မေႃႇၵျူး:en-headword
This module is used for English headword-line templates. This module currently implements {{en-noun}}, {{en-proper noun}}, {{en-verb}}, {{en-adj}}, {{en-adv}}, {{en-intj}} and {{en-con}} (for conjunctions). See the documentation of those templates for more information. Other English headword templates are in the process of being converted to use this module.
The module is always invoked the same way, by passing a single parameter to the "show" function. This parameter is the name of the part of speech, but in plural (examples given are for nouns, and for adjective forms respectively):
{{#invoke:en-headword|show|nouns}} {{#invoke:en-headword|show|adjective forms}}
The template will, by default, accept the following parameters (specific parts of speech may accept or require others):
|head=
,|head2=
,|head3=
, ...- Override the headword display; used to add links to individual words in a multiword term.
|id=
- Sense ID for linking to this headword. See {{senseid}} for more information.
|nolink=1
or|nolinkhead=1
- Don't link individual words in the headword of a multiword term. Useful for foreign or otherwise unanalyzable terms like a posteriori and yabba dabba doo where the expression functions as a whole in English but the individual parts are not English words.
|splithyph=1
- Indicate that automatic splitting and linking of words should split on hyphens in multiword expressions with spaces in them, even if the hyphenated component would normally be linked as-is or with hyphens converted to spaces. See #Autosplitting below.
|nosplithyph=1
- Indicate that automatic splitting and linking of words should not split on hyphens in multiword expressions with spaces in them, even if this would normally happen. See #Autosplitting below.
|hyphspace=1
- Indicate that hyphenated components should be linked as a whole using the space-separated equivalent, even if this would not normally happen (i.e. because the space-separated equivalent is not defined as an English term). See #Autosplitting below.
|nosuffix=1
- Prevent terms beginning with a hyphen from being interpreted as suffixes. See #Suffix handling below.
|nomultiwordcat=1
- Prevent multiword terms (those with spaces or with hyphens in the middle) from being added to Category:English multiword terms.
|pagename=
- Override the page name used to compute default values of various sorts. Useful when testing, for documentation pages, etc.
|sort=
- Sort key. Rarely needs to be specified, as it is normally automatically generated.
Autosplitting
[မႄးထတ်း]All templates using this module use an intelligent autosplitting algorithm to link portions of multipart and multiword expressions, as follows:
- If there are spaces in the term but no apostrophes or hyphens, the module will automatically split and link distinct space-separated words, similarly to {{head}}; hence, absent without leave will be linked as
[[absent]] [[without]] [[leave]]
. - If there are spaces and apostrophes but no hyphens, the module will likewise split and link distinct space-separated words, but may also split up words with apostrophes in them. Specifically:
- If a word ends in
's
, the part before the's
will be linked as a word, and the's
will be linked separately to -'s, on the assumption that the's
is functioning as a possessive. For example, Abel's impossibility theorem will be linked as[[Abel]][[-'s|'s]] [[impossibility]] [[theorem]]
. (An exception is made for one's, someone's, he's, she's and it's, which are linked as-is without splitting.) - If a word ends in
'
, the apostrophe will be linked to -' (on the assumption that the'
is functioning as a plural possessive, similarly to above), and the part before will be separately linked. If the part before ends in ans
, the module converts it to its singular equivalent and looks that up to see if it exists and has a definition as an English term. If so, the term is linked to the singular form; otherwise, it is linked to the plural form. (Converting to the singular means that-ies
becomes-y
;-es
is dropped aftersh
,ch
andx
; and otherwises
is dropped.) For example, flies' graveyard will be linked as[[fly|flies]][[-'|']] [[graveyard]]
because fly exists as an English term, but Achilles' heel will be linked as[[Achilles]][[-'|']] [[heel]]
because Achille does not exist as an English term. - All other terms containing apostrophes are linked unsplit.
- If a word ends in
- If there are hyphens in the term but no spaces or apostrophes, the hyphenated components will be linked individually. For example, beggar-thy-neighbor will be linked as
[[beggar]]-[[thy]]-[[neighbor]]
.- An exception to this occurs with certain recognized prefixes, which are linked with the hyphen included in the prefix. For example, Afro-American is linked as
[[Afro-]][[American]]
and co-occurrence is linked as[[co-]][[occurrence]]
, because Afro- and co- are in the list of recognized prefixes. (For the full list, see below.)
- An exception to this occurs with certain recognized prefixes, which are linked with the hyphen included in the prefix. For example, Afro-American is linked as
- If there are hyphens and apostrophes but no spaces, the effect is similar to the situation with spaces and apostrophes. For example, beggar's-lice is linked as
[[beggar]][[-'s|'s]]-[[lice]]
. - If there are both hyphens and spaces, the space-separated components that do not have hyphens will be linked separately, as above. Any hyphen-separated components may be linked in one of three ways:
- If
|hyphspace=1
is specified or the hyphen-separated component exists as an English term when the hyphens are converted to spaces, it will be linked to that term. For example, closed-circuit television will be linked as[[closed circuit|closed-circuit]] [[television]]
because closed circuit exists as an English term. (In this case, closed-circuit also exists but is approximately a soft redirect to closed circuit, as is often the case with such attributive compounds. This is why we prefer the space-separated variant.) - If
|nosplithyph=1
is specified or the hyphen-separated component exists as an English term in its unmodified form but not when the hyphens are converted to spaces, it will be linked as an unmodified whole. For example, coin-operated laundry will be linked as[[coin-operated]] [[laundry]]
because coin-operated exists as an English term but coin operated does not. (An example that requires|nosplithyph=1
is close-up lens, where the default algorithm would incorrectly link the first component to close up. Here, close up [a verb] and close-up [an adjective] both exist but refer to different things.) - If
|splithyph=1
is specified or the hyphen-separated component does not exist as an English term (either unmodified or when the hyphens are converted to spaces), each hyphenated component is linked separately. Examples where this happens are adult-onset diabetes (linked as[[adult]]-[[onset]] [[diabetes]]
) and Bombieri-Friedlander-Iwaniec theorem linked as[[Bombieri]]-[[Friedlander]]-[[Iwaniec]] [[theorem]]
). Note that when separately linking hyphenated components, prefixes are recognized and handled specially, as documented below.
