View source for Module:etymology

local export = {}

local m_inline = require("Module:inline")
local m_languages = require("Module:languages")
local m_links = require("Module:links")
local m_parameters = require("Module:parameters")
local plaenk_alt = mw.loadData("Module:lnk-pro-morph/data").alt

local new_pos_data = {
	["adjective"] = {
		categories = {"adjectivisations"},
	},
	["noun"] = {
		categories = {"nominalisations"},
	},
	["verb"] = {
		categories = {"verbalisations"},
	},
}
local new_pos_aliases = {
	["adj"] = "adjective",
	["n"] = "noun",
	["v"] = "verb",
}

local no_term_params = {
    [1] = {required = true},
    ["nocap"] = {type = "boolean"},
    ["nc"] = {alias_of = "nocap"},
    ["notext"] = {type = "boolean"},
    ["nocat"] = {type = "boolean"},
}
local no_term_data = {
    ["onomatopoeic"] = {
        categories = {"onomatopoeias"},
        text = true,
    },
    ["unknown"] = {
        categories = {"terms with unknown etymologies"},
        text = function(nocap) return (nocap and "u" or "U") .. "nknown" end,
    },
    ["uncertain"] = {
        categories = {"terms with uncertain etymologies"},
        text = function(nocap) return (nocap and "u" or "U") .. "ncertain" end,
    },
}

local single_term_params = {
    [1] = {required = true},
    [2] = {required = true},
    [3] = {alias_of = "alt"},
    [4] = {alias_of = "t"},
    ["alt"] = {},
    ["t"] = {},
    ["anchor"] = {},
    ["a"] = {alias_of = "anchor"},
    ["pos"] = {},
    ["newpos"] = {},
    ["nocap"] = {type = "boolean"},
    ["nc"] = {alias_of = "nocap"},
    ["notext"] = {type = "boolean"},
    ["nocat"] = {type = "boolean"},
}
local single_term_data = {
    ["abbreviation"] = {
        text = true,
        categories = {"abbreviations"}
    },
    ["back-formation"] = {
        text = true,
        preposition = "from",
        categories = {"back-formations"}
    },
    ["clipping"] = {
        text = true,
        categories = {"clippings"},
    },
    ["contraction"] = {
    	text = true,
    	categories = {"contractions"},
    },
    ["deadjectival"] = {
    	text = true,
    	preposition = "from",
    	categories = {"deadjectivals"},
    },
    ["denominal"] = {
		text = true,
		preposition = "from",
		categories = {"denominals"},
    },
    ["deverbal"] = {
    	text = true,
    	preposition = "from",
    	categories = {"deverbals"},
    },
    ["doublet"] = {
        text = true,
        categories = {"doublets"},
    },
    ["ellipsis"] = {
        text = true,
        categories = {"ellipses"},
    },
    ["initialism"] = {
        text = true,
        categories = {"initialisms"},
    },
    ["rebracketing"] = {
        text = true,
        categories = {"rebracketings"},
    },
    ["reduplication"] = {
        text = true,
        categories = {"reduplications"},
    },
}

local multi_term_params = {
	[1] = {required = true},
	[2] = {list = true},
	["alt"] = {list = true, allow_holes = true},
	["anchor"] = {list = true, allow_holes = true},
    ["a"] = {alias_of = "anchor"},
	["t"] = {list = true, allow_holes = true},
	["pos"] = {list = true, allow_holes = true},
	["nocap"] = {type = "boolean"},
	["nc"] = {alias_of = "nocap"},
	["notext"] = {type = "boolean"},
	["nocat"] = {type = "boolean"},
}
local multi_term_data = {
	["blend"] = {
		text = true,
		categories = {"blends"},
	},
}

