dest = ("%s:%s"):format(otherlang, langdest) otherlang = nil elseif otherlang then otherlang = get_lang(otherlang, true, "allow etym") dest = langdest end end end if not subterm then if modspec == "?" or modspec == "!" then subterm = "$" dest = modspec elseif modspec == "..." or modspec == "...?" then subterm = "$" dest = " " .. modspec elseif modspec:find("^[A-Z]$") then -- X, Y, etc. by themselves are unlinked, to help with snowclones subterm = modspec dest = "_" else subterm = modspec dest = "~" end end if subterm == "^" then linked_term = dest:gsub("_", " ") .. linked_term elseif subterm == "$" then linked_term = linked_term .. dest:gsub("_", " ") else if subterm:find("[", nil, true) then error(("Subterm '%s' in modifier spec '%s' cannot have brackets in it"):format( escape_wikicode(subterm), escape_wikicode(modspec))) end local escaped_subterm = pattern_escape(subterm) local subterm_re = "%[%[" .. escaped_subterm:gsub("(%%?[ ',%-])", "%%]*%1%%[*") .. "%]%]" local expanded_dest if dest:find("~", nil, true) then expanded_dest = dest:gsub("~", replacement_escape(subterm)) else expanded_dest = dest end if otherlang then expanded_dest = expanded_dest .. "#" .. otherlang:getCanonicalName() end local subterm_replacement if expanded_dest == "_" then subterm_replacement = subterm if id then error("Can't supply <id:...> with an unlinked subterm") end if otherlang then error("Can't supply prefixed language with an unlinked subterm") end elseif id or otherlang then if id and expanded_dest:find("[", nil, true) then error("Can't supply <id:...> with destination with embedded brackets") end subterm_replacement = require(links_module).language_link { lang = otherlang or lang, term = expanded_dest, alt = subterm, id = id, } elseif expanded_dest:find("[", nil, true) then -- Use the destination directly if it has brackets in it (e.g. to put brackets around parts of a word). subterm_replacement = expanded_dest elseif expanded_dest == subterm then subterm_replacement = "[[" .. subterm .. "]]" else subterm_replacement = "[[" .. expanded_dest .. "|" .. subterm .. "]]" end local escaped_subterm_replacement = replacement_escape(subterm_replacement) local replaced_linked_term = ugsub(linked_term, subterm_re, escaped_subterm_replacement) if replaced_linked_term == linked_term then mw.log(("Attempted to replace %s with %s in %s"):format(subterm_re, escaped_subterm_replacement, linked_term)) error(("Subterm '%s' could not be located in %slinked expression %s, or replacement same as subterm"):format( subterm, j > 1 and "intermediate " or "", escape_wikicode(linked_term))) else linked_term = replaced_linked_term end end end return linked_term end return export 8uip0he2epyv5rdmi0wwc1j9o7grekr 89200 89199 2026-09-02T18:53:14Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Module:headword_utilities]] 89199 Scribunto text/plain local export = {} local fun_is_callable_module = "Module:fun/isCallable" local languages_module = "Module:languages" local links_module = "Module:links" local parse_utilities_module = "Module:parse utilities" local string_pattern_escape_module = "Module:string/patternEscape" local string_replacement_escape_module = "Module:string/replacementEscape" local string_utilities_module = "Module:string utilities" local table_module = "Module:table" local dump = mw.dumpObject local unpack = unpack or table.unpack -- Lua 5.2 compatibility local insert = table.insert local concat = table.concat local remove = table.remove local sort = table.sort local function deepEquals(...) deepEquals = require(table_module).deepEquals return deepEquals(...) end local function escape_wikicode(...) escape_wikicode = require(parse_utilities_module).escape_wikicode return escape_wikicode(...) end local function extend(...) extend = require(table_module).extend return extend(...) end local function get_lang(...) get_lang = require(languages_module).getByCode return get_lang(...) end local function insert_if_not(...) insert_if_not = require(table_module).insertIfNot return insert_if_not(...) end local function is_callable(...) is_callable = require(fun_is_callable_module) return is_callable(...) end local function parse_inline_modifiers(...) parse_inline_modifiers = require(parse_utilities_module).parse_inline_modifiers return parse_inline_modifiers(...) end local function pattern_escape(...) pattern_escape = require(string_pattern_escape_module) return pattern_escape(...) end local function replacement_escape(...) replacement_escape = require(string_replacement_escape_module) return replacement_escape(...) 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 term_contains_top_level_html(...) term_contains_top_level_html = require(parse_utilities_module).term_contains_top_level_html return term_contains_top_level_html(...) end local function ugsub(...) ugsub = require(string_utilities_module).gsub return ugsub(...) end local function umatch(...) umatch = require(string_utilities_module).match return umatch(...) end local param_mods = { id = {}, -- disabled when `is_head = true` q = {type = "qualifier"}, qq = {type = "qualifier"}, l = {type = "labels"}, ll = {type = "labels"}, -- [[Module:headword]] expects part references in `.refs`. ref = {item_dest = "refs", type = "references", store = "insert-flattened"}, } local optional_param_mods = { g = {item_dest = "genders", type = "genders"}, alt = {}, lang = {type = "language"}, sc = {type = "script"}, t = {item_dest = "gloss"}, gloss = {}, pos = {}, lit = {}, tr = {}, ts = {}, face = {}, nolinkinfl = {type = "boolean"}, } local optional_headword_param_mods = { sc = {type = "script"}, tr = {}, ts = {}, } --[==[ Parse a single inflection or headword form or list of such forms. In either case, inline modifiers may be attached. `data` is an object with the following fields: * `val`: The raw value to parse. Required. * `paramname`: The name of the parameter from which the value was taken; used in error messages. Required. * `is_head`: We are parsing a headword parameter (a value which goes into the `heads` field of `data`). This changes the allowed modifiers, disabling the `id` modifier and only allowing a subset of optional modifiers. * `frob`: An optional function of one value to apply to the form after inline modifiers have been removed (i.e. to apply to the `.term` field of the returned object). * `include_mods`: List of extra inline modifiers to include, besides the default ones (see below). Each list item is either a string specifying a recognized extra inline modifier (see `optional_param_mods` in the code), or a two-item list of modifier name and modifier spec, where the spec should follow the syntax for modifier specs in `parse_inline_modifiers` in [[Module:parse utilities]]. * `exclude_mods`: List of default inline modifiers to not include. * `splitchar`: If specified, the value in `val` can be a list of forms to parse, separated by the value of `splitchar` (which is a Lua pattern, as in `parse_inline_modifiers` in [[Module:parse utilities]]). Most commonly, `splitchar` is a single comma and the values are comma-separated (in this case, splitting will not happen if a space follows the comma). * `parse_lang_prefix`: If specified, allow a language prefix to precede a form, and if found, store into the `.lang` field of the returned object. * `preserve_splitchar`, `delimiter_key`, `escape_fun`, `unescape_fun`, `pre_normalize_modifiers`: As in `parse_inline_modifiers` in [[Module:parse utilities]]. Returns an object suitable for storing as one element of one of the lists in `headdata.inflections`, where `headdata` is the structure passed to [[Module:headword]]. If `splitchar` is specified, howeve, the return value is a list of such objects. The following default inline modifiers are currently recognized: * `q`: Left qualifier. * `qq`: Right qualifier. * `l`: Comma-separated list of left labels. No space should follow the comma. * `ll`: Comma-separated list of right labels. No space should follow the comma. * `ref`: Reference or references. See {{tl|IPA}} for the syntax. * `id`: Sense ID, in case there are multiple senses. See {{tl|l}}. The following are the recognized additional inline modifiers: * `g`: Comma-separated list of genders. * `alt`: Display text. * `lang`: Language code of language of the form, if different from the language of the headword. * `sc`: Script code of script of the form. Almost never needed. * `t`: Gloss for the form. * `gloss`: Gloss for the form (alias for `t`). * `pos`: Part of speech of the form. * `lit`: Literal meaning of the form. * `tr`: Manual transliteration of the form. * `ts`: Transcription of the form, for languages where the transliteration differs markedly from the pronunciation. * `face`: Face to display the form in, e.g. {"hypothetical"} for a hypothetical form (unlinkable and displayed in italics). * `nolinkinfl`: Make the form unlinkable. ]==] function export.parse_term_with_modifiers(data) local paramname, val, frob = data.paramname, data.val, data.frob local function generate_obj(term, parse_err) if frob then term = frob(term, parse_err) end if data.parse_lang_prefix and term:find(":") then return require(parse_utilities_module).generate_obj_maybe_parsing_lang_prefix { term = term, paramname = paramname, parse_lang_prefix = true, parse_err = parse_err, } else return {term = term} end end -- Check for inline modifier, e.g. מרים<tr:Miryem>. But exclude top-level HTML entry with <span ...>, -- <sup> or similar in it. if (val:find("<", nil, true) or data.splitchar) and not term_contains_top_level_html(val) and -- don't parse inline modifiers if is_head and the value begins with a ~ (link modifier syntax) (not data.is_head or not val:find("^~")) then local param_mods = param_mods if data.is_head then param_mods = shallow_copy(param_mods) param_mods.id = nil end if data.include_mods or data.exclude_mods then if not data.is_head then -- already copied when data.is_head param_mods = shallow_copy(param_mods) end if data.include_mods then local optional_mods = data.is_head and optional_headword_param_mods or optional_param_mods for _, mod in ipairs(data.include_mods) do if type(mod) == "table" then if #mod ~= 2 then error(("Internal error: Modifier spec %s in `include_mods` should be of length 2"):format( dump(mod))) end local modkey, modvalue = unpack(mod) param_mods[modkey] = modvalue elseif not optional_mods[mod] then error(("Internal error: Unrecognized modifier spec %s in `include_mods`"):format( dump(mod))) else param_mods[mod] = optional_mods[mod] end end end if data.exclude_mods then for _, mod in ipairs(data.exclude_mods) do if not param_mods[mod] then error(("Internal error: Modifier spec %s in `exclude_mods` not found among existing modifiers" ):format(dump(mod))) else param_mods[mod] = nil end end end end return parse_inline_modifiers(val, { paramname = paramname, param_mods = param_mods, generate_obj = generate_obj, splitchar = data.splitchar, preserve_splitchar = data.preserve_splitchar, delimiter_key = data.delimiter_key, escape_fun = data.escape_fun, unescape_fun = data.unescape_fun, pre_normalize_modifiers = data.pre_normalize_modifiers, }) else local retval = generate_obj(val) if data.splitchar then retval = {retval} end return retval end end --[==[ Parse a list of inflection forms that may have inline modifiers attached. `data` is an object with the following fields: * `forms`: The list of raw values to parse. Required. * `paramname`: The name of the first parameter from which the value was taken; used in error messages. If this is a two-element list, the first element is the first parameter and the second element is the prefix of the remaining parameters. Parameter names that are numbers are handled correctly, as are those with \1 in it marking where the parameter index goes. Required. * `qualifiers`: If specified, a possibly gappy list of left qualifiers to add to the parsed terms (for compatibility purposes). * `splitchar`: As in `parse_term_with_modifiers()`. The resulting per-term lists will be flattened. * `frob`, `include_mods`, `exclude_mods`, `is_head`, `preserve_splitchar`, `parse_lang_prefix`, `delimiter_key`, `escape_fun`, `unescape_fun`, `pre_normalize_modifiers`: As in `parse_term_with_modifiers()`. Returns a list of objects, suitable for storing as one of the lists in `headdata.inflections` (once a label is added), where `headdata` is the structure passed to [[Module:headword]]. ]==] function export.parse_term_list_with_modifiers(data) local paramname, forms = data.paramname, data.forms local qualifiers = data.qualifiers local first, restpref if type(paramname) == "table" then first = paramname[1] restpref = paramname[2] else first = paramname restpref = paramname end local terms = {} data = shallow_copy(data) for i, val in ipairs(forms) do data.paramname = i == 1 and first or type(restpref) == "number" and restpref + i - 1 or restpref:find("\1", nil, true) and restpref:gsub("\1", tostring(i)) or restpref .. i data.val = val local parsed = export.parse_term_with_modifiers(data) if qualifiers and qualifiers[i] then if data.splitchar then for _, term in ipairs(parsed) do term.q = {qualifiers[i]} end else parsed.q = {qualifiers[i]} end end if data.splitchar then extend(terms, parsed) else terms[i] = parsed end end return terms end --[==[ Construct a link to [[Appendix:Glossary]] for `entry`. If `text` is specified, it is the display text; otherwise, `entry` is used. ]==] function export.glossary_link(entry, text) text = text or entry return "[[Appendix:Glossary#" .. entry .. "|" .. text .. "]]" end function export.replace_glossary_links_in_label(label) if label:find("<<", nil, true) then label = label:gsub("<<(.-)|(.-)>>", export.glossary_link):gsub("<<(.-)>>", export.glossary_link) end return label end function export.insert_fixed_inflection(data) local headdata, origterm, label = data.headdata, data.originating_term, data.label local inflobj = data.inflobj or headdata inflobj.inflections = inflobj.inflections or {} if not origterm then insert(inflobj.inflections, { label = export.replace_glossary_links_in_label(label) }) else if origterm.id then error(("It doesn't make sense to pass in an ID '%s' for label '%s' in conjunction with a term value '%s'" ):format(origterm.id, label, origterm.term)) end -- Preserve qualifiers, labels, references origterm.term = nil origterm.label = export.replace_glossary_links_in_label(label) insert(inflobj.inflections, origterm) end end --[==[ Insert previously-parsed terms into an `inflections` field. The `inflections` field will be initialized if needed. `data` is an object with the following fields: * `headdata`: The headword structure passed to [[Module:headword]]. Required. * `inflobj`: The object whose `inflections` field the terms are inserted into. Defaults to `headdata`. Only needs to be set for nested inflections, which are specified for an inflection object rather than the headword data structure as a whole. * `terms`: The list of parsed terms. If {nil} or omitted, nothing happens unless `request` is set. * `label`: The label that the inflections are given; any parts of the label surrounded in <<...>> are linked to the glossary. (If the contents of <<...> contain a | in them, they are a two-part link.) Required. * `no_label`: If the term is {"-"} and there are no other terms, insert a fixed label with this value. Defaults to {"no "} plus the label. * `usually_no_label`: If the term is {"-"} and there are other terms, insert a fixed label with this value. Defaults to {"usually no "} plus the label. * `accel`: If specified, a full accelerator object to add to the inflections. * `request`: If specified and no terms are given, insert a label with a request for inflections to be given. * `enable_auto_translit`: If specified and terms are given, display automatic transliteration of the terms. The return value indicates whether the inflection exists and how many terms are in it. It is an object with the following fields: * `exists`: {"yes"} if one or more terms were specified; {"no"} if the value was given as {"-"}; {"usually no"} if the first value was given as {"-"} but additional terms were supplied; otherwise {nil}, indicating that the status is unspecified. * `numterms`: Number of terms in the inflection. Will be 0 unless `exists` has the value {"yes"} or {"usually no"}. * `request`: True if no terms were specified but a term request was inserted into the inflection (because `data.request` was specified). Otherwise {nil}. ]==] function export.insert_inflection(data) local headdata, terms, label = data.headdata, data.terms, data.label local inflobj = data.inflobj or headdata local retval = {} if terms and terms[1] then if terms[1].term == "-" then if terms[2] then export.insert_fixed_inflection { headdata = headdata, inflobj = inflobj, originating_term = terms[1], label = data.usually_no_label or "usually no " .. label, } remove(terms, 1) retval.numterms = #terms retval.exists = "usually no" else export.insert_fixed_inflection { headdata = headdata, inflobj = inflobj, originating_term = terms[1], label = data.no_label or "no " .. label, } retval.numterms = 0 retval.exists = "no" return retval end else retval.numterms = #terms retval.exists = "yes" end if data.check_missing then error("check_missing support removed; use checkredlinks=true in [[Module:headword]]") end terms.label = export.replace_glossary_links_in_label(label) if data.accel then terms.accel = data.accel end terms.enable_auto_translit = data.enable_auto_translit inflobj.inflections = inflobj.inflections or {} insert(inflobj.inflections, terms) elseif data.request then inflobj.inflections = inflobj.inflections or {} insert(inflobj.inflections, { label = export.replace_glossary_links_in_label(label), request = true, }) retval.numterms = 0 -- retval.exists == nil retval.request = true else retval.numterms = 0 -- retval.exists == nil end return retval end --[==[ Parse raw arguments from `forms` for inline modifiers, and insert the resulting terms (which should not require significant additional processing) into `headdata.inflections`. `data` is an object with the following fields: * `forms`: The list of raw values to parse. If {nil} or omitted, nothing happens. * `headdata`: The headword structure passed to [[Module:headword]]. Required. * `paramname`: As in `parse_term_list_with_modifiers()`. Required. * `label`: As in `insert_inflection()`. Required. * `qualifiers`, `frob`, `include_mods`, `exclude_mods`, `is_head`, `splitchar`, `preserve_splitchar`, `delimiter_key`, `escape_fun`, `unescape_fun`, `pre_normalize_modifiers`: As in `parse_term_list_with_modifiers()`. * `accel`: As in `insert_inflection()`. Return value is as in `insert_inflection()`. ]==] function export.parse_and_insert_inflection(data) local forms = data.forms if forms and forms[1] then data = shallow_copy(data) data.forms = forms data.terms = export.parse_term_list_with_modifiers(data) return