- If
Special prefix handling
[မႄးထတ်း]As described above, when splitting hyphenated components, if a component is not the last component and looks like one of the following prefixes, the following hyphen will be included inside of the link.
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
Suffix handling
[မႄးထတ်း]If the term begins with a hyphen (-
), it is assumed to be a suffix rather than a base form, and is categorized into Category:English suffixes and Category:English POS-forming suffixes rather than Category:English POSs (e.g. Category:English noun-forming suffixes rather than Category:English nouns). This can be overridden using |nosuffix=1
. (An example where this is necessary is -ussification, which refers to a linguistic process of blending words with the suffix -ussy but is not itself a suffix.)
Link modifications
[မႄးထတ်း]The default behavior described above under #Autosplitting is sufficient in most circumstances, but some multiword terms need special linking behavior to handle things like inflected terms (e.g. those ending in -ing or -s), capitalized terms, multiword subexpressions, etc. One way to handle that is to use |head=
and spell out the entire headword, appropriately linked, effectively ignoring the default linking behavior. But this can be awkward for long multiword terms. For cases like this, a shortcut syntax is provided to apply link modifications on top of the autolinked term. To enable this, put a tilde (~
) at the beginning of the value specified to |head=
, followed by the changes to individual words.
For example, for the term acute necrotising ulcerative gingivitis, we would like to link necrotising to necrotise. This can be done as follows:
- {{en-noun}}
or more compactely as
- {{en-noun}}
This is equivalent to writing {{en-noun}}, but shorter. In general, syntax of the form prefix[from:to]
is equivalent to writing prefixfrom:prefixto
, and says to replace prefixfrom
with prefixto
in the default output produced by the #Autosplitting mechanism described above.
The same syntax works on the beginning of a word, which is especially useful when linking to the lowercase equivalent of a capitalized term. For example, for admiral of the Swiss Navy, use the following to link Navy to navy:
- {{en-noun}}
This is equivalent to writing {{en-noun}} but shorter.
Modifications need to match full words, but can be applied to multiple words. A ~
on the right-hand side is a shortcut that stands for the left-hand side, which is especially useful when multiple words are given on the left-hand side, and causes the words to be linked together. For example, for acute respiratory distress syndrome, to link respiratory distress as a single entity, use the following:
- {{en-noun}}
which is equivalent to {{en-noun}}. The right-hand side need not consist solely of a tilde, but can contain other surrounding text. For example, for Charlie Brown Christmas tree, use the following to link to the Wikipedia entry for Charlie Brown:
- {{en-noun}}
This is equivalent to writing {{en-noun}}.
Multiple modifications can be specified, separated by a semicolon (optionally with surrounding spaces). For example, for Admiral of the Fleet, use:
- {{en-noun}}
This is equivalent to writing {{en-noun}}.
local export = {}
local pos_functions = {}
local force_cat = false -- for testing; if true, categories appear in non-mainspace pages
local m_links = require("Module:links")
local table_module = "Module:table"
local headword_utilities_module = "Module:headword utilities"
local string_utilities_module = "Module:string utilities"
local lang = require("Module:languages").getByCode("en")
local langname = lang:getCanonicalName()
local rsplit = mw.text.split
local function glossary_link(entry, text)
text = text or entry
return "[[Appendix:Glossary#" .. entry .. "|" .. text .. "]]"
end
local function track(page)
require("Module:debug/track")("en-headword/" .. page)
return true
end
-- The main entry point.
-- This is the only function that can be invoked from a template.
function export.show(frame)
local poscat = frame.args[1] or error("Part of speech has not been specified. Please pass parameter 1 to the module invocation.")
local params = {
["head"] = {list = true},
["id"] = {},
["json"] = {type = "boolean"},
["sort"] = {},
["splithyph"] = {type = "boolean"},
["nosplithyph"] = {type = "boolean"},
["hyphspace"] = {type = "boolean"},
["nolink"] = {type = "boolean"},
["nolinkhead"] = {type = "boolean", alias_of = "nolink"},
["nosuffix"] = {type = "boolean"},
["nomultiwordcat"] = {type = "boolean"},
["pagename"] = {}, -- for testing
}
local pos_data = pos_functions[poscat]
if pos_data then
for key, val in pairs(pos_data.params) do
params[key] = val
end
end
local args = require("Module:parameters").process(frame:getParent().args, params, nil, "en-headword", "show")
local pagename = args.pagename or mw.loadData("Module:headword/data").pagename -- Accounts for unsupported titles.
local user_specified_heads = args.head
local heads = user_specified_heads
local autohead
if args.nolink or not pagename:find("[ '%-]") then
autohead = pagename
else
local m_headutil = require(headword_utilities_module)
local en_no_split_apostrophe_words = require("Module:table/listToSet") {
"one's",
"someone's",
"he's",
"she's",
"it's",
}
local en_include_hyphen_prefixes = require("Module:table/listToSet") {
-- 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("%-") 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$")
if base then
return "[[" .. base .. "]][[-'s|'s]]"
end
base = word:match("^(.*)'$")
if base then
if base:find("s$") then
local sg = require(string_utilities_module).singularize(base)
if is_english(sg) then
return "[[" .. sg .. "|" .. base .. "]][[-'|']]"
end
end
return "[[" .. base .. "]][[-'|']]"
end
return "[[" .. word .. "]]"
end
autohead = m_headutil.add_links_to_multiword_term(pagename, {
split_hyphen_when_space = en_split_hyphen_when_space,
split_apostrophe = en_split_apostrophe,
no_split_apostrophe_words = en_no_split_apostrophe_words,
include_hyphen_prefixes = en_include_hyphen_prefixes,
})
end
if #heads == 0 then
heads = {autohead}
else
for i, head in ipairs(heads) do
if head:find("^~") then
head = require(headword_utilities_module).apply_link_modifiers(autohead, head:sub(2))
heads[i] = head
end
if head == autohead then
track("redundant-head")
end
end
end
local data = {
lang = lang,
pos_category = poscat,
categories = {},
heads = heads,
user_specified_heads = user_specified_heads,
no_redundant_head_cat = #user_specified_heads == 0,
inflections = {},
nomultiwordcat = args.nomultiwordcat,
sort_key = args.sort,
pagename = args.pagename,
-- This is always set, and in the case of unsupported titles, it's the displayed version (e.g. 'C|N>K' instead of
-- 'Unsupported titles/C through N to K').