local derived_term_params = {
    [1] = {required = true},
    [2] = {required = true},
    [3] = {required = true},
    [4] = {alias_of = "alt"},
    [5] = {alias_of = "t"},
    ["alt"] = {},
    ["anchor"] = {},
    ["a"] = {alias_of = "anchor"},
    ["t"] = {},
    ["pos"] = {},
    ["nocap"] = {type = "boolean"},
    ["nc"] = {alias_of = "nocap"},
    ["notext"] = {type = "boolean"},
    ["nocat"] = {type = "boolean"},
}
local derived_term_data = {
    ["borrowed"] = {
        categories = {"terms borrowed from"},
    },
    ["calque"] = {
        text = true,
        categories = {"terms calqued from"},
    },
    ["derived"] = {
        categories = {"terms derived from"},
    },
    ["inherited"] = {
        categories = {"terms inherited from"},
    },
    ["learned borrowing"] = {
        text = true,
        categories = {"terms borrowed from", "learned borrowings from"},
    },
    ["orthographic borrowing"] = {
        text = true,
        categories = {"terms borrowed from", "orthographic borrowings from"},
    },
    ["semantic loan"] = {
        text = true,
        categories = {"terms derived from", "semantic loans from"},
    },
    ["phono-semantic matching"] = {
        text = true,
        categories = {"terms derived from", "phono-semantic matchings from"},
    },
    ["transliteration"] = {
        text = true,
        categories = {"terms derived from", "transliterations of"},
    },
}

local affix_params = {
    [1] = {required = true},
    [2] = {list = true},
    ["type"] = {},
    ["t"] = {list = true, allow_holes = true},
    ["l"] = {list = true, allow_holes = true},
    ["alt"] = {list = true, allow_holes = true},
    ["anchor"] = {list = true, allow_holes = true},
    ["a"] = {alias_of = "anchor"},
    ["pos"] = {list = true, allow_holes = true},
    ["noaff"] = {list = true, allow_holes = true, type = "boolean"},
    ["root"] = {alias_of = "noaff"},
    ["nolink"] = {list = true, allow_holes = true, type = "boolean"},
    ["hypo"] = {list = true, allow_holes = true, type = "boolean"},
    ["hypothetical"] = {alias_of = "hypo"},
    ["notext"] = {type = "boolean"},
    ["nocap"] = {type = "boolean"},
    ["nc"] = {alias_of = "nocap"},
    ["nocat"] = {type = "boolean"},
}
local affix_data = {
    ["affix"] = {},
    ["surface analysis"] = {
        text = function(nocap) return (nocap and "b" or "B") .. "y [[Appendix:Glossary#surface analysis|surface analysis]], " end
    }
}
local affix_delimiter = {
    ["-"] = true,
    ["·"] = true,
}
local affix_delimiter_pattern = "[%-·]"
local affix_compound_data = {
	["alliterative"] = {
		categories = {"alliterative compounds"},
	},
	["antonymous"] = {
		categories = {"antonymous compounds"},
	},
	["bahuvrihi"] = {
		categories = {"bahuvrihi compounds"},
	},
	["coordinative"] = {
		categories = {"coordinative compounds"},
	},
	["descriptive"] = {
		categories = {"descriptive compounds"},
	},
	["determinative"] = {
		categories = {"determinative compounds"},
	},
	["dvandva"] = {
		categories = {"dvandva compounds"},
	},
	["endocentric"] = {
		categories = {"endocentric compounds"},
	},
	["exocentric"] = {
		categories = {"exocentric compounds"},
	},
	["karmadharaya"] = {
		categories = {"karmadharaya compounds"},
	},
	["rhyming"] = {
		categories = {"rhyming compounds"},
	},
	["synonymous"] = {
		categories = {"synonymous compounds"},
	},
	["tatpurusa"] = {
		categories = {"tatpurusa compounds"},
	},
}
local affix_compound_aliases = {
	allit = "alliterative",
	ant = "antonymous",
	bahu = "bahuvrihi",
	bv = "bahuvrihi",
	coord = "coordinative",
	desc = "descriptive",
	det = "determinative",
	dva = "dvandva",
	endo = "endocentric",
	exo = "exocentric",
	karma = "karmadharaya",
	kd = "karmadharaya",
	rhy = "rhyming",
	syn = "synonymous",
	tat = "tatpurusa",
	tp = "tatpurusa",
}

local function is_circumfix(word)
	return mw.ustring.find(word, affix_delimiter_pattern .. " " .. affix_delimiter_pattern) ~= nil
end

local function is_infix(word)
    return affix_delimiter[mw.ustring.sub(word, 1, 1)] and affix_delimiter[mw.ustring.sub(word, -1)]
end