export.insert_inflection(data) end return { numterms = 0 } end --[==[ Combine two sets of qualifiers or labels. If either is {nil}, just return the other, and if both are {nil}, return {nil}. ]==] function export.combine_qualifiers_or_labels(quals1, quals2) if not quals1 and not quals2 then return nil end if not quals1 then return quals2 end if not quals2 then return quals1 end local combined = shallow_copy(quals1) for _, note in ipairs(quals2) do insert_if_not(combined, note) end return combined end --[==[ Combine the qualifiers, labels, references and ID's of two term objects. `destobj` is the "destination term object" into which the combined properties are written, and `srcobj` is the "source object" into which the properties are merged. `destobj` is side-effected (but the lists inside of `destobj` are not); if this is undesirable, make sure to shallow-copy `destobj` first. If both objects have values for a given qualifier, label or reference, the values of `destobj` come first. If both objects have a value for `id`, the values must match or an error is thrown; otherwise, the resulting value of `id` comes from whichever one is defined. '''NOTE:''' This may not be the correct behavior when deduplicating a list of term objects. See `insert_termobj_combining_duplicates` for a different approach. ]==] function export.combine_termobj_qualifiers_labels(destobj, srcobj) destobj.q = export.combine_qualifiers_or_labels(destobj.q, srcobj.q) destobj.qq = export.combine_qualifiers_or_labels(destobj.qq, srcobj.qq) destobj.l = export.combine_qualifiers_or_labels(destobj.l, srcobj.l) destobj.ll = export.combine_qualifiers_or_labels(destobj.ll, srcobj.ll) destobj.refs = export.combine_qualifiers_or_labels(destobj.refs, srcobj.refs) if destobj.id and srcobj.id and destobj.id ~= srcobj.id then -- FIXME: We probably want to pass in an error function error(("Can't specify two different ID's %s and %s when combining objects"):format(srcobj.id, destobj.id)) end destobj.id = destobj.id or srcobj.id return destobj end function export.termobj_has_qualifiers_or_labels(obj) return obj.q and obj.q[1] or obj.qq and obj.qq[1] or obj.l and obj.l[1] or obj.ll and obj.ll[1] or obj.refs and obj.refs[1] end local function one_ancillary_property_equal(prop1, prop2) local prop1_is_nil = not prop1 or not prop1[1] local prop2_is_nil = not prop2 or not prop2[1] if prop1_is_nil and prop2_is_nil then return true end if prop1_is_nil or prop2_is_nil then return false end return deepEquals(prop1, prop2) end function export.termobj_ancillary_properties_equal(obj1, obj2) return one_ancillary_property_equal(obj1.q, obj2.q) and one_ancillary_property_equal(obj1.qq, obj2.qq) and one_ancillary_property_equal(obj1.l, obj2.l) and one_ancillary_property_equal(obj1.ll, obj2.ll) and one_ancillary_property_equal(obj1.refs, obj2.refs) and obj1.id == obj2.id end function export.convert_termobj_to_formobj(termobj) local formobj = { form = termobj.term, translit = termobj.tr, } local footnotes local function mods_to_footnote(mod_prefix, mod_vals) if mod_vals and mod_vals[1] then footnotes = footnotes or {} for _, val in ipairs(mod_vals) do insert(footnotes, "[" .. mod_prefix .. ":" .. val .. "]") end end end mods_to_footnote("q", termobj.q) mods_to_footnote("qq", termobj.qq) mods_to_footnote("l", termobj.l) mods_to_footnote("ll", termobj.ll) mods_to_footnote("ref", termobj.refs) mods_to_footnote("id", termobj.id and {termobj.id} or nil) formobj.footnotes = footnotes return formobj end local recognized_multi_mods = { q = "q", qq = "qq", l = "l", ll = "ll", ref = "refs", } local recognized_single_mods = { id = "id", } function export.add_footnote_to_termobj(termobj, footnote) local stripped_footnote = footnote:match("^%[(.*)%]$") if not stripped_footnote then error("Internal error: Footnote should be surrounded by brackets at this stage: " .. footnote) end local prefix, rest = stripped_footnote:match("^([a-z]+):(.+)$") local field, is_multi if prefix then if recognized_multi_mods[prefix] then field = recognized_multi_mods[prefix] is_multi = true elseif recognized_single_mods[prefix] then field = recognized_single_mods[prefix] is_multi = false end end if not field then rest = stripped_footnote field = "l" is_multi = true end if is_multi then if not termobj[field] then termobj[field] = {} end insert(termobj[field], rest) else if termobj[field] and termobj[field] ~= rest then error(("Can't set two values for '%s': '%s' and '%s'"):format(field, termobj[field], rest)) end termobj[field] = rest end end function export.convert_formobj_to_termobj(formobj) local termobj = { term = formobj.form, tr = formobj.translit, } if formobj.footnotes then for _, footnote in ipairs(formobj.footnotes) do export.add_footnote_to_termobj(termobj, footnote) end end return termobj end local function extract_termobj_field_modifiers(fieldval) return fieldval:match("^([*+]?)(.*)$") end function export.remove_termobj_field_modifiers(termobj) local function remove_field_modifiers(field) if termobj[field] and termobj[field][1] then local any_field_modifiers = false for _, val in ipairs(termobj[field]) do local field_mods, _ = extract_termobj_field_modifiers(val) if field_mods ~= "" then any_field_modifiers = true break end end local new_field = {} if any_field_modifiers then for _, val in ipairs(termobj[field]) do local _, field_without_mods = extract_termobj_field_modifiers(val) insert_if_not(new_field, field_without_mods) end termobj[field] = new_field end end end remove_field_modifiers("q") remove_field_modifiers("qq") remove_field_modifiers("l") remove_field_modifiers("ll") remove_field_modifiers("refs") end function export.insert_termobj_combining_duplicates(destobjs, termobj) for _, destobj in ipairs(destobjs) do if destobj.term == termobj.term and destobj.tr == termobj.tr then -- Form already present; maybe combine footnotes. local function combine_field_values(field) if termobj[field] and termobj[field][1] then -- Check to see if there are existing values with *; if so, remove them. if destobj[field] and destobj[field][1] then local any_values_with_asterisk = false for _, val in ipairs(destobj[field]) do local field_mods, _ = extract_termobj_field_modifiers(val) if field_mods:find("%*") then any_values_with_asterisk = true break end end if any_values_with_asterisk then local filtered_values = {} for _, val in ipairs(destobj[field]) do local field_mods, _ = extract_termobj_field_modifiers(val) if not field_mods:find("%*") then insert(filtered_values, val) end end if filtered_values[1] then destobj[field] = filtered_values else destobj[field] = nil end end end local any_footnotes_with_plus = false for _, val in ipairs(termobj[field]) do local field_mods, _ = extract_termobj_field_modifiers(val) if field_mods:find("%+") then any_footnotes_with_plus = true break end end if any_footnotes_with_plus then if not destobj[field] then destobj[field] = {} else destobj[field] = shallow_copy(destobj[field]) end for _, val in ipairs(termobj[field]) do local already_seen = false local field_mods, field_without_mods = extract_termobj_field_modifiers(val) if field_mods:find("%+") then for _, existing_val in ipairs(destobj[field]) do local _, existing_field_without_mods = extract_termobj_field_modifiers(existing_val) if existing_field_without_mods == field_without_mods then already_seen = true break end end if not already_seen then insert(destobj[field], val) end end end end end end combine_field_values("q") combine_field_values("qq") combine_field_values("l") combine_field_values("ll") combine_field_values("refs") if destobj.id and termobj.id and destobj.id ~= termobj.id then -- FIXME: We probably want to pass in an error function error(("Can't specify two different ID's %s and %s when combining objects"):format(termobj.id, destobj.id)) end destobj.id = destobj.id or termobj.id return end end insert(destobjs, termobj) end export.allowed_special_indicators = { ["first"] = true, ["first-second"] = true, ["first-last"] = true, ["second"] = true, ["last"] = true, ["each"] = true, ["+"] = true, -- requests the default behavior with preposition handling } --[==[ Check for special indicators (values such as {"+first"} or {"+first-last"} that are used in a `pl`, `f`, etc. argument and indicate how to inflect a multiword term). If `form` is such an indicator, the return value is `form` minus the initial `+` sign; otherwise, if form begins with a `+` sign, an error is thrown; otherwise the return value is nil. ]==] function export.get_special_indicator(form, noerror) if form:find("^%+") then form = form:gsub("^%+", "") if not export.allowed_special_indicators[form] then if noerror then return nil end local indicators = {} for indic, _ in pairs(export.allowed_special_indicators) do insert(indicators, "+" .. indic) end sort(indicators) error("Special inflection indicator beginning with '+' can only be " .. mw.text.listToText(indicators) .. ": +" .. form) end return form end return nil end local function add_endings(bases, endings) local retval = {} if type(bases) ~= "table" then bases = {bases} end if type(endings) ~= "table" then endings = {endings} end for _, base in ipairs(bases) do for _, ending in ipairs(endings) do insert(retval, base .. ending) end end return retval end --[==[ Inflect a possibly multiword or hyphenated term `form` using the function `inflect`, which is a function of one argument that is called on a single word to inflect and should return either the inflected word or a list of inflected words. `special` indicates how to inflect the multiword term and should be e.g. {"first"} to inflect only the first word, {"first-last"} to inflect the first and last words, {"each"} to inflect each word, etc. See `allowed_special_indicators` above for the possibilities. If `special` is `+`, or is omitted and the term is multiword (i.e. containing a space character), and `prepositions` is supplied, the function checks for multiword or hyphenated terms containing the prepositions in `prepositions`, e.g. Italian [[senso di marcia]] or [[medaglia d'oro]] or Portuguese [[tartaruga-do-mar]]. If such a term is found, only the first word is inflected. Otherwise, the default is {"first-last"}. `prepositions` is a list of Lua patterns matching prepositions. The patterns will automatically have the separator character (space or hyphen) added to the left side but not the right side, so they should contain a space character (which will automatically be converted to the appropriate separator) on the right side unless the preposition is joined on the right side with an apostrophe. Examples of preposition patterns for Italian are {"di "}, {"sull'"} and {"d?all[oae] "} (which matches {"dallo "}, {"dalle "}, {"alla "}, etc.). The return value is always either a list of inflected multiword or hyphenated terms, or nil if `special` is omitted and `form` is not multiword. (If `special` is specified and `form` is not multiword or hyphenated, an error results.) ]==] function export.handle_multiword(form, special, inflect, prepositions, sep) sep = sep or form:find(" ") and " " or "%-" local raw_sep = sep == " " and " " or "-" -- Used to add regex version of separator in the replacement portion of ugsub() or :gsub() local sep_replacement = sep == " " and " " or "%%-" -- Given a Lua pattern, replace space with the appropriate separator. local function hack_re(re) if sep == " " then return re end return (re:gsub(" ", sep_replacement)) end if special == "first" then local first, rest = form:match(hack_re("^(.-)( .*)$")) if not first then error("Special indicator 'first' can only be used with a multiword term: " .. form) end return add_endings(inflect(first), rest) elseif special == "second" then local first, second, rest = form:match(hack_re("^([^ ]+ )([^ ]+)( .*)$")) if not first then error("Special indicator 'second' can only be used with a term with three or more words: " .. form) end return add_endings(add_endings({first}, inflect(second)), rest) elseif special == "first-second" then local first, space, second, rest = form:match(hack_re("^([^ ]+)( )([^ ]+)( .*)$")) if not first then error("Special indicator 'first-second' can only be used with a term with three or more words: " .. form) end return add_endings(add_endings(add_endings(inflect(first), space), inflect(second)), rest) elseif special == "each" then local terms = split(form, sep) if #terms < 2 then error("Special indicator 'each' can only be used with a multiword term: " .. form) end for i, term in ipairs(terms) do terms[i] = inflect(term) if i > 1 then terms[i] = add_endings(raw_sep, terms[i]) end end local result = "" for _, term in ipairs(terms) do result = add_endings(result, term) end return result elseif special == "first-last" then local first, middle, last = form:match(hack_re("^(.-)( .* )(.-)$")) if not first then first, middle, last = form:match(hack_re("^(.-)( )(.*)$")) end if not first then error("Special indicator 'first-last' can only be used with a multiword term: " .. form) end return add_endings(add_endings(inflect(first), middle), inflect(last)) elseif special == "last" then local rest, last = form:match(hack_re("^(.* )(.-)$")) if not rest then error("Special indicator 'last' can only be used with a multiword term: " .. form) end return add_endings(rest, inflect(last)) elseif special and special ~= "+" then error("Unrecognized special=" .. special) end -- Only do default behavior if special indicator '+' explicitly given or separator is space; otherwise we will -- break existing behavior with hyphenated words. if (special == "+" or sep == " ") and form:find(sep) then if prepositions then -- check for prepositions in the middle of the word; do it this way so we can handle -- more than one word before the preposition (and usually inflect each word) for _, prep in ipairs(prepositions) do local first, space_prep_rest = umatch(form, hack_re("^(.-)( " .. prep .. ".*)$")) if first then return add_endings(inflect(first), space_prep_rest) end end end -- multiword or hyphenated expressions default to first-last; we need to pass in the separator to avoid -- problems with multiword terms containing hyphens in the individual words return export.handle_multiword(form, "first-last", inflect, prepositions, sep) end return nil end local function link_hyphen_split_component(word, data) if data.link_hyphen_split_component then return data.link_hyphen_split_component(word) else return "[[" .. word .. "]]" end end -- Default function to split a word on apostrophes. Don't split apostrophes at the beginning or end of a word (e.g. -- [['ndrangheta]] or [[po']]). Handle multiple apostrophes correctly, e.g. [[l'altr'ieri]] -> [[l']][altr']][[ieri]]. function export.default_split_apostrophe(word, data) local apostrophe_parts = split(word, "'", true, true) local linked_apostrophe_parts = {} local apostrophes_at_beginning = "" local i = 1 -- Apostrophes at beginning get attached to the first word after (which will always exist but may -- be blank if the word consists only of apostrophes). while i < #apostrophe_parts do -- <, not <=, in case the word consists only of apostrophes local apostrophe_part = apostrophe_parts[i] i = i + 1 if apostrophe_part == "" then apostrophes_at_beginning = apostrophes_at_beginning .. "'" else break end end apostrophe_parts[i] = apostrophes_at_beginning .. apostrophe_parts[i] -- Now, do the remaining parts. A blank part indicates more than one apostrophe in a row; we join -- all of them to the preceding word. while i <= #apostrophe_parts do local apostrophe_part = apostrophe_parts[i] if apostrophe_part == "" then linked_apostrophe_parts[#linked_apostrophe_parts] = linked_apostrophe_parts[#linked_apostrophe_parts] .. "'" elseif i == #apostrophe_parts then insert(linked_apostrophe_parts, apostrophe_part) else insert(linked_apostrophe_parts, apostrophe_part .. "'") end i = i + 1 end for j, tolink in ipairs(linked_apostrophe_parts) do linked_apostrophe_parts[j] = link_hyphen_split_component(tolink, data) end return concat(linked_apostrophe_parts) end --[=[ Auto-add links to a word that should not have spaces but may have hyphens and/or apostrophes. We split off final punctuation, then split on hyphens if `data.split_hyphen` is given, and also split on apostrophes if `data.split_apostrophe` is given. We only split on hyphens if they are in the middle of the word, not at the beginning or end (hyphens at the beginning or end indicate suffixes or prefixes, respectively). `include_hyphen_prefixes`, if given, is a set of prefixes (not including the final hyphen) where we should include the final hyphen in the prefix. Hence, e.g. if "anti" is in the set, a Portuguese word like [[anti-herói]] "anti-hero" will be split [[anti-]][[herói]] (whereas a word like [[código-fonte]] "source code" will be split as [[código]]-[[fonte]]). If `data.split_apostrophe` is specified, we split on apostrophes unless `data.no_split_apostrophe_words` is given and the word is in the specified set, such as French [[c'est]] and [[quelqu'un]]. If `data.split_apostrophe` is true, the default algorithm applies, which splits on all apostrophes except those at the beginning and end of a word (as in Italian [['ndrangheta]] or [[po']]), and includes the apostrophe in the link to its left (so we auto-split French [[l'eau]] as [[l']][[eau]] and [[l'altr'ieri]] as [[l']][altr']][[ieri]]). If `data.split_apostrophe` is specified but not `true`, it should be a function of one argument that does custom apostrophe-splitting. The argument is the word to split, and the return value should be the split and linked word. ]=] local function add_single_word_links(space_word, data, term_has_spaces) local space_word_no_punct, punct local punct_pattern = data.punctuation if punct_pattern and is_callable(punct_pattern) then space_word_no_punct, punct = punct_pattern(space_word) else if punct_pattern == nil then punct_pattern = "[,;:?!]" end space_word_no_punct, punct = umatch(space_word, "^(.