displayed_pagename = pagename,
id = args.id,
force_cat_output = force_cat,
}
local is_suffix = false
if not args.nosuffix and pagename:find("^%-") and not pagename:find("^%-%-") and poscat ~= "suffix forms" then
is_suffix = true
data.pos_category = "suffixes"
local singular_poscat = require("Module:string utilities").singularize(poscat)
table.insert(data.categories, singular_poscat .. " " .. langname .. "-forming suffixes")
table.insert(data.inflections, {label = singular_poscat .. "-forming suffix"})
end
if pos_data then
pos_data.func(args, data, is_suffix)
end
local extra_categories = {}
if pagename:find("[Qq][^Uu]") or pagename:find("[Qq]$") then
table.insert(data.categories, langname .. " words containing Q not followed by U")
end
-- mw.ustring.toNFD performs decomposition, so letters that decompose
-- to an ASCII vowel and a diacritic, such as é, are counted as vowels and
-- do not need to be included in the pattern.
if not mw.ustring.find(mw.ustring.lower(mw.ustring.toNFD(pagename)), "[aeiouyæœø]") then
table.insert(data.categories, langname .. " words without vowels")
end
if pagename:find("yre$") then
table.insert(data.categories, langname .. ' words ending in "-yre"')
end
if not pagename:find(" ") and mw.ustring.len(pagename) >= 25 then
table.insert(extra_categories, "Long " .. langname .. ' words')
end
if pagename:find("^[^aeiou ]*a[^aeiou ]*e[^aeiou ]*i[^aeiou ]*o[^aeiou ]*u[^aeiou ]*$") then
table.insert(data.categories, langname .. ' words that use all vowels in alphabetical order')
end
if args.json then
return require("Module:JSON").toJSON(data)
end
return require("Module:headword").full_headword(data)
.. (#extra_categories > 0
and require("Module:utilities").format_categories(extra_categories, lang, args.sort)
or "")
end
-- This function does the common work between adjectives and adverbs
local function make_comparatives(params, data)
local comp_parts = {label = glossary_link("ဢၼ်တႅၵ်ႈၼိူင်း"), accel = {form = "ဢၼ်တႅၵ်ႈၼိူင်း"}}
local sup_parts = {label = glossary_link("ဢၼ်ၶိုၵ်ႉသုတ်း"), accel = {form = "ဢၼ်ၶိုၵ်ႉသုတ်း"}}
local pagename = data.displayed_pagename
if #params == 0 then
table.insert(params, {"more"})
end
-- To form the stem, replace -(e)y with -i and remove a final -e.
local stem = pagename:gsub("([^aeiou])e?y$", "%1i"):gsub("e$", "")
-- Go over each parameter given and create a comparative and superlative form
for i, val in ipairs(params) do
local comp = val[1]
local comp_qual = val[2]
local sup = val[3]
local sup_qual = val[4]
local comp_part, sup_part
if comp == "more" and pagename ~= "many" and pagename ~= "much" then
comp_part = "[[more]] " .. pagename
sup_part = "[[most]] " .. pagename
elseif comp == "further" and pagename ~= "far" then
comp_part = "[[further]] " .. pagename
sup_part = "[[furthest]] " .. pagename
elseif comp == "er" then
comp_part = stem .. "er"
sup_part = stem .. "est"
elseif comp == "-" or sup == "-" then
-- Allowing '-' makes it more flexible to not have some forms
if comp ~= "-" then
comp_part = comp
end
if sup ~= "-" then
sup_part = sup
end
else
-- If the full comparative was given, but no superlative, then
-- create it by replacing the ending -er with -est.
if not sup then
if comp:find("er$") then
sup = comp:gsub("er$", "est")
else
error("The superlative of \"" .. comp .. "\" cannot be generated automatically. Please provide it with the \"sup" .. (i == 1 and "" or i) .. "=\" parameter.")
end
end
comp_part = comp
sup_part = sup
end
if comp_part then
table.insert(comp_parts, {term = comp_part, q = {comp_qual}})
end
if sup_part then
table.insert(sup_parts, {term = sup_part, q = {sup_qual}})
end
end
table.insert(data.inflections, comp_parts)
table.insert(data.inflections, sup_parts)
end
local function make_heads_definite(args, data)
if args.def == "~" then
local newheads = {}
for i, head in ipairs(data.heads) do
table.insert(newheads, head)
table.insert(newheads, "the " .. head)
end
data.heads = newheads
else
for i, head in ipairs(data.heads) do
data.heads[i] = "the " .. head
end
end
end
pos_functions["ၵမ်ႉႁိၵ်ႈ"] = {
params = {
[1] = {list = true, allow_holes = true},
["def"] = {},
["the"] = {alias_of = "def"},
["comp_qual"] = {list = "comp\1_qual", allow_holes = true},
["sup"] = {list = true, allow_holes = true},
["sup_qual"] = {list = "sup\1_qual", allow_holes = true},
},
func = function(args, data)
local shift = 0
local is_not_comparable = false
local is_comparative_only = false
if args.def then
make_heads_definite(args, data)
end
-- If the first parameter is ?, then don't show anything, just return.
if args[1][1] == "?" then
return
-- If the first parameter is -, then move all parameters up one position.
elseif args[1][1] == "-" then
shift = 1
is_not_comparable = true
-- If the only argument is +, then remember this and clear parameters
elseif args[1][1] == "+" and args[1].maxindex == 1 then
shift = 1
is_comparative_only = true
end
-- Gather all the comparative and superlative parameters.
local params = {}
for i = 1, args[1].maxindex - shift do
local comp = args[1][i + shift]
local comp_qual = args["comp_qual"][i + shift]
local sup = args["sup"][i]
local sup_qual = args["sup_qual"][i + shift]
if comp or sup then
table.insert(params, {comp, comp_qual, sup, sup_qual})
end
end
if shift == 1 then
-- If the first parameter is "-" but there are no parameters,
-- then show "not comparable" only and return.
