Module:User:Benwing2/category tree/poscatboiler/data/lang-specific/ja-ryu

local export = {}

local Hira = require("Module:scripts").getByCode("Hira")
local m_numeric = require("Module:ConvertNumeric")
local rmatch = mw.ustring.match


function export.add_labels(labels, lang)
	labels = {
		description = "{{{langname}}} {{m|ja|連体詞||], attributive|tr=れんたいし, rentaishi|sc=Jpan}}. Modifies nouns, doesn’t conjugate, and doesn’t ].",
		parents = {{name = "{{{langcat}}}", raw = true}},
	}

	labels = {
		description = "The ] (], ]) form of " .. (lang:getCode() == "ryu" and "an" or "a") .. " {{{langname}}} word is a ] representation of that word. " ..
		"Wiktionary represents {{{langname}}}-language segments in three ways: in normal form (with ], if appropriate), in ] " ..
		"form (this differs from kanji form only when the segment contains kanji), and in ] form.",
		additional = (lang:getCode() == "ja" and "For more information, see ].\n\n" or "") .. "''See also'' ]",
		toc_template = "categoryTOC-hiragana",
		parents = {
			{name = "{{{langcat}}}", raw = true},
			"Category:Hiragana script characters",
		}
	}

	labels = {
		description = "{{{langname}}} historical ].",
		additional = "''See also'' ]",
		toc_template = "categoryTOC-hiragana",
		parents = {
			"hiragana",
			{name = "{{{langcat}}}", raw = true},
			"Category:Hiragana script characters",
		}
	}

	labels = {
		description = "{{{langname}}} words and terms with katakana forms, sorted by conventional katakana sequence. Katakana is used primarily for transliterations of foreign words, including old Chinese Hanzi not used in ].",
		additional = "Entries in this category are made by {{{langname}}} POS templates, ], etc.",
		toc_template = "categoryTOC-katakana",
		parents = {
			{name = "{{{langcat}}}", raw = true},
			"Category:Katakana script characters",
		}
	}

	labels = {
		description = "{{{langname}}} terms which combine ] and ] characters, potentially with ] too.",
		parents = {
			{name = "{{{langcat}}}", raw = true},
			"hiragana",
			"katakana",
		},
	}

	labels = {
		intro = "{{wikipedia|Honorific speech in Japanese}}",
		description = "{{{langname}}} ]s.",
		parents = {{name = "{{{langcat}}}", raw = true}},
	}

	labels = {
		description = "{{{langname}}} humble terms, or {{ja-r|謙譲語|けんじょうご}}, which is a type of honorific speech in {{{langname}}} that lowers the speaker in relation to the listener.",
		parents = "honorifics",
	}

	labels = {
		description = "{{{langname}}} respectful terms, or {{ja-r|尊敬語|そんけいご}}, which is a type of honorific speech in {{{langname}}} that elevates the listener in relation to the speaker.",
		parents = "honorifics",
	}

	labels = {
		description = "{{{langname}}} kanji categorized by reading.",
		parents = {{name = "Han characters", sort = "reading"}},
	}

	labels = {
		description = "{{{langname}}} categories grouped with regard to the readings of the kanji with which they are spelled.",
		parents = {{name = "{{{langcat}}}", raw = true}},
	}

	labels = {
		description = "{{{langname}}} categories with terms grouped by their reading patterns.",
		parents = {{name = "{{{langcat}}}", raw = true}},
	}

	for reading_pattern, link in pairs {
		 = "{{ja-r|熟字訓|じゅくじくん}}",
		 = "{{ja-r|重%箱%読み|じゅう%ばこ%よみ}}, a reading pattern in {{{langname}}} compounds where kanji read with ''on'yomi'' are followed by kanji read with ''kun'yomi''",
		 = "{{ja-r|訓読み|くんよみ}}",
		 = "{{ja-r|名乗り|なのり}}",
		 = "{{ja-r|音%読み|おん%よみ}}",
		 = "{{ja-r|湯%桶%読み|ゆ%とう%よみ}}, a reading pattern in {{{langname}}} compounds where kanji read with ''kun'yomi'' are followed by kanji read with ''on'yomi''",
	} do
		labels = {
			description = "{{{langname}}} terms exhibiting " .. link .. ".",
			breadcrumb = reading_pattern,
			parents = {{name = "terms by reading pattern", sort = reading_pattern}},
		}
	end

	labels = {
		description = "{{{langname}}} categories with terms grouped with regard to the types of readings of the kanji with which " ..
		"they are spelled, broadly those of Chinese origin (''on'' readings) and those of {{{langname}}} origin (''kun'' readings).",
		parents = {{name = "{{{langcat}}}", raw = true}},
	}

	local on_continuation = [=[


	Categories of terms with more specific types of ''on'' readings can be found in the following categories:
	* ]
	* ]
	* ]
	* ]
	]=]