*)(" .. punct_pattern .. ")$") end space_word_no_punct = space_word_no_punct or space_word punct = punct or "" local words if space_word_no_punct:sub(1, 1) == "-" or space_word_no_punct:sub(-1) == "-" then -- don't split prefixes and suffixes words = {space_word_no_punct} else local splitter if term_has_spaces then splitter = data.split_hyphen_when_space else splitter = data.split_hyphen_when_no_space end if is_callable(splitter) then words = splitter(space_word_no_punct) if type(words) == "string" then return words .. punct end end end if not words then local split_hyphen if term_has_spaces then split_hyphen = data.split_hyphen_when_space else split_hyphen = data.split_hyphen_when_no_space if split_hyphen == nil then -- default to true; use `false` to avoid this split_hyphen = true end end if split_hyphen then words = split(space_word_no_punct, "-", true, true) else words = {space_word_no_punct} end end local linked_words = {} for j, word in ipairs(words) do if j < #words and data.include_hyphen_prefixes and data.include_hyphen_prefixes[word] then word = "[[" .. word .. "-]]" elseif j > 1 and data.include_hyphen_suffixes and data.include_hyphen_suffixes[word] then word = "[[-" .. word .. "]]" else -- Don't split on apostrophes if the word is in `no_split_apostrophe_words`. if (not data.no_split_apostrophe_words or not data.no_split_apostrophe_words[word]) and data.split_apostrophe and word:find("'", nil, true) then if data.split_apostrophe == true then word = export.default_split_apostrophe(word, data) else -- custom apostrophe splitter/linker word = data.split_apostrophe(word) end elseif word ~= "" then -- avoid -[[]]- (e.g. f--k) word = link_hyphen_split_component(word, data) end if j < #words then word = word .. "-" end end insert(linked_words, word) end return concat(linked_words) .. punct end --[=[ Auto-add links to a multiword term. `data` contains fields customizing how to do this. By default we proceed as follows: (1) If the term already has embedded links in it, they are left unchanged. (2) Otherwise, if there are spaces present, we split on spaces and link each word separately. (3) If a given space-separated component ends in punctuation (defaulting to [,;:?!]), it is separated off, the remainder of the algorithm run, and the punctuation pasted back on. (4) If there are hyphens in a given space-separated component, we may link each hyphenated term separately depending on the settings in `data`. Normally the hyphens are not included in the linked terms, but this can be overridden for specific prefixes and/or suffixes. By default, if there are spaces in the multiword term, we do not link hyphenated components (because of cases like "boire du petit-lait" where "petit-lait" should be linked as a whole), but do so otherwise (e.g. for "avant-avant-hier"); this can overridden for cases like "croyez-le ou non". Cases where only some of the hyphens should be split can always be handled by explicitly specifying the head (e.g. "Nord-Pas-de-Calais" given as head=[[Nord]]-[[Pas-de-Calais]]). (5) If there are apostrophes in a given component, we may link each apostrophe-separated term separately depending on the settings in `data`, including the apostrophe in the link to its left (so we split "de l'eau" as "[[de]] [[l']][[eau]]"). The settings in `data` are as follows: `split_hyphen_when_no_space`: Whether to split on hyphens when the term has no spaces. Defaults to true if set to `nil`. This can be a function of one argument, to implement a custom splitting algorithm for hyphen-separated terms. If this returns [FIXME: FINISH ME ...] If `data.split_apostrophe` is specified, we split on apostrophes unless `data.no_split_apostrophe_words` is given and the word is in the specified set, such as French [[c'est]] and [[quelqu'un]]. If `data.split_apostrophe` is true, the default algorithm applies, which splits on all apostrophes except those at the beginning and end of a word (as in Italian [['ndrangheta]] or [[po']]), and includes the apostrophe in the link to its left (so we auto-split French [[l'eau]] as [[l']][[eau]] and [[l'altr'ieri]] as [[l']][altr']][[ieri]]). If `data.split_apostrophe` is specified but not `true`, it should be a function of one argument that does custom apostrophe-splitting. The argument is the word to split, and the return value should be the split and linked word. We don't always split on hyphens because of cases like "boire du petit-lait" where "petit-lait" should be linked as a whole, but provide the option to do it for cases like "croyez-le ou non". If there's no space, however, then it makes sense to split on hyphens by `no_split_apostrophe_words` and `include_hyphen_prefixes` allow for special-case handling of particular words and are as described in the comment above add_single_word_links(). ]=] function export.add_links_to_multiword_term(term, data) if term:match("[%[%]]") then return term end local words = split(term, " ", true, true) local term_has_spaces = #words > 1 local linked_words = {} for _, word in ipairs(words) do insert(linked_words, add_single_word_links(word, data, term_has_spaces)) end local retval = concat(linked_words, " ") -- If we ended up with a single link consisting of the entire term, -- remove the link. return retval:match("^%[%[([^%[%]]*)%]%]$") or retval end local function canonicalize_begin_end_spec(spec) local from, to = spec:match("^(.-):(.*)$") if not from then from = spec to = "" end return from, to end --[==[ Given a `linked_term` that is the output of add_links_to_multiword_term(), apply modifications as given in `modifier_spec` to change the link destination of subterms (normally single-word non-lemma forms; sometimes collections of adjacent words). This is usually used to link non-lemma forms to their corresponding lemma, but can also be used to replace a span of adjacent separately-linked words to a single multiword lemma. The format of `modifier_spec` is one or more semicolon-separated subterm specs, where each such spec is of the form SUBTERM:DEST, where SUBTERM is one or more words in the `linked_term` but without brackets in them, and DEST is the corresponding link destination to link the subterm to. Any occurrence of ~ in DEST is replaced with SUBTERM. Alternatively, a single modifier spec can be of the form BEGIN[FROM:TO], which is equivalent to writing BEGINFROM:BEGINTO (see example below). For example, given the source phrase [[il bue che dice cornuto all'asino]] "the pot calling the kettle black" (literally "the ox that calls the donkey horned/cuckolded"), the result of calling add_links_to_multiword_term() is [[il]] [[bue]] [[che]] [[dice]] [[cornuto]] [[all']][[asino]]. With a modifier_spec of 'dice:dire', the result is [[il]] [[bue]] [[che]] [[dire|dice]] [[cornuto]] [[all']][[asino]]. Here, based on the modifier spec, the non-lemma form [[dice]] is replaced with the two-part link [[dire|dice]]. Another example: given the source phrase [[chi semina vento raccoglie tempesta]] "sow the wind, reap the whirlwind" (literally (he) who sows wind gathers [the] tempest"). The result of calling add_links_to_multiword_term() is [[chi]] [[semina]] [[vento]] [[raccoglie]] [[tempesta]], and with a modifier_spec of 'semina:~re; raccoglie:~re', the result is [[chi]] [[seminare|semina]] [[vento]] [[raccogliere|raccoglie]] [[tempesta]]. Here we use the ~ notation to stand for the non-lemma form in the destination link. A more complex example is [[se non hai altri moccoli puoi andare a letto al buio]], which becomes [[se]] [[non]] [[hai]] [[altri]] [[moccoli]] [[puoi]] [[andare]] [[a]] [[letto]] [[al]] [[buio]] after calling add_links_to_multiword_term(). With the following modifier_spec: 'hai:avere; altr[i:o]; moccol[i:o]; puoi: potere; andare a letto:~; al buio:~', the result of applying the spec is [[se]] [[non]] [[avere|hai]] [[altro|altri]] [[moccolo|moccoli]] [[potere|puoi]] [[andare a letto]] [[al buio]]. Here, we rely on the alternative notation mentioned above for e.g. 'altr[i:o]', which is equivalent to 'altri:altro', and link multiword subterms using e.g. 'andare a letto:~'. (The code knows how to handle multiword subexpressions properly, and if the link text and destination are the same, only a single-part link is formed.) ]==] function export.apply_link_modifiers(linked_term, modifier_spec, lang) local split_modspecs = split(modifier_spec, "%s*;%s*") for j, modspec in ipairs(split_modspecs) do local id if modspec:find("<") then local rest rest, id = modspec:match("^(.*)<id:(.-)>$") if rest then modspec = rest end end local subterm, dest, otherlang local begin_spec, rest, end_spec = modspec:match("^%[(.-)%]([^:]*)%[(.-)%]$") if begin_spec then local begin_from, begin_to = canonicalize_begin_end_spec(begin_spec) local end_from, end_to = canonicalize_begin_end_spec(end_spec) subterm = begin_from .. rest .. end_from dest = begin_to .. rest .. end_to end if not subterm then rest, end_spec = modspec:match("^([^:]*)%[(.-)%]$") if rest then local end_from, end_to = canonicalize_begin_end_spec(end_spec) subterm = rest .. end_from dest = rest .. end_to end end if not subterm then begin_spec, rest = modspec:match("^%[(.-)%]([^:]*)$") if begin_spec then local begin_from, begin_to = canonicalize_begin_end_spec(begin_spec) subterm = begin_from .. rest dest = begin_to .. rest end end if not subterm then subterm, dest = modspec:match("^(.-)%s*:%s*(.*)$") if subterm and subterm ~= "^" and subterm ~= "$" then local langdest -- Parse off an initial language code (e.g. 'en:Higgs', 'la:minūtia' or 'grc:σκατός'). Also handle -- Wikipedia prefixes ('w:Abatemarco' or 'w:it:Colle Val d'Elsa'). otherlang, langdest = dest:match("^([A-Za-z0-9._-]+):([^ ].*)$") if otherlang == "w" then local foreign_wikipedia, foreign_term = langdest:match("^([A-Za-z0-9._-]+):([^ ].*)$") if foreign_wikipedia then otherlang = otherlang .. ":" .. foreign_wikipedia langdest = foreign_term end dest = ("%s:%s"):format(otherlang, langdest) otherlang = nil elseif otherlang then otherlang = get_lang(otherlang, true, "allow etym") dest = langdest end end end if not subterm then if modspec == "?" or modspec == "!" then subterm = "$" dest = modspec elseif modspec == "..." or modspec == "...?" then subterm = "$" dest = " " .. modspec elseif modspec:find("^[A-Z]$") then -- X, Y, etc. by themselves are unlinked, to help with snowclones subterm = modspec dest = "_" else subterm = modspec dest = "~" end end if subterm == "^" then linked_term = dest:gsub("_", " ") .. linked_term elseif subterm == "$" then linked_term = linked_term .. dest:gsub("_", " ") else if subterm:find("[", nil, true) then error(("Subterm '%s' in modifier spec '%s' cannot have brackets in it"):format( escape_wikicode(subterm), escape_wikicode(modspec))) end local escaped_subterm = pattern_escape(subterm) local subterm_re = "%[%[" .. escaped_subterm:gsub("(%%?[ ',%-])", "%%]*%1%%[*") .. "%]%]" local expanded_dest if dest:find("~", nil, true) then expanded_dest = dest:gsub("~", replacement_escape(subterm)) else expanded_dest = dest end if otherlang then expanded_dest = expanded_dest .. "#" .. otherlang:getCanonicalName() end local subterm_replacement if expanded_dest == "_" then subterm_replacement = subterm if id then error("Can't supply <id:...> with an unlinked subterm") end if otherlang then error("Can't supply prefixed language with an unlinked subterm") end elseif id or otherlang then if id and expanded_dest:find("[", nil, true) then error("Can't supply <id:...> with destination with embedded brackets") end subterm_replacement = require(links_module).language_link { lang = otherlang or lang, term = expanded_dest, alt = subterm, id = id, } elseif expanded_dest:find("[", nil, true) then -- Use the destination directly if it has brackets in it (e.g. to put brackets around parts of a word). subterm_replacement = expanded_dest elseif expanded_dest == subterm then subterm_replacement = "[[" .. subterm .. "]]" else subterm_replacement = "[[" .. expanded_dest .. "|" .. subterm .. "]]" end local escaped_subterm_replacement = replacement_escape(subterm_replacement) local replaced_linked_term = ugsub(linked_term, subterm_re, escaped_subterm_replacement) if replaced_linked_term == linked_term then mw.log(("Attempted to replace %s with %s in %s"):format(subterm_re, escaped_subterm_replacement, linked_term)) error(("Subterm '%s' could not be located in %slinked expression %s, or replacement same as subterm"):format( subterm, j > 1 and "intermediate " or "", escape_wikicode(linked_term))) else linked_term = replaced_linked_term end end end return linked_term end return export 8uip0he2epyv5rdmi0wwc1j9o7grekr 10 18091 89201 2026-04-15T09:13:21Z en>Surjection 0 Changed protection settings for "[[Template:dercat]]": Highly visible template/module ([Edit=Allow only autopatrollers] (indefinite) [Move=Allow only autopatrollers] (indefinite)) 89201 wikitext text/x-wiki <includeonly>{{#invoke:dercat|dercat}}</includeonly><noinclude>{{documentation}}</noinclude> 5cd5b1zpnezetgb5ibnpc1g49queo4o 89202 89201 2026-09-02T18:53:47Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Template:dercat]] 89201 wikitext text/x-wiki <includeonly>{{#invoke:dercat|dercat}}</includeonly><noinclude>{{documentation}}</noinclude> 5cd5b1zpnezetgb5ibnpc1g49queo4o 828 18092 89203 2025-12-03T00:04:59Z en>Saph 0 generate actual categories for bad audios, much more accessible for the avg user 89203 Scribunto text/plain local export = {} local headword_data_module = "Module:headword/data" local IPA_module = "Module:IPA" local labels_module = "Module:labels" local links_module = "Module:links" local parameters_module = "Module:parameters" local qualifier_module = "Module:qualifier" local references_module = "Module:references" local string_utilities_module = "Module:string utilities" local table_module = "Module:table" local template_styles_module = "Module:TemplateStyles" local utilities_module = "Module:utilities" local audio_styles_css = "audio/styles.css" local function track(page) require("Module:debug/track")("audio/" .. page) return true end local function wrap_qualifier_css(text, suffix) return require(qualifier_module).wrap_qualifier_css(text, suffix) end --[==[ Display a box that can be used to play an audio file. `data` is a table containing the following fields: * `lang` ('''required'''): language object for the audio files; * `file` ('''required'''): file containing the audio; * `caption`: Caption to display before the audio box; normally {"Audio"}, and does not usually need to be changed; * `nocaption`: If specified, don't display the caption; * `q`: {nil} or a list of left regular qualifier strings, formatted using {format_qualifier()} in [[Module:qualifier]] and displayed before the audio box and after the caption (and any accent qualifiers); * `qq`: {nil} or a list of right regular qualifier strings, displayed directly after the audio box (and after any accent qualifiers); * `a`: {nil} or a list of left accent qualifier strings, formatted using {format_qualifiers()} in [[Module:accent qualifier]] and displayed before the audio box and after the caption; * `aa`: {nil} or a list of right accent qualifier strings, displayed directly after the homophone in question; * `refs`: {nil} or a list of references or reference specs to add directly after the audio box; the value of a list item is either a string containing the reference text (typically a call to a citation template such as {{tl|cite-book}}, or a template wrapping such a call), or an object with fields `text` (the reference text), `name` (the name of the reference, as in {{cd|<nowiki><ref name="foo">...</ref></nowiki>}} or {{cd|<nowiki><ref name="foo" /></nowiki>}}) and/or `group` (the group of the reference, as in {{cd|<nowiki><ref name="foo" group="bar">...</ref></nowiki>}} or {{cd|<nowiki><ref name="foo" group="bar"/></nowiki>}}); this uses a parser function to format the reference appropriately and insert a footnote number that hyperlinks to the actual reference, located in the {{cd|<nowiki><references /></nowiki>}} section; * `text`: Text of the audio snippet; if specified, should be an object of the form passed to {full_link()} in [[Module:links]], including a `lang` field containing the language of the text (usually the same as `data.lang`); displayed before the audio box, after any regular and accent qualifiers; * `IPA`: IPA of the audio snippet, or a list of IPA specs; if specified, should be surrounded by slashes or brackets, and will be processed using {format_IPA_multiple()} in [[Module:IPA]] and displayed before the audio box, after any regular and accent qualifiers and after the text of the audio snippet, if given; * `nocat`: If true, suppress categorization; * `sort`: Sort key for categorization. ]==] function export.format_audio(data) local langname = data.lang:getFullName() local cats = { langname .. " terms with audio pronunciation" } local function format_a(a) if a and a[1] then return require(labels_module).show_labels { lang = data.lang, labels = a, mode = "accent", nocat = true, open = false, close = false, no_track_already_seen = true, } end return nil end local function format_q(q) if q and q[1] then return require(qualifier_module).format_qualifier(q, false, false) end return nil end local function make_td_if(text) if text == "" then return text end return "<td>" .. text .. "</td>" end -- Generate the full text preceding the audio box. local pretext_parts = {} local function ins(text) table.insert(pretext_parts, text) end local formatted_accent_labels, formatted_qualifiers, formatted_text, formatted_ipa formatted_accent_labels = format_a(data.a) formatted_qualifiers = format_q(data.q) if data.text then formatted_text = require(links_module).full_link(data.text, "term", true) end if data.IPA then local ipa_cats local ipa = data.IPA if type(ipa) == "string" then ipa = {ipa} end local ipa_items = {} for _, ipa_item in ipairs(ipa) do table.insert(ipa_items, {pron = ipa_item}) end formatted_ipa, ipa_cats = require(IPA_module).format_IPA_multiple(data.lang, ipa_items, nil, "no count", "raw") if ipa_cats[1] then require(table_module).extend(cats, ipa_cats) end end local has_qual = formatted_accent_labels or formatted_qualifiers if not data.nocaption then -- Track uses of caption (3=). Over time as we eliminate most of them, we can use this to find and -- eliminate the remainder. if data.caption then track("caption") end ins(data.caption or "Audio") if has_qual then ins(" " .. wrap_qualifier_css("(", "brac")) end end if formatted_accent_labels then ins(formatted_accent_labels) if formatted_qualifiers then ins(wrap_qualifier_css(",", "comma") .. " ") end end if formatted_qualifiers then ins(formatted_qualifiers) end if has_qual then if not data.nocaption then ins(wrap_qualifier_css(")", "brac")) end end if (formatted_text or formatted_ipa) and (has_qual or not data.nocaption) then ins(wrap_qualifier_css(";", "semicolon") .. " ") end if formatted_text then ins(formatted_text) if formatted_ipa then ins(" ") end end ins(formatted_ipa) if not data.nocaption then ins(wrap_qualifier_css(":", "colon")) end local pretext = make_td_if(table.concat(pretext_parts)) -- Generate the full text following the audio box. local posttext_parts = {} local function ins(text) table.insert(posttext_parts, text) end local formatted_post_accent_labels = format_a(data.aa) local formatted_post_qualifiers = format_q(data.qq) local formatted_references = data.refs and require(references_module).format_references(data.refs) or nil if formatted_references then ins(formatted_references) end if formatted_post_accent_labels or formatted_post_qualifiers then if formatted_references then ins(" ") end ins(wrap_qualifier_css("(", "brac")) if formatted_post_accent_labels then ins(formatted_post_accent_labels) if formatted_post_qualifiers then ins(wrap_qualifier_css(",", "comma") .. " ") end end if formatted_post_qualifiers then ins(formatted_post_qualifiers) end ins(wrap_qualifier_css(")", "brac")) end if data.bad then table.insert(cats, langname .. " terms with nonstandard or incorrect audio pronunciations") ins(" " .. require(qualifier_module).wrap_css("Note: this pronunciation may be nonstandard or incorrect: " .. data.bad, "bad-audio-note")) end local posttext = make_td_if(table.concat(posttext_parts)) local template = [=[ <tr>%s<td class="audiofile">[[File:%s|noicon|175px]]</td><td class="audiometa" style="font-size: 80%%;">([[:File:%s|file]])</td>%s</tr>]=] local text = template:format(pretext, data.file, data.file, posttext) text = '<table class="audiotable" style="vertical-align: middle; display: inline-block; list-style: none; line-height: 1em; border-collapse: collapse; margin: 0;">' .. text .. "</table>" local stylesheet = require(template_styles_module)(audio_styles_css) local categories = data.nocat and "" or cats[1] and require(utilities_module).format_categories(cats, data.lang, data.sort) or "" return stylesheet .. text .. categories end --[==[ FIXME: Old entry point for formatting multiple audios in a single table. Not used anywhere and needs rewriting to the standard of format_audio(). Meant to be called from a module. `data` is a table containing the following fields: <pre> { lang = LANGUAGE_OBJECT, audios = {{file = "FILENAME", qualifiers = nil or {"QUALIFIER", "QUALIFIER", ...