-- If there are parameters, then show "not generally comparable"
-- before the forms.
if #params == 0 then
if is_not_comparable then
table.insert(data.inflections, {label = "ဢမ်ႇၸႂ်ႈ " .. glossary_link("ဢၼ်တႅၵ်ႈၼိူင်းလႆႈ")})
table.insert(data.categories, " ၵမ်ႉႁိၵ်ႈ " .. langname .. " ဢၼ်တႅၵ်ႈၼိူင်းဢမ်ႇလႆႈ")
return
end
if is_comparative_only then
table.insert(data.inflections, {label = glossary_link("ဢၼ်တႅၵ်ႈၼိူင်း") .. " form only"})
table.insert(data.categories," ၵမ်ႉႁိၵ်ႈ " .. langname .. " ဢၼ်တႅၵ်ႈၼိူင်းလၢႆလၢႆၵူၺ်း")
return
end
else
table.insert(data.inflections, {label = "ၵမ်ႉၼမ် ဢမ်ႇၸႂ်ႈ " .. glossary_link("ဢၼ်တႅၵ်ႈၼိူင်းလႆႈ")})
end
end
-- Process the parameters
make_comparatives(params, data)
end
}
pos_functions["ၵမ်ႉသၢင်ႈ"] = {
params = {
[1] = {list = true, allow_holes = true},
["comp_qual"] = {list = "comp\1_qual", allow_holes = true},
["sup"] = {list = true, allow_holes = true},
["sup_qual"] = {list = "sup\1_qual", allow_holes = true},
},
func = function(args, data)
local shift = 0
-- If the first parameter is ?, then don't show anything, just return.
if args[1][1] == "?" then
return
-- If the first parameter is -, then move all parameters up one position.
elseif args[1][1] == "-" then
shift = 1
end
-- Gather all the comparative and superlative parameters.
local params = {}
for i = 1, args[1].maxindex - shift do
local comp = args[1][i + shift]
local comp_qual = args["comp_qual"][i + shift]
local sup = args["sup"][i]
local sup_qual = args["sup_qual"][i + shift]
if comp or sup then
table.insert(params, {comp, comp_qual, sup, sup_qual})
end
end
if shift == 1 then
-- If the first parameter is "-" but there are no parameters,
-- then show "not comparable" only and return. If there are parameters,
-- then show "not generally comparable" before the forms.
if #params == 0 then
table.insert(data.inflections, {label = "ဢမ်ႇၸႂ်ႈ " .. glossary_link("ဢၼ်တႅၵ်ႈၼိူင်းလႆႈ")})
table.insert(data.categories,"ၵမ်ႉသၢင်ႈ " .. langname .. " ၸိူဝ်းတႅၵ်ႈၼိူင်းဢမ်ႇလႆႈ")
return
else
table.insert(data.inflections, {label = "not generally " .. glossary_link("comparable")})
end
end
-- Process the parameters
make_comparatives(params, data)
end
}
pos_functions["ၵပ်း"] = {
params = {
[1] = { alias_of = "head" },
},
func = function(args, data)
end,
}
pos_functions["တိုၼ်ႇ"] = {
params = {
[1] = { alias_of = "head" },
},
func = function(args, data)
end,
}
local function default_plural(noun)
local new_pl
if noun:find("[sxz]$") or noun:find("[cs]h$") then
new_pl = noun .. "es"
elseif noun:find("[^aeiou]y$") then
new_pl = noun:gsub("y$", "i") .. "es"
else
new_pl = noun .. "s"
end
return (new_pl
:gsub("\\([:#])", "\\\\%1")
:gsub("[:#]", "\\%0"))
end
local function canonicalize_plural(pl, stem, pagename)
local can_pl
if pl == "s" then
can_pl = stem .. "s"
elseif pl == "es" then
can_pl = stem .. "es"
elseif pl == "+" then
return default_plural(pagename)
else
return nil
end
return (can_pl
:gsub("\\([:#])", "\\\\%1")
:gsub("[:#]", "\\%0"))
end
local function do_nouns(args, data, is_proper)
local pagename = data.displayed_pagename
local function gather_inflections_with_quals(infl_field, qual_field, label)
-- Gather all the plural parameters from the numbered parameters.
local infls = {}
if label then
infls.label = label
end
for i, infl in ipairs(args[infl_field]) do
local qual = args[qual_field][i]
if qual then
table.insert(infls, {term = infl, q = {qual}})
else
table.insert(infls, infl)
end
end
return infls
end
if args.def then
make_heads_definite(args, data)
end
local plurals = gather_inflections_with_quals(1, "plqual")
if plurals[1] == "p" then
-- plurale tantum
if #plurals > 1 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
if #args.sg > 0 then
table.insert(data.inflections, {label = "normally plural"})
table.insert(data.inflections, gather_inflections_with_quals("sg", "sgqual", "singular"))
else
table.insert(data.inflections, {label = "plural only"})
end
if #args.attr > 0 then
table.insert(data.inflections, gather_inflections_with_quals("attr", "attrqual", "attributive"))
end
return
end
local need_default_plural = not is_proper
if plurals[1] == "-" then
-- Uncountable noun; may occasionally have a plural
table.remove(plurals, 1) -- Remove the "-"
table.insert(data.categories,"ႁိၵ်ႈ" .. langname .. " ၸိူဝ်းၼပ်ႉဢၢၼ်ႇဢမ်ႇလႆႈ")
-- If plural forms were given explicitly, then show "usually"
if #plurals > 0 then
table.insert(data.inflections, {label = "ၵမ်ႉၼမ် " .. glossary_link("ၼပ်ႉဢၢၼ်ႇဢမ်ႇလႆႈ")})
table.insert(data.categories,"ႁိၵ်ႈ" .. langname .. " ၸိူဝ်းၼပ်ႉဢၢၼ်ႇလႆႈ")
else
table.insert(data.inflections, {label = glossary_link("ၼပ်ႉဢၢၼ်ႇဢမ်ႇလႆႈ")})
end
need_default_plural = false
elseif plurals[1] == "~" then
-- Mixed countable/uncountable noun, always has a plural
table.remove(plurals, 1) -- Remove the "~"
table.insert(data.inflections, {label = glossary_link("ၼပ်ႉဢၢၼ်ႇလႆႈ") .. " လႄႈ " .. glossary_link("ၼပ်ႉဢၢၼ်ႇဢမ်ႇလႆႈ")})
table.insert(data.categories,"ႁိၵ်ႈ" .. langname .. " ၸိူဝ်းၼပ်ႉဢၢၼ်ႇဢမ်ႇလႆႈ")
table.insert(data.categories, "ႁိၵ်ႈ" .. langname .. " ၸိူဝ်းၼပ်ႉဢၢၼ်ႇလႆႈ")
-- If no plural was given, add a default one now
if #plurals == 0 then
plurals = {default_plural(pagename)}
end
elseif is_proper then
-- For proper nouns, the default is uncountable
table.insert(data.categories,"ႁိၵ်ႈ" .. langname .. " ၸိူဝ်းၼပ်ႉဢၢၼ်ႇဢမ်ႇလႆႈ")
else
-- For common nouns, the default is countable, has a plural
table.insert(data.categories, "ႁိၵ်ႈ" .. langname .. " ၸိူဝ်းၼပ်ႉဢၢၼ်ႇလႆႈ")
end
-- Plural is unknown
if plurals[1] == "?" then
table.remove(plurals, 1) -- Remove the "?"