local function is_prefix(word)
    return affix_delimiter[mw.ustring.sub(word, -1)]
end

local function is_suffix(word)
    return affix_delimiter[mw.ustring.sub(word, 1, 1)]
end

local function format_etymology(out, categories)
    for _, category in ipairs(categories) do
        out = out .. "[[Category:" .. category .. "]]"
    end
    return out
end

local function format_solo_text(label, text_data, nocap_arg)
    if not text_data then return "" end
    if type(text_data) == "function" then return text_data(nocap_arg) end
    return "[[Appendix:Glossary#" .. label .. "|" .. (nocap_arg and label or mw.ustring.gsub(label, "^%l", string.upper)) .. "]]"
end

local function format_prefixed_text(label, text_data, nocap_arg, preposition)
    if not text_data then return "" end
    if type(text_data) == "function" then return text_data(nocap_arg) end
    return "[[Appendix:Glossary#" .. label .. "|" .. (nocap_arg and label or mw.ustring.gsub(label, "^%l", string.upper)) .. "]] " .. (preposition or "of") .. " "
end

local function hydrate_category(category, language_to, language_from, newpos)
    local new_category = language_to.name .. " " .. category
    if language_from then new_category = new_category .. " " .. language_from.name end
    return new_category
end

local function hydrate_categories(categories, language_to, language_from, newpos)
    local new_categories = {}
    for i, category in ipairs(categories) do
        new_categories[i] = language_to.name .. " " .. category .. (language_from and (" " .. language_from.name) or "")
    end
    if newpos then
    	newpos = new_pos_aliases[newpos] or newpos
    	local newpos_data = new_pos_data[newpos]
    	for _, category in ipairs(newpos_data["categories"]) do table.insert(new_categories, language_to.name .. " " .. category) end
    end
    return new_categories
end

local function no_term_etymology(template, frame)
    local data, args = no_term_data[template], m_parameters.process(frame:getParent().args, no_term_params)
    local out, categories = "", {}
    local language = m_languages.get_by_code(args[1])
    if not args["nocat"] then categories = hydrate_categories(data["categories"], language) end
    if not args["notext"] then out = format_solo_text(template, data["text"], args["nocap"]) end
    return format_etymology(out, categories)
end

local function single_term_etymology(template, frame)
    local data, args = single_term_data[template], m_parameters.process(frame:getParent().args, single_term_params)
    local out, categories = "", {}
    local language = m_languages.get_by_code(args[1])
    if not args["notext"] then out = out .. format_prefixed_text(template, data["text"], args["nocap"], data["preposition"]) end
    out = out .. m_links.full_link({
        term = args[2],
        language = language,
        alt = args["alt"],
        gloss = args["t"],
        pos = args["pos"],
        anchor = args["anchor"],
        nobold = true,
    }, "term")
    if not args["nocat"] then categories = hydrate_categories(data["categories"], language, nil, args["newpos"]) end
    return format_etymology(out, categories)
end

local function multi_term_etymology(template, frame)
	local data, args = multi_term_data[template], m_parameters.process(frame:getParent().args, multi_term_params)
	local out, categories = "", {}
	local language = m_languages.get_by_code(args[1])
	if not args["notext"] then out = out .. format_prefixed_text(template, data["text"], args["nocap"]) end
	for i, word in ipairs(args[2]) do
		local i_term, i_args = m_inline.parse(word)
		if i > 1 then out = out .. " + " end
		out = out .. m_links.full_link({
			term = i_term,
			language = language,
			alt = args["alt"][i] or i_args["alt"],
			gloss = args["t"][i] or i_args["t"],
			pos = args["pos"][i] or i_args["pos"],
			anchor = args["anchor"][i] or i_args["anchor"] or i_args["a"],
		}, "term")
	end
	if not args["nocat"] then categories = hydrate_categories(data["categories"], language, nil, args["preposition"]) end
	return format_etymology(out, categories)
end

