Module:User:Erutuon/grc-decl/sandbox/decl/staticdata

--	Abbreviations and codes used in declension names:	- accent		- nothing: oxytone		- pax: paroxytone		- prx: proparoxytone		- con: perispomenon	- contraction		- con: contracted (perispomenon)		- open: uncontracted	- the characteristics of the ending (historically the stem)		- 1st declension			- alp: long alpha			- eta: eta			- als: long alpha, but short alpha in nominative and accusative singular			- als: eta, but short alpha in nominative and accusative singular		- 3rd declension			- pure: begins with ι or υ throughout			- weak: begins with ι or υ alternating with ε or η

local export = { adjinflections = {}, adjinflections_con = {}, }

local module_path = 'Module:grc-decl/sandbox/decl/staticdata' local m_paradigms = mw.loadData(module_path .. "/paradigms")

local ustring = mw.ustring local U = ustring.char local sub = ustring.sub local gsub = ustring.gsub local toNFC = ustring.toNFC local toNFD = ustring.toNFD

--	Creates a version of an ending with macrons removed, and with macrons or breves switched.	The switching is pointless for endings containing more than one macron or breve:	for instance, ᾱσᾰ becomes ᾰσᾱ. local function mess_with_length(text) -- local text = toNFD(text) local macron = U(0x304) local breve = U(0x306) local unmarked_length = gsub(text, "[" .. macron .. breve .. "]", "") local opposite_length = gsub(		text,		"[" .. macron .. breve .. "]",		{ [macron] = breve, [breve] = macron }	) return unmarked_length, opposite_length end

-- Adds a key with macrons or breves removed to the table. local function add_unmarked_form(list, key, value) if type(list) ~= "table" then return list end -- key = toNFD(key) local unmarked_length, opposite_length = mess_with_length(key) if unmarked_length ~= key and not list[opposite_length] then list[unmarked_length] = list[unmarked_length] or value end list[key] = list[key] or value return list end

--[=[	Noun declension categories that Module:grc-decl/sandbox/decl uses to determine the declension category of the nominative and genitive forms that are supplied to the template. ]=] local infl_categories = { ['1st-alp'] = m_paradigms.alp_pax, ['1st-eta'] = m_paradigms.eta_pax, ['1st-als'] = m_paradigms.als_prx, ['1st-ets'] = m_paradigms.ets_prx, ['1st-M-alp'] = m_paradigms.M_alp_pax, ['1st-M-alp2'] = m_paradigms.M_alp_con, -- add 2 for unique key ['1st-M-eta'] = m_paradigms.M_eta_pax, ['2nd'] = m_paradigms.second, ['2nd-con'] = m_paradigms.second_con, ['2nd-N'] = m_paradigms.second_N, ['2nd-N-con'] = m_paradigms.second_N_con, ['2nd-att'] = m_paradigms.second_att, ['2nd-N-att'] = m_paradigms.second_N_att_prx, ['3rd-εσ'] = m_paradigms.es_adj, ['3rd-εσ-open'] = m_paradigms.es_adj_prx_open, ['3rd-N-εσ'] = m_paradigms.N_es_adj, ['3rd-N-ος'] = m_paradigms.N_es_prx, ['3rd-N-ος-open'] = m_paradigms.N_es_prx_open, ['3rd-N-ᾰσ'] = m_paradigms.N_as_prx, ['3rd-N-ᾰσ-open'] = m_paradigms.N_as_prx_open, ['3rd-κλῆς'] = m_paradigms.kles, ['3rd-κλῆς-open'] = m_paradigms.kles_open, -- add 2 for unique key ['3rd-weak-ι'] = m_paradigms.weak_i_prx, ['3rd-weak-υ'] = m_paradigms.weak_u, ['3rd-N-weak-ι'] = m_paradigms.N_weak_i_prx, ['3rd-N-weak-υ'] = m_paradigms.N_weak_u, ['3rd-pure-ι'] = m_paradigms.pure_i_prx, ['3rd-N-pure-ι'] = m_paradigms.N_pure_i_prx, ['3rd-pure-υ-long'] = m_paradigms.pure_u_long_prx, ['3rd-pure-υ'] = m_paradigms.pure_u, ['3rd-N-pure-υ'] = m_paradigms.N_pure_u_prx, ['3rd-ευς'] = m_paradigms.eus, ['3rd-ευς-con'] = m_paradigms.eus_con, ['3rd-οι'] = m_paradigms.oi, }