-- Not desired; see [[Wiktionary:Tea_room/2021/August#"Plural unknown or uncertain"]]
-- table.insert(data.inflections, {label = "plural unknown or uncertain"})
table.insert(data.categories,"ႁိၵ်ႈ" .. langname .. " ၸိူဝ်းဢမ်ႇႁူႉ ဢမ်ႇၼၼ် တူဝ်ၼမ်ဢမ်ႇၼႄႉၼွၼ်း")
if #plurals > 0 then
error("Can't specify explicit plurals along with '?' for unknown/uncertain plural")
end
return
end
-- Plural is not attested
if plurals[1] == "!" then
table.remove(plurals, 1) -- Remove the "!"
table.insert(data.inflections, {label = "plural not attested"})
table.insert(data.categories,"ႁိၵ်ႈ" .. langname .. " ၸိူဝ်းတူဝ်ၼပ်ဢမ်ႇလပ်ႉလွင်း")
if #plurals > 0 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) return
if #plurals == 0 then
if need_default_plural then
plurals = {default_plural(pagename)}
else
return
end
end
-- There are plural forms to show, so show them
local pl_parts = {label = "တူဝ်ၼမ်", accel = {form = "p"}}
local function check_ies(pl, stem)
local newplural, nummatches = stem:gsub("([^aeiou])y$","%1ies")
return nummatches > 0 and pl == newplural
end
local stem = pagename
local irregular = false
for i, pl in ipairs(plurals) do
local canon_pl = canonicalize_plural(pl, stem, pagename)
if canon_pl then
table.insert(pl_parts, canon_pl)
elseif type(pl) == "table" then
canon_pl = canonicalize_plural(pl.term, stem, pagename)
if canon_pl then
table.insert(pl_parts, {term = canon_pl, q = pl.q})
end
end
if not canon_pl then
table.insert(pl_parts, pl)
if type(pl) == "table" then
pl = pl.term
end
local check_pl = m_links.get_link_page(pl, lang)
if not stem:find(" ") and not (check_pl == stem .. "s" or check_pl == stem .. "es" or check_ies(check_pl, stem)) then
irregular = true
if check_pl == stem then
table.insert(data.categories,"ႁိၵ်ႈ" .. langname .. " ၸိူဝ်းထဵင်ဢမ်ႇလႆႈ")
end
end
end
end
if irregular then
table.insert(data.categories,"ႁိၵ်ႈ" .. langname .. "ၸိူဝ်း တူဝ်ၼမ်ဢမ်ႇၸႂ်ႈပၵတိ")
end
table.insert(data.inflections, pl_parts)
end
-- Return the parameters to be used for nouns and proper nouns. Currently the same.
local function get_noun_params(is_proper)
return {
[1] = {list = true, disallow_holes = true},
["def"] = {},
["the"] = {alias_of = "def"},
["pl\1qual"] = {list = true, allow_holes = true},
-- The following four only used for pluralia tantum (1=p)
["sg"] = {list = true, disallow_holes = true},
["sg\1qual"] = {list = true, allow_holes = true},
["attr"] = {list = true, disallow_holes = true},
["attr\1qual"] = {list = true, allow_holes = true},
}
end
pos_functions["ႁိၵ်ႈ"] = {
params = get_noun_params(false),
func = do_nouns,
}
pos_functions["ႁိၵ်ႈသုၼ်ႇလဵဝ်"] = {
params = get_noun_params("is proper"),
func = function(args, data) return do_nouns(args, data, "is proper") end,
}
local function base_default_verb_forms(verb)
local s_form = default_plural(verb)
local ing_form, ed_form
local vowel = "aeiouáéíóúàèìòùâêîôûäëïöüæœø"
local ulvowel = vowel .. "AEIOUÁÉÍÓÚÀÈÌÒÙÂÊÎÔÛÄËÏÖÜÆŒØ"
-- (1) Check for C*VC verbs.
--
-- flip -> flipping/flipped, strum -> strumming/strummed, nag -> nagging/nagged, etc.
-- Do not include words with final -y, e.g. 'stay' (staying/stayed), 'toy' (toying/toyed),
-- or with final -w, e.g. 'flow' (flowing/flowed), or with final -h, e.g. 'ah' (ahing/ahed),
-- or with final -x, e.g. 'box' (boxing/boxed), or ending in an uppercase consonant,
-- e.g. 'XOR' (XORing/XORed), 'OK' (OKing/OKed). Check specially for initial y- as a consonant,
-- e.g. 'yip' (yipping/yipped), otherwise treat y as a vowel, so we don't trigger on 'hyphen'
-- but do trigger on 'gyp'.
local last_cons = mw.ustring.match(verb, "^[Yy][" .. vowel .. "y]([^A-Z" .. vowel .. "ywxh])$")
if not last_cons then
last_cons = mw.ustring.match(verb, "^[^" .. ulvowel .. "yY]*[" .. ulvowel .. "yY]([^A-Z" .. vowel .. "ywxh])$")
end
if last_cons then
ing_form = verb .. last_cons .. "ing"
ed_form = verb .. last_cons .. "ed"
else
-- (2) Generate -ing form.
