From 9e341cc7f9020eda842c8ad0f1f57bf6326eb89d Mon Sep 17 00:00:00 2001 From: aarne Date: Thu, 18 Dec 2008 14:21:28 +0000 Subject: [PATCH] Hindi and Thai compile in 1.5 --- next-lib/src/hindi/AdjectiveHin.gf | 36 ++ next-lib/src/hindi/AdverbHin.gf | 20 + next-lib/src/hindi/AllHin.gf | 3 + next-lib/src/hindi/AllHinAbs.gf | 1 + next-lib/src/hindi/BackwardHin.gf | 76 ++++ next-lib/src/hindi/CatHin.gf | 90 +++++ next-lib/src/hindi/ConjunctionHin.gf | 37 ++ next-lib/src/hindi/GrammarHin.gf | 21 + next-lib/src/hindi/IdiomHin.gf | 30 ++ next-lib/src/hindi/IrregHin.gf | 181 +++++++++ next-lib/src/hindi/LangHin.gf | 10 + next-lib/src/hindi/LexiconHin.gf | 375 ++++++++++++++++++ next-lib/src/hindi/MorphoHin.gf | 196 ++++++++++ next-lib/src/hindi/NounHin.gf | 121 ++++++ next-lib/src/hindi/NumeralHin.gf | 95 +++++ next-lib/src/hindi/ParadigmsHin.gf | 556 +++++++++++++++++++++++++++ next-lib/src/hindi/PhraseHin.gf | 24 ++ next-lib/src/hindi/QuestionHin.gf | 55 +++ next-lib/src/hindi/RelativeHin.gf | 54 +++ next-lib/src/hindi/ResHin.gf | 300 +++++++++++++++ next-lib/src/hindi/SentenceHin.gf | 66 ++++ next-lib/src/hindi/StructuralHin.gf | 132 +++++++ next-lib/src/hindi/VerbHin.gf | 50 +++ next-lib/src/thai/AdjectiveTha.gf | 36 ++ next-lib/src/thai/AdverbTha.gf | 23 ++ next-lib/src/thai/AllTha.gf | 3 + next-lib/src/thai/AllThaAbs.gf | 3 + next-lib/src/thai/CatTha.gf | 77 ++++ next-lib/src/thai/ConjunctionTha.gf | 45 +++ next-lib/src/thai/ExtraTha.gf | 52 +++ next-lib/src/thai/GrammarTha.gf | 21 + next-lib/src/thai/IdiomTha.gf | 30 ++ next-lib/src/thai/IrregTha.gf | 181 +++++++++ next-lib/src/thai/LangTha.gf | 10 + next-lib/src/thai/LexiconTha.gf | 368 ++++++++++++++++++ next-lib/src/thai/MorphoTha.gf | 198 ++++++++++ next-lib/src/thai/NounTha.gf | 76 ++++ next-lib/src/thai/NumeralTha.gf | 66 ++++ next-lib/src/thai/OverloadTha.gf | 1 + next-lib/src/thai/ParadigmsTha.gf | 496 ++++++++++++++++++++++++ next-lib/src/thai/PhraseTha.gf | 24 ++ next-lib/src/thai/QuestionTha.gf | 41 ++ next-lib/src/thai/RelativeTha.gf | 48 +++ next-lib/src/thai/ResTha.gf | 484 +++++++++++++++++++++++ next-lib/src/thai/SentenceTha.gf | 61 +++ next-lib/src/thai/StringsTha.gf | 87 +++++ next-lib/src/thai/StructuralTha.gf | 110 ++++++ next-lib/src/thai/VerbTha.gf | 51 +++ 48 files changed, 5121 insertions(+) create mode 100644 next-lib/src/hindi/AdjectiveHin.gf create mode 100644 next-lib/src/hindi/AdverbHin.gf create mode 100644 next-lib/src/hindi/AllHin.gf create mode 100644 next-lib/src/hindi/AllHinAbs.gf create mode 100644 next-lib/src/hindi/BackwardHin.gf create mode 100644 next-lib/src/hindi/CatHin.gf create mode 100644 next-lib/src/hindi/ConjunctionHin.gf create mode 100644 next-lib/src/hindi/GrammarHin.gf create mode 100644 next-lib/src/hindi/IdiomHin.gf create mode 100644 next-lib/src/hindi/IrregHin.gf create mode 100644 next-lib/src/hindi/LangHin.gf create mode 100644 next-lib/src/hindi/LexiconHin.gf create mode 100644 next-lib/src/hindi/MorphoHin.gf create mode 100644 next-lib/src/hindi/NounHin.gf create mode 100644 next-lib/src/hindi/NumeralHin.gf create mode 100644 next-lib/src/hindi/ParadigmsHin.gf create mode 100644 next-lib/src/hindi/PhraseHin.gf create mode 100644 next-lib/src/hindi/QuestionHin.gf create mode 100644 next-lib/src/hindi/RelativeHin.gf create mode 100644 next-lib/src/hindi/ResHin.gf create mode 100644 next-lib/src/hindi/SentenceHin.gf create mode 100644 next-lib/src/hindi/StructuralHin.gf create mode 100644 next-lib/src/hindi/VerbHin.gf create mode 100644 next-lib/src/thai/AdjectiveTha.gf create mode 100644 next-lib/src/thai/AdverbTha.gf create mode 100644 next-lib/src/thai/AllTha.gf create mode 100644 next-lib/src/thai/AllThaAbs.gf create mode 100644 next-lib/src/thai/CatTha.gf create mode 100644 next-lib/src/thai/ConjunctionTha.gf create mode 100644 next-lib/src/thai/ExtraTha.gf create mode 100644 next-lib/src/thai/GrammarTha.gf create mode 100644 next-lib/src/thai/IdiomTha.gf create mode 100644 next-lib/src/thai/IrregTha.gf create mode 100644 next-lib/src/thai/LangTha.gf create mode 100644 next-lib/src/thai/LexiconTha.gf create mode 100644 next-lib/src/thai/MorphoTha.gf create mode 100644 next-lib/src/thai/NounTha.gf create mode 100644 next-lib/src/thai/NumeralTha.gf create mode 100644 next-lib/src/thai/OverloadTha.gf create mode 100644 next-lib/src/thai/ParadigmsTha.gf create mode 100644 next-lib/src/thai/PhraseTha.gf create mode 100644 next-lib/src/thai/QuestionTha.gf create mode 100644 next-lib/src/thai/RelativeTha.gf create mode 100644 next-lib/src/thai/ResTha.gf create mode 100644 next-lib/src/thai/SentenceTha.gf create mode 100644 next-lib/src/thai/StringsTha.gf create mode 100644 next-lib/src/thai/StructuralTha.gf create mode 100644 next-lib/src/thai/VerbTha.gf diff --git a/next-lib/src/hindi/AdjectiveHin.gf b/next-lib/src/hindi/AdjectiveHin.gf new file mode 100644 index 000000000..b016a2b9d --- /dev/null +++ b/next-lib/src/hindi/AdjectiveHin.gf @@ -0,0 +1,36 @@ +concrete AdjectiveHin of Adjective = CatHin ** open ResHin, Prelude in { + + lin + + PositA a = a ; + +-- ComparA a np = { +-- s = \\_ => a.s ! AAdj Compar ++ "than" ++ np.s ! Nom ; +-- isPre = False +-- } ; +-- +---- $SuperlA$ belongs to determiner syntax in $Noun$. +-- +-- ComplA2 a np = { +-- s = \\_ => a.s ! AAdj Posit ++ a.c2 ++ np.s ! Acc ; +-- isPre = False +-- } ; +-- +-- ReflA2 a = { +-- s = \\ag => a.s ! AAdj Posit ++ a.c2 ++ reflPron ! ag ; +-- isPre = False +-- } ; +-- +-- SentAP ap sc = { +-- s = \\a => ap.s ! a ++ sc.s ; +-- isPre = False +-- } ; +-- +-- AdAP ada ap = { +-- s = \\a => ada.s ++ ap.s ! a ; +-- isPre = ap.isPre +-- } ; +-- +-- UseA2 a = a ; +-- +} diff --git a/next-lib/src/hindi/AdverbHin.gf b/next-lib/src/hindi/AdverbHin.gf new file mode 100644 index 000000000..96ac3242f --- /dev/null +++ b/next-lib/src/hindi/AdverbHin.gf @@ -0,0 +1,20 @@ +concrete AdverbHin of Adverb = CatHin ** open ResHin, Prelude in { +-- +-- lin +-- PositAdvAdj a = {s = a.s ! AAdv} ; +-- ComparAdvAdj cadv a np = { +-- s = cadv.s ++ a.s ! AAdv ++ "than" ++ np.s ! Nom +-- } ; +-- ComparAdvAdjS cadv a s = { +-- s = cadv.s ++ a.s ! AAdv ++ "than" ++ s.s +-- } ; +-- +-- PrepNP prep np = {s = prep.s ++ np.s ! Acc} ; +-- +-- AdAdv = cc2 ; +-- +-- SubjS = cc2 ; +-- +-- AdnCAdv cadv = {s = cadv.s ++ "than"} ; +-- +} diff --git a/next-lib/src/hindi/AllHin.gf b/next-lib/src/hindi/AllHin.gf new file mode 100644 index 000000000..e0256e481 --- /dev/null +++ b/next-lib/src/hindi/AllHin.gf @@ -0,0 +1,3 @@ +--# -path=.:../abstract:../common:prelude + +concrete AllHin of AllHinAbs = LangHin ; diff --git a/next-lib/src/hindi/AllHinAbs.gf b/next-lib/src/hindi/AllHinAbs.gf new file mode 100644 index 000000000..89cdaef54 --- /dev/null +++ b/next-lib/src/hindi/AllHinAbs.gf @@ -0,0 +1 @@ +abstract AllHinAbs = Lang ; \ No newline at end of file diff --git a/next-lib/src/hindi/BackwardHin.gf b/next-lib/src/hindi/BackwardHin.gf new file mode 100644 index 000000000..c880bcbfb --- /dev/null +++ b/next-lib/src/hindi/BackwardHin.gf @@ -0,0 +1,76 @@ +--concrete BackwardHin of Backward = CatHin ** open ResHin in { +-- +-- flags