}}, ...}, caption = nil or "CAPTION" } </pre> Here: * `lang` is a language object. * `audios` is the list of audio files to display. FILENAME is the name of the audio file without a namespace. QUALIFIER is a qualifier string to display after the specific audio file in question, formatted using {format_qualifier()} in [[Module:qualifier]]. * `caption`, if specified, adds a caption before the audio file. ]==] function export.format_multiple_audios(data) local audiocats = { data.lang:getFullName() .. " terms with audio pronunciation" } local rows = { } local caption = data.caption for _, audio in ipairs(data.audios) do local qualifiers = audio.qualifiers local function repl(key) if key == "file" then return audio.file elseif key == "caption" then if not caption then return "" end return "<td rowspan=" .. #data.audios .. ">" .. caption .. ":</td>" elseif key == "qualifiers" then if not qualifiers or not qualifiers[1] then return "" end return "<td>" .. require(qualifier_module).format_qualifier(qualifiers) .. "</td>" end end local template = [=[ <tr>{{{caption}}} <td class="audiofile">[[File:{{{file}}}|noicon|175px]]</td> <td class="audiometa" style="font-size: 80%;">([[:File:{{{file}}}|file]])</td> {{{qualifiers}}}</tr>]=] local text = (mw.ustring.gsub(template, "{{{([a-z0-9_:]+)}}}", repl)) table.insert(rows, text) caption = nil end local function repl(key) if key == "rows" then return table.concat(rows, "\n") end end local template = [=[ <table class="audiotable" style="vertical-align: middle; display: inline-block; list-style: none; line-height: 1em; border-collapse: collapse;"> {{{rows}}} </table> ]=] local stylesheet = require(template_styles_module)(audio_styles_css) local text = mw.ustring.gsub(template, "{{{([a-z0-9_:]+)}}}", repl) local categories = data.nocat and "" or #audiocats > 0 and require(utilities_module).format_categories(audiocats, data.lang, data.sort) or "" -- remove newlines due to HTML generator bug in MediaWiki(?) - newlines in tables cause list items to not end correctly text = mw.ustring.gsub(text, "\n", "") return stylesheet .. text .. categories end --[==[ Construct the `text` object passed into {format_audio()}, from raw-ish arguments (essentially, the output of {process()} in [[Module:parameters]]). On entry, `args` contains the following fields: * `lang` ('''required'''): Language object. * `text`: Text. If this isn't defined and neither are any of `gloss`, `tr`, `ts`, `pos`, `lit` or `genders`, the function returns {nil}. * `gloss`: Gloss of text. * `tr`: Manual transliteration of text. * `ts`: Transcription of text. * `pos`: Part of speech of text. * `lit`: Literal meaning of text. * `genders`: List of gender/number spec(s) of text. * `sc`: Optional script object of text (rarely needs to be set). * `pagename`: Pagename; used in place of `text` when `text` is unset but other text-related parameters are set. If not specified, taken from the actual pagename. ]==] function export.construct_audio_textobj(args) local textobj if args.text or args.gloss or args.tr or args.ts or args.pos or args.lit or args.genders and args.genders[1] then local text = args.text or args.pagename or mw.loadData("Module:headword/data").pagename textobj = { lang = args.lang, alt = wrap_qualifier_css("“", "quote") .. text .. wrap_qualifier_css("”", "quote"), gloss = args.gloss, tr = args.tr, ts = args.ts, pos = args.pos, lit = args.lit, genders = args.genders, sc = args.sc, } end return textobj end --[==[ Entry point for {{tl|audio}} template. ]==] function export.show(frame) local parent_args = frame:getParent().args local compat = parent_args.lang local offset = compat and 0 or 1 local params = { [compat and "lang" or 1] = {required = true, type = "language", default = "en"}, [1 + offset] = {required = true, default = "Example.ogg"}, [2 + offset] = {}, ["q"] = {type = "qualifier"}, ["qq"] = {type = "qualifier"}, ["a"] = {type = "labels"}, ["aa"] = {type = "labels"}, ["ref"] = {type = "references"}, ["IPA"] = {sublist = true}, ["text"] = {}, ["t"] = {}, ["gloss"] = {alias_of = "t"}, ["tr"] = {}, ["ts"] = {}, ["pos"] = {}, ["lit"] = {}, ["g"] = {sublist = true}, ["sc"] = {type = "script"}, ["bad"] = {}, ["nocat"] = {type = "boolean"}, ["sort"] = {}, ["pagename"] = {}, } local args = require(parameters_module).process(parent_args, params) local lang = args[compat and "lang" or 1] -- Needed in construct_audio_textobj(). args.lang = lang local textobj = export.construct_audio_textobj(args) local caption = args[2 + offset] local nocaption if caption == "-" then caption = nil nocaption = true end if caption then -- Remove final colon if given, to avoid two colons. caption = caption:gsub(":$", "") end local data = { lang = lang, file = args[1 + offset], caption = caption, nocaption = nocaption, q = args.q, qq = args.qq, a = args.a, aa = args.aa, refs = args.ref, text = textobj, IPA = args.IPA, bad = args.bad, nocat = args.nocat, sort = args.sort, } return export.format_audio(data) end return export akem8fdq9thdv6zx89zo3kwaky8a452 89204 89203 2026-09-02T18:54:21Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Module:audio]] 89203 Scribunto text/plain local export = {} local headword_data_module = "Module:headword/data" local IPA_module = "Module:IPA" local labels_module = "Module:labels" local links_module = "Module:links" local parameters_module = "Module:parameters" local qualifier_module = "Module:qualifier" local references_module = "Module:references" local string_utilities_module = "Module:string utilities" local table_module = "Module:table" local template_styles_module = "Module:TemplateStyles" local utilities_module = "Module:utilities" local audio_styles_css = "audio/styles.css" local function track(page) require("Module:debug/track")("audio/" .. page) return true end local function wrap_qualifier_css(text, suffix) return require(qualifier_module).wrap_qualifier_css(text, suffix) end --[==[ Display a box that can be used to play an audio file. `data` is a table containing the following fields: * `lang` ('''required'''): language object for the audio files; * `file` ('''required'''): file containing the audio; * `caption`: Caption to display before the audio box; normally {"Audio"}, and does not usually need to be changed; * `nocaption`: If specified, don't display the caption; * `q`: {nil} or a list of left regular qualifier strings, formatted using {format_qualifier()} in [[Module:qualifier]] and displayed before the audio box and after the caption (and any accent qualifiers); * `qq`: {nil} or a list of right regular qualifier strings, displayed directly after the audio box (and after any accent qualifiers); * `a`: {nil} or a list of left accent qualifier strings, formatted using {format_qualifiers()} in [[Module:accent qualifier]] and displayed before the audio box and after the caption; * `aa`: {nil} or a list of right accent qualifier strings, displayed directly after the homophone in question; * `refs`: {nil} or a list of references or reference specs to add directly after the audio box; the value of a list item is either a string containing the reference text (typically a call to a citation template such as {{tl|cite-book}}, or a template wrapping such a call), or an object with fields `text` (the reference text), `name` (the name of the reference, as in {{cd|<nowiki><ref name="foo">...</ref></nowiki>}} or {{cd|<nowiki><ref name="foo" /></nowiki>}}) and/or `group` (the group of the reference, as in {{cd|<nowiki><ref name="foo" group="bar">...</ref></nowiki>}} or {{cd|<nowiki><ref name="foo" group="bar"/></nowiki>}}); this uses a parser function to format the reference appropriately and insert a footnote number that hyperlinks to the actual reference, located in the {{cd|<nowiki><references /></nowiki>}} section; * `text`: Text of the audio snippet; if specified, should be an object of the form passed to {full_link()} in [[Module:links]], including a `lang` field containing the language of the text (usually the same as `data.lang`); displayed before the audio box, after any regular and accent qualifiers; * `IPA`: IPA of the audio snippet, or a list of IPA specs; if specified, should be surrounded by slashes or brackets, and will be processed using {format_IPA_multiple()} in [[Module:IPA]] and displayed before the audio box, after any regular and accent qualifiers and after the text of the audio snippet, if given; * `nocat`: If true, suppress categorization; * `sort`: Sort key for categorization. ]==] function export.format_audio(data) local langname = data.lang:getFullName() local cats = { langname .. " terms with audio pronunciation" } local function format_a(a) if a and a[1] then return require(labels_module).show_labels { lang = data.lang, labels = a, mode = "accent", nocat = true, open = false, close = false, no_track_already_seen = true, } end return nil end local function format_q(q) if q and q[1] then return require(qualifier_module).format_qualifier(q, false, false) end return nil end local function make_td_if(text) if text == "" then return text end return "<td>" .. text .. "</td>" end -- Generate the full text preceding the audio box. local pretext_parts = {} local function ins(text) table.insert(pretext_parts, text) end local formatted_accent_labels, formatted_qualifiers, formatted_text, formatted_ipa formatted_accent_labels = format_a(data.a) formatted_qualifiers = format_q(data.q) if data.text then formatted_text = require(links_module).full_link(data.text, "term", true) end if data.IPA then local ipa_cats local ipa = data.IPA if type(ipa) == "string" then ipa = {ipa} end local ipa_items = {} for _, ipa_item in ipairs(ipa) do table.insert(ipa_items, {pron = ipa_item}) end formatted_ipa, ipa_cats = require(IPA_module).format_IPA_multiple(data.lang, ipa_items, nil, "no count", "raw") if ipa_cats[1] then require(table_module).extend(cats, ipa_cats) end end local has_qual = formatted_accent_labels or formatted_qualifiers if not data.nocaption then -- Track uses of caption (3=). Over time as we eliminate most of them, we can use this to find and -- eliminate the remainder. if data.caption then track("caption") end ins(data.caption or "Audio") if has_qual then ins(" " .. wrap_qualifier_css("(", "brac")) end end if formatted_accent_labels then ins(formatted_accent_labels) if formatted_qualifiers then ins(wrap_qualifier_css(",", "comma") .. " ") end end if formatted_qualifiers then ins(formatted_qualifiers) end if has_qual then if not data.nocaption then ins(wrap_qualifier_css(")", "brac")) end end if (formatted_text or formatted_ipa) and (has_qual or not data.nocaption) then ins(wrap_qualifier_css(";", "semicolon") .. " ") end if formatted_text then ins(formatted_text) if formatted_ipa then ins(" ") end end ins(formatted_ipa) if not data.nocaption then ins(wrap_qualifier_css(":", "colon")) end local pretext = make_td_if(table.concat(pretext_parts)) -- Generate the full text following the audio box. local posttext_parts = {} local function ins(text) table.insert(posttext_parts, text) end local formatted_post_accent_labels = format_a(data.aa) local formatted_post_qualifiers = format_q(data.qq) local formatted_references = data.refs and require(references_module).format_references(data.refs) or nil if formatted_references then ins(formatted_references) end if formatted_post_accent_labels or formatted_post_qualifiers then if formatted_references then ins(" ") end ins(wrap_qualifier_css("(", "brac")) if formatted_post_accent_labels then ins(formatted_post_accent_labels) if formatted_post_qualifiers then ins(wrap_qualifier_css(",", "comma") .. " ") end end if formatted_post_qualifiers then ins(formatted_post_qualifiers) end ins(wrap_qualifier_css(")", "brac")) end if data.bad then table.insert(cats, langname .. " terms with nonstandard or incorrect audio pronunciations") ins(" " .. require(qualifier_module).wrap_css("Note: this pronunciation may be nonstandard or incorrect: " .. data.bad, "bad-audio-note")) end local posttext = make_td_if(table.concat(posttext_parts)) local template = [=[ <tr>%s<td class="audiofile">[[File:%s|noicon|175px]]</td><td class="audiometa" style="font-size: 80%%;">([[:File:%s|file]])</td>%s</tr>]=] local text = template:format(pretext, data.file, data.file, posttext) text = '<table class="audiotable" style="vertical-align: middle; display: inline-block; list-style: none; line-height: 1em; border-collapse: collapse; margin: 0;">' .. text .. "</table>" local stylesheet = require(template_styles_module)(audio_styles_css) local categories = data.nocat and "" or cats[1] and require(utilities_module).format_categories(cats, data.lang, data.sort) or "" return stylesheet .. text .. categories end --[==[ FIXME: Old entry point for formatting multiple audios in a single table. Not used anywhere and needs rewriting to the standard of format_audio(). Meant to be called from a module. `data` is a table containing the following fields: <pre> { lang = LANGUAGE_OBJECT, audios = {{file = "FILENAME", qualifiers = nil or {"QUALIFIER", "QUALIFIER", ...}}, ...}, caption = nil or "CAPTION" } </pre> Here: * `lang` is a language object. * `audios` is the list of audio files to display. FILENAME is the name of the audio file without a namespace. QUALIFIER is a qualifier string to display after the specific audio file in question, formatted using {format_qualifier()} in [[Module:qualifier]]. * `caption`, if specified, adds a caption before the audio file. ]==] function export.format_multiple_audios(data) local audiocats = { data.lang:getFullName() .. " terms with audio pronunciation" } local rows = { } local caption = data.caption for _, audio in ipairs(data.audios) do local qualifiers = audio.qualifiers local function repl(key) if key == "file" then return audio.file elseif key == "caption" then if not caption then return "" end return "<td rowspan=" .. #data.audios .. ">" .. caption .. ":</td>" elseif key == "qualifiers" then if not qualifiers or not qualifiers[1] then return "" end return "<td>" .. require(qualifier_module).format_qualifier(qualifiers) .. "</td>" end end local template = [=[ <tr>{{{caption}}} <td class="audiofile">[[File:{{{file}}}|noicon|175px]]</td> <td class="audiometa" style="font-size: 80%;">([[:File:{{{file}}}|file]])</td> {{{qualifiers}}}</tr>]=] local text = (mw.ustring.gsub(template, "{{{([a-z0-9_:]+)}}}", repl)) table.insert(rows, text) caption = nil end local function repl(key) if key == "rows" then return table.concat(rows, "\n") end end local template = [=[ <table class="audiotable" style="vertical-align: middle; display: inline-block; list-style: none; line-height: 1em; border-collapse: collapse;"> {{{rows}}} </table> ]=] local stylesheet = require(template_styles_module)(audio_styles_css) local text = mw.ustring.gsub(template, "{{{([a-z0-9_:]+)}}}", repl) local categories = data.nocat and "" or #audiocats > 0 and require(utilities_module).format_categories(audiocats, data.lang, data.sort) or "" -- remove newlines due to HTML generator bug in MediaWiki(?) - newlines in tables cause list items to not end correctly text = mw.ustring.gsub(text, "\n", "") return stylesheet .. text .. categories end --[==[ Construct the `text` object passed into {format_audio()}, from raw-ish arguments (essentially, the output of {process()} in [[Module:parameters]]). On entry, `args` contains the following fields: * `lang` ('''required'''): Language object. * `text`: Text. If this isn't defined and neither are any of `gloss`, `tr`, `ts`, `pos`, `lit` or `genders`, the function returns {nil}. * `gloss`: Gloss of text. * `tr`: Manual transliteration of text. * `ts`: Transcription of text. * `pos`: Part of speech of text. * `lit`: Literal meaning of text. * `genders`: List of gender/number spec(s) of text. * `sc`: Optional script object of text (rarely needs to be set). * `pagename`: Pagename; used in place of `text` when `text` is unset but other text-related parameters are set. If not specified, taken from the actual pagename. ]==] function export.construct_audio_textobj(args) local textobj if args.text or args.gloss or args.tr or args.ts or args.pos or args.lit or args.genders and args.genders[1] then local text = args.text or args.pagename or mw.loadData("Module:headword/data").pagename textobj = { lang = args.lang, alt = wrap_qualifier_css("“", "quote") .. text .. wrap_qualifier_css("”", "quote"), gloss = args.gloss, tr = args.tr, ts = args.ts, pos = args.pos, lit = args.lit, genders = args.genders, sc = args.sc, } end return textobj end --[==[ Entry point for {{tl|audio}} template. ]==] function export.show(frame) local parent_args = frame:getParent().args local compat = parent_args.lang local offset = compat and 0 or 1 local params = { [compat and "lang" or 1] = {required = true, type = "language", default = "en"}, [1 + offset] = {required = true, default = "Example.ogg"}, [2 + offset] = {}, ["q"] = {type = "qualifier"}, ["qq"] = {type = "qualifier"}, ["a"] = {type = "labels"}, ["aa"] = {type = "labels"}, ["ref"] = {type = "references"}, ["IPA"] = {sublist = true}, ["text"] = {}, ["t"] = {}, ["gloss"] = {alias_of = "t"}, ["tr"] = {}, ["ts"] = {}, ["pos"] = {}, ["lit"] = {}, ["g"] = {sublist = true}, ["sc"] = {type = "script"}, ["bad"] = {}, ["nocat"] = {type = "boolean"}, ["sort"] = {}, ["pagename"] = {}, } local args = require(parameters_module).process(parent_args, params) local lang = args[compat and "lang" or 1] -- Needed in construct_audio_textobj(). args.lang = lang local textobj = export.construct_audio_textobj(args) local caption = args[2 + offset] local nocaption if caption == "-" then caption = nil nocaption = true end if caption then -- Remove final colon if given, to avoid two colons. caption = caption:gsub(":$", "") end local data = { lang = lang, file = args[1 + offset], caption = caption, nocaption = nocaption, q = args.q, qq = args.qq, a = args.a, aa = args.aa, refs = args.ref, text = textobj, IPA = args.IPA, bad = args.bad, nocat = args.nocat, sort = args.sort, } return export.format_audio(data) end return export akem8fdq9thdv6zx89zo3kwaky8a452 10 18093 89205 2026-04-15T01:41:33Z en>Polomo 0 Changed protection settings for "[[Template:etymon]]": Highly visible template/module ([Edit=Allow only autopatrollers] (indefinite) [Move=Allow only template editors and administrators] (indefinite)) 89205 wikitext text/x-wiki <includeonly>{{#invoke:etymon|main}}</includeonly><noinclude>{{documentation}}</noinclude> 5t3xwy26hk5cfvmzk7ny2oqnf3fvdz7 89206 89205 2026-09-02T18:54:31Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Template:etymon]] 89205 wikitext text/x-wiki <includeonly>{{#invoke:etymon|main}}</includeonly><noinclude>{{documentation}}</noinclude> 5t3xwy26hk5cfvmzk7ny2oqnf3fvdz7 10 18094 89207 2026-08-05T15:06:48Z en>Tc14Hd 0 Use {{word}} instead of {{catlangname}} and add id= parameter 89207 wikitext text/x-wiki {{#invoke:checkparams|error}}<!