-- (2a) lie -> lying, untie -> untying, etc.
local stem = verb:match("^(.*)ie$")
if stem then
ing_form = stem .. "ying"
else
-- (2b) argue -> arguing, sprue -> spruing, dialogue -> dialoguing, etc.
stem = verb:match("^(.*)ue$")
if stem then
ing_form = stem .. "uing"
else
stem = mw.ustring.match(verb, "^(.*[" .. ulvowel .. "yY][^" .. vowel .. "y]+)e$")
if stem then
-- (2c) baptize -> baptizing, rake -> raking, type -> typing, parse -> parsing, etc.
-- (ending in vowel + consonant(s) + -e); but not referee -> refereeing,
-- backhoe -> backhoeing, redye -> redyeing (ending in some other vowel + -e or in -ye);
-- and not be -> being (no vowel before the consonant preceding the -e)
ing_form = stem .. "ing"
else
-- (2d) regular verbs
ing_form = verb .. "ing"
end
end
end
-- (3) Generate -ed form.
if verb:find("e$") then
-- (3a) baptize -> baptized, rake -> raked, parse -> parsed, free -> freed, hoe -> hoed
ed_form = verb .. "d"
else
stem = mw.ustring.match(verb, "^(.*[^" .. ulvowel .. "yY])y$")
if stem then
-- (3b) marry -> married, levy -> levied, try -> tried, etc.; but not toy -> toyed
ed_form = stem .. "ied"
else
-- (3c) regular verbs
ed_form = verb .. "ed"
end
end
end
ing_form = ing_form
:gsub("\\([:#])", "\\\\%1")
:gsub("[:#]", "\\%0")
ed_form = ed_form
:gsub("\\([:#])", "\\\\%1")
:gsub("[:#]", "\\%0")
return s_form, ing_form, ed_form
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("^(.-)( .*)$")
local first_s_form, first_ing_form, first_ed_form = base_default_verb_forms(first)
return full_s_form, full_ing_form, full_ed_form, first_s_form .. rest, first_ing_form .. rest, first_ed_form .. rest
else
return full_s_form, full_ing_form, full_ed_form, nil, nil, nil
end
end
pos_functions["သၢင်ႈ"] = {
params = {
[1] = {list = "pres_3sg", allow_holes = true},
["pres_3sg_qual"] = {list = "pres_3sg\1_qual", allow_holes = true},
[2] = {list = "pres_ptc", allow_holes = true},
["pres_ptc_qual"] = {list = "pres_ptc\1_qual", allow_holes = true},
[3] = {list = "past", allow_holes = true},
["past_qual"] = {list = "past\1_qual", allow_holes = true},
[4] = {list = "past_ptc", allow_holes = true},
["past_ptc_qual"] = {list = "past_ptc\1_qual", allow_holes = true},
["noautolinkverb"] = {type = "boolean"},
},
func = function(args, data)
-- Get parameters
local par1 = args[1][1]
local par2 = args[2][1]
local par3 = args[3][1]
local par4 = args[4][1]
local pres_3sgs, pres_ptcs, pasts, past_ptcs
local pagename = data.displayed_pagename
------------------------------------------- UTILITY FUNCTIONS #1 ------------------------------------------
-- These functions are used directly in the <> format as well as in the utility functions #2 below.
local function compute_double_last_cons_stem(verb)
local last_cons = verb:match("([bcdfghjklmnpqrstvwxyzBCDFGHJKLMNPQRSTVWXYZ])$")
if not last_cons then
error("Verb stem '" .. verb .. "' must end in a consonant to use ++")
end
return verb .. last_cons
end
local function compute_plusplus_s_form(verb, default_s_form)
if verb:find("[sz]$") then
-- regas -> regasses, derez -> derezzes
return compute_double_last_cons_stem(verb) .. "es"
else
return default_s_form
end
end
------------------------------------------- UTILITY FUNCTIONS #2 ------------------------------------------
-- These functions are used in both in the separate-parameter format and in the override params such as past_ptc2=.
local new_default_s, new_default_ing, new_default_ed, split_default_s, split_default_ing, split_default_ed =
default_verb_forms(pagename)
local function compute_double_last_cons_stem_of_split_verb(verb, ending)
local first, rest = verb:match("^(.-)( .*)$")
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)
if form == nil then
error("Verb '" .. pagename .. "' must have a space in it to use * or ++*")
end
return form
end
local function sub_tilde(form)
if not form then
return nil
end
local retval = form:gsub("~", pagename) -- discard second return value
return retval
end
local function canonicalize_s_form(form)
if form == "+" then
return new_default_s
elseif form == "*" then
return check_non_nil_star_form(split_default_s)
elseif form == "++" then
return compute_plusplus_s_form(pagename, new_default_s)
elseif form == "++*" then
if pagename:find("^[^ ]*[sz] ") then
return compute_double_last_cons_stem_of_split_verb(pagename, "es")
else
return check_non_nil_star_form(split_default_s)
end
else
return sub_tilde(form)
end
end
local function canonicalize_ing_form(form)
if form == "+" then
return new_default_ing
elseif form == "*" then
return check_non_nil_star_form(split_default_ing)
elseif form == "++" then
return compute_double_last_cons_stem(pagename) .. "ing"
elseif form == "++*" then
return compute_double_last_cons_stem_of_split_verb(pagename, "ing")
else
return sub_tilde(form)
end
end
local function canonicalize_ed_form(form)
if form == "+" then
return new_default_ed
elseif form == "*" then
return check_non_nil_star_form(split_default_ed)
elseif form == "++" then
return compute_double_last_cons_stem(pagename) .. "ed"
elseif form == "++*" then
return compute_double_last_cons_stem_of_split_verb(pagename, "ed")
else
return sub_tilde(form)
end
end
--------------------------------- MAIN PARSING/CONJUGATING CODE --------------------------------
local past_ptcs_given
if par1 and par1:find("<") then
-------------------------- ANGLE-BRACKET FORMAT --------------------------
if par2 or par3 or par4 then
error("Can't specify 2=, 3= or 4= when 1= contains angle brackets: " .. par1)
end
-- In the angle bracket format, we always copy the full past tense specs to the past participle
-- specs if none of the latter are given, so act as if the past participle is always given.