optimize=all_subs ; +-- +-- lin +-- +---- A repository of obsolete constructs, needed for backward compatibility. +---- They create spurious ambiguities if used in combination with Lang. +-- +---- from Verb 19/4/2008 +-- +-- ComplV2 v np = insertObj (\\_ => v.c2 ++ np.s ! Acc) (predV v) ; +-- ComplV3 v np np2 = +-- insertObj (\\_ => v.c2 ++ np.s ! Acc ++ v.c3 ++ np2.s ! Acc) (predV v) ; +-- ComplV2V v np vp = +-- insertObj (\\a => infVP v.isAux vp a) +-- (insertObj (\\_ => v.c2 ++ np.s ! Acc) (predV v)) ; +-- ComplV2S v np s = +-- insertObj (\\_ => conjThat ++ s.s) +-- (insertObj (\\_ => v.c2 ++ np.s ! Acc) (predV v)) ; +-- ComplV2Q v np q = +-- insertObj (\\_ => q.s ! QIndir) +-- (insertObj (\\_ => v.c2 ++ np.s ! Acc) (predV v)) ; +-- ComplV2A v np ap = +-- insertObj (\\_ => v.c2 ++ np.s ! Acc ++ ap.s ! np.a) (predV v) ; +-- +-- ReflV2 v = insertObj (\\a => v.c2 ++ reflPron ! a) (predV v) ; +-- +---- from Sentence 19/4/2008 +-- +-- SlashV2 np v2 = +-- mkClause (np.s ! Nom) np.a (predV v2) ** {c2 = v2.c2} ; +-- +-- SlashVVV2 np vv v2 = +-- mkClause (np.s ! Nom) np.a +-- (insertObj (\\a => infVP vv.isAux (predV v2) a) (predVV vv)) ** +-- {c2 = v2.c2} ; +-- +---- from Noun 19/4/2008 +-- +-- NumInt n = {s = n.s ; n = Pl} ; +-- OrdInt n = {s = n.s ++ "th"} ; --- DEPRECATED +-- +-- DetSg quant ord = { +-- s = quant.s ! Sg ++ ord.s ; +-- n = Sg +-- } ; +-- +-- DetPl quant num ord = { +-- s = quant.s ! num.n ++ num.s ++ ord.s ; +-- n = num.n +-- } ; +-- +-- NoNum = {s = []; n = Pl } ; +-- +-- DefArt = {s = \\_ => artDef} ; +-- +-- IndefArt = { +-- s = table { +-- Sg => artIndef ; +-- Pl => [] +-- } +-- } ; +-- +-- MassDet = {s = \\_ => []} ; +-- +-- +-- +---- from Structural 19/4/2008 +-- +-- that_NP = regNP "that" Sg ; +-- these_NP = regNP "these" Pl ; +-- this_NP = regNP "this" Sg ; +-- those_NP = regNP "those" Pl ; +-- +--} diff --git a/next-lib/src/hindi/CatHin.gf b/next-lib/src/hindi/CatHin.gf new file mode 100644 index 000000000..0d32318d7 --- /dev/null +++ b/next-lib/src/hindi/CatHin.gf @@ -0,0 +1,90 @@ +concrete CatHin of Cat = CommonX ** open ResHin, Prelude in { + + flags optimize=all_subs ; + + lincat +-- +---- Tensed/Untensed +-- +-- S = {s : Str} ; +-- QS = {s : QForm => Str} ; +-- RS = {s : Agr => Str ; c : Case} ; -- c for it clefts +-- SSlash = {s : Str ; c2 : Str} ; +-- +---- Sentence +-- + Cl = ResHin.Clause ; +-- ClSlash = { +-- s : ResHin.Tense => Anteriority => CPolarity => Order => Str ; +-- c2 : Str +-- } ; +-- Imp = {s : CPolarity => ImpForm => Str} ; +-- +---- Question +-- +-- QCl = {s : ResHin.Tense => Anteriority => CPolarity => QForm => Str} ; +-- IP = {s : Case => Str ; n : Number} ; +-- IComp = {s : Str} ; +-- IDet = {s : Str ; n : Number} ; +-- IQuant = {s : Number => Str} ; +-- +---- Relative +-- +-- RCl = { +-- s : ResHin.Tense => Anteriority => CPolarity => Agr => Str ; +-- c : Case +-- } ; +-- RP = {s : RCase => Str ; a : RAgr} ; +-- +---- Verb +-- + VP = ResHin.VPH ; + VPSlash = ResHin.VPHSlash ; +-- Comp = {s : Agr => Str} ; +-- +---- Adjective +-- + AP = ResHin.Adjective ; +-- +---- Noun +-- + CN = ResHin.Noun ; + NP = ResHin.NP ; + Pron = {s : PronCase => Str ; a : Agr} ; + Det = {s : Gender => Case => Str ; n : Number} ; +-- Predet, Ord = {s : Str} ; + Num = {s : Str ; n : Number} ; +-- Card = {s : Str; n : Number} ; + Quant = {s : Number => Gender => Case => Str} ; + Art = {s : Str} ; +-- +---- Numeral +-- +-- Numeral = {s : CardOrd => Str ; n : Number} ; +-- Digits = {s : CardOrd => Str ; n : Number ; tail : DTail} ; +-- +---- Structural +-- +-- Conj = {s1,s2 : Str ; n : Number} ; +-----b Conj = {s : Str ; n : Number} ; +-----b DConj = {s1,s2 : Str ; n : Number} ; +-- Subj = {s : Str} ; +-- Prep = {s : Str} ; +-- +---- Open lexical classes, e.g. Lexicon +-- + V, VS, VQ, VA = Verb ; -- = {s : VForm => Str} ; + V2, V2A, V2Q, V2S = Verb ** {c2 : Compl} ; +-- V3 = Verb ** {c2, c3 : Str} ; +-- VV = {s : VVForm => Str ; isAux : Bool} ; +-- V2V = Verb ** {c2 : Str ; isAux : Bool} ; +-- + A = ResHin.Adjective ; --- {s : Gender => Number => Case => Str} ; +-- A2 = {s : AForm => Str ; c2 : Str} ; +-- + N = {s : Number => Case => Str ; g : Gender} ; +-- N2 = {s : Number => Case => Str ; g : Gender} ** {c2 : Str} ; +-- N3 = {s : Number => Case => Str ; g : Gender} ** {c2,c3 : Str} ; + PN = {s : Case => Str ; g : Gender} ; +-- +} diff --git a/next-lib/src/hindi/ConjunctionHin.gf b/next-lib/src/hindi/ConjunctionHin.gf new file mode 100644 index 000000000..4a2e56096 --- /dev/null +++ b/next-lib/src/hindi/ConjunctionHin.gf @@ -0,0 +1,37 @@ +concrete ConjunctionHin of Conjunction = + CatHin ** open ResHin, Coordination, Prelude in { +-- +-- flags optimize=all_subs ; +-- +-- lin +-- +-- ConjS = conjunctDistrSS ; +-- +-- ConjAdv = conjunctDistrSS ; +-- +-- ConjNP conj ss = conjunctDistrTable Case conj ss ** { +-- a = conjAgr (agrP3 conj.n) ss.a +-- } ; +-- +-- ConjAP conj ss = conjunctDistrTable Agr conj ss ** { +-- isPre = ss.isPre +-- } ; +-- +---- These fun's are generated from the list cat's. +-- +-- BaseS = twoSS ; +-- ConsS = consrSS comma ; +-- BaseAdv = twoSS ; +-- ConsAdv = consrSS comma ; +-- BaseNP x y = twoTable Case x y ** {a = conjAgr x.a y.a} ; +-- ConsNP xs x = consrTable Case comma xs x ** {a = conjAgr xs.a x.a} ; +-- BaseAP x y = twoTable Agr x y ** {isPre = andB x.isPre y.isPre} ; +-- ConsAP xs x = consrTable Agr comma xs x ** {isPre = andB xs.isPre x.isPre} ; +-- +-- lincat +-- [S] = {s1,s2 : Str} ; +-- [Adv] = {s1,s2 : Str} ; +-- [NP] = {s1,s2 : Case => Str ; a : Agr} ; +-- [AP] = {s1,s2 : Agr => Str ; isPre : Bool} ; +-- +} diff --git a/next-lib/src/hindi/GrammarHin.gf b/next-lib/src/hindi/GrammarHin.gf new file mode 100644 index 000000000..21367747b --- /dev/null +++ b/next-lib/src/hindi/GrammarHin.gf @@ -0,0 +1,21 @@ +--# -path=.:../abstract:../common:prelude + +concrete GrammarHin of Grammar = + NounHin, + VerbHin, + AdjectiveHin, + AdverbHin, + NumeralHin, + SentenceHin, + QuestionHin, + RelativeHin, + ConjunctionHin, + PhraseHin, + TextX, + StructuralHin, + IdiomHin + ** { + +flags startcat = Phr ; unlexer = text ; lexer = text ; + +} diff --git a/next-lib/src/hindi/IdiomHin.gf b/next-lib/src/hindi/IdiomHin.gf new file mode 100644 index 000000000..3f012f514 --- /dev/null +++ b/next-lib/src/hindi/IdiomHin.gf @@ -0,0 +1,30 @@ +concrete IdiomHin of Idiom = CatHin ** open Prelude, ResHin in { +-- +-- flags optimize=all_subs ; +-- +-- lin +-- ImpersCl vp = mkClause "it" (agrP3 Sg) vp ; +-- GenericCl vp = mkClause "one" (agrP3 Sg) vp ; +-- +-- CleftNP np rs = mkClause "it" (agrP3 Sg) +-- (insertObj (\\_ => rs.s ! np.a) +-- (insertObj (\\_ => np.s ! rs.c) (predAux auxBe))) ; +-- +-- CleftAdv ad s = mkClause "it" (agrP3 Sg) +-- (insertObj (\\_ => optStr conjThat ++ s.s) +-- (insertObj (\\_ => ad.s) (predAux auxBe))) ; +-- +-- ExistNP np = +-- mkClause "there" (agrP3 (fromAgr np.a).n) +-- (insertObj (\\_ => np.s ! Acc) (predAux auxBe)) ; +-- +-- ExistIP ip = +-- mkQuestion (ss (ip.s ! Nom)) +-- (mkClause "there" (agrP3 ip.n) (predAux auxBe)) ; +-- +-- ProgrVP vp = insertObj (\\a => vp.ad ++ vp.prp ++ vp.s2 ! a) (predAux