-- Used by the function that generates the list of declension categories on the documentation page. export.conversion = { ['1st-alp'] = 'alp_pax', ['1st-eta'] = 'eta_pax', ['1st-als'] = 'als_prx', ['1st-ets'] = 'ets_prx', ['1st-M-alp'] = 'M_alp_pax', ['1st-M-alp2'] = 'M_alp_con', ['1st-M-eta'] = 'M_eta_pax', ['2nd'] = 'second', ['2nd-con'] = 'second_con', ['2nd-N'] = 'second_N', ['2nd-N-con'] = 'second_N_con', ['2nd-att'] = 'second_att', ['2nd-N-att'] = 'second_N_att_prx', ['3rd-εσ'] = 'es_adj', ['3rd-εσ-open'] = 'es_adj_prx_open', ['3rd-N-εσ'] = 'N_es_adj', ['3rd-N-ος'] = 'N_es_prx', ['3rd-N-ος-open'] = 'N_es_prx_open', ['3rd-N-ᾰσ'] = 'N_as_prx', ['3rd-κλῆς'] = 'kles', ['3rd-κλῆς-open'] = 'kles_open', ['3rd-weak-ι'] = 'weak_i_prx', ['3rd-weak-υ'] = 'weak_u', ['3rd-N-weak-ι'] = 'N_weak_i_prx', ['3rd-N-weak-υ'] = 'N_weak_u', ['3rd-pure-ι'] = 'pure_i_prx', ['3rd-N-pure-ι'] = 'N_pure_i_prx', ['3rd-pure-υ-long'] = 'pure_u_long_prx', ['3rd-pure-υ'] = 'pure_u', ['3rd-N-pure-υ'] = 'N_pure_u_prx', ['3rd-ευς'] = 'eus', ['3rd-ευς-con'] = 'eus_con', ['3rd-οι'] = 'oi', }

infl_info = {} export.ambig_forms = {}

-- Constructs a table for nouns with the same structure as the one for adjectives. local strip_tone = require("Module:grc-accent").strip_tone for name, decl in pairs(infl_categories) do	local nom = strip_tone(sub(decl.NS, 2)) local gen = strip_tone(sub(strip_tone(decl.GS), 2)) if not infl_info[nom] then infl_info[nom] = {} end if type(infl_info[nom]) == "table" then if infl_info[nom][gen] then error('Conflict in noun declensions; two declensions with nominative ' .. nom ..					' and genitive ' .. gen .. '.') end name = gsub(name, "%d$", "") infl_info[nom][gen] = name end end

for nominative, list in pairs(infl_info) do	local unmarked_length, opposite_length = mess_with_length(nominative) local data_for_opposite if unmarked_length ~= opposite_length then data_for_opposite = infl_info[opposite_length] or infl_info[toNFC(opposite_length)] if not data_for_opposite then infl_info[unmarked_length] = nominative end end if type(list) == "table" then local new_list = {} for gen, name in pairs(list) do			local unmarked_length, opposite_length = mess_with_length(gen) if unmarked_length ~= gen and not list[opposite_length] then new_list[unmarked_length] = name end new_list[gen] = name end list = new_list local combined_gens = {} if data_for_opposite and not infl_info[unmarked_length] then local is_ambig = false for gen, name in pairs(list) do				combined_gens[gen] = name end for gen, name in pairs(data_for_opposite or {}) do				local gen_unmarked_length, gen_opposite_length = mess_with_length(gen) if list[gen] or list[gen_opposite_length] then --						If there are two declension types with the same nominative						and genitive endings aside from the length of the nominative,						then strip the macron or breve from both forms and record them. is_ambig = true end combined_gens = add_unmarked_form(combined_gens, gen, name) end if is_ambig then --					Remove length marks and record all the nominative–genitive pairs					that were found. local nom = mess_with_length(nominative) for gen, name in pairs(combined_gens) do					gen = mess_with_length(gen) export.ambig_forms[nom] = export.ambig_forms[nom] or {} export.ambig_forms[nom][gen] = true end else infl_info[unmarked_length] = combined_gens end end infl_info[nominative] = list end end