-- There is a separate check to see if the past tense and past participle are identical, in any case.
past_ptcs_given = true
local iut = require("Module:inflection utilities")
-- (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 = iut.parse_balanced_segment_run(inside, "[", "]")
local comma_separated_groups = iut.split_alternating_runs(segments, ",")
if #comma_separated_groups > 4 then
error("Too many comma-separated parts in indicator spec: " .. angle_bracket_spec)
end
local function fetch_qualifiers(separated_group)
local qualifiers
for j = 2, #separated_group - 1, 2 do
if separated_group[j + 1] ~= "" then
error("Extraneous text after bracketed qualifiers: '" .. table.concat(separated_group) .. "'")
end
if not qualifiers then
qualifiers = {}
end
table.insert(qualifiers, separated_group[j])
end
return qualifiers
end
local function fetch_specs(comma_separated_group)
if not comma_separated_group then
return {{}}
end
local specs = {}
local colon_separated_groups = iut.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 == "++*" then
error("* and ++* not allowed inside of indicator specs: " .. angle_bracket_spec)
end
if form == "" then
form = nil
end
table.insert(specs, {form = form, q = fetch_qualifiers(colon_separated_group)})
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])
for _, spec in ipairs(s_specs) do
if spec.form == "++" and #ing_specs == 1 and not ing_specs[1].form and not ing_specs[1].q
and #ed_specs == 1 and not ed_specs[1].form and not ed_specs[1].q then
ing_specs[1].form = "++"
ed_specs[1].form = "++"
break
end
end
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(par1, 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 = m_links.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 def_s_form, def_ing_form, def_ed_form = base_default_verb_forms(base.lemma)
local function process_specs(slot, specs, default_form, canonicalize_plusplus)
for _, spec in ipairs(specs) do
local form = spec.form
if not form or form == "+" then
form = default_form
elseif form == "++" then
form = canonicalize_plusplus()
end
-- If there's a ~ in the form, substitute it with the lemma,
-- but make sure to first replace % in the lemma with %% so that
-- it doesn't get interpreted as a capture replace expression.
if form:find("~") then
-- Assign to a var because gsub returns multiple values.
local subbed_lemma = base.lemma:gsub("%%", "%%%%")
form = form:gsub("~", subbed_lemma)
end
-- 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 form ~= "-" then
iut.insert_form(base.forms, slot, {form = form, footnotes = spec.q})
end
end
end
process_specs("s_form", base.s_specs, def_s_form,
function() return compute_plusplus_s_form(base.lemma, def_s_form) end)
process_specs("ing_form", base.ing_specs, def_ing_form,
function() return compute_double_last_cons_stem(base.lemma) .. "ing" end)
process_specs("ed_form", base.ed_specs, def_ed_form,
function() return compute_double_last_cons_stem(base.lemma) .. "ed" end)
-- If the -en spec is completely missing, substitute the -ed spec in its entirely.
-- Otherwise, if individual -en forms are missing or use +, we will substitute the
-- default -ed form, as with the -ed spec.
local en_specs = base.en_specs
if #en_specs == 1 and not en_specs[1].form and not en_specs[1].q then
en_specs = base.ed_specs
end
process_specs("en_form", en_specs, def_ed_form,
function() return compute_double_last_cons_stem(base.lemma) .. "ed" end)
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_forms(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 #forms == 0 then
forms = {{form = "-"}}
end
return forms
end
pres_3sgs = fetch_forms("s_form")
pres_ptcs = fetch_forms("ing_form")
pasts = fetch_forms("ed_form")
past_ptcs = fetch_forms("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 #data.user_specified_heads == 0 and alternant_multiword_spec.saw_bracket then
data.heads = {}
for _, lemma_obj in ipairs(alternant_multiword_spec.forms.lemma_linked) do
local lemma = lemma_obj.form
local footnotes = lemma_obj.footnotes
local quals, refs
if footnotes then
quals, refs = require("Module:inflection utilities").fetch_headword_qualifiers_and_references(footnotes)
end
table.insert(data.heads, {term = lemma, q = quals, refs = refs})
end
end
else
-------------------------- SEPARATE-PARAM FORMAT --------------------------
local pres_3sg, pres_ptc, past
if par1 and not par2 and not par3 then
-- Use of a single parameter other than "++", "*" or "++*" is now the "legacy" format,
-- and no longer supported.
if par1 == "es" or par1 == "ies" or par1 == "d" then
error("Legacy parameter 1=es/ies/d no longer supported, just use 'en-verb' without params")
elseif par1 == "++" or par1 == "*" or par1 == "++*" then
pres_3sg = canonicalize_s_form(par1)
pres_ptc = canonicalize_ing_form(par1)
past = canonicalize_ed_form(par1)
else
error("Legacy parameter 1=STEM no longer supported, just use 'en-verb' without params")
end
else
if par3 then
track("xxx3")
elseif par2 then
track("xxx2")
end
end
if not pres_3sg or not pres_ptc or not past then
-- Either all three should be set above, or none of them.