auxBe) ; +-- +-- ImpPl1 vp = {s = "let's" ++ infVP True vp (AgP1 Pl)} ; +-- +} + diff --git a/next-lib/src/hindi/IrregHin.gf b/next-lib/src/hindi/IrregHin.gf new file mode 100644 index 000000000..ed8ccd392 --- /dev/null +++ b/next-lib/src/hindi/IrregHin.gf @@ -0,0 +1,181 @@ +----# -path=.:prelude:../abstract:../common +-- +--concrete IrregHin of IrregHinAbs = CatHin ** open ParadigmsHin in { +-- +--flags optimize=values ; +-- +-- lin +-- awake_V = irregV "awake" "awoke" "awoken" ; +-- bear_V = irregV "bear" "bore" "born" ; +-- beat_V = irregV "beat" "beat" "beat" ; +-- become_V = irregV "become" "became" "become" ; +-- begin_V = irregV "begin" "began" "begun" ; +-- bend_V = irregV "bend" "bent" "bent" ; +-- beset_V = irregV "beset" "beset" "beset" ; +-- bet_V = irregDuplV "bet" "bet" "bet" ; +-- bid_V = irregDuplV "bid" (variants {"bid" ; "bade"}) (variants {"bid" ; "bidden"}) ; +-- bind_V = irregV "bind" "bound" "bound" ; +-- bite_V = irregV "bite" "bit" "bitten" ; +-- bleed_V = irregV "bleed" "bled" "bled" ; +-- blow_V = irregV "blow" "blew" "blown" ; +-- break_V = irregV "break" "broke" "broken" ; +-- breed_V = irregV "breed" "bred" "bred" ; +-- bring_V = irregV "bring" "brought" "brought" ; +-- broadcast_V = irregV "broadcast" "broadcast" "broadcast" ; +-- build_V = irregV "build" "built" "built" ; +-- burn_V = irregV "burn" (variants {"burned" ; "burnt"}) (variants {"burned" ; "burnt"}) ; +-- burst_V = irregV "burst" "burst" "burst" ; +-- buy_V = irregV "buy" "bought" "bought" ; +-- cast_V = irregV "cast" "cast" "cast" ; +-- catch_V = irregV "catch" "caught" "caught" ; +-- choose_V = irregV "choose" "chose" "chosen" ; +-- cling_V = irregV "cling" "clung" "clung" ; +-- come_V = irregV "come" "came" "come" ; +-- cost_V = irregV "cost" "cost" "cost" ; +-- creep_V = irregV "creep" "crept" "crept" ; +-- cut_V = irregDuplV "cut" "cut" "cut" ; +-- deal_V = irregV "deal" "dealt" "dealt" ; +-- dig_V = irregDuplV "dig" "dug" "dug" ; +-- dive_V = irregV "dive" (variants {"dived" ; "dove"}) "dived" ; +-- do_V = mk5V "do" "does" "did" "done" "doing" ; +-- draw_V = irregV "draw" "drew" "drawn" ; +-- dream_V = irregV "dream" (variants {"dreamed" ; "dreamt"}) (variants {"dreamed" ; "dreamt"}) ; +-- drive_V = irregV "drive" "drove" "driven" ; +-- drink_V = irregV "drink" "drank" "drunk" ; +-- eat_V = irregV "eat" "ate" "eaten" ; +-- fall_V = irregV "fall" "fell" "fallen" ; +-- feed_V = irregV "feed" "fed" "fed" ; +-- feel_V = irregV "feel" "felt" "felt" ; +-- fight_V = irregV "fight" "fought" "fought" ; +-- find_V = irregV "find" "found" "found" ; +-- fit_V = irregDuplV "fit" "fit" "fit" ; +-- flee_V = irregV "flee" "fled" "fled" ; +-- fling_V = irregV "fling" "flung" "flung" ; +-- fly_V = irregV "fly" "flew" "flown" ; +-- forbid_V = irregDuplV "forbid" "forbade" "forbidden" ; +-- forget_V = irregDuplV "forget" "forgot" "forgotten" ; +-- forgive_V = irregV "forgive" "forgave" "forgiven" ; +-- forsake_V = irregV "forsake" "forsook" "forsaken" ; +-- freeze_V = irregV "freeze" "froze" "frozen" ; +-- get_V = irregDuplV "get" "got" "gotten" ; +-- give_V = irregV "give" "gave" "given" ; +-- go_V = mk5V "go" "goes" "went" "gone" "going" ; +-- grind_V = irregV "grind" "ground" "ground" ; +-- grow_V = irregV "grow" "grew" "grown" ; +-- hang_V = irregV "hang" "hung" "hung" ; +-- have_V = mk5V "have" "has" "had" "had" "having" ; +-- hear_V = irregV "hear" "heard" "heard" ; +-- hide_V = irregV "hide" "hid" "hidden" ; +-- hit_V = irregDuplV "hit" "hit" "hit" ; +-- hold_V = irregV "hold" "held" "held" ; +-- hurt_V = irregV "hurt" "hurt" "hurt" ; +-- keep_V = irregV "keep" "kept" "kept" ; +-- kneel_V = irregV "kneel" "knelt" "knelt" ; +-- knit_V = irregDuplV "knit" "knit" "knit" ; +-- know_V = irregV "know" "knew" "know" ; +-- lay_V = irregV "lay" "laid" "laid" ; +-- lead_V = irregV "lead" "led" "led" ; +-- leap_V = irregV "leap" (variants {"leaped" ; "lept"}) (variants {"leaped" ; "lept"}) ; +-- learn_V = irregV "learn" (variants {"learned" ; "learnt"}) (variants {"learned" ; "learnt"}) ; +-- leave_V = irregV "leave" "left" "left" ; +-- lend_V = irregV "lend" "lent" "lent" ; +-- let_V = irregDuplV "let" "let" "let" ; +-- lie_V = irregV "lie" "lay" "lain" ; +-- light_V = irregV "light" (variants {"lighted" ; "lit"}) "lighted" ; +-- lose_V = irregV "lose" "lost" "lost" ; +-- make_V = irregV "make" "made" "made" ; +-- mean_V = irregV "mean" "meant" "meant" ; +-- meet_V = irregV "meet" "met" "met" ; +-- misspell_V = irregV "misspell" (variants {"misspelled" ; "misspelt"}) (variants {"misspelled" ; "misspelt"}) ; +-- mistake_V = irregV "mistake" "mistook" "mistaken" ; +-- mow_V = irregV "mow" "mowed" (variants {"mowed" ; "mown"}) ; +-- overcome_V = irregV "overcome" "overcame" "overcome" ; +-- overdo_V = mk5V "overdo" "overdoes" "overdid" "overdone" "overdoing" ; +-- overtake_V = irregV "overtake" "overtook" "overtaken" ; +-- overthrow_V = irregV "overthrow" "overthrew" "overthrown" ; +-- pay_V = irregV "pay" "paid" "paid" ; +-- plead_V = irregV "plead" "pled" "pled" ; +-- prove_V = irregV "prove" "proved" (variants {"proved" ; "proven"}) ; +-- put_V = irregDuplV "put" "put" "put" ; +-- quit_V = irregDuplV "quit" "quit" "quit" ; +-- read_V = irregV "read" "read" "read" ; +-- rid_V = irregDuplV "rid" "rid" "rid" ; +-- ride_V = irregV "ride" "rode" "ridden" ; +-- ring_V = irregV "ring" "rang" "rung" ; +-- rise_V = irregV "rise" "rose" "risen" ; +-- run_V = irregDuplV "run" "ran" "run" ; +-- saw_V = irregV "saw" "sawed" (variants {"sawed" ; "sawn"}) ; +-- say_V = irregV "say" "said" "said" ; +-- see_V = irregV "see" "saw" "seen" ; +-- seek_V = irregV "seek" "sought" "sought" ; +-- sell_V = irregV "sell" "sold" "sold" ; +-- send_V = irregV "send" "sent" "sent" ; +-- set_V = irregDuplV "set" "set" "set" ; +-- sew_V = irregV "sew" "sewed" (variants {"sewed" ; "sewn"}) ; +-- shake_V = irregV "shake" "shook" "shaken" ; +-- shave_V = irregV "shave" "shaved" (variants {"shaved" ; "shaven"}) ; +-- shear_V = irregV "shear" "shore" "shorn" ; +-- shed_V = irregDuplV "shed" "shed" "shed" ; +-- shine_V = irregV "shine" "shone" "shone" ; +-- shoe_V = irregV "shoe" "shoed" (variants {"shoed" ; "shod"}) ; +-- shoot_V = irregV "shoot" "shot" "shot" ; +-- show_V = irregV "show" "showed" (variants {"showed" ; "shown"}) ; +-- shrink_V = irregV "shrink" "shrank" "shrunk" ; +-- shut_V = irregDuplV "shut" "shut" "shut" ; +-- sing_V = irregV "sing" "sang" "sung" ; +-- sink_V = irregV "sink" "sank" "sunk" ; +-- sit_V = irregDuplV "sit" "sat" "sat" ; +-- sleep_V = irregV "sleep" "slept" "slept" ; +-- slay_V = irregV "slay" "slew" "slain" ; +-- slide_V = irregV "slide" "slid" "slid" ; +-- sling_V = irregV "sling" "slung" "slung" ; +-- slit_V = irregDuplV "slit" "slit" "slit" ; +-- smite_V = irregV "smite" "smote" "smitten" ; +-- sow_V = irregV "sow" "sowed" (variants {"sowed" ; "sown"}) ; +-- speak_V = irregV "speak" "spoke" "spoken" ; +-- speed_V = irregV "speed" "sped" "sped" ; +-- spend_V = irregV "spend" "spent" "spent" ; +-- spill_V = irregV "spill" (variants {"spilled" ; "spilt"}) (variants {"spilled" ; "spilt"}) ; +-- spin_V = irregDuplV "spin" "spun" "spun" ; +-- spit_V = irregDuplV "spit" (variants {"spit" ; "spat"}) "spit" ; +-- split_V = irregDuplV "split" "split" "split" ; +-- spread_V = irregV "spread" "spread" "spread" ; +-- spring_V = irregV "spring" (variants {"sprang" ; "sprung"}) "sprung" ; +-- stand_V = irregV "stand" "stood" "stood" ; +-- steal_V = irregV "steal" "stole" "stolen" ; +-- stick_V = irregV "stick" "stuck" "stuck" ; +-- sting_V = irregV "sting" "stung" "stung" ; +-- stink_V = irregV "stink" "stank" "stunk" ; +-- stride_V = irregV "stride" "strod" "stridden" ; +-- strike_V = irregV "strike" "struck" "struck" ; +-- string_V = irregV "string" "strung" "strung" ; +-- strive_V = irregV "strive" "strove" "striven" ; +-- swear_V = irregV "swear" "swore" "sworn" ; +-- sweep_V = irregV "sweep" "swept" "swept" ; +-- swell_V = irregV "swell" "swelled" (variants {"swelled" ; "swollen"}) ; +-- swim_V = irregDuplV "swim" "swam" "swum" ; +-- swing_V = irregV "swing" "swung" "swung" ; +-- take_V = irregV "take" "took" "taken" ; +-- teach_V = irregV "teach" "taught" "taught" ; +-- tear_V = irregV "tear" "tore" "torn" ; +-- tell_V = irregV "tell" "told" "told" ; +-- think_V = irregV "think" "thought" "thought" ; +-- thrive_V = irregV "thrive" (variants {"thrived" ; "throve"}) "thrived" ; +-- throw_V = irregV "throw" "threw" "thrown" ; +-- thrust_V = irregV "thrust" "thrust" "thrust" ; +-- tread_V = irregV "tread" "trod" "trodden" ; +-- understand_V = irregV "understand" "understood" "understood" ; +-- uphold_V = irregV "uphold" "upheld" "upheld" ; +-- upset_V = irregDuplV "upset" "upset" "upset" ; +-- wake_V = irregV "wake" "woke" "woken" ; +-- wear_V = irregV "wear" "wore" "worn" ; +-- weave_V = irregV "weave" (variants {"weaved" ; "wove"}) (variants {"weaved" ; "woven"}) ; +-- wed_V = irregDuplV "wed" "wed" "wed" ; +-- weep_V = irregV "weep" "wept" "wept" ; +-- wind_V = irregV "wind" "wound" "wound" ; +-- win_V = irregDuplV "win" "won" "won" ; +-- withhold_V = irregV "withhold" "withheld" "withheld" ; +-- withstand_V = irregV "withstand" "withstood" "withstood" ; +-- wring_V = irregV "wring" "wrung" "wrung" ; +-- write_V = irregV "write" "wrote" "written" ; +--} diff --git a/next-lib/src/hindi/LangHin.gf b/next-lib/src/hindi/LangHin.gf new file mode 100644 index 000000000..d043d04f2 --- /dev/null +++ b/next-lib/src/hindi/LangHin.gf @@ -0,0 +1,10 @@ +--# -path=.:../abstract:../common:prelude + +concrete LangHin of Lang = + GrammarHin, + LexiconHin + ** { + +flags startcat = Phr ; unlexer=unwords ; lexer=words ; + +} diff --git a/next-lib/src/hindi/LexiconHin.gf b/next-lib/src/hindi/LexiconHin.gf new file mode 100644 index 000000000..dfa32d661 --- /dev/null +++ b/next-lib/src/hindi/LexiconHin.gf @@ -0,0 +1,375 @@ +--# -path=.:prelude + +concrete LexiconHin of Lexicon = CatHin ** + open ParadigmsHin, Prelude in { + + flags + optimize=values ; + + lin +-- airplane_N = regN "airplane" ; +-- answer_V2S = mkV2S (regV "answer") toP ; +-- apartment_N = regN "apartment" ; + apple_N = mkN "seb" ; +-- art_N = regN "art" ; +-- ask_V2Q = mkV2Q (regV "ask") noPrep ; +-- baby_N = regN "baby" ; +-- bad_A = mkADeg "bad" "worse" "worst" "badly" ; +-- bank_N = regN "bank" ; +-- beautiful_A = compoundADeg (regA "beautiful") ; +-- become_VA = mkVA (irregV "become" "became" "become") ; +-- beer_N = regN "beer" ; +-- beg_V2V = mkV2V (regDuplV "beg") noPrep toP ; + big_A = mkA "baRA" ; +-- bike_N = regN "bike" ; +-- bird_N = regN "bird" ; +-- black_A = regADeg "black" ; +-- blue_A = regADeg "blue" ; +-- boat_N = regN "boat" ; +-- book_N = regN "book" ; +-- boot_N = regN "boot" ; +-- boss_N = mkN human (regN "boss") ; + boy_N = mkN "lar.kA" ; + bread_N = mkN "rot.I" ; +-- break_V2 = dirV2 (irregV "break" "broke" "broken") ; +-- broad_A = regADeg "broad" ; +-- brother_N2 = mkN2 (mkN masculine (mkN "brother")) (mkPrep "of") ; +-- brown_A = regADeg "brown" ; +-- butter_N = regN "butter" ; +-- buy_V2 = dirV2 (irregV "buy" "bought" "bought") ; +-- camera_N = regN "camera" ; +-- cap_N = regN "cap" ; +-- car_N = regN "car" ; +-- carpet_N = regN "carpet" ; +-- cat_N = regN "cat" ; +-- ceiling_N = regN "ceiling" ; +-- chair_N = regN "chair" ; +-- cheese_N = regN "cheese" ; +-- child_N = mk2N "child" "children" ; +-- church_N = regN "church" ; +-- city_N = regN "city" ; +-- clean_A = regADeg "clean" ; +-- clever_A = regADeg "clever" ; +-- close_V2 = dirV2 (regV "close") ; +-- coat_N = regN "coat" ; +-- cold_A = regADeg "cold" ; +-- come_V = (irregV "come" "came" "come") ; +-- computer_N = regN "computer" ; +-- country_N = regN "country" ; +-- cousin_N = mkN human (regN "cousin") ; +-- cow_N = regN "cow" ; +-- die_V = (regV "die") ; +-- dirty_A = regADeg "dirty" ; +-- distance_N3 = mkN3 (regN "distance") fromP toP ; +-- doctor_N = mkN human (regN "doctor") ; +-- dog_N = regN "dog" ; +-- door_N = regN "door" ; +-- drink_V2 = dirV2 (irregV "drink" "drank" "drunk") ; +-- easy_A2V = mkA2V (regA "easy") forP ; + eat_V2 = mkV2 "KA" ; +-- empty_A = regADeg "empty" ; +-- enemy_N = regN "enemy" ; +-- factory_N = regN "factory" ; +-- father_N2 = mkN2 (mkN masculine (mkN "father")) (mkPrep "of") ; +-- fear_VS = mkVS (regV "fear") ; +-- find_V2 = dirV2 (irregV "find" "found" "found") ; +-- fish_N = mk2N "fish" "fish" ; +-- floor_N = regN "floor" ; +-- forget_V2 = dirV2 (irregDuplV "forget" "forgot" "forgotten") ; +-- fridge_N = regN "fridge" ; +-- friend_N = mkN human (regN "friend") ; +-- fruit_N = regN "fruit" ; +-- fun_AV = mkAV (regA "fun") ; +-- garden_N = regN "garden" ; + girl_N = mkN "lar.kI" ; +-- glove_N = regN "glove" ; +-- gold_N = regN "gold" ; + good_A = mkA "a-cCA" ; + go_V = mkV "cal" ; +-- green_A = regADeg "green" ; +-- harbour_N = regN "harbour" ; +-- hate_V2 = dirV2 (regV "hate") ; +-- hat_N = regN "hat" ; +-- have_V2 = dirV2 (mk5V "have" "has" "had" "had" "having") ; +-- hear_V2 = dirV2 (irregV "hear" "heard" "heard") ; +-- hill_N = regN "hill" ; +-- hope_VS = mkVS (regV "hope") ; +-- horse_N = regN "horse" ; +-- hot_A = duplADeg "hot" ; +-- house_N = regN "house" ; +-- important_A = compoundADeg (regA "important") ; +-- industry_N = regN "industry" ; +-- iron_N = regN "iron" ; +-- king_N = mkN masculine (regN "king") ; +-- know_V2 = dirV2 (irregV "know" "knew" "known") ; +-- lake_N = regN "lake" ; +-- lamp_N = regN "lamp" ; +-- learn_V2 = dirV2 (regV "learn") ; +-- leather_N = regN "leather" ; +-- leave_V2 = dirV2 (irregV "leave" "left" "left") ; +-- like_V2 = dirV2 (regV "like") ; +-- listen_V2 = prepV2 (regV "listen") toP ; + live_V = mkV "CU" ; ---- touch +-- long_A = regADeg "long" ; +-- lose_V2 = dirV2 (irregV "lose" "lost" "lost") ; +-- love_N = regN "love" ; +-- love_V2 = dirV2 (regV "love") ; +-- man_N = mkN masculine (mk2N "man" "men") ; +-- married_A2 = mkA2 (regA "married") toP ; +-- meat_N = regN "meat" ; +-- milk_N = regN "milk" ; +-- moon_N = regN "moon" ; +-- mother_N2 = mkN2 (mkN feminine (mkN "mother")) (mkPrep "of") ; +-- mountain_N = regN "mountain" ; +-- music_N = regN "music" ; +-- narrow_A = regADeg "narrow" ; +-- new_A = regADeg "new" ; +-- newspaper_N = regN "newspaper" ; +-- oil_N = regN "oil" ; +-- old_A = regADeg "old" ; +-- open_V2 = dirV2 (regV "open") ; +-- paint_V2A = mkV2A (regV "paint") noPrep ; +-- paper_N = regN "paper" ; +-- paris_PN = mkPN (mkN nonhuman (mkN "Paris")) ; +-- peace_N = regN "peace" ; +-- pen_N = regN "pen" ; +-- planet_N = regN "planet" ; +-- plastic_N = regN "plastic" ; +-- play_V2 = dirV2 (regV "play") ; +-- policeman_N = mkN masculine (mkN "policeman" "policemen") ; +-- priest_N = mkN human (regN "priest") ; +-- probable_AS = mkAS (regA "probable") ; +-- queen_N = mkN feminine (regN "queen") ; +-- radio_N = regN "radio" ; +-- rain_V0 = mkV0 (regV "rain") ; +-- read_V2 = dirV2 (irregV "read" "read" "read") ; + red_A = mkA "lAl" ; +-- religion_N = regN "religion" ; +-- restaurant_N = regN "restaurant" ; +-- river_N = regN "river" ; +-- rock_N = regN "rock" ; +-- roof_N = regN "roof" ; +-- rubber_N = regN "rubber" ; +-- run_V = (irregDuplV "run" "ran" "run") ; +-- say_VS = mkVS (irregV "say" "said" "said") ; +-- school_N = regN "school" ; +-- science_N = regN "science" ; +-- sea_N = regN "sea" ; +-- seek_V2 = dirV2 (irregV "seek" "sought" "sought") ; +-- see_V2 = dirV2 (irregV "see" "saw" "seen") ; +-- sell_V3 = dirV3 (irregV "sell" "sold" "sold") toP ; +-- send_V3 = dirV3 (irregV "send" "sent" "sent") toP ; +-- sheep_N = mk2N "sheep" "sheep" ; +-- ship_N = regN "ship" ; +-- shirt_N = regN "shirt" ; +-- shoe_N = regN "shoe" ; +-- shop_N = regN "shop" ; +-- short_A = regADeg "short" ; +-- silver_N = regN "silver" ; +-- sister_N = mkN2 (mkN feminine (mkN "sister")) (mkPrep "of") ; +-- sleep_V = (irregV "sleep" "slept" "slept") ; +-- small_A = regADeg "small" ; +-- snake_N = regN "snake" ; +-- sock_N = regN "sock" ; +-- speak_V2 = dirV2 (irregV "speak" "spoke" "spoken") ; +-- star_N = regN "star" ; +-- steel_N = regN "steel" ; +-- stone_N = regN "stone" ; +-- stove_N = regN "stove" ; +-- student_N = mkN human (regN "student") ; +-- stupid_A = regADeg "stupid" ; +-- sun_N = regN "sun" ; +-- switch8off_V2 = dirV2 (partV (regV "switch") "off") ; +-- switch8on_V2 = dirV2 (partV (regV "switch") "on") ; +-- table_N = regN "table" ; +-- talk_V3 = mkV3 (regV "talk") toP aboutP ; +-- teacher_N = mkN human (regN "teacher") ; +-- teach_V2 = dirV2 (irregV "teach" "taught" "taught") ; +-- television_N = regN "television" ; +-- thick_A = regADeg "thick" ; +-- thin_A = duplADeg "thin" ; +-- train_N = regN "train" ; +-- travel_V = (regDuplV "travel") ; +-- tree_N = regN "tree" ; +-- ---- trousers_N = regN "trousers" ; +-- ugly_A = regADeg "ugly" ; +-- understand_V2 = dirV2 (irregV "understand" "understood" "understood") ; +-- university_N = regN "university" ; +-- village_N = regN "village" ; +-- wait_V2 = prepV2 (regV "wait") forP ; +-- walk_V = (regV "walk") ; +-- warm_A = regADeg "warm" ; +-- war_N = regN "war" ; +-- watch_V2 = dirV2 (regV "watch") ; +-- water_N = regN "water" ; +-- white_A = regADeg "white" ; +-- window_N = regN "window" ; +-- wine_N = regN "wine" ; +-- win_V2 = dirV2 (irregDuplV "win" "won" "won") ; +-- woman_N = mkN feminine (mk2N "woman" "women") ; +-- wonder_VQ = mkVQ (regV "wonder") ; +-- wood_N = regN "wood" ; +-- write_V2 = dirV2 (irregV "write" "wrote" "written") ; +-- yellow_A = regADeg "yellow" ; +-- young_A = regADeg "young" ; +-- +-- do_V2 = dirV2 (mk5V "do" "does" "did" "done" "doing") ; +-- now_Adv = mkAdv "now" ; +-- already_Adv = mkAdv "already" ; +-- song_N = regN "song" ; +-- add_V3 = dirV3 (regV "add") toP ; +-- number_N = regN "number" ; +-- put_V2 = prepV2 (irregDuplV "put" "put" "put") noPrep ; +-- stop_V = regDuplV "stop" ; +-- jump_V = regV "jump" ; +-- +-- left_Ord = ss "left" ; +-- right_Ord = ss "right" ; +-- far_Adv = mkAdv "far" ; +-- correct_A = (regA "correct") ; +-- dry_A = regA "dry" ; +-- dull_A = regA "dull" ; +-- full_A = regA "full" ; +-- heavy_A = regA "heavy" ; +-- near_A = regA "near" ; +-- rotten_A = (regA "rotten") ; +-- round_A = regA "round" ; +-- sharp_A = regA "sharp" ; +-- smooth_A = regA "smooth" ; +-- straight_A = regA "straight" ; +-- wet_A = regA "wet" ; ---- +-- wide_A = regA "wide" ; +-- animal_N = regN "animal" ; +-- ashes_N = regN "ash" ; -- FIXME: plural only? +-- back_N = regN "back" ; +-- bark_N = regN "bark" ; +-- belly_N = regN "belly" ; +-- blood_N = regN "blood" ; +-- bone_N = regN "bone" ; +-- breast_N = regN "breast" ; +-- cloud_N = regN "cloud" ; +-- day_N = regN "day" ; +-- dust_N = regN "dust" ; +-- ear_N = regN "ear" ; +-- earth_N = regN "earth" ; +-- egg_N = regN "egg" ; +-- eye_N = regN "eye" ; +-- fat_N = regN "fat" ; +-- feather_N = regN "feather" ; +-- fingernail_N = regN "fingernail" ; +-- fire_N = regN "fire" ; +-- flower_N = regN "flower" ; +-- fog_N = regN "fog" ; +-- foot_N = mk2N "foot" "feet" ; +-- forest_N = regN "forest" ; +-- grass_N = regN "grass" ; +-- guts_N = regN "gut" ; -- FIXME: no singular +-- hair_N = regN "hair" ; +-- hand_N = regN "hand" ; +-- head_N = regN "head" ; +-- heart_N = regN "heart" ; +-- horn_N = regN "horn" ; +-- husband_N = mkN masculine (regN "husband") ; +-- ice_N = regN "ice" ; +-- knee_N = regN "knee" ; +-- leaf_N = mk2N "leaf" "leaves" ; +-- leg_N = regN "leg" ; +-- liver_N = regN "liver" ; +-- louse_N = mk2N "louse" "lice" ; +-- mouth_N = regN "mouth" ; +-- name_N = regN "name" ; +-- neck_N = regN "neck" ; +-- night_N = regN "night" ; +-- nose_N = regN "nose" ; +-- person_N = mkN human (regN "person") ; +-- rain_N = regN "rain" ; +-- road_N = regN "road" ; +-- root_N = regN "root" ; +-- rope_N = regN "rope" ; +-- salt_N = regN "salt" ; +-- sand_N = regN "sand" ; +-- seed_N = regN "seed" ; +-- skin_N = regN "skin" ; +-- sky_N = regN "sky" ; +-- smoke_N = regN "smoke" ; +-- snow_N = regN "snow" ; +-- stick_N = regN "stick" ; +-- tail_N = regN "tail" ; +-- tongue_N = regN "tongue" ; +-- tooth_N = mk2N "tooth" "teeth" ; +-- wife_N = mkN feminine (mk2N "wife" "wives") ; +-- wind_N = regN "wind" ; +-- wing_N = regN "wing" ; +-- worm_N = regN "worm" ; +-- year_N = regN "year" ; +-- blow_V = IrregHin.blow_V ; +-- breathe_V = dirV2 (regV "breathe") ; +-- burn_V = IrregHin.burn_V ; +-- dig_V = IrregHin.dig_V ; +-- fall_V = IrregHin.fall_V ; +-- float_V = regV "float" ; +-- flow_V = regV "flow" ; +-- fly_V = IrregHin.fly_V ; +-- freeze_V = IrregHin.freeze_V ; +-- give_V3 = dirV3 give_V toP ; +-- laugh_V = regV "laugh" ; +-- lie_V = IrregHin.lie_V ; +-- play_V = regV "play" ; + sew_V = mkV "sI" ; +-- sing_V = IrregHin.sing_V ; +-- sit_V = IrregHin.sit_V ; +-- smell_V = regV "smell" ; +-- spit_V = IrregHin.spit_V ; +-- stand_V = IrregHin.stand_V ; +-- swell_V = IrregHin.swell_V ; +-- swim_V = IrregHin.swim_V ; +-- think_V = IrregHin.think_V ; +-- turn_V = regV "turn" ; +-- vomit_V = regV "vomit" ; +-- +-- bite_V2 = dirV2 IrregHin.bite_V ; +-- count_V2 = dirV2 (regV "count") ; +-- cut_V2 = dirV2 IrregHin.cut_V ; +-- fear_V2 = dirV2 (regV "fear") ; +-- fight_V2 = dirV2 fight_V ; + hit_V2 = mkV2 (mkV "mAr") "ko" ; +-- hold_V2 = dirV2 hold_V ; +-- hunt_V2 = dirV2 (regV "hunt") ; +-- kill_V2 = dirV2 (regV "kill") ; +-- pull_V2 = dirV2 (regV "pull") ; +-- push_V2 = dirV2 (regV "push") ; +-- rub_V2 = dirV2 (regDuplV "rub") ; +-- scratch_V2 = dirV2 (regV "scratch") ; +-- split_V2 = dirV2 split_V ; +-- squeeze_V2 = dirV2 (regV "squeeze") ; +-- stab_V2 = dirV2 (regDuplV "stab") ; +-- suck_V2 = dirV2 (regV "suck") ; +-- throw_V2 = dirV2 throw_V ; +-- tie_V2 = dirV2 (regV "tie") ; +-- wash_V2 = dirV2 (regV "wash") ; +-- wipe_V2 = dirV2 (regV "wipe") ; +-- +---- other_A = regA "other" ; +-- +-- grammar_N = regN "grammar" ; +-- language_N = regN "language" ; +-- rule_N = regN "rule" ; +-- +---- added 4/6/2007 + john_PN = mkPN "jon" ; +-- question_N = regN "question" ; +-- ready_A = regA "ready" ; +-- reason_N = regN "reason" ; +-- today_Adv = mkAdv "today" ; +-- uncertain_A = regA "uncertain" ; +-- +--oper +-- aboutP = mkPrep "about" ; +-- atP = mkPrep "at" ; +-- forP = mkPrep "for" ; +-- fromP = mkPrep "from" ; +-- inP = mkPrep "in" ; +-- onP = mkPrep "on" ; +-- toP = mkPrep "to" ; +-- +} diff --git a/next-lib/src/hindi/MorphoHin.gf b/next-lib/src/hindi/MorphoHin.gf new file mode 100644 index 000000000..0cb00fde5 --- /dev/null +++ b/next-lib/src/hindi/MorphoHin.gf @@ -0,0 +1,196 @@ +--# -path=.:../../prelude +-- +----1 A Simple Hindi Resource Morphology +---- +---- Aarne Ranta 2002 -- 2005 +---- +---- This resource morphology contains definitions needed in the resource +---- syntax. To build a lexicon, it is better to use $ParadigmsHin$, which +---- gives a higher-level access to this module. +-- +resource MorphoHin = ResHin ** open Prelude, (Predef=Predef) in { +-- +-- flags optimize=all ; +-- +----2 Phonology +---- +---- To regulate the use of endings for both nouns, adjectives, and verbs: +-- +--oper +-- y2ie : Str -> Str -> Str = \fly,s -> +-- let y = last (init fly) in +-- case y of { +-- "a" => fly + s ; +-- "e" => fly + s ; +-- "o" => fly + s ; +-- "u" => fly + s ; +-- _ => init fly + "ie" + s +-- } ; +-- +-- +----2 Nouns +---- +---- For conciseness and abstraction, we define a worst-case macro for +---- noun inflection. It is used for defining special case that +---- only need one string as argument. +-- +--oper +-- CommonNoun : Type = {s : Number => Case => Str} ; +-- +-- nounGen : Str -> CommonNoun = \dog -> case last dog of { +-- "y" => nounY "dog" ; +-- "s" => nounS (init "dog") ; +-- _ => nounReg "dog" +-- } ; +-- +---- These are auxiliaries to $nounGen$. +-- +-- nounReg : Str -> CommonNoun = \dog -> +-- mkNoun dog (dog + "s") (dog + "'s") (dog + "s'"); +-- nounS : Str -> CommonNoun = \kiss -> +-- mkNoun kiss (kiss + "es") (kiss + "'s") (kiss + "es'") ; +-- nounY : Str -> CommonNoun = \fl -> +-- mkNoun (fl + "y") (fl + "ies") (fl + "y's") (fl + "ies'") ; +-- +-- +----3 Proper names +---- +---- Regular proper names are inflected with "'s" in the genitive. +-- +-- nameReg : Str -> Gender -> {s : Case => Str} = \john,g -> +-- {s = table {Gen => john + "'s" ; _ => john} ; g = g} ; +-- +----2 Determiners +-- +-- mkDeterminer : Number -> Str -> {s : Str ; n : Number} = \n,s -> +-- {s = s ; n = n} ; +-- +----2 Pronouns +---- +---- Here we define personal pronouns. +---- +---- We record the form "mine" and the gender for later use. +-- +-- Pronoun : Type = +-- {s : Case => Str ; a : Agr} ; +-- +-- mkPronoun : (_,_,_,_ : Str) -> Number -> Person -> Gender -> Pronoun = +-- \I,me,my,mine,n,p,g -> +-- {s = table {Nom => I ; Acc => me ; Gen => my} ; +-- a = toAgr n p g +-- } ; +-- +-- human : Gender = Masc ; --- doesn't matter +-- +-- pronI = mkPronoun "I" "me" "my" "mine" Sg P1 human ; +-- pronYouSg = mkPronoun "you" "you" "your" "yours" Sg P2 human ; -- verb agr OK +-- pronHe = mkPronoun "he" "him" "his" "his" Sg P3 Masc ; +-- pronShe = mkPronoun "she" "her" "her" "hers" Sg P3 Fem ; +-- pronIt = mkPronoun "it" "it" "its" "it" Sg P3 Neutr ; +-- +-- pronWe = mkPronoun "we" "us" "our" "ours" Pl P1 human ; +-- pronYouPl = mkPronoun "you" "you" "your" "yours" Pl P2 human ; +-- pronThey = mkPronoun "they" "them" "their" "theirs" Pl P3 human ; --- +-- +-- +----2 Adjectives +---- +---- To form the adjectival and the adverbial forms, two strings are needed +---- in the worst case. (First without degrees.) +-- +-- Adjective = {s : AForm => Str} ; +-- +---- However, most adjectives can be inflected using the final character. +---- N.B. this is not correct for "shy", but $mkAdjective$ has to be used. +-- +-- regAdjective : Str -> Adjective = \free -> +-- let +-- e = last free ; +-- fre = init free ; +-- freely = case e of { +-- "y" => fre + "ily" ; +-- _ => free + "ly" +-- } ; +-- fre = case e of { +-- "e" => fre ; +-- "y" => fre + "i" ; +-- _ => free +-- } +-- in +-- mkAdjective free (fre + "er") (fre + "est") freely ; +-- +---- Many adjectives are 'inflected' by adding a comparison word. +-- +-- adjDegrLong : Str -> Adjective = \ridiculous -> +-- mkAdjective +-- ridiculous +-- ("more" ++ ridiculous) +-- ("most" ++ ridiculous) +-- ((regAdjective ridiculous).s ! AAdv) ; +-- +-- +----3 Verbs +---- +---- The worst case needs five forms. (The verb "be" is treated separately.) +-- +-- mkVerb4 : (_,_,_,_: Str) -> Verb = \go,goes,went,gone -> +-- let going = case last go of { +-- "e" => init go + "ing" ; +-- _ => go + "ing" +-- } +-- in +-- mkVerb go goes went gone going ; +-- +---- This is what we use to derive the irregular forms in almost all cases +-- +-- mkVerbIrreg : (_,_,_ : Str) -> Verb = \bite,bit,bitten -> +-- let bites = case last bite of { +-- "y" => y2ie bite "s" ; +-- "s" => init bite + "es" ; +-- _ => bite + "s" +-- } +-- in mkVerb4 bite bites bit bitten ; +-- +---- This is used to derive regular forms. +-- +-- mkVerbReg : Str -> Verb = \soak -> +-- let +-- soaks = case last soak of { +-- "y" => y2ie soak "s" ; +-- "s" => init soak + "es" ; +-- _ => soak + "s" +-- } ; +-- soaked = case last soak of { +-- "e" => init soak + "s" ; +-- _ => soak + "ed" +-- } +-- in +-- mkVerb4 soak soaks soaked soaked ; +-- +-- verbGen : Str -> Verb = \kill -> case last kill of { +-- "y" => verbP3y (init kill) ; +-- "e" => verbP3e (init kill) ; +-- "s" => verbP3s (init kill) ; +-- _ => regVerbP3 kill +-- } ; +-- +---- These are just auxiliary to $verbGen$. +-- +-- regVerbP3 : Str -> Verb = \walk -> +-- mkVerbIrreg walk (walk + "ed") (walk + "ed") ; +-- verbP3s : Str -> Verb = \kiss -> +-- mkVerb4 kiss (kiss + "es") (kiss + "ed") (kiss + "ed") ; +-- verbP3e : Str -> Verb = \love -> +-- mkVerb4 love (love + "s") (love + "d") (love + "d") ; +-- verbP3y : Str -> Verb = \cr -> +-- mkVerb4 (cr + "y") (cr + "ies") (cr + "ied") (cr + "ied") ; +-- +----- The particle always appears right after the verb. +-- +-- verbPart : Verb -> Str -> Verb = \v,p -> +-- {s = \\f => v.s ! f ++ p ; isRefl = v.isRefl} ; +-- +-- verbNoPart : Verb -> Verb = \v -> verbPart v [] ; +-- +-- +} diff --git a/next-lib/src/hindi/NounHin.gf b/next-lib/src/hindi/NounHin.gf new file mode 100644 index 000000000..6db90c867 --- /dev/null +++ b/next-lib/src/hindi/NounHin.gf @@ -0,0 +1,121 @@ +concrete NounHin of Noun = CatHin ** open ResHin, Prelude in { + + flags optimize=all_subs ; + + lin + DetCN det cn = { + s = \\c => det.s ! cn.g ! npcase2case c ++ toNP (cn.s ! det.n) c ; + a = agrP3 