-- Validate template parameters -->{{#if:{{{nobox|}}}||<table class="wikitable floatright PIE-word" style="text-align: center;"><!-- --><tr><th>PIE word</th></tr><!-- --><tr><td>{{l|ine-pro|*{{{2|<noinclude>deywós</noinclude>}}}|id={{{id|}}}}}</td></tr><!-- --></table><!-- -->}}{{word|{{{1|<noinclude>und</noinclude>}}}|ine|{{{2|<noinclude>deywós</noinclude>}}}|id={{{id|}}}}}<!-- --><noinclude>{{documentation}}</noinclude> 6yesa77e8h00ekdsroy4xn84xiag8vm 89208 89207 2026-09-02T18:54:38Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Template:PIE_word]] 89207 wikitext text/x-wiki {{#invoke:checkparams|error}}<!-- Validate template parameters -->{{#if:{{{nobox|}}}||<table class="wikitable floatright PIE-word" style="text-align: center;"><!-- --><tr><th>PIE word</th></tr><!-- --><tr><td>{{l|ine-pro|*{{{2|<noinclude>deywós</noinclude>}}}|id={{{id|}}}}}</td></tr><!-- --></table><!-- -->}}{{word|{{{1|<noinclude>und</noinclude>}}}|ine|{{{2|<noinclude>deywós</noinclude>}}}|id={{{id|}}}}}<!-- --><noinclude>{{documentation}}</noinclude> 6yesa77e8h00ekdsroy4xn84xiag8vm 10 18095 89210 2026-08-06T22:50:51Z en>Surjection 0 Protected "[[Template:word]]": Highly visible template/module ([Edit=Allow only autopatrollers] (indefinite) [Move=Allow only autopatrollers] (indefinite)) 89210 wikitext text/x-wiki {{#invoke:checkparams|error}}<!-- Validate template parameters --><includeonly><!-- -->{{dercat|{{{1}}}|{{{2}}}-pro}}<!-- -->{{catlangname|{{{1}}}|terms derived from the {{langname|{{{2}}}-pro}} word *{{{3}}} {{#if:{{{id|}}}|({{{id|}}})}}}}<!-- --></includeonly><noinclude>{{documentation}}</noinclude> m10f212p7gp92lt2izg32uef1w2spnu 89211 89210 2026-09-02T18:55:08Z Altronic 4137 Chaidh 1 mhùthadh ion-phortachadh o [[:en:Template:word]] 89210 wikitext text/x-wiki {{#invoke:checkparams|error}}<!-- Validate template parameters --><includeonly><!-- -->{{dercat|{{{1}}}|{{{2}}}-pro}}<!-- -->{{catlangname|{{{1}}}|terms derived from the {{langname|{{{2}}}-pro}} word *{{{3}}} {{#if:{{{id|}}}|({{{id|}}})}}}}<!-- --></includeonly><noinclude>{{documentation}}</noinclude> m10f212p7gp92lt2izg32uef1w2spnu 828 18096 89212 2026-08-31T08:14:55Z en>Surjection 0 89212 Scribunto text/plain --[=[ This module implements the {{etymon}} template for structured etymology data on Wiktionary. It enables the creation of etymology trees and text by parsing etymon chains, scraping linked pages for their own {{etymon}} data, and recursively building a tree of derivational relationships. Authors: - Original implementation: [[User:Ioaxxere]] - Full refactor (September 2025): [[User:Fenakhay]] ([[Special:Diff/86717746]]) Modules: - [[Module:etymon]]: main module handling parsing, validation, tree building, and page scraping - [[Module:etymon/data]]: keyword definitions, configuration, and status constants - [[Module:etymon/tree]]: etymology tree rendering - [[Module:etymon/text]]: etymology text generation - [[Module:etymon/categories]]: category generation logic - [[Module:etymon/tracking]]: tracking ]=] local export = {} local __state = { cached_etymon_args = {}, cached_etymon_pages = {}, cached_descendants_checks = {}, senseid_parent_etymon = {}, available_etymon_ids = {}, single_etymons = {}, entry_title = nil, entry_lang_code = nil, current_page_has_inline_etymology = false, current_page_has_redundant_etymology = false, used_idless_etymon = false, toplevel_has_inline_etymology = false, toplevel_redundant_etymology = false, toplevel_idless_etymon = false, has_mismatched_id = false, linked_page_multiple_etymons_idless = false, linked_page_partial_etymology_sections = false, partial_etymology_targets = {}, skip_partial_etymology_category = false, max_depth_reached = 0, total_nodes = 0, language_count = {}, toplevel_keyword_stats = {}, id_stats = nil, warnings = {}, } local function reset_invocation_state() __state.current_page_has_inline_etymology = false __state.current_page_has_redundant_etymology = false __state.used_idless_etymon = false __state.toplevel_has_inline_etymology = false __state.toplevel_redundant_etymology = false __state.toplevel_idless_etymon = false __state.has_mismatched_id = false __state.linked_page_multiple_etymons_idless = false __state.linked_page_partial_etymology_sections = false __state.max_depth_reached = 0 __state.total_nodes = 0 __state.language_count = {} __state.toplevel_keyword_stats = {} __state.warnings = {} end local M = require("Module:module loader").init({ require = { data = "Module:etymon/data", tree = "Module:etymon/tree", text = "Module:etymon/text", categories = "Module:etymon/categories", tracking = "Module:etymon/tracking", descendants = "Module:etymon/descendants", anchors = "Module:anchors", etydate = "Module:etydate", etymology = "Module:etymology", families = "Module:families", languages = "Module:languages", languages_errorgetby = "Module:languages/errorGetBy", links = "Module:links", pages = "Module:pages", parameters = "Module:parameters", string_utilities = "Module:string utilities", template_parser = "Module:template parser", utilities = "Module:utilities", debug = "Module:debug", en_utilities = "Module:en-utilities", parse_utilities = "Module:parse utilities", references = "Module:references", template_styles = "Module:TemplateStyles", script_utilities = "Module:script utilities", JSON = "Module:JSON", yesno = "Module:yesno", }, loadData = { headword_data = "Module:headword/data", parameters_data = "Module:parameters/data", text_allowed = "Module:etymon/data/text_allowed", }, }) local Util = {} function Util.format_error(message, preview_only) if preview_only and not M.pages.is_preview() then return nil end return '<span class="error">' .. message .. '</span>' end function Util.add_warning(message, preview_only) local formatted = Util.format_error(message, preview_only) if formatted then table.insert(__state.warnings, formatted) end end function Util.is_text_param_allowed_for_lang(lang) if not lang or type(lang) ~= "table" then return false end local types = lang.getTypes and lang:getTypes() if types and types.family then local code = lang.getCode and lang:getCode() return code and M.text_allowed.families[code] == true end local full_code = lang.getFullCode and lang:getFullCode() if full_code and M.text_allowed.langs[full_code] then return true end if lang.inFamily then for family_code in pairs(M.text_allowed.families) do if lang:inFamily(family_code) then return true end end end return false end function Util.get_lang(code, no_error) if no_error then return M.languages.getByCode(code, nil, true) end return M.languages.getByCode(code, nil, true) or M.languages_errorgetby.code(code, true, true) end -- Match a term language against a text=:lang stop target (supports etymology-only codes). function Util.lang_matches_stop_code(term_lang, stop_code) if not term_lang or not stop_code or stop_code == "" then return false end local stop_lang = Util.get_lang(stop_code, true) if not stop_lang then return false end if term_lang:getCode() == stop_lang:getCode() then return true end if stop_lang:getFullCode() == stop_lang:getCode() then return term_lang:getFullCode() == stop_lang:getCode() end return false end function Util.get_family(code) return M.families.getByCode(code) end function Util.get_lang_exception(lang) -- Families have no language-specific exceptions if lang.getTypes and lang:getTypes().family then return nil end local code = lang:getCode() local lang_exceptions = M.data.config.lang_exceptions if lang_exceptions[code] then return lang_exceptions[code] end for norm_code, exc in pairs(lang_exceptions) do if exc.normalize_to and code == exc.normalize_to then return exc end if exc.normalize_from_families then local should_normalize = false for _, family in ipairs(exc.normalize_from_families) do if lang:inFamily(family) then should_normalize = true break end end if should_normalize and exc.normalize_exclude_families then for _, family in ipairs(exc.normalize_exclude_families) do if lang:inFamily(family) then should_normalize = false break end end end if should_normalize then local ret = {} for k, v in pairs(exc) do ret[k] = v end ret.suppress_tr = nil return ret end end end return nil end function Util.get_norm_lang(lang) local exc = Util.get_lang_exception(lang) if exc and exc.normalize_to then return M.languages.getByCode(exc.normalize_to) end return lang end function Util.resolve_context_lang(lang, node_args) if type(node_args) ~= "table" then return lang end if node_args.status == M.data.STATUS.INLINE then return lang end if not (lang.hasType and lang:hasType("etymology-only")) then return lang end local full = lang.getFull and lang:getFull() if not full or full:getCode() == lang:getCode() then return lang end if full.hasAncestor and full:hasAncestor(lang) then return lang end return full end -- Add default values for boolean modifiers (e.g., <unc> becomes <unc:1>) -- This is needed because Module:parse utilities expects boolean modifiers to have explicit values function Util.add_boolean_defaults(str, param_mods) local result = str for name, spec in pairs(param_mods) do if spec.type == "boolean" then -- Replace <name> with <name:1> (but not <name:...> which already has a value) result = result:gsub("<" .. name .. ">", "<" .. name .. ":1>") end end return result end local REQUEST_TEMPLATE_PARAM_MODS = { rfe = { nocat = { type = "boolean" }, sort = {}, y = {}, m = {}, fragment = {}, section = {}, box = { type = "boolean" }, noes = { type = "boolean" }, }, etystub = { nocat = { type = "boolean" }, sort = {}, nocap = { type = "boolean" }, nodot = { type = "boolean" }, }, } function Util.expand_request_template(frame, template_name, param_value, lang_code) local param_mods = REQUEST_TEMPLATE_PARAM_MODS[template_name] local with_defaults = Util.add_boolean_defaults(param_value, param_mods) local parsed = M.parse_utilities.parse_inline_modifiers(with_defaults, { param_mods = param_mods, generate_obj = function(text) if M.yesno(text, false) then return { is_boolean = true } end return { text = text } end, }) local template_args = { [1] = lang_code } for name in pairs(param_mods) do template_args[name] = parsed[name] end if not parsed.is_boolean then template_args[2] = parsed.text end return " " .. frame:expandTemplate({ title = template_name, args = template_args, }) end -- Centralized term formatting: handles suppress_term (-), unknown_term (empty/+), and regular terms function Util.format_term(term, is_toplevel, opts) opts = opts or {} -- suppress_term (-) returns nil if term.suppress_term then return nil end local lang = term.lang local exc = Util.get_lang_exception(lang) if is_toplevel then local display_text = term.alt or term.title or "" local sc = term.sc or lang:findBestScript(display_text) local bold_text = tostring(mw.html.create("strong") :addClass("selflink") :wikitext(display_text)) return M.script_utilities.tag_text(bold_text, lang, sc, "term") end local link_params = { lang = lang } link_params.term = not term.unknown_term and term.title or nil link_params.alt = term.alt link_params.id = (not term.unknown_term and term.id and term.id ~= "") and term.id or nil if not (exc and exc.suppress_tr) then link_params.tr = term.tr link_params.ts = term.ts else link_params.suppress_tr = true end link_params.lit = (opts.lit ~= "suppress") and term.lit or nil if opts.gloss ~= "suppress" then link_params.gloss = term.t end if term.g and term.g ~= "" then local genders = M.string_utilities.split(term.g, ",") for i = 1, #genders do genders[i] = M.string_utilities.trim(genders[i]) end link_params.genders = genders end if opts.pos ~= "suppress" then link_params.pos = term.pos link_params.ng = term.ng link_params.infl = term.infl end if exc and exc.suppress_tr then link_params.lit = nil end local show_qualifiers if opts.tree_ql ~= "suppress" then if term.q then link_params.q = term.q end if term.qq then link_params.qq = term.qq end if term.l then link_params.l = term.l end if term.ll then link_params.ll = term.ll end show_qualifiers = term.q or term.qq or term.l or term.ll end return M.links.full_link(link_params, "term", nil, show_qualifiers and true or nil) end local __is_content_page_cached function Util.is_content_page() if __is_content_page_cached == nil then __is_content_page_cached = M.pages.is_content_page(mw.title.getCurrentTitle()) end return __is_content_page_cached end local __page_data_cached function Util.get_page_data() if not __page_data_cached then __page_data_cached = M.headword_data.page end return __page_data_cached end -- Extract base keyword from param (without modifiers) local function get_keyword_base(param) if type(param) ~= "string" then return nil end local base = param:match("^:?([^<]+)") or param:gsub("^:", "") return base end local function is_keyword(param, allow_colon_less) if type(param) ~= "string" then return false end local keywords = M.data.keywords if param:sub(1, 1) == ":" then local base = get_keyword_base(param) return keywords[base] ~= nil end if allow_colon_less then local base = get_keyword_base(param) return keywords[base] ~= nil end return false end local function get_keyword(param, allow_colon_less) if type(param) ~= "string" then return nil end local keywords = M.data.keywords if param:sub(1, 1) == ":" then return get_keyword_base(param) end if allow_colon_less then local base = get_keyword_base(param) if keywords[base] then return base end end return nil end local function normalize_keyword(keyword) if keyword:sub(1, 1) == ":" then return keyword end return ":" .. keyword end -- Resolve keyword (possibly an alias) to its canonical form. Used only at input boundaries local function get_canonical_keyword(keyword) if not keyword then return keyword end return M.data.keyword_canonical[keyword] or keyword end local function is_affix_group_keyword(keyword) local config = keyword and M.data.keywords[keyword] return config and config.affix_categories or false end local function reject_removed_surf_keyword(param) local base = get_keyword_base(param) if base == "surf" then error("The `:surf` keyword has been removed. Use `<surf>` on a formation keyword instead (e.g. `:af<surf>`, `:bor<surf>`).") end end local function copy_keyword_info(source) local copy = {} for k, v in pairs(source) do copy[k] = v end return copy end local function lowercase_glossary_display(text) return text:gsub("(%[%[Appendix:Glossary#[^|]+|)([^%]])([^%]]*)%]%]", function(prefix, first, rest) return prefix .. mw.ustring.lower(first) .. rest .. "]]" end) end local function surf_should_keep_formation_phrase(base) if not base.phrase then return false end if base.glossary then return true end return not (base.phrase == "from" and (base.text == "From" or base.text == "from")) end -- Runtime overrides when <surf> is present on a keyword. local function get_effective_keyword_info(keyword, modifiers) local base = M.data.keywords[keyword] if not base or not modifiers or not modifiers.surf then return base end local effective = copy_keyword_info(base) local surf_text = "By [[Appendix:Glossary#surface_analysis|surface analysis]]," local surf_phrase = "by surface analysis," effective.new_sentence = true effective.invisible = "tree" if surf_should_keep_formation_phrase(base) then effective.phrase = surf_phrase .. " " .. base.phrase if base.text then effective.text = surf_text .. " " .. lowercase_glossary_display(base.text) else effective.text = surf_text .. " " .. base.phrase end else effective.text = surf_text effective.phrase = surf_phrase end return effective end -- Build text/phrase for nominalization with <g:code> (uses data module for codes only). local function get_nominalization_label_for_g(code) if not code or code == "" then return nil end local codes = M.data.nominalization_g_codes local adj = codes[code] if not adj and #code == 2 then local gender_adj = codes[code:sub(1, 1)] local number_adj = codes[code:sub(2, 2)] if gender_adj and number_adj then adj = gender_adj .. " " .. number_adj end end if not adj then return nil end local text = adj:gsub("^%l", function(c) return string.upper(c) end) .. " [[Appendix:Glossary#nominalization|nominalization]] of" local phrase = M.en_utilities.add_indefinite_article(adj .. " [[Appendix:Glossary#nominalization|nominalization]] of", false) return { text = text, phrase = phrase } end local EtymonParser = {} -- Keyword modifier definitions EtymonParser.keyword_param_mods = { unc = { type = "boolean" }, ref = {}, text = { restrict = { keywords = { "from", "derived" } } }, lit = { restrict = { affix_group = true } }, conj = {}, -- conjunction for alternatives: "and", "or", "and/or", etc. g = { restrict = { keywords = { "nominalization" } } }, surf = { type = "boolean" }, senseid = { restrict = { keywords = { "semantic loan" } } }, } -- Term modifier definitions EtymonParser.etymon_param_mods = { id = {}, t = {}, tr = {}, ts = {}, q = {}, qq = {}, l = {}, ll = {}, pos = {}, ng = {}, alt = {}, g = {}, infl = { type = "form of tags" }, ety = {}, lit = {}, unc = { type = "boolean" }, ref = {}, aftype = { restrict = { affix_group = true } }, postype = {}, bor = { type = "boolean", restrict = { affix_group = true } }, slbor = { type = "boolean", restrict = { affix_group = true } }, lbor = { type = "boolean", restrict = { affix_group = true } }, } local function get_clean_param_mods(param_mods) local clean = {} for mod_name, mod_def in pairs(param_mods) do clean[mod_name] = {} for key, value in pairs(mod_def) do if key ~= "restrict" then clean[mod_name][key] = value end end end return clean end function