local infl_info_adj = { ['ος'] = { ['ᾱ'] = '1&2-alp', ['η'] = '1&2-eta', ['ον'] = '2nd', },	['ους'] = { ['ᾱ'] = '1&2-alp-con', ['η'] = '1&2-eta-con', ['ουσᾰ'] = '1&3-ουντ', },	['ῠς'] = { ['ειᾰ'] = '1&3-ups' }, --	['υς'] = 'ῠς', ['ως'] = { ['υιᾰ'] = '1&3-οτ', ['ων'] = '2nd-att', },	['ᾱς'] = { ['αινᾰ'] = '1&3-ᾰν', ['ᾱσᾰ'] = '1&3-ᾰντ', },	['ην'] = { ['εινᾰ'] = '1&3-εν' }, ['εις'] = { ['εισᾰ'] = '1&3-εντ', ['εσσᾰ'] = '1&3-εσσ', },	['ων'] = { ['ουσᾰ'] = '1&3-οντ', ['ωσᾰ'] = '1&3-ωντ', },	['ῡς'] = { ['ῡσᾰ'] = '1&3-ῠντ' }, ['ης'] = { ['ες'] = '3rd-εσ' }, }

export.ambig_forms_adj = {} --	[] = ,

for masculine, list in pairs(infl_info_adj) do	if type(list) == "table" then -- Decompose masculine and feminine (or neuter) endings. local new_fems = {} local fems_changed = false for feminine, name in pairs(list) do			new_feminine = toNFD(feminine) if new_feminine ~= feminine then fems_changed = true end new_fems[new_feminine] = name end list = new_fems new_masculine = toNFD(masculine) if fems_changed or new_masculine ~= masculine then -- Delete existing entry infl_info_adj[masculine] = nil -- Create decomposed one. infl_info_adj[new_masculine] = list end elseif type(list) == "string" then if not infl_info_adj[list] then error(masculine .. ' has been redirected to the form ' .. list .. ', which does not exist.') end end end

for masculine, list in pairs(infl_info_adj) do	local unmarked_length, opposite_length = mess_with_length(masculine) local data_for_opposite if unmarked_length ~= opposite_length then data_for_opposite = infl_info_adj[opposite_length] or infl_info_adj[toNFC(opposite_length)] if not data_for_opposite then infl_info_adj[unmarked_length] = masculine end end local new_list = {} for fem, name in pairs(list) do		local unmarked_length, opposite_length = mess_with_length(fem) if unmarked_length ~= fem and not list[opposite_length] then new_list[unmarked_length] = name end new_list[fem] = name end list = new_list local combined_fems = {} if data_for_opposite and not infl_info_adj[unmarked_length] then for fem, name in pairs(list) do			combined_fems[fem] = name end for fem, name in pairs(data_for_opposite or {}) do			if list[fem] then --					If there are declension types with the same masculine and					feminine endings, aside from the length (not currently true),					then strip the macron or breve and log them. local key = mess_with_length(masculine) local value = mess_with_length(fem) export.ambig_forms_adj[key] = value combined_fems = nil break end combined_fems = add_unmarked_form(combined_fems, fem, name) end if combined_fems then infl_info_adj[unmarked_length] = combined_fems end end infl_info_adj[masculine] = list end

--	Mapping from adjectival declension codes to the declension type used for the masculine (1),	feminine (2), and neuter (3), along with any suffixes added to the stem in the masculine	and neuter (a1) or feminine (a2).

export.headers = { ['1&2'] = 'First and second declension', ['1&3'] = 'First and third declension', ['2nd'] = 'Second declension', ['3rd'] = 'Third declension', ['Attic'] = 'Attic second declension', ['irreg'] = 'Irregular declension', }

for k, header in pairs(m_paradigms.headers) do	if export.headers[k] then if export.headers[k] ~= header then error('Two headers with same name: ' .. export.header[k] .. ', ' .. header .. '.') end else export.headers[k] = header end end