cn.g det.n + } ; + + UsePN pn = {s = \\c => toNP pn.s c ; a = agrP3 pn.g Sg} ; + UsePron p = {s = \\c => p.s ! np2pronCase c ; a = p.a} ; +-- +-- PredetNP pred np = { +-- s = \\c => pred.s ++ np.s ! c ; +-- a = np.a +-- } ; +-- +-- PPartNP np v2 = { +-- s = \\c => np.s ! c ++ v2.s ! VPPart ; +-- a = np.a +-- } ; +-- +-- RelNP np rs = { +-- s = \\c => np.s ! c ++ "," ++ rs.s ! np.a ; +-- a = np.a +-- } ; +-- +-- AdvNP np adv = { +-- s = \\c => np.s ! c ++ adv.s ; +-- a = np.a +-- } ; +-- +-- DetQuantOrd quant num ord = { +-- s = quant.s ! num.n ++ num.s ++ ord.s ; +-- n = num.n +-- } ; + + DetQuant quant num = { + s = \\g,c => quant.s ! num.n ! g ! c ++ num.s ; + n = num.n + } ; + +-- DetNP det = { +-- s = \\c => det.s ; ---- case +-- a = agrP3 det.n +-- } ; +-- + PossPron p = {s = \\_,_,_ => p.s ! PPoss} ; + + NumSg = {s = []; n = Sg} ; + NumPl = {s = []; n = Pl} ; + +-- NumCard n = n ** {hasCard = True} ; +-- +-- NumDigits n = {s = n.s ! NCard ; n = n.n} ; +-- OrdDigits n = {s = n.s ! NOrd} ; +-- +-- NumNumeral numeral = {s = numeral.s ! NCard; n = numeral.n} ; +-- OrdNumeral numeral = {s = numeral.s ! NOrd} ; +-- +-- AdNum adn num = {s = adn.s ++ num.s ; n = num.n} ; +-- +-- OrdSuperl a = {s = a.s ! AAdj Superl} ; +-- +-- DetArtOrd art num ord = { +-- s = art.s ! num.hasCard ! num.n ++ num.s ++ ord.s ; +-- n = num.n +-- } ; +-- +-- DetArtCard art card = { +-- s = art.s ! True ! card.n ++ card.s ; +-- n = card.n +-- } ; + + DefArt = {s = \\_,_,_ => []} ; + IndefArt = {s = \\_,_,_ => []} ; + +-- MassNP cn = { +-- s = cn.s ! Sg ; +-- a = agrP3 Sg +-- } ; + + UseN n = n ; +-- UseN2 n = n ; +-- +-- Use2N3 f = { +-- s = \\n,c => f.s ! n ! Nom ; +-- g = f.g ; +-- c2 = f.c2 +-- } ; +-- +-- Use3N3 f = { +-- s = \\n,c => f.s ! n ! Nom ; +-- g = f.g ; +-- c2 = f.c3 +-- } ; +-- +-- ComplN2 f x = {s = \\n,c => f.s ! n ! Nom ++ f.c2 ++ x.s ! c ; g = f.g} ; +-- ComplN3 f x = { +-- s = \\n,c => f.s ! n ! Nom ++ f.c2 ++ x.s ! c ; +-- g = f.g ; +-- c2 = f.c3 +-- } ; + + AdjCN ap cn = { + s = \\n,c => ap.s ! cn.g ! n ! c ++ cn.s ! n ! c ; + g = cn.g + } ; + +-- RelCN cn rs = { +-- s = \\n,c => cn.s ! n ! c ++ rs.s ! agrgP3 n cn.g ; +-- g = cn.g +-- } ; +-- AdvCN cn ad = {s = \\n,c => cn.s ! n ! c ++ ad.s ; g = cn.g} ; +-- +-- SentCN cn sc = {s = \\n,c => cn.s ! n ! c ++ sc.s ; g = cn.g} ; +-- +-- ApposCN cn np = {s = \\n,c => cn.s ! n ! Nom ++ np.s ! c ; g = cn.g} ; +-- +} diff --git a/next-lib/src/hindi/NumeralHin.gf b/next-lib/src/hindi/NumeralHin.gf new file mode 100644 index 000000000..ae4a01ac9 --- /dev/null +++ b/next-lib/src/hindi/NumeralHin.gf @@ -0,0 +1,95 @@ +concrete NumeralHin of Numeral = CatHin ** open ResHin in { +-- +--lincat +-- Digit = {s : DForm => CardOrd => Str} ; +-- Sub10 = {s : DForm => CardOrd => Str ; n : Number} ; +-- Sub100 = {s : CardOrd => Str ; n : Number} ; +-- Sub1000 = {s : CardOrd => Str ; n : Number} ; +-- Sub1000000 = {s : CardOrd => Str ; n : Number} ; +-- +--lin num x = x ; +--lin n2 = let two = mkNum "two" "twelve" "twenty" "second" in +-- {s = \\f,c => case of { +-- => "twelfth" ; +-- _ => two.s ! f ! c +-- } +-- } ; +-- +--lin n3 = mkNum "three" "thirteen" "thirty" "third" ; +--lin n4 = mkNum "four" "fourteen" "forty" "fourth" ; +--lin n5 = mkNum "five" "fifteen" "fifty" "fifth" ; +--lin n6 = regNum "six" ; +--lin n7 = regNum "seven" ; +--lin n8 = mkNum "eight" "eighteen" "eighty" "eighth" ; +--lin n9 = mkNum "nine" "nineteen" "ninety" "ninth" ; +-- +--lin pot01 = mkNum "one" "eleven" "ten" "first" ** {n = Sg} ; +--lin pot0 d = d ** {n = Pl} ; +--lin pot110 = regCardOrd "ten" ** {n = Pl} ; +--lin pot111 = regCardOrd "eleven" ** {n = Pl} ; +--lin pot1to19 d = {s = d.s ! teen} ** {n = Pl} ; +--lin pot0as1 n = {s = n.s ! unit} ** {n = n.n} ; +--lin pot1 d = {s = d.s ! ten} ** {n = Pl} ; +--lin pot1plus d e = { +-- s = \\c => d.s ! ten ! NCard ++ "-" ++ e.s ! unit ! c ; n = Pl} ; +--lin pot1as2 n = n ; +--lin pot2 d = {s = \\c => d.s ! unit ! NCard ++ mkCard c "hundred"} ** {n = Pl} ; +--lin pot2plus d e = { +-- s = \\c => d.s ! unit ! NCard ++ "hundred" ++ "and" ++ e.s ! c ; n = Pl} ; +--lin pot2as3 n = n ; +--lin pot3 n = { +-- s = \\c => n.s ! NCard ++ mkCard c "thousand" ; n = Pl} ; +--lin pot3plus n m = { +-- s = \\c => n.s ! NCard ++ "thousand" ++ m.s ! c ; n = Pl} ; +-- +---- numerals as sequences of digits +-- +-- lincat +-- Dig = TDigit ; +-- +-- lin +-- IDig d = d ** {tail = T1} ; +-- +-- IIDig d i = { +-- s = \\o => d.s ! NCard ++ commaIf i.tail ++ i.s ! o ; +-- n = Pl ; +-- tail = inc i.tail +-- } ; +-- +-- D_0 = mkDig "0" ; +-- D_1 = mk3Dig "1" "1st" Sg ; +-- D_2 = mk2Dig "2" "2nd" ; +-- D_3 = mk2Dig "3" "3rd" ; +-- D_4 = mkDig "4" ; +-- D_5 = mkDig "5" ; +-- D_6 = mkDig "6" ; +-- D_7 = mkDig "7" ; +-- D_8 = mkDig "8" ; +-- D_9 = mkDig "9" ; +-- +-- oper +-- commaIf : DTail -> Str = \t -> case t of { +-- T3 => "," ; +-- _ => [] +-- } ; +-- +-- inc : DTail -> DTail = \t -> case t of { +-- T1 => T2 ; +-- T2 => T3 ; +-- T3 => T1 +-- } ; +-- +-- mk2Dig : Str -> Str -> TDigit = \c,o -> mk3Dig c o Pl ; +-- mkDig : Str -> TDigit = \c -> mk2Dig c (c + "th") ; +-- +-- mk3Dig : Str -> Str -> Number -> TDigit = \c,o,n -> { +-- s = table {NCard => c ; NOrd => o} ; +-- n = n +-- } ; +-- +-- TDigit = { +-- n : Number ; +-- s : CardOrd => Str +-- } ; +-- +} diff --git a/next-lib/src/hindi/ParadigmsHin.gf b/next-lib/src/hindi/ParadigmsHin.gf new file mode 100644 index 000000000..7844caf6e --- /dev/null +++ b/next-lib/src/hindi/ParadigmsHin.gf @@ -0,0 +1,556 @@ +--# -path=.:../abstract:../../prelude:../common +-- +----1 Hindi Lexical Paradigms + +resource ParadigmsHin = open + (Predef=Predef), + Prelude, + MorphoHin, + CatHin + in { + +--2 Parameters + +oper + masculine : Gender ; + feminine : Gender ; + + masculine = Masc ; feminine = Fem ; --i + + +--2 Nouns + + mkN = overload { + mkN : Str -> N + = \s -> regNoun s ** {lock_N = <>} ; + mkN : Str -> Gender -> N + = \s,g -> reggNoun s g ** {lock_N = <>} ; + mkN : (x1,_,_,_,_,x6 : Str) -> Gender -> N + = \sd,so,sv,pd,po,pv,g -> mkNoun sd so sv pd po pv g ** {lock_N = <>} ; + } ; + + mkPN = overload { + mkPN : Str -> PN = \s -> let n = regNoun s in {s = n.s ! Sg ; g = n.g ; lock_PN = <>} ; + mkPN : N -> Gender -> PN = \n,g -> {s = n.s ! Sg ; g = g ; lock_PN = <>} ; + } ; + +--2 Adjectives + + mkA = overload { + mkA : Str -> A + = \s -> regAdjective s ** {lock_A = <>} ; + mkA : (x1,_,x3 : Str) -> A + = \msd,m,f -> mkAdjective msd m f ** {lock_A = <>} ; + } ; + +--2 Verbs + + mkV = overload { + mkV : Str -> V + = \s -> regVerb s ** {lock_V = <>} ; + mkV : (x1,_,_,_,_,_,_,_,_,_,_,_,_,_,x15 : Str) -> V + = \inf,stem,ims,imp,ifs,ifp,pms,pmp,pfs,pfp,ss1,ss2,sp2,sp3,r -> + mkVerb inf stem ims imp ifs ifp pms pmp pfs pfp ss1 ss2 sp2 sp3 r ** + {lock_V = <>} ; + } ; + + + mkV2 = overload { + mkV2 : Str -> V2 + = \s -> regVerb s ** {c2 = {s = [] ; c = VTrans} ; lock_V2 = <>} ; + mkV2 : V -> V2 + = \v -> v ** {c2 = {s = [] ; c = VTrans} ; lock_V2 = <>} ; + mkV2 : V -> Str -> V2 + = \v,p -> v ** {c2 = {s = p ; c = VTransPost} ; lock_V2 = <>} ; + } ; + + +----3 Compound nouns +---- +---- A compound