EtymonParser.check_modifier_restrictions(modifiers, current_keyword, param_mods) for mod_name, mod_value in pairs(modifiers) do -- Only check restrictions if the modifier has a non-false/nil value if mod_value then local mod_def = param_mods[mod_name] if mod_def and mod_def.restrict then if mod_def.restrict.affix_group then if not is_affix_group_keyword(current_keyword) then local mod_display = mod_value == true and "<" .. mod_name .. ">" or "<" .. mod_name .. ":" .. tostring(mod_value) .. ">" error("The modifier `" .. mod_display .. "` is only allowed for affix-group keywords (e.g. `:af`, `:blend`, `:univ`).") end elseif mod_def.restrict.keywords then local allowed_keywords = mod_def.restrict.keywords local is_allowed = false for _, allowed_keyword in ipairs(allowed_keywords) do if current_keyword == allowed_keyword then is_allowed = true break end end if not is_allowed then local keyword_list = {} for _, kw in ipairs(allowed_keywords) do table.insert(keyword_list, ":" .. kw) end local keyword_str = table.concat(keyword_list, #keyword_list == 2 and " or " or ", ") if #keyword_list > 2 then -- Replace last comma with "or" keyword_str = keyword_str:gsub(", ([^,]+)$", " or %1") end local mod_display = mod_value == true and "<" .. mod_name .. ">" or "<" .. mod_name .. ":" .. tostring(mod_value) .. ">" error("The modifier `" .. mod_display .. "` is only allowed for the keyword" .. (#keyword_list > 1 and "s " or " ") .. keyword_str .. ".") end end end end end end local TERM_RULE_DISALLOW = { suppress = { field = "suppress_term", label = "suppressed" }, unknown = { field = "unknown_term", label = "unknown" }, family = { field = "is_family", label = "family" }, } function EtymonParser.check_etymon_limits(count, limits, label, opts) if not limits then return end opts = opts or {} local min_etymons = limits.min_etymons if min_etymons == nil and not opts.skip_default_min then min_etymons = 1 end if min_etymons and count < min_etymons then if min_etymons > 1 then error("Detected " .. label .. " group with fewer than " .. min_etymons .. " etymons.") else error("Detected " .. label .. " with no etymons.") end end if limits.max_etymons and count > limits.max_etymons then local unit = (limits.max_etymons == 1) and "etymon" or "etymons" error("Detected " .. label .. " with more than " .. limits.max_etymons .. " " .. unit .. ".") end end function EtymonParser.check_term_rules(etymon_data, entry_lang, rules, label) label = label or "term" if rules and rules.disallow then local disallowed = {} for _, typ in ipairs(rules.disallow) do local spec = TERM_RULE_DISALLOW[typ] if spec and etymon_data[spec.field] then table.insert(disallowed, spec.label) end end if #disallowed > 0 then error(label .. " does not support " .. mw.text.listToText(disallowed, "or") .. " etymons.") end end if etymon_data.is_family then if rules and rules.family == "disallowed" then error(label .. " does not support family codes" .. (rules.family_suffix or ".")) elseif not etymon_data.suppress_term then error("Family codes require suppressed term (use family:-).") end end if rules then if rules.require_term and (not etymon_data.term or etymon_data.term == "") then error(label .. " requires a term for each listed form.") end if rules.entry_lang then if Util.get_norm_lang(etymon_data.lang):getFullCode() ~= Util.get_norm_lang(entry_lang):getFullCode() then error(label .. " terms must be in the entry language (" .. entry_lang:getFullCode() .. "), got '" .. etymon_data.lang:getFullCode() .. "'.") end end if rules.ancestor_check then M.etymology.check_ancestor(entry_lang, etymon_data.lang) end elseif etymon_data.is_family and not etymon_data.suppress_term then error("Family codes require suppressed term (use family:-).") end end function EtymonParser.check_keyword_term(etymon_data, entry_lang, keyword) local config = M.data.keywords[keyword] EtymonParser.check_term_rules(etymon_data, entry_lang, config and config.term_rules, "`:" .. keyword .. "`") end function EtymonParser.check_supplement_term(etymon_data, entry_lang, supplement_type) local config = M.data.supplements[supplement_type] EtymonParser.check_term_rules(etymon_data, entry_lang, config and config.term_rules, "|" .. supplement_type .. "=") end -- Parse keyword with modifiers (e.g., ":bor<unc>" or ":bor<ref:{{R:example}}>") function EtymonParser.parse_keyword_modifiers(param) if type(param) ~= "string" then return nil, {} end local base_keyword = get_keyword_base(param) if not base_keyword then return nil, {} end local canonical_keyword = get_canonical_keyword(base_keyword) -- Check if there are any modifiers if not param:find("<", 1, true) then return canonical_keyword, {} end -- Parse modifiers using the same mechanism as etymon parsing local rest_with_defaults = Util.add_boolean_defaults(param, EtymonParser.keyword_param_mods) local function generate_obj(ignored) return {} end local parsed = M.parse_utilities.parse_inline_modifiers(rest_with_defaults:gsub("^:?[^<]+", ""), { param_mods = get_clean_param_mods(EtymonParser.keyword_param_mods), generate_obj = generate_obj }) local modifiers = { unc = parsed.unc or false, ref = parsed.ref, text = parsed.text, lit = parsed.lit, conj = parsed.conj, g = parsed.g, surf = parsed.surf or false, senseid = parsed.senseid, } -- Validate modifiers against restrictions EtymonParser.check_modifier_restrictions(modifiers, canonical_keyword, EtymonParser.keyword_param_mods) return canonical_keyword, modifiers end local function normalize_keyword_param(keyword_with_mods) local trimmed = M.string_utilities.trim(keyword_with_mods) reject_removed_surf_keyword(trimmed:match("^:") and trimmed or (":" .. trimmed)) local base = get_keyword_base(trimmed) if not base or not M.data.keywords[base] then error("Invalid keyword '" .. trimmed .. "' in inline etymology") end local canonical_base = get_canonical_keyword(base) local without_colon = trimmed:gsub("^:", "") local mods_part = without_colon:sub(#base + 1) local kw_param = normalize_keyword(canonical_base .. mods_part) EtymonParser.parse_keyword_modifiers(kw_param) return kw_param end local function get_keyword_mod_names() local names = {} for mod_name in pairs(EtymonParser.keyword_param_mods) do names[mod_name] = true end return names end local function parse_inline_ety_run(ety_string) local body = ety_string or "" if body == "" then error("Empty inline etymology") end local keyword_mod_names = get_keyword_mod_names() local pos = 1 local len = #body local function parse_err(msg) error(msg .. " in inline etymology: '" .. body .. "'") end local function peek_double() return body:sub(pos, pos + 1) == "<<" end local function mod_name_from_unwrapped(unwrapped) return unwrapped:match("^<([^:>]+)") end local function is_keyword_mod(unwrapped) local name = mod_name_from_unwrapped(unwrapped) return name and keyword_mod_names[name] or false end local function read_double_bracket() if not peek_double() then return nil end local start = pos pos = pos + 2 while pos <= len - 1 do if body:sub(pos, pos + 1) == ">>" then local token = body:sub(start, pos + 1) pos = pos + 2 return token, token:sub(2, -2) end pos = pos + 1 end parse_err("Unmatched <<") end local function read_angle_cell() if body:sub(pos, pos) ~= "<" or peek_double() then return nil end local open = pos pos = pos + 1 local depth = 1 local i = pos while i <= len do local ch = body:sub(i, i) if ch == "<" then depth = depth + 1 elseif ch == ">" then depth = depth - 1 if depth == 0 then local inner = body:sub(open + 1, i - 1) pos = i + 1 return inner end end i = i + 1 end parse_err("Unmatched <") end local function read_bare_run() local start = pos while pos <= len and body:sub(pos, pos) ~= "<" do pos = pos + 1 end return body:sub(start, pos - 1) end local function absorb_double_keyword_mods(keyword_str) while peek_double() do local saved = pos local _, unwrapped = read_double_bracket() if is_keyword_mod(unwrapped) then keyword_str = keyword_str .. unwrapped else pos = saved break end end return keyword_str end local kw_start = pos while pos <= len and body:sub(pos, pos) ~= "<" do pos = pos + 1 end local keyword = body:sub(kw_start, pos - 1) if keyword:match("^%s*$") then parse_err("Missing keyword") end keyword = absorb_double_keyword_mods(keyword) local cells = {} while pos <= len do if peek_double() then local _, unwrapped = read_double_bracket() if is_keyword_mod(unwrapped) then parse_err("Unexpected keyword modifier " .. unwrapped .. " outside of a keyword") end table.insert(cells, "+" .. unwrapped) elseif body:sub(pos, pos) == "<" then local inner = read_angle_cell() if inner ~= "" then table.insert(cells, inner) end else local bare = read_bare_run() if bare ~= "" then if bare:sub(1, 1) ~= ":" then parse_err("Unexpected bare text '" .. bare .. "' (use :keyword for nested keywords in inline etymology)") end if not is_keyword(bare, true) then parse_err("Invalid keyword '" .. bare .. "' in inline etymology") end table.insert(cells, absorb_double_keyword_mods(bare)) end end end return { keyword = keyword, cells = cells, } end function EtymonParser.inline_ety_to_pipe(ety_string) local run = parse_inline_ety_run(ety_string) if not run.keyword or run.keyword:match("^%s*$") then return "|" end local pipe_parts = { normalize_keyword_param(M.string_utilities.trim(run.keyword)) } for _, segment in ipairs(run.cells) do if is_keyword(segment, true) then table.insert(pipe_parts, normalize_keyword_param(segment)) else table.insert(pipe_parts, segment) end end return "|" .. table.concat(pipe_parts, "|") .. "|" end function EtymonParser.pipe_to_inline_ety(pipe_string) local cells = {} for cell in pipe_string:gmatch("([^|]+)") do if cell ~= "" then table.insert(cells, cell) end end if #cells == 0 then return "" end local inline_parts = {} for index, cell in ipairs(cells) do local base = get_keyword_base(cell) if base and M.data.keywords[base] then local without_colon = cell:gsub("^:", "") local kw_base, mods = without_colon:match("^([^<]+)(.*)$") local inline_kw = (kw_base or without_colon) .. (mods or ""):gsub("<([^>]+)>", "<<%1>>") if index > 1 then inline_kw = ":" .. inline_kw end table.insert(inline_parts, inline_kw) elseif cell:sub(1, 1) == "+" then local mod = cell:sub(2) if mod:match("^<.->$") then mod = mod:sub(2, -2) end table.insert(inline_parts, "<<" .. mod .. ">>") else table.insert(inline_parts, "<" .. cell .. ">") end end return table.concat(inline_parts, "") end function EtymonParser.parse_inline_ety(ety_string, context_lang) local run = parse_inline_ety_run(ety_string) local keyword = M.string_utilities.trim(run.keyword) reject_removed_surf_keyword(":" .. keyword) if not is_keyword(keyword, true) then error("Invalid keyword '" .. keyword .. "' in inline etymology <ety:" .. keyword .. "...>") end local args = { context_lang:getCode(), normalize_keyword_param(keyword) } for _, segment in ipairs(run.cells) do if is_keyword(segment, true) then table.insert(args, normalize_keyword_param(segment)) else table.insert(args, segment) end end return args end function EtymonParser.parse_etymon(param, context_lang) if is_keyword(param) then return nil end if type(param) ~= "string" then return nil end local lang, rest local is_family = false local before_bracket = param:match("^([^<]*)") or param local lang_code, rest_match = before_bracket:match("^([a-zA-Z][a-zA-Z0-9._-]*):(.*)$") if lang_code then local potential_lang = Util.get_lang(lang_code, true) if potential_lang then lang = potential_lang rest = param:sub(#lang_code + 2) else local potential_family = Util.get_family(lang_code) if potential_family then lang = potential_family rest = param:sub(#lang_code + 2) is_family = true else lang = context_lang rest = param end end else lang = context_lang rest = param end M.tracking.track_term(rest) if rest == "" or rest == "+" then return { lang = lang, term = nil, unknown_term = true, is_family = is_family, } end if rest == "-" then return { lang = lang, term = nil, suppress_term = true, is_family = is_family, } end if not rest:find("<", 1, true) then return { lang = lang, term = M.string_utilities.trim(rest), is_family = is_family, } end local term_text = rest:match("^([^<]*)") or "" local is_unknown = (term_text == "" or term_text == "+") local is_suppress = (term_text == "-") local function generate_obj(ignored_term) return { term = (is_unknown or is_suppress) and nil or M.string_utilities.trim(term_text) } end local rest_with_defaults = Util.add_boolean_defaults(rest, EtymonParser.etymon_param_mods) local parsed_obj = M.parse_utilities.parse_inline_modifiers(rest_with_defaults, { param_mods = get_clean_param_mods(EtymonParser.etymon_param_mods), generate_obj = generate_obj }) if parsed_obj.id and parsed_obj.id:match("^!") then parsed_obj.id = parsed_obj.id:sub(2) parsed_obj.override = true end parsed_obj.lang = lang parsed_obj.is_family = is_family if is_unknown then parsed_obj.unknown_term = true elseif is_suppress then parsed_obj.suppress_term = true end return parsed_obj end function EtymonParser.validate(lang, args, id, title, pos, starts_with_lang_code) -- id is now optional, so only validate if provided if id then if mw.ustring.len(id) < 2 then error("The `id` parameter must have at least two characters.") end if id == title or id == Util.get_page_data().pagename then error("The `id` parameter must not be the same as the page title.") end end local valid_pos = { prefix = true, suffix = true, interfix = true, infix = true, root = true, word = true } if pos and not valid_pos[pos] then error("Unknown value provided for `pos`. Valid values: " .. table.concat(require("Module:table").keysToList(valid_pos), ", ") .. ".") end local current_keyword = "from" local current_keyword_explicit = false local keyword_etymons = {} local keywords = M.data.keywords local function checkKeyword() local config = keywords[current_keyword] if current_keyword == "from" and not current_keyword_explicit and #keyword_etymons == 0 then keyword_etymons = {} return end EtymonParser.check_etymon_limits(#keyword_etymons, config, "`:" .. current_keyword .. "`") keyword_etymons = {} end local start_index = starts_with_lang_code and 2 or 1 for i = start_index, #args do local param = args[i] if type(param) ~= "string" then elseif param:sub(1, 1) == ":" and not is_keyword(param) then reject_removed_surf_keyword(param) error("Invalid keyword '" .. param .. "'. Did you mean a valid keyword like ':bor', ':inh', etc.?") elseif is_keyword(param) then checkKeyword() current_keyword = get_canonical_keyword(get_keyword(param)) current_keyword_explicit = true else local etymon_data = EtymonParser.parse_etymon(param, lang) if etymon_data then table.insert(keyword_etymons, param) EtymonParser.check_keyword_term(etymon_data, lang, current_keyword) -- Check modifier restrictions EtymonParser.check_modifier_restrictions(etymon_data, current_keyword, EtymonParser.etymon_param_mods) -- postype must be "root" or "word" local VALID_POSTYPES = { root = true, word = true } if etymon_data.postype and not VALID_POSTYPES[etymon_data.postype] then error("Invalid <postype:" .. etymon_data.postype .. ">; must be \"root\" or \"word\".") end if etymon_data.ety then local inline_args = EtymonParser.parse_inline_ety(etymon_data.ety, etymon_data.lang) EtymonParser.validate(etymon_data.lang, inline_args, nil, nil, nil, true) end else table.insert(keyword_etymons, param) end end end checkKeyword() end local DataRetriever = {} local function format_etymon_id_hint(id_data, idx) local id = type(id_data) == "table" and id_data.id or id_data local pos = type(id_data) == "table" and id_data.pos if id and id ~= "" and id ~= "*" then return '"' .. id .. '"' end if pos and pos ~= "" then return "unnamed (|pos=" .. pos .. "|)" end return "etymon #" .. idx .. " (no |id= on page)" end local function etymon_target_page_link(page, norm_lang) return M.links.full_link({ term = page, lang = norm_lang, no_generate_forms = true, }, "term") end -- Summarize {{etymon}} id slots on a linked page for preview warnings. local function summarize_available_etymon_ids(ids) local id_list = {} local all_idless = true local target_has_idless = false local any_pos = false for i, id_data in ipairs(ids) do local id = type(id_data) == "table" and id_data.id or id_data local pos = type(id_data) == "table" and id_data.pos if id and id ~= "" and id ~= "*" then all_idless = false else target_has_idless = true end if pos and pos ~= "" then any_pos = true end table.insert(id_list, format_etymon_id_hint(id_data, i)) end return { id_list = id_list, all_idless = all_idless, target_has_idless = target_has_idless, any_pos = any_pos, count = #ids, options_text = mw.text.listToText(id_list), } end local function ambiguous_etymon_suggestion(page_link, summary) if summary.all_idless then if summary.any_pos then return " None set `|id=` yet; add a unique `|id=` to each on " .. page_link .. ", then `<id:identifier>` after the term here. Section order / hints: " .. summary.options_text .. "." end return " None set `|id=` yet; add a unique `|id=` to each {{etymon}} in that section from top to bottom, then `<id:identifier>` after the term here (same value as `|id=`)." end return " Specify which one with `<id:identifier>` after the term. Options: " .. summary.options_text .. "." end local function warn_ambiguous_etymon_link(page, norm_lang, ids, is_toplevel) local page_link = etymon_target_page_link(page, norm_lang) local summary = summarize_available_etymon_ids(ids) if is_toplevel and summary.target_has_idless then __state.linked_page_multiple_etymons_idless = true end local lang_name = norm_lang:getCanonicalName() local lead = "Etymology link to " .. page_link .. " is ambiguous (" .. summary.count .. " {{etymon}} templates for " .. lang_name .. ")." Util.add_warning(lead .. ambiguous_etymon_suggestion(page_link, summary), true) end local function is_mismatched_explicit_id(base_key, cached_args, parent_etymon) return cached_args == M.data.STATUS.MISSING and not parent_etymon and #(__state.available_etymon_ids[base_key] or {}) > 0 end local function maybe_flag_partial_etymology_reference(base_key, etymon_data, cached_args, is_toplevel) if not is_toplevel or __state.skip_partial_etymology_category then return end if not __state.partial_etymology_targets[base_key] then return end if etymon_data.id and type(cached_args) == "table" then return end __state.linked_page_partial_etymology_sections = true end local function is_nonlemma_etymon_template(template_args) return template_args and M.yesno(template_args.nl, false) end local function warn_mismatched_explicit_id(page, norm_lang, base_key, etymon_id) local page_link = etymon_target_page_link(page, norm_lang) local summary = summarize_available_etymon_ids(__state.available_etymon_ids[base_key] or {}) local lang_name = norm_lang:getCanonicalName() local lead = "Etymology link to " .. page_link .. " uses `<id:" .. etymon_id .. ">`, but no {{etymon}} on that page has `|id=" .. etymon_id .. "|` for " .. lang_name .. "." Util.add_warning(lead .. " Valid IDs: " .. summary.options_text .. ".", true) end -- Given an etymon data, scrape its page and cache the result in the global state object. function DataRetriever.cache_page_etymons(etymon_page, etymon_title, key, etymon_lang, etymon_id, redirected_from, descendants_is_toplevel) local content = etymon_title:getContent() if not content then __state.cached_etymon_args[key] = M.data.STATUS.REDLINK return end -- Check if the linked page is a redirect. If it is, the template parsing -- code below will be effectively skipped, and `scrape_page` will be called -- again on the redirect target (see the bottom of this function) local lang_section_for_descendants = nil local redirect_target = etymon_title.redirect_target if not redirect_target then content = M.pages.get_section(content, etymon_lang:getFullName(), 2) if not content then __state.cached_etymon_args[key] = M.data.STATUS.MISSING return end lang_section_for_descendants = content end local etymon_lang_code = etymon_lang:getFullCode() local lang_page_key = etymon_lang_code .. ":" .. etymon_page local found_templates_for_lang = {} local found_ids = {} local get_node_class = M.template_parser.class_else_type -- Look for all {{etymon}} templates within the page content using the template parser -- This way the same page is never parsed more than once -- Build a map from senseids to their parent etymonids. local active_etymon_args = nil local etymology_section_count = 0 local etymology_sections_with_etymon = 0 local current_etymology_has_etymon = false local current_etymology_has_nonlemma = false local function finalize_current_etymology_section() if etymology_section_count == 0 then return end if current_etymology_has_etymon or current_etymology_has_nonlemma then etymology_sections_with_etymon = etymology_sections_with_etymon + 1 end current_etymology_has_etymon = false current_etymology_has_nonlemma = false end for node in M.template_parser.parse(content):iterate_nodes() do local node_class = get_node_class(node) if node_class == "heading" then -- A new L2 or etymology section acts as a barrier: an {{etymon}} usage -- used previously cannot be the parent of any subsequent senseids. -- Note that we don't have to check for L2s due to the usage of `M.pages.get_section` above. if node:get_name():find("^Etymology") then finalize_current_etymology_section() etymology_section_count = etymology_section_count + 1 active_etymon_args = nil end elseif node_class == "template" then local template_name = node:get_name() if template_name == "etymon" then local template_args = node:get_arguments() -- Check if this etymon is for our language if template_args[1] == etymon_lang_code then if is_nonlemma_etymon_template(template_args) then if etymology_section_count > 0 then current_etymology_has_nonlemma = true end else if etymology_section_count > 0 then current_etymology_has_etymon = true end table.insert(found_templates_for_lang, template_args) if template_args.id then local etymon_key = lang_page_key .. ":" .. template_args.id __state.cached_etymon_args[etymon_key] = template_args __state.cached_etymon_pages[etymon_key] = tostring(etymon_page) table.insert(found_ids, template_args.id) active_etymon_args = template_args else -- Store idless etymon with default key local etymon_key = lang_page_key .. ":*" __state.cached_etymon_args[etymon_key] = template_args __state.cached_etymon_pages[etymon_key] = tostring(etymon_page) table.insert(found_ids, "*") active_etymon_args = template_args end end end elseif active_etymon_args and template_name == "senseid" then local template_args = node:get_arguments() -- This should always be true for proper usages of {{senseid}}. if template_args[1] == etymon_lang_code and template_args[2] then local sense_id_key = lang_page_key .. ":" .. template_args[2] __state.senseid_parent_etymon[sense_id_key] = active_etymon_args __state.cached_etymon_pages[sense_id_key] = tostring(etymon_page) end end end end finalize_current_etymology_section() if lang_section_for_descendants and etymology_section_count > 1 and etymology_sections_with_etymon > 0 and etymology_sections_with_etymon < etymology_section_count then __state.partial_etymology_targets[lang_page_key] = true end if descendants_is_toplevel and lang_section_for_descendants and #found_templates_for_lang > 0 then M.descendants.cache_page_checks({ lang_section = lang_section_for_descendants, etymon_lang_code = etymon_lang_code, found_templates_for_lang = found_templates_for_lang, entry_title = __state.entry_title, entry_lang_code = __state.entry_lang_code, entry_lang = __state.entry_lang_code and Util.get_lang(__state.entry_lang_code, true) or nil, cached_descendants_checks = __state.cached_descendants_checks, lang_page_key = lang_page_key, redirected_from = redirected_from, }) end local id_data_list = {} for _, args in ipairs(found_templates_for_lang) do local id = args.id or "*" table.insert(id_data_list, { id = id, pos = args.pos }) end __state.available_etymon_ids[lang_page_key] = id_data_list if #found_templates_for_lang == 1 then __state.single_etymons[lang_page_key] = found_templates_for_lang[1] end if redirected_from and __state.available_etymon_ids[lang_page_key] then __state.available_etymon_ids[redirected_from] = __state.available_etymon_ids[redirected_from] or {} for _, id_data in ipairs(__state.available_etymon_ids[lang_page_key]) do table.insert(__state.available_etymon_ids[redirected_from], id_data) end end if __state.cached_etymon_args[key] ~= nil or __state.senseid_parent_etymon[key] ~= nil then -- All done! return elseif redirect_target and not redirected_from then -- Try scraping the redirect. etymon_page = redirect_target.prefixedText DataRetriever.cache_page_etymons(etymon_page, redirect_target, lang_page_key .. ":" .. etymon_id, etymon_lang, etymon_id, lang_page_key, descendants_is_toplevel) __state.cached_etymon_args[key] = __state.cached_etymon_args[etymon_lang_code .. ":" .. etymon_page .. ":" .. etymon_id] else __state.cached_etymon_args[key] = M.data.STATUS.MISSING end end local function has_linkable_term(etymon_data) if etymon_data.is_family or etymon_data.suppress_term or etymon_data.unknown_term then return false end local term = etymon_data.term if term == nil or term == "" then return false end return M.string_utilities.trim(term) ~= "" end local function record_term_id_tracking(etymon_data) if not has_linkable_term(etymon_data) then return end local term_page = M.links.get_link_page(etymon_data.term, etymon_data.lang) M.tracking.record_term_id_usage(__state.id_stats, etymon_data, term_page) end -- Given an etymon object, scrape its page (if necessary) and return its own etymon arguments as well as the page name. function DataRetriever.get_etymon_args(etymon_data, is_toplevel) if not has_linkable_term(etymon_data) then return M.data.STATUS.MISSING, nil, nil, nil end local page = M.links.get_link_page(etymon_data.term, etymon_data.lang) local norm_lang = Util.get_norm_lang(etymon_data.lang) local base_key = norm_lang:getFullCode() .. ":" .. page if etymon_data.id then local key = base_key .. ":" .. etymon_data.id local cached_args = __state.cached_etymon_args[key] or __state.senseid_parent_etymon[key] if cached_args == nil then local title = mw.title.new(page) if not title then error('Invalid page title "' .. page .. '" encountered.') end DataRetriever.cache_page_etymons(page, title, key, norm_lang, etymon_data.id, nil, is_toplevel) end cached_args = __state.cached_etymon_args[key] or __state.senseid_parent_etymon[key] -- refresh -- Get etymon_id from parent if this was resolved via senseid local parent_etymon = __state.senseid_parent_etymon[key] local resolved_etymon_id = parent_etymon and parent_etymon.id local descendants_check = M.descendants.get_lookup_check({ cached_descendants_checks = __state.cached_descendants_checks, is_toplevel = is_toplevel, base_key = base_key, lookup = { explicit_id = etymon_data.id, parent_etymon = parent_etymon, }, }) if is_toplevel and descendants_check == nil then local title = mw.title.new(page) if title then DataRetriever.cache_page_etymons(page, title, key, norm_lang, etymon_data.id, nil, true) descendants_check = M.descendants.get_lookup_check({ cached_descendants_checks = __state.cached_descendants_checks, is_toplevel = true, base_key = base_key, lookup = { explicit_id = etymon_data.id, parent_etymon = parent_etymon, }, }) end end local mismatched_id = is_mismatched_explicit_id(base_key, cached_args, parent_etymon) if mismatched_id and is_toplevel then __state.has_mismatched_id = true M.tracking.record_mismatched_id_usage(__state.id_stats, norm_lang, page, etymon_data.id) warn_mismatched_explicit_id(page, norm_lang, base_key, etymon_data.id) end maybe_flag_partial_etymology_reference(base_key, etymon_data, cached_args, is_toplevel) return cached_args, __state.cached_etymon_pages[key], resolved_etymon_id, descendants_check else __state.used_idless_etymon = true if is_toplevel then __state.toplevel_idless_etymon = true end if __state.available_etymon_ids[base_key] == nil then local title = mw.title.new(page) if not title then error('Invalid page title "' .. page .. '" encountered.') end DataRetriever.cache_page_etymons(page, title, base_key .. ":*", norm_lang, "*", nil, is_toplevel) end local ids = __state.available_etymon_ids[base_key] or {} local count = #ids -- Try to filter by postype if available and we have multiple candidates if count > 1 and etymon_data.postype then local matching_ids = {} for _, id_data in ipairs(ids) do if id_data.pos == etymon_data.postype then table.insert(matching_ids, id_data) end end if #matching_ids == 1 then local matched_id = matching_ids[1].id local matched_key = base_key .. ":" .. matched_id M.tracking.record_idless_resolution(__state.id_stats, norm_lang, page, "postype") local descendants_check = M.descendants.get_lookup_check({ cached_descendants_checks = __state.cached_descendants_checks, is_toplevel = is_toplevel, base_key = base_key, lookup = { id = matched_id }, }) if is_toplevel and descendants_check == nil then local title = mw.title.new(page) if title then DataRetriever.cache_page_etymons(page, title, base_key .. ":*", norm_lang, "*", nil, true) descendants_check = M.descendants.get_lookup_check({ cached_descendants_checks = __state.cached_descendants_checks, is_toplevel = true, base_key = base_key, lookup = { id = matched_id }, }) end end local matched_args = __state.cached_etymon_args[matched_key] maybe_flag_partial_etymology_reference(base_key, etymon_data, matched_args, is_toplevel) return matched_args, __state.cached_etymon_pages[matched_key], nil, descendants_check end end if count == 1 then local only_id_data = ids[1] local only_id = (type(only_id_data) == "table" and only_id_data.id) or only_id_data or "*" M.tracking.record_idless_resolution(__state.id_stats, norm_lang, page, "single") local descendants_check = M.descendants.get_lookup_check({ cached_descendants_checks = __state.cached_descendants_checks, is_toplevel = is_toplevel, base_key = base_key, lookup = { id_data = only_id_data }, }) if is_toplevel and descendants_check == nil then local title = mw.title.new(page) if title then DataRetriever.cache_page_etymons(page, title, base_key .. ":*", norm_lang, "*", nil, true) descendants_check = M.descendants.get_lookup_check({ cached_descendants_checks = __state.cached_descendants_checks, is_toplevel = true, base_key = base_key, lookup = { id_data = only_id_data }, }) end end local single_args = __state.single_etymons[base_key] maybe_flag_partial_etymology_reference(base_key, etymon_data, single_args, is_toplevel) return single_args, __state.cached_etymon_pages[base_key .. ":" .. only_id], nil, descendants_check elseif count > 1 then M.tracking.record_idless_resolution(__state.id_stats, norm_lang, page, "ambiguous") warn_ambiguous_etymon_link(page, norm_lang, ids, is_toplevel) maybe_flag_partial_etymology_reference(base_key, etymon_data, M.data.STATUS.AMBIGUOUS, is_toplevel) return M.data.STATUS.AMBIGUOUS, nil, nil, nil else M.tracking.record_idless_resolution(__state.id_stats, norm_lang, page, "missing") maybe_flag_partial_etymology_reference(base_key, etymon_data, M.data.STATUS.MISSING, is_toplevel) return M.data.STATUS.MISSING, nil, nil, nil end end end local function keyword_invisible_in_tree(keyword_info) if not keyword_info then return false end local inv = keyword_info.invisible return inv == "all" or inv == true or inv == "tree" end -- True when the node has at least one top-level child container visible in the tree. local function node_has_visible_tree_children(node) for _, container in ipairs(node.children or {}) do if not keyword_invisible_in_tree(container.keyword_info) then return true end end return false end -- Count visible term nodes in the tree. local function get_visible_tree_depth(node, skip_child_rendering) local max_depth = 1 if skip_child_rendering or not node then return max_depth end for _, container in ipairs(node.children or {}) do local keyword_info = container.keyword_info if not keyword_invisible_in_tree(keyword_info) then local skip_grandchildren = keyword_info and keyword_info.no_child_categories for _, term in ipairs(container.terms or {}) do if term.is_duplicate then if term.original_has_children then max_depth = math.max(max_depth, 2) end else max_depth = math.max(max_depth, 1 + get_visible_tree_depth(term, skip_grandchildren)) end end end end return max_depth end local function as_param_list(val) if val == nil then return {} end if type(val) == "table" then return val end if type(val) == "string" and val ~= "" then return { val } end return {} end local TreeBuilder = {} local function parse_etymon_references(refs_text) if not refs_text or refs_text == "" then return "" end return M.references.parse_references(refs_text) end local function parse_tree_references(node) if node.ref then node.parsed_ref = parse_etymon_references(node.ref) end if node.children then for _, container in ipairs(node.children) do if container.terms then for _, term in ipairs(container.terms) do parse_tree_references(term) end end end end if node.supplements then for _, supplement in ipairs(node.supplements) do if supplement.terms then for _, term in ipairs(supplement.terms) do parse_tree_references(term) end end end end end -- Build a unique key for deduplication in the seen table function TreeBuilder.build_key(lang, title, args) local norm_lang_code = Util.get_norm_lang(lang):getFullCode() local is_table = type(args) == "table" local id = (is_table and args.id) or "" if title then return norm_lang_code .. ":" .. M.links.get_link_page(title, lang) .. ":" .. id end if is_table and args.status == M.data.STATUS.INLINE then local content_parts = {} for i = 1, #args do content_parts[i] = tostring(args[i]) end return norm_lang_code .. ":*:" .. id .. "\0" .. table.concat(content_parts, "\0") end return norm_lang_code .. ":*:" .. id end -- Copy parsed etymon modifiers onto a tree/supplement term node. function TreeBuilder.apply_etymon_fields(term, etymon_data) term.id = etymon_data.id term.t = etymon_data.t term.tr = etymon_data.tr term.ts = etymon_data.ts term.alt = etymon_data.alt term.g = etymon_data.g term.pos = etymon_data.pos term.ng = etymon_data.ng term.infl = etymon_data.infl term.ref = etymon_data.ref term.is_uncertain = etymon_data.unc term.lit = etymon_data.lit term.q = etymon_data.q term.qq = etymon_data.qq term.l = etymon_data.l term.ll = etymon_data.ll term.suppress_term = etymon_data.suppress_term term.unknown_term = etymon_data.unknown_term term.is_family = etymon_data.is_family term.override = etymon_data.override term.aftype = etymon_data.aftype term.postype = etymon_data.postype term.bor = etymon_data.bor term.lbor = etymon_data.lbor term.slbor = etymon_data.slbor end function TreeBuilder.build_supplement_term(etymon_data, entry_lang, supplement_type) EtymonParser.check_supplement_term(etymon_data, entry_lang, supplement_type) local term = { lang = etymon_data.lang, title = etymon_data.term, children = {}, status = M.data.STATUS.OK, } TreeBuilder.apply_etymon_fields(term, etymon_data) return term end function TreeBuilder.build_supplement_terms(entry_lang, supplement_type, param_value) local terms = {} for _, term_param in ipairs(as_param_list(param_value)) do if type(term_param) == "string" and term_param ~= "" then local etymon_data = EtymonParser.parse_etymon(term_param, entry_lang) if etymon_data then table.insert(terms, TreeBuilder.build_supplement_term(etymon_data, entry_lang, supplement_type)) end end end return terms end -- Attach a |param= supplement defined in etymon_data.supplements (e.g. doublet=). function TreeBuilder.append_term_supplement(data_tree, entry_lang, supplement_type, param_value) local config = M.data.supplements[supplement_type] if not config then error("Unknown supplement '" .. tostring(supplement_type) .. "'.") end local terms = TreeBuilder.build_supplement_terms(entry_lang, supplement_type, param_value) if #terms == 0 then return end data_tree.supplements = data_tree.supplements or {} table.insert(data_tree.supplements, { type = supplement_type, config = config, terms = terms, }) M.tracking.record_keyword_usage(__state.toplevel_keyword_stats, supplement_type, entry_lang, entry_lang, true) end function TreeBuilder.build(lang, title, args, seen, depth, stop_recursion) seen = seen or {} depth = depth or 0 local is_toplevel = (depth == 0) if depth > __state.max_depth_reached then __state.max_depth_reached = depth end __state.total_nodes = __state.total_nodes + 1 local lang_code = lang:getCode() __state.language_count[lang_code] = (__state.language_count[lang_code] or 0) + 1 local current_id = (type(args) == "table" and args.id) or "" local key = TreeBuilder.build_key(lang, title, args) local node = { lang = lang, title = title, id = current_id, args = args, children = {}, status = M.data.STATUS.OK } if type(args) ~= "table" or seen[key] then node.status = args or M.data.STATUS.MISSING -- Mark as duplicate if we've seen this node before if seen[key] then node.is_duplicate = true node.duplicate_key = key local original_node = seen[key] if type(original_node) == "table" and original_node.children and #original_node.children > 0 then node.original_has_children = true end end return node end node.status = args.status or M.data.STATUS.OK seen[key] = node -- If stop_recursion is set, skip parsing children but check for visible children if stop_recursion then local keywords = M.data.keywords local has_visible_children = false for i = 2, #args do local param = args[i] if type(param) == "string" then local keyword_base = get_keyword_base(param) if keyword_base and keywords[keyword_base] then local _, kw_modifiers = EtymonParser.parse_keyword_modifiers(param:sub(1, 1) == ":" and param or (":" .. param)) if not keyword_invisible_in_tree(get_effective_keyword_info(keyword_base, kw_modifiers)) then has_visible_children = true break end elseif param:sub(1, 