	local on_desc = ", which is a type of {{ja-r|音%読み|おん%よみ}} or {{ja-r|音|おん}} reading"
	for reading_type, reading_desc in pairs {
		 = "a {{ja-r|呉%音|ご%おん}} reading" .. on_desc,
		 = "a {{ja-r|漢%音|かん%おん}} reading" .. on_desc,
		 = "a {{ja-l|慣用音|かんようおん|kan'yōon}} reading" .. on_desc,
		 = "a {{ja-r|訓%読み|くん%よみ}} or {{ja-r|訓|くん}} reading",
		 = "a {{ja-r|名%乗り|な%のり}} reading, which is a type of reading used for people and places",
		 = "an {{ja-r|音%読み|おん%よみ}} or {{ja-r|音|おん}} reading",
		 = "a {{ja-r|唐%音|とう%おん}} reading" .. on_desc,
		 = "a {{ja-r|宋%音|そう%おん}} reading" .. on_desc,
	} do
		labels = {
			description = "{{{langname}}} categories with terms that are spelled with one or more kanji which exhibit " .. reading_desc .. ".",
			additional = reading_type == "on" and on_continuation,
			breadcrumb = reading_type,
			parents = {{name = "terms by kanji reading type", sort = reading_pattern}},
		}
	end

	labels = {
		intro = "{{wikipedia|Ateji}}",
		description = "{{{langname}}} terms containing one or more ] ({{m|ja|当て字}}), which are ] used to represent sounds rather than meanings (though meaning may have some influence on which kanji are chosen).",
		parents = {{name = "{{{langcat}}}", raw = true}},
	}

	labels = {
		description = "{{{langname}}} terms containing one or more ] ({{m|ja|熟字訓}}), which are ] used to represent meanings rather than sounds.",
		parents = {{name = "{{{langcat}}}", raw = true}},
	}

	labels = {
		description = "{{{langname}}} words with multiple pronunciations (hence multiple ] spellings).",
		parents = {{name = "{{{langcat}}}", raw = true}},
	}

	labels = {
		description = "{{{langname}}} categories grouped with regard to the number of morae in their kanji readings.",
		parents = {{name = "{{{langcat}}}", raw = true}},
	}
end



function export.add_handlers(handlers, lang, m_lang, m_lang_kanji_readings)
	-- FIXME: Only works for 0 through 19.
	local word_to_number = {}
	for k, v in pairs(m_numeric.ones_position) do
		word_to_number = k
	end

	local kana_capture = "(+)"

	local periods = {
		historical = true,
		ancient = true,
	}

	local function get_period_text_and_reading_type_link(period, reading_type)
		period = period ~= "" and period or nil
		if period and not periods then
			return nil
		end
		local period_text = period and period .. " " or nil

		-- Allow periods (historical or ancient) by themselves; they will parse as reading types.
		if not period and periods then
			return nil, reading_type
		end

		local reading_label = m_lang_kanji_readings.accented_labels_to_labels
		if not reading_label then
			return nil
		end
		reading_label = m_lang_kanji_readings.labels
		local reading_type_link = "]"
		return period_text, reading_type_link
	end
			
	local function get_tagged_reading(reading)
		return require("Module:script utilities").tag_text(reading, lang, m_lang_kanji_readings.get_script(reading))
	end

	local function get_reading_link(reading, historical)
		return m_lang_kanji_readings.plain_link {
			term = reading, tr = m_lang.kana_to_romaji(string.gsub(reading, '%-', ''), historical and { hist = true })
		}
	end

	local function is_on_subtype(reading_type)
		return reading_type:find(".on$")
	end


	table.insert(handlers, function(data)
		local count, plural = data.label:match("^terms written with (.+) Han script character(s?)$")
		if count then
			-- Make sure 'one' goes with singular and other numbers with plural.
			if (count == "one") ~= (plural == "") then
				return nil
			end
			local num = word_to_number
			if not num then
				return nil
			end
			return {
				description = "{{{langname}}} terms written with " .. count .. " Han script character" .. plural .. " (also known as ]).",
				breadcrumb = num,
				parents = {{name = "character counts", sort = num}},
			}
		end
	end)


	table.insert(handlers, function(data)
		local count, plural = data.label:match("^kanji readings with (.+) mora(e?)$")
		if count then
			-- Make sure 'one' goes with singular and other numbers with plural.
			if (count == "one") ~= (plural == "") then
				return nil
			end
			local num = word_to_number
			if not num then
				return nil
			end
			return {
				description = "{{{langname}}} kanji readings containing " .. count .. " mora" .. plural .. ".",
				breadcrumb = num,
				parents = {{name = "kanji readings by number of morae", sort = num}},
			}
		end
	end)


	table.insert(handlers, function(data)
		local label_pref, period, reading_type, reading = rmatch(data.label, "^(kanji with (-) ?(+) reading )" .. kana_capture .. "$")
		if period then
			local period_text, reading_type_link = get_period_text_and_reading_type_link(period, reading_type)
			if not reading_type_link then
				return nil
			end