noun is an uninflected string attached to an inflected noun, +---- such as "baby boom", "chief executive officer". +-- +-- mkN : Str -> N -> N +-- } ; +-- +-- +----3 Relational nouns +---- +---- Relational nouns ("daughter of x") need a preposition. +-- +-- mkN2 : N -> Prep -> N2 ; +-- +---- The most common preposition is "of", and the following is a +---- shortcut for regular relational nouns with "of". +-- +-- regN2 : Str -> N2 ; +-- +---- Use the function $mkPrep$ or see the section on prepositions below to +---- form other prepositions. +---- +---- Three-place relational nouns ("the connection from x to y") need two prepositions. +-- +-- mkN3 : N -> Prep -> Prep -> N3 ; +-- +-- +-- +----3 Proper names and noun phrases +---- +---- Proper names, with a regular genitive, are formed from strings. +-- +-- mkPN : overload { +-- +-- mkPN : Str -> PN ; +-- +---- Sometimes a common noun can be reused as a proper name, e.g. "Bank" +-- +-- mkPN : N -> PN +-- } ; +-- +-- +----2 Adjectives +-- +-- mkA : overload { +-- +---- For regular adjectives, the adverbial and comparison forms are derived. This holds +---- even for cases with the variations "happy - happily - happier - happiest", +---- "free - freely - freer - freest", and "rude - rudest". +-- +-- mkA : (happy : Str) -> A ; +-- +---- However, the duplication of the final consonant cannot be predicted, +---- but a separate case is used to give the comparative +-- +-- mkA : (fat,fatter : Str) -> A ; +-- +---- As many as four forms may be needed. +-- +-- mkA : (good,better,best,well : Str) -> A +-- } ; +-- +---- To force comparison to be formed by "more - most", +---- the following function is used: +-- +-- compoundA : A -> A ; -- -/more/most ridiculous +-- +-- +-- +----3 Two-place adjectives +---- +---- Two-place adjectives need a preposition for their second argument. +-- +-- mkA2 : A -> Prep -> A2 ; +-- +-- +-- +----2 Adverbs +-- +---- Adverbs are not inflected. Most lexical ones have position +---- after the verb. Some can be preverbal (e.g. "always"). +-- +-- mkAdv : Str -> Adv ; +-- mkAdV : Str -> AdV ; +-- +---- Adverbs modifying adjectives and sentences can also be formed. +-- +-- mkAdA : Str -> AdA ; +-- +----2 Prepositions +---- +---- A preposition as used for rection in the lexicon, as well as to +---- build $PP$s in the resource API, just requires a string. +-- +-- mkPrep : Str -> Prep ; +-- noPrep : Prep ; +-- +---- (These two functions are synonyms.) +-- +----2 Verbs +---- +-- +---- Verbs are constructed by the function $mkV$, which takes a varying +---- number of arguments. +-- +-- mkV : overload { +-- +---- The regular verb function recognizes the special cases where the last +---- character is "y" ("cry-cries" but "buy-buys") or a sibilant +---- ("kiss-"kisses", "jazz-jazzes", "rush-rushes", "munch - munches", +---- "fix - fixes"). +-- +-- mkV : (cry : Str) -> V ; +-- +---- Give the present and past forms for regular verbs where +---- the last letter is duplicated in some forms, +---- e.g. "rip - ripped - ripping". +-- +-- mkV : (stop, stopped : Str) -> V ; +-- +---- There is an extensive list of irregular verbs in the module $IrregularHin$. +---- In practice, it is enough to give three forms, +---- e.g. "drink - drank - drunk". +-- +-- mkV : (drink, drank, drunk : Str) -> V ; +-- +---- Irregular verbs with duplicated consonant in the present participle. +-- +-- mkV : (run, ran, run, running : Str) -> V ; +-- +---- Except for "be", the worst case needs five forms: the infinitive and +---- the third person singular present, the past indicative, and the +---- past and present participles. +-- +-- mkV : (go, goes, went, gone, going : Str) -> V +-- }; +-- +---- Verbs with a particle. +---- The particle, such as in "switch on", is given as a string. +-- +-- partV : V -> Str -> V ; +-- +---- Reflexive verbs. +---- By default, verbs are not reflexive; this function makes them that. +-- +-- reflV : V -> V ; +-- +----3 Two-place verbs +---- +---- Two-place verbs need a preposition, except the special case with direct object. +---- (transitive verbs). Notice that a particle comes from the $V$. +-- +-- mkV2 : overload { +-- mkV2 : V -> Prep -> V2 ; -- believe in +-- mkV2 : V -> V2 -- kill +-- }; +-- +----3 Three-place verbs +---- +---- Three-place (ditransitive) verbs need two prepositions, of which +---- the first one or both can be absent. +-- +-- mkV3 : V -> Prep -> Prep -> V3 ; -- speak, with, about +-- dirV3 : V -> Prep -> V3 ; -- give,_,to +-- dirdirV3 : V -> V3 ; -- give,_,_ +-- +----3 Other complement patterns +---- +---- Verbs and adjectives can take complements such as sentences, +---- questions, verb phrases, and adjectives. +-- +-- mkV0 : V -> V0 ; +-- mkVS : V -> VS ; +-- mkV2S : V -> Prep -> V2S ; +-- mkVV : V -> VV ; +-- mkV2V : V -> Prep -> Prep -> V2V ; +-- mkVA : V -> VA ; +-- mkV2A : V -> Prep -> V2A ; +-- mkVQ : V -> VQ ; +-- mkV2Q : V -> Prep -> V2Q ; +-- +-- mkAS : A -> AS ; +-- mkA2S : A -> Prep -> A2S ; +-- mkAV : A -> AV ; +-- mkA2V : A -> Prep -> A2V ; +-- +---- Notice: Categories $V0, AS, A2S, AV, A2V$ are just $A$. +---- $V0$ is just $V$; the second argument is treated as adverb. +-- +-- V0 : Type ; +-- AS, A2S, AV, A2V : Type ; +-- +----. +----2 Definitions of paradigms +---- +---- The definitions should not bother the user of the API. So they are +---- hidden from the document. +-- +-- Gender = MorphoHin.Gender ; +-- Number = MorphoHin.Number ; +-- Case = MorphoHin.Case ; +-- human = Masc ; +-- nonhuman = Neutr ; +-- masculine = Masc ; +-- feminine = Fem ; +-- singular = Sg ; +-- plural = Pl ; +-- nominative = Nom ; +-- genitive = Gen ; +-- +-- Preposition : Type = Str ; -- obsolete +-- +-- regN = \ray -> +-- let rays = add_s ray +-- in +-- mk2N ray rays ; +-- +-- +-- add_s : Str -> Str = \w -> case w of { +-- _ + ("io" | "oo") => w + "s" ; -- radio, bamboo +-- _ + ("s" | "z" | "x" | "sh" | "ch" | "o") => w + "es" ; -- bus, hero +-- _ + ("a" | "o" | "u" | "e") + "y" => w + "s" ; -- boy +-- x + "y" => x + "ies" ; -- fly +-- _ => w + "s" -- car +-- } ; +-- +-- mk2N = \man,men -> +-- let mens = case last men of { +-- "s" => men + "'" ; +-- _ => men + "'s" +-- } +-- in +-- mk4N man men (man + "'s") mens ; +-- +-- mk4N = \man,men,man's,men's -> +-- mkNoun man man's men men's ** {g = Neutr ; lock_N = <>} ; +-- +-- genderN g man = {s = man.s ; g = g ; lock_N = <>} ; +-- +-- compoundN s n = {s = \\x,y => s ++ n.s ! x ! y ; g=n.g ; lock_N = <>} ; +-- +-- mkPN = overload { +-- mkPN : Str -> PN = regPN ; +-- mkPN : N -> PN = nounPN +-- } ; +-- +-- +-- mkN2 = \n,p -> n ** {lock_N2 = <> ; c2 = p.s} ; +-- regN2 n = mkN2 (regN n) (mkPrep "of") ; +-- mkN3 = \n,p,q -> n ** {lock_N3 = <> ; c2 = p.s ; c3 = q.s} ; +-- +----3 Relational common noun phrases +---- +---- In some cases, you may want to make a complex $CN$ into a +---- relational noun (e.g. "the old town hall of"). +-- +-- cnN2 : CN -> Prep -> N2 ; +-- cnN3 : CN -> Prep -> Prep -> N3 ; +-- +---- This is obsolete. +-- cnN2 = \n,p -> n ** {lock_N2 = <> ; c2 = p.s} ; +-- cnN3 = \n,p,q -> n ** {lock_N3 = <> ; c2 = p.s ; c3 = q.s} ; +-- +-