1) ~= ":" then -- It's a term (not a keyword), so there are visible children has_visible_children = true break end end end node.has_visible_children = has_visible_children return node end -- Parse args into keyword containers local current_keyword = "from" local current_keyword_modifiers = {} local current_container = nil local function ensure_container() if not current_container or current_container.keyword ~= current_keyword then local keyword_info = get_effective_keyword_info(current_keyword, current_keyword_modifiers) current_container = { keyword = current_keyword, keyword_info = keyword_info, keyword_modifiers = current_keyword_modifiers, terms = {}, } table.insert(node.children, current_container) -- Override keyword text/phrase for nominalization with <g:code> if current_keyword_modifiers.g and current_keyword == "nominalization" then local labels = get_nominalization_label_for_g(current_keyword_modifiers.g) if not labels then local codes = {} for c in pairs(M.data.nominalization_g_codes) do table.insert(codes, c) end table.sort(codes) error("Invalid <g:" .. tostring(current_keyword_modifiers.g) .. ">. Supported codes for nominalization: " .. table.concat(codes, ", ")) end current_container.keyword_info = copy_keyword_info(keyword_info) current_container.keyword_info.text = labels.text current_container.keyword_info.phrase = labels.phrase end end return current_container end local parse_context_lang = Util.resolve_context_lang(lang, args) for i = 2, #args do local param = args[i] if is_keyword(param) then local keyword, modifiers = EtymonParser.parse_keyword_modifiers(param) if not keyword then error("Invalid keyword '" .. param .. "'.") end current_keyword = keyword current_keyword_modifiers = modifiers current_container = nil -- Force new container for new keyword elseif type(param) == "string" and param:sub(1, 1) == ":" then reject_removed_surf_keyword(param) error("Invalid keyword '" .. param .. "'. Did you mean a valid keyword like ':bor', ':inh', etc.?") elseif type(param) == "string" then local etymon_data = EtymonParser.parse_etymon(param, parse_context_lang) if etymon_data then -- Track keyword usage at top level M.tracking.record_keyword_usage(__state.toplevel_keyword_stats, current_keyword, lang, etymon_data.lang, is_toplevel) local term_node = {} local container -- Handle suppress_term (-) and unknown_term (empty or +) directly if etymon_data.suppress_term or etymon_data.unknown_term then container = ensure_container() if etymon_data.ety then local inline_args = EtymonParser.parse_inline_ety(etymon_data.ety, etymon_data.lang) inline_args.id = etymon_data.id inline_args.status = M.data.STATUS.INLINE term_node = TreeBuilder.build(etymon_data.lang, nil, inline_args, seen, depth + 1) else term_node = { lang = etymon_data.lang, children = {}, status = M.data.STATUS.OK, } end TreeBuilder.apply_etymon_fields(term_node, etymon_data) else -- Regular term: fetch arguments from page record_term_id_tracking(etymon_data) local etymon_args, page_of, resolved_etymon_id, descendants_check = DataRetriever.get_etymon_args(etymon_data, is_toplevel) -- Check for <ety> inline parameter doesn't override the scraped arguments, unless the latter are missing if etymon_data.ety then if etymon_args == M.data.STATUS.REDLINK or etymon_args == M.data.STATUS.MISSING then __state.current_page_has_inline_etymology = true if is_toplevel then __state.toplevel_has_inline_etymology = true end local inline_args = EtymonParser.parse_inline_ety(etymon_data.ety, etymon_data.lang) -- Track inline ety keywords too local inline_keyword = get_keyword(inline_args[2], true) if inline_keyword and #inline_args >= 3 then local inline_etymon = EtymonParser.parse_etymon(inline_args[3], etymon_data.lang) if inline_etymon then M.tracking.record_keyword_usage(__state.toplevel_keyword_stats, inline_keyword, etymon_data.lang, inline_etymon.lang, is_toplevel) end end inline_args.id = etymon_data.id inline_args.status = M.data.STATUS.INLINE etymon_args = inline_args term_node.page_of = __state.cached_etymon_pages[key] -- term node is on the same page as the parent else -- Scraped arguments exist, <ety> is redundant and ignored __state.current_page_has_redundant_etymology = true if is_toplevel then __state.toplevel_redundant_etymology = true end end end -- Ensure container exists before checking keyword info container = ensure_container() -- Check if current keyword has no_child_categories - if so, stop recursion local keyword_info = container.keyword_info local should_stop_recursion = (stop_recursion or (keyword_info and keyword_info.no_child_categories)) term_node = TreeBuilder.build(etymon_data.lang, etymon_data.term, etymon_args, seen, depth + 1, should_stop_recursion) term_node.target_key = Util.get_norm_lang(etymon_data.lang):getFullCode() .. ":" .. M.links.get_link_page(etymon_data.term, etymon_data.lang) term_node.etymon_id = resolved_etymon_id -- The actual etymon id when resolved via senseid term_node.page_of = page_of TreeBuilder.apply_etymon_fields(term_node, etymon_data) term_node.missing_descendants_header, term_node.missing_descendants_entry = M.descendants.get_term_sync_flags(current_keyword, term_node.status, descendants_check) end table.insert(container.terms, term_node) end end end return node end -- Convert etymology tree to JSON-serializable table local function tree_to_json(node) local obj = { term = node.title, lang = node.lang:getCode(), lang_name = node.lang:getCanonicalName(), id = (node.id and node.id ~= "") and node.id or nil, status = node.status, is_uncertain = node.is_uncertain or nil, is_duplicate = node.is_duplicate or nil, gloss = node.t, transliteration = node.tr, transcription = node.ts, alt = node.alt, g = node.g, pos = node.pos, ng = node.ng, infl = node.infl, children = {}, } for _, container in ipairs(node.children or {}) do local keyword_info = container.keyword_info if keyword_info then local container_obj = { keyword = container.keyword, keyword_label = keyword_info.text, keyword_abbrev = keyword_info.abbrev, is_group = keyword_info.is_group or nil, is_invisible = keyword_info.invisible or nil, is_uncertain = (container.keyword_modifiers and container.keyword_modifiers.unc) or nil, terms = {}, } for _, term in ipairs(container.terms or {}) do table.insert(container_obj.terms, tree_to_json(term)) end table.insert(obj.children, container_obj) end end return obj end -- Build and return the etymology data tree for a given term. function export.get_tree(lang, title, args, options) options = options or {} __state.entry_title = title __state.entry_lang_code = lang:getCode() __state.id_stats = M.tracking.new_id_stats() __state.skip_partial_etymology_category = options.skip_partial_etymology_category == true if options.validate then EtymonParser.validate(lang, args, options.id, title, options.pos, false) end local lang_code = lang:getCode() local start_index = (args[1] == lang_code) and 2 or 1 local tree_args = { [1] = lang_code, id = options.id or args.id } for i = start_index, #args do table.insert(tree_args, args[i]) end __state.cached_etymon_args[lang_code .. ":" .. title .. ":" .. (tree_args.id or "")] = tree_args local ety_data_tree = TreeBuilder.build(lang, title, tree_args) parse_tree_references(ety_data_tree) if options.json then return M.JSON.toJSON(tree_to_json(ety_data_tree)) end return ety_data_tree end -- Given a language code, page name and optionally the id= parameter, -- render the tree and only the etymology tree for the relevant page. -- Fetches and parses the corresponding {{etymon}} from the requested page, -- and any further pages needed to render the tree. -- Parameters can be passed either through the #invoke or as -- template parameters *through* an #invoke. function export.render_tree_for_etymon_on_page(frame) local frame_args = frame.args local parent_args = frame:getParent().args local langcode = frame_args[1] or parent_args[1] local pagename = frame_args[2] or parent_args[2] local id = frame_args["id"] or parent_args["id"] local display_title = frame_args["title"] or parent_args["title"] local parsed_title = mw.title.new(pagename, 0) local title if parsed_title.namespace == 0 then title = M.pages.safe_page_name(parsed_title) elseif parsed_title.namespace == 118 then title = "*" .. M.pages.safe_page_name(parsed_title) else error("Unsupported namespace for render_tree_for_etymon_on_page: " .. parsed_title.namespace) end local lang = Util.get_lang(langcode) __state.entry_title = title __state.entry_lang_code = lang:getCode() __state.id_stats = M.tracking.new_id_stats() -- Construct etymon_data for DataRetriever.get_args. local etymon_data = { lang = lang, term = title, id = id } local args, pagename = DataRetriever.get_etymon_args(etymon_data, true) if args == M.data.STATUS.MISSING then error("The etymon template was not found (language " .. langcode .. ", title '" .. title .. "'" .. (id and ", ID '" .. id .. "'" or ", no ID given") .. "). Page contents may have changed in the interim.") end local tree_title = display_title or title if lang:stripDiacritics(M.links.remove_links(tree_title)) ~= lang:stripDiacritics(M.links.remove_links(title)) then M.tracking.track_title_pagename_mismatch(lang) end reset_invocation_state() local ety_data_tree = export.get_tree(lang, tree_title, args, { validate = true, id = id, }) local output = {} table.insert(output, M.template_styles("Module:etymon/styles.css")) table.insert(output, M.tree.render({ data_tree = ety_data_tree, format_term_func = function(term, is_toplevel) return Util.format_term(term, is_toplevel, { gloss = "suppress", pos = "suppress", lit = "suppress", tree_ql = "suppress", }) end, })) return table.concat(output) end function export.main(frame) local parent_args = frame:getParent().args local args = M.parameters.process(parent_args, M.parameters_data.etymon) local lang = args[1] local etymon_args = args[2] local id = args.id local title = args.title local text = args.text local tree = args.tree local etydate = args.etydate local doublet = args.doublet local rfe = args.rfe local etystub = args.etystub local is_nonlemma = M.yesno(args.nl, false) local page_data = Util.get_page_data() if not title then title = page_data.pagename if page_data.namespace == "Reconstruction" then title = "*" .. title end end local entry_pagename = page_data.pagename if page_data.namespace == "Reconstruction" then entry_pagename = "*" .. entry_pagename end if lang:stripDiacritics(M.links.remove_links(title)) ~= lang:stripDiacritics(M.links.remove_links(entry_pagename)) then M.tracking.track_title_pagename_mismatch(lang) end local current_L2 = M.pages.get_current_L2() if current_L2 then local norm_lang = Util.get_norm_lang(lang) local norm_name = norm_lang:getCanonicalName() if current_L2 ~= norm_name then local lang_desc = lang:getCode() .. " (" .. lang:getCanonicalName() .. ")" if norm_lang:getCode() ~= lang:getCode() then lang_desc = lang_desc .. ", normalized to " .. norm_lang:getCode() .. " (" .. norm_name .. ")" end error("Language '" .. lang_desc .. "' does not match the L2 header (" .. current_L2 .. ").") end end reset_invocation_state() local ety_data_tree = export.get_tree(lang, title, etymon_args, { validate = true, pos = args.pos, id = id, json = args.json, skip_partial_etymology_category = is_nonlemma, }) if args.json then return ety_data_tree end local output = {} local text_allowlist_mode = M.text_allowed.default_mode or "off" if text and text_allowlist_mode ~= "off" and not Util.is_text_param_allowed_for_lang(lang) then local msg = "Etymology texts (parameter <code>text=</code>) are not allowed for " .. lang:getFullName() .. "; see [[Template:etymon#Text allowlist|Template:etymon § Text allowlist]] for the list of languages that may use the <code>text=</code> parameter." if text_allowlist_mode == "error" then error(msg) else Util.add_warning(msg, true) end end local lang_exc = Util.get_lang_exception(lang) if lang_exc and lang_exc.disallow then local disallow = lang_exc.disallow local error_text = " for " .. lang:getFullName() if disallow.ref then error_text = error_text .. "; see " .. disallow.ref else error_text = error_text .. "." end if tree and disallow.tree then error("Etymology trees are not allowed" .. error_text) end if text and disallow.text then error("Etymology texts are not allowed" .. error_text) end end if etydate then local etydate_param_mods = { ref = { list = true, type = "references", allow_holes = true }, refn = { list = true, allow_holes = true }, nocap = { type = "boolean" }, } local function generate_etydate_obj(etydate_text) local etydate_specs = {} for spec in etydate_text:gmatch("[^,]+") do table.insert(etydate_specs, mw.text.trim(spec)) end return { [1] = etydate_specs } end local parsed_etydate = M.parse_utilities.parse_inline_modifiers(etydate, { param_mods = etydate_param_mods, generate_obj = generate_etydate_obj }) local etydate_args = { [1] = parsed_etydate[1], nocap = parsed_etydate.nocap or false, } ety_data_tree.supplements = ety_data_tree.supplements or {} table.insert(ety_data_tree.supplements, { type = "etydate", etydate_text = M.etydate.format_etydate(etydate_args, { omit_refs = true }), etydate_refs = (parsed_etydate.ref and #parsed_etydate.ref > 0) and parsed_etydate.ref or nil, }) end TreeBuilder.append_term_supplement(ety_data_tree, lang, "doublet", doublet) if ety_data_tree.supplements then parse_tree_references(ety_data_tree) end local has_visible_children = node_has_visible_tree_children(ety_data_tree) -- Suppress trees for multiword entries and one-step chains local visible_tree_depth = get_visible_tree_depth(ety_data_tree) local is_trivial_tree = visible_tree_depth <= 2 local is_multiword = title:find("%s") ~= nil or title:find("_") ~= nil if tree and (is_multiword or is_trivial_tree) then tree = false end if tree then table.insert(output, M.template_styles("Module:etymon/styles.css")) table.insert(output, M.tree.render({ data_tree = ety_data_tree, format_term_func = function(term, is_toplevel) return Util.format_term(term, is_toplevel, { gloss = "suppress", pos = "suppress", lit = "suppress", tree_ql = "suppress", }) end, })) end local tree_disallowed = lang_exc and lang_exc.disallow and lang_exc.disallow.tree local ety_tree_json = M.JSON.toJSON(tree_to_json(ety_data_tree)) local anchor = M.anchors.etymonid(lang, id, { no_tree = args.notree, title = title, empty_tree = (not has_visible_children) or tree_disallowed, ety_tree_json = ety_tree_json, }) table.insert(output, anchor) local text_stop_lang_missing = nil if text then local max_depth, stop_at_blue_link, stop_at_lang, stop_at_lang_or_bluelink if text == "++" then max_depth, stop_at_blue_link = false, false elseif text == "+" then max_depth, stop_at_blue_link = 1, false elseif text == "*" then max_depth, stop_at_blue_link = false, true elseif text:match("^:[^*]+%*$") then -- Stop at a specific language OR first bluelink after it, e.g., ":ota*" -- If the target language is a redlink, continue to the first bluelink local lang_code = text:match("^:([^*]+)%*$") if lang_code and lang_code ~= "" then local lang_obj = Util.get_lang(lang_code, true) if lang_obj then stop_at_lang_or_bluelink = lang_code else Util.add_warning('Invalid language code "' .. lang_code .. '" in text parameter. Showing full chain instead.') max_depth, stop_at_blue_link = false, false end else Util.add_warning('Empty language code in text parameter. Showing full chain instead.') max_depth, stop_at_blue_link = false, false end elseif text:sub(1, 1) == ":" then -- Stop at a specific language, e.g., ":ar" stops at first Arabic term local lang_code = text:sub(2) if lang_code ~= "" then -- Validate the language code local lang_obj = Util.get_lang(lang_code, true) if lang_obj then stop_at_lang = lang_code else Util.add_warning('Invalid language code "' .. lang_code .. '" in text parameter. Showing full chain instead.') max_depth, stop_at_blue_link = false, false -- default to ++ end else Util.add_warning('Empty language code in text parameter. Showing full chain instead.') max_depth, stop_at_blue_link = false, false -- default to ++ end else local num = tonumber(text) if num and num >= 1 then max_depth, stop_at_blue_link = num, false else error('Invalid text value "' .. text .. '". Valid values are: "++" (full chain), "+" (first step only), "*" (until first blue link), a number (max steps), ":lang" (stop at language), or ":lang*" (stop at language or first bluelink if redlink)') end end local text_output, text_render_meta = M.text.render({ data_tree = ety_data_tree, format_term_func = Util.format_term, lang_matches_stop_code = Util.lang_matches_stop_code, max_depth = max_depth, stop_at_blue_link = stop_at_blue_link, curr_page = page_data.pagename, nodot = args.nodot, dot = args.dot, stop_at_lang = stop_at_lang, stop_at_lang_or_bluelink = stop_at_lang_or_bluelink, }) table.insert(output, text_output) if stop_at_lang and text_render_meta and not text_render_meta.stop_lang_reached then M.tracking.track_text_stop_lang_missing(lang, stop_at_lang) text_stop_lang_missing = stop_at_lang end end if rfe then table.insert(output, Util.expand_request_template(frame, "rfe", rfe, lang:getCode())) end if etystub then table.insert(output, Util.expand_request_template(frame, "etystub", etystub, lang:getCode())) end if is_nonlemma then table.insert(output, " " .. frame:expandTemplate({ title = "nonlemma", args = {}, })) end local categories = {} if Util.is_content_page() then M.tracking.track_tree_metrics({ max_depth_reached = __state.max_depth_reached, total_nodes = __state.total_nodes, language_count = __state.language_count, lang = lang, }) categories = M.categories.build({ data_tree = ety_data_tree, page_lang = lang, available_etymon_ids = __state.available_etymon_ids, senseid_parent_etymon = __state.senseid_parent_etymon, get_norm_lang_func = Util.get_norm_lang, lang_exc = lang_exc, suppress_categories = lang_exc and lang_exc.suppress_categories, nocat = args.nocat, tree = tree, text = text, exnihilo = args.exnihilo, toplevel_has_inline_etymology = __state.toplevel_has_inline_etymology, toplevel_redundant_etymology = __state.toplevel_redundant_etymology, toplevel_idless_etymon = __state.toplevel_idless_etymon, has_mismatched_id = __state.has_mismatched_id, linked_page_multiple_etymons_idless = __state.linked_page_multiple_etymons_idless, linked_page_partial_etymology_sections = __state.linked_page_partial_etymology_sections, text_stop_lang_missing = text_stop_lang_missing, }) M.tracking.track_keywords(__state.toplevel_keyword_stats, lang) M.tracking.track_page_id(lang, id) M.tracking.track_ids(__state.id_stats, lang) end if #categories > 0 then table.insert(output, M.categories.format(categories, lang)) end if __state.warnings then for i, warning in ipairs(__state.warnings) do table.insert(output, (i == 1 and "\n" or "") .. warning .. "\n") end end return table.concat(output) end return export d0thspkud5zawi6og8iirurr2pt34q1