--First-and-second-declension adjectives export.adjinflections['1&2-alp'] = { '2nd', '1st-alp', '2nd-N', adeclheader = '1&2' } export.adjinflections['1&2-alp-con'] = { '2nd', '1st-alp', '2nd-N', a1 = 'ε', a2 = 'ε', adeclheader = '1&2' } -- avoid collisions export.adjinflections['1&2-eta'] = { '2nd', '1st-eta', '2nd-N', adeclheader = '1&2' } export.adjinflections['1&2-eta-con'] = { '2nd', '1st-alp', '2nd-N', a1 = 'ε', a2 = 'ε', adeclheader = '1&2' }

-- First-and-third-declension adjectives export.adjinflections['1&3-ups'] = { '3rd-weak-υ', '1st-als', '3rd-N-weak-υ', a2 = 'ει', adeclheader = '1&3' } export.adjinflections['1&3-ᾰν'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'ᾰν', a2 = 'αιν', adeclheader = '1&3' } export.adjinflections['1&3-εν'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'εν', a2 = 'ειν', adeclheader = '1&3' } export.adjinflections['1&3-εσσ'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'εντ', a2 = 'εσσ', adeclheader = '1&3' }

--	Participles (subtype of first-and-third) export.adjinflections['1&3-ουντ'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'οντ', a2 = 'ουσ', adeclheader = '1&3' } -- perfect active: export.adjinflections['1&3-οτ'] = { '3rd-cons', '1st-als', '3rd-N-cons', a1 = 'οτ', a2 = 'υι', adeclheader = '1&3' } -- From roots with reflex of *h₁, and in the aorist passive: export.adjinflections['1&3-εντ'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'εντ', a2 = 'εισ', adeclheader = '1&3' } -- From roots with reflex of *h₂, and 1st aorist active: export.adjinflections['1&3-ᾰντ'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'ᾰντ', a2 = 'ᾱσ', adeclheader = '1&3' } -- From roots with reflex of *h₃, and thematic active: export.adjinflections['1&3-οντ'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'οντ', a2 = 'ουσ', adeclheader = '1&3' } -- έω/όω contract; this category does not appear in infl_info_adj, because it -- has to be assigned specifically in Module:grc-decl/sandbox/decl, if -- masculine singular has acute on ultima (is perispomenon): export.adjinflections['1&3-οντ-con'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'ουντ', a2 = 'ουσ', adeclheader = '1&3' } -- νῡμι active: export.adjinflections['1&3-ῠντ'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'ῠντ', a2 = 'ῡσ', adeclheader = '1&3' } -- άω contract: export.adjinflections['1&3-ωντ'] = { '3rd-cons', '1st-ets', '3rd-N-cons', a1 = 'ωντ', a2 = 'ωσ', adeclheader = '1&3' }

-- Masculine and feminine identical export.adjinflections['2nd'] = { '2nd', nil, '2nd-N', adeclheader = '2nd' } export.adjinflections['2nd-att'] = { '2nd-att', nil, '2nd-N-att', adeclheader = 'Attic' } export.adjinflections['3rd-cons'] = { '3rd-cons', nil, '3rd-N-cons', adeclheader = '3rd' } export.adjinflections['3rd-εσ-open'] = { '3rd-εσ-open', nil, '3rd-N-εσ-open', adeclheader = '3rd' }

export.adjinflections['1&2-alp-con'] = { '2nd-con', '1st-alp', '2nd-N-con', adeclheader = '1&2' } export.adjinflections['1&2-eta-con'] = { '2nd-con', '1st-eta', '2nd-N-con', adeclheader = '1&2' } export.adjinflections['3rd-εσ'] = { '3rd-εσ', nil, '3rd-N-εσ', adeclheader = '3rd' }

export.adjinflections_con['1&2-alp-con'] = export.adjinflections['1&2-alp-con'] export.adjinflections_con['1&2-eta-con'] = export.adjinflections['1&2-eta-con'] export.adjinflections_con['3rd-εσ'] = export.adjinflections['3rd-εσ']

export.infl_info = {} export.infl_info.noun = infl_info export.infl_info.adj = infl_info_adj

return export