local function derived_term_etymology(template, frame)
    local data, args = derived_term_data[template], m_parameters.process(frame:getParent().args, derived_term_params)
    local out, categories = "", {}
    local language_to = m_languages.get_by_code(args[1])
    local language_from = m_languages.get_by_code(args[2])
    if (not args["notext"]) and (not data["silent"]) then out = out .. format_prefixed_text(template, data["text"], args["nocap"], data["preposition"]) end
    out = out .. m_links.full_link({
        term = args[3],
        language = language_from,
        alt = args["alt"],
        gloss = args["t"],
        pos = args["pos"],
        anchor = args["anchor"],
        showlanguage = true,
        nobold = true,
    }, "term")
    if not args["nocat"] then categories = hydrate_categories(data["categories"], language_to, language_from) end
    return format_etymology(out, categories)
end

local function affix_etymology(template, frame)
    local data, args = affix_data[template], m_parameters.process(frame:getParent().args, affix_params)
    local pre_out, categories = {}, {}
    local language_to = m_languages.get_by_code(args[1])
    local n_parts, n_affixes = 0, 0
    for i, term in ipairs(args[2]) do
        local i_term, i_args = m_inline.parse(term)
        if args[1] == "lnk-pro" and plaenk_alt[i_term] then i_term = plaenk_alt[i_term] end
        n_parts = n_parts + 1
        local language_from = nil
        if args["l"][i] or i_args["l"] then language_from = m_languages.get_by_code(args["l"][i] or i_args["l"]) end
        local cite_term = i_term
        if (language_from and language_from.proto) or ((not language_from) and language_to.proto) then cite_term = "&#42;" .. cite_term end
        if not args["nocat"] then
            if args["noaff"][i] or i_args["noaff"] then
                -- this is a marked non-affix, don't let it be classified as one!
            elseif language_from then
                table.insert(categories, language_to.name .. " terms derived from " .. language_from.name)
                if is_infix(i_term) or is_prefix(i_term) or is_suffix(i_term) or is_circumfix(i_term) then n_affixes = n_affixes + 1 end
            elseif is_infix(i_term) then
                table.insert(categories, language_to.name .. " terms infixed with " .. cite_term)
                n_affixes = n_affixes + 1
            elseif is_prefix(i_term) then
                table.insert(categories, language_to.name .. " terms prefixed with " .. cite_term)
                n_affixes = n_affixes + 1
            elseif is_suffix(i_term) then
                table.insert(categories, language_to.name .. " terms suffixed with " .. cite_term)
                n_affixes = n_affixes + 1
            elseif is_circumfix(i_term) then
            	table.insert(categories, language_to.name .. " terms circumfixed with " .. cite_term)
            	n_affixes = n_affixes + 1
            end
        end
        table.insert(pre_out,  m_links.full_link({
	            term = i_term,
	            language = language_from or language_to,
	            alt = args["alt"][i] or i_args["alt"],
	            gloss = args["t"][i] or i_args["t"],
	            pos = args["pos"][i] or i_args["pos"],
	            anchor = args["anchor"][i] or i_args["anchor"] or i_args["a"],
	            --nolink = args["nolink"][i] or i_args["nolink"],
	            --hypo = args["hypo"][i] or i_args["hypo"],
	            showlanguage = (language_from and true),
	            nobold = true,
	        }, "term"))
    end
    if (not args["nocat"]) and (n_parts > 1 and n_affixes == 0) then
        table.insert(categories, language_to.name .. " compound terms")
        if args["type"] then
        	local affix_type = affix_compound_data[affix_compound_aliases[args["type"]] or args["type"]]
        	if affix_type then
        		for _, category in ipairs(affix_type["categories"]) do
        			table.insert(categories, language_to.name .. " " .. category)
        		end
        	end
        end
    end
    local out = table.concat(pre_out, " + ")
    if not args["notext"] then
        out = format_prefixed_text(template, data["text"], args["nocap"], data["preposition"]) .. out
    end
    return format_etymology(out, categories)
end

function export.show(frame)
    local template = frame.args[1]
    if no_term_data[template] then return no_term_etymology(template, frame) end
    if single_term_data[template] then return single_term_etymology(template, frame) end
    if multi_term_data[template] then return multi_term_etymology(template, frame) end
    if derived_term_data[template] then return derived_term_etymology(template, frame) end
    if affix_data[template] then return affix_etymology(template, frame) end
    error("No such sub-template type is defined!")
end

return export