assert(not pres_3sg and not pres_ptc and not past)
if par1 then
pres_3sg = canonicalize_s_form(par1)
else
pres_3sg = new_default_s
end
if par2 then
pres_ptc = canonicalize_ing_form(par2)
else
pres_ptc = new_default_ing
end
if par3 then
past = canonicalize_ed_form(par3)
else
past = new_default_ed
end
end
if par4 then
past_ptcs_given = true
past_ptc = canonicalize_ed_form(par4)
else
past_ptc = past
end
pres_3sgs = {{form = pres_3sg}}
pres_ptcs = {{form = pres_ptc}}
pasts = {{form = past}}
past_ptcs = {{form = past_ptc}}
end
------------------------------------------- HANDLE OVERRIDES ------------------------------------------
local pres_3sg_infls, pres_ptc_infls, past_infls, past_ptc_infls
local function strip_brackets(qualifiers)
if not qualifiers then
return nil
end
local stripped_qualifiers = {}
for _, qualifier in ipairs(qualifiers) do
local stripped_qualifier = qualifier:match("^%[(.*)%]$")
if not stripped_qualifier then
error("Internal error: Qualifier should be surrounded by brackets at this stage: " .. qualifier)
end
table.insert(stripped_qualifiers, stripped_qualifier)
end
return stripped_qualifiers
end
local function collect_forms(label, accel_form, defaults, overrides, override_qualifiers, canonicalize)
if defaults[1].form == "-" then
return {label = "no " .. label}
else
local into_table = {label = label, accel = {form = accel_form}}
local maxindex = math.max(#defaults, overrides.maxindex)
local qualifiers = override_qualifiers[1] and {override_qualifiers[1]} or strip_brackets(defaults[1].footnotes)
table.insert(into_table, {term = defaults[1].form, q = qualifiers})
-- Present 3rd singular
for i = 2, maxindex do
local override_form = canonicalize(overrides[i])
if override_form then
-- If there is an override such as past_ptc2=..., only use the qualifier specified
-- using an override (past_ptc2_qual=...), if any; it doesn't make sense to combine
-- an override form with a qualifier specified inside of angle brackets.
table.insert(into_table, {term = override_form, q = {override_qualifiers[i]}})
elseif defaults[i] then
-- If the form comes from inside angle brackets, allow any override qualifier
-- (past_ptc2_qual=...) to override any qualifier specified inside of angle brackets.
-- FIXME: Maybe we should throw an error here if both exist.
local qualifiers = override_qualifiers[i] and {override_qualifiers[i]} or strip_brackets(defaults[i].footnotes)
table.insert(into_table, {term = defaults[i].form, q = qualifiers})
end
end
return into_table
end
end
local pres_3sg_infls = collect_forms("third-person singular simple present", "s-verb-form",
pres_3sgs, args[1], args.pres_3sg_qual, canonicalize_s_form)
local pres_ptc_infls = collect_forms("present participle", "ing-form",
pres_ptcs, args[2], args.pres_ptc_qual, canonicalize_ing_form)
local past_infls = collect_forms("simple past", "spast",
pasts, args[3], args.past_qual, canonicalize_ed_form)
local past_ptc_infls = collect_forms("past participle", "past|part",
past_ptcs, args[4], args.past_ptc_qual, canonicalize_ed_form)
-- Are the past forms identical to the past participle forms? If so, we use a single
-- combined "simple past and past participle" label on the past tense forms.
-- We check for two conditions: Either no past participle forms were given at all, or
-- they were given but are identical in every way (all forms and qualifiers) to the past
-- tense forms. The former "no explicit past participle forms" check is important in the
-- "separate-parameter" format; if past tense overrides are given and no past participle
-- forms given, the past tense overrides should apply to the past participle as well.
-- In the angle-bracket format, it's expected that all forms and qualifiers are specified
-- using that format, and we explicitly copy past tense forms and qualifiers to past
-- participle ones if the latter are omitted, so we disable to "no explicit past participle
-- forms" check.
if args[4].maxindex > 0 or args.past_ptc_qual.maxindex > 0 then
past_ptcs_given = true
end
local identical = true
-- For the past and past participle to be identical, there must be
-- the same number of inflections, and each inflection must match
-- in term and qualifiers.
if #past_infls ~= #past_ptc_infls then
identical = false
else
for key, val in ipairs(past_infls) do
if past_ptc_infls[key].term ~= val.term then
identical = false
break
else
local quals1 = past_ptc_infls[key].q
local quals2 = val.q
if (not not quals1) ~= (not not quals2) then
-- one is nil, the other is not
identical = false
elseif quals1 and quals2 then
-- qualifiers present in both; each qualifier must match
if #quals1 ~= #quals2 then
identical = false
else
for k, v in ipairs(quals1) do
if v ~= quals2[k] then
identical = false
break
end
end
end
end
if not identical then
break
end
end
end
end
-- Insert the forms
table.insert(data.inflections, pres_3sg_infls)
table.insert(data.inflections, pres_ptc_infls)
if not past_ptcs_given or identical then
if past_ptcs[1].form == "-" then
past_infls.label = "no simple past or past participle"
else
past_infls.label = "simple past and past participle"
past_infls.accel = {form = "ed-form"}
end
table.insert(data.inflections, past_infls)
else
table.insert(data.inflections, past_infls)
table.insert(data.inflections, past_ptc_infls)
end
if pagename:find(" ") then
-- Check for placeholder "it"
local words = rsplit(pagename, " ")
for i, word in ipairs(words) do
if word == "it" or word == "its" or word == "it's" then
table.insert(data.categories, langname .. ' terms with placeholder "it"')
break
end
end
-- Check for phrasal verbs
local phrasal_particles = require("Module:table/listToSet") {
-- NOTE: This should only contain common adverbial particles, 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_particle_words = require("Module:table/listToSet") {
"it",
"one",
"oneself",
"someone",
}
local base = pagename
local seen_particles = {}
-- Only consider a verb to be phrasal if it consists of a single base verb followed exclusively by either
-- particles from `phrasal_particles` or placeholder words from `allowed_non_particle_words`, where at
-- least one following word is from `phrasal_particles` (hence [[can it]] is not a phrasal verb).
while true do
local prev, particle = base:match("^(.+) (.-)$")
if not prev then
break
end
if phrasal_particles[particle] then
table.insert(seen_particles, particle)
elseif allowed_non_particle_words[particle] then
-- do nothing
else
break
end
base = prev
end
if not base:find(" ") and #seen_particles > 0 then
table.insert(data.categories, langname .. " phrasal verbs")
for i = #seen_particles, 1, -1 do
table.insert(data.categories, langname .. " phrasal verbs with particle (" .. seen_particles[i] ..
")")
end
end
end
end
}
return export