			-- Compute parents.
			local parents = {
				{name = "kanji by " .. (period_text or "") .. reading_type .. " reading", sort = m_lang.jsort(reading)}
			}
			if is_on_subtype(reading_type) then
				table.insert(parents, {name = "kanji with " .. (period_text or "") .. "on reading " .. reading, sort = reading_type})
			elseif period_text then
				table.insert(parents, {name = "kanji with " .. period_text .. "reading " .. reading, sort = reading_type})
			end
			if not period_text then
				table.insert(parents, {name = "kanji read as " .. reading, sort = reading_type})
			end

			local tagged_reading = get_tagged_reading(reading)
			return {
				description = "{{{langname}}} ] with the " .. (period_text or "") .. reading_type_link .. " reading " ..
					get_reading_link(reading, period_text) .. ".",
				displaytitle = "{{{langname}}} " .. label_pref .. tagged_reading,
				breadcrumb = tagged_reading,
				parents = parents,
			}
		end
	end)


	table.insert(handlers, function(data)
		local period, reading_type = rmatch(data.label, "^kanji by (-) ?(+) reading$")
		if period then
			local period_text, reading_type_link = get_period_text_and_reading_type_link(period, reading_type)
			if not reading_type_link then
				return nil
			end

			-- Compute parents.
			local parents = {
				is_on_subtype(reading_type) and {name = "kanji by " .. (period_text or "") .. "on reading", sort = reading_type} or
				period_text and {name = "kanji by " .. reading_type .. " reading", sort = period} or
				{name = "kanji by reading", sort = reading_type}
			}
			if period_text then
				table.insert(parents, {name = "kanji by " .. period_text .. "reading", sort = reading_type})
			end

			-- Compute description.
			local description = "{{{langname}}} ] categorized by " .. (period_text or "") .. reading_type_link .. " reading."
			return {
				description = description,
				breadcrumb = (period_text or "") .. reading_type,
				parents = parents,
			}
		end
	end)


	table.insert(handlers, function(data)
		local label_pref, reading = rmatch(data.label, "^(kanji read as )" .. kana_capture .. "$")
		if reading then
			local parents = {{name = "kanji by reading", sort = m_lang.jsort(reading)}}

			local tagged_reading = get_tagged_reading(reading)
			return {
				description = "{{{langname}}} ] read as " .. get_reading_link(reading, period_text) .. ".",
				displaytitle = "{{{langname}}} " .. label_pref .. tagged_reading,
				breadcrumb = tagged_reading,
				parents = parents,
			}
		end
	end)


	table.insert(handlers, function(data)
		local label_pref, reading = rmatch(data.label, "^(terms spelled with kanji read as )" .. kana_capture .. "$")
		if reading then
			-- Compute parents.
			local sort_key = m_lang.jsort(reading)
			local mora_count = m_lang.count_morae(reading)
			local mora_count_words = m_numeric.spell_number(tostring(mora_count))
			local parents = {
				{name = "terms by kanji readings", sort = sort_key},
				{name = "kanji readings with " .. mora_count_words .. " mora" .. (mora_count > 1 and "e" or ""), sort = sort_key},
				{name = "kanji read as " .. reading, sort = " "},
			}

			local tagged_reading = get_tagged_reading(reading)
			return {
				description = "{{{langname}}} terms that contain kanji that exhibit a reading of " .. get_reading_link(reading) ..
				" in those terms prior to any sound changes.",
				displaytitle = "{{{langname}}} " .. label_pref .. tagged_reading,
				breadcrumb = tagged_reading,
				parents = parents,
			}
		end
	end)


	table.insert(handlers, function(data)
		local kanji, reading = rmatch(data.label, "^terms spelled with (.) read as " .. kana_capture .. "$")
		if not kanji then
			return nil
		end
		local params = {
			 = {list = true},
		}
		local args = require("Module:parameters").process(data.args, params)
		if #args == 0 then
			error("For categories of the form \"" .. data.lang:getCanonicalName() ..
				" terms spelled with KANJI read as READING\", at least one reading type (e.g. \"kun\" or \"on\") must be specified using 1=, 2=, 3=, etc.")
		end
		local reading_types = {}
		for _, reading_type in ipairs(args) do
			table.insert(reading_types, "<i>" .. reading_type .. "</i> reading")
		end
		reading_types = require("Module:table").serialCommaJoin(reading_types, {conj = "or"})

		local parents = {
			{name = "terms spelled with " .. kanji, sort = m_lang.jsort(reading)},
			-- FIXME, using the kanji directly as the sort key is what it did before but maybe we should call ]
			-- to get the radical/stroke sort key
			{name = "terms spelled with kanji read as " .. reading, sort = kanji},
		}
		for _, reading_type in ipairs(args) do
			table.insert(parents, {name = "terms spelled with kanji with " .. reading_type .. " readings", sort = m_lang.jsort(reading)})
		end
		local tagged_kanji = get_tagged_reading(kanji)
		local tagged_reading = get_tagged_reading(reading)
		return {
			description = "{{{langname}}} terms spelled with {{{{{langcode}}}-l|" .. kanji .. "}} with its " ..
				reading_types .. " of " .. get_reading_link(reading) .. ".",
			displaytitle = "{{{langname}}} terms spelled with " .. tagged_kanji .. " read as " .. tagged_reading,
			breadcrumb = "read as " .. tagged_reading,
			parents = parents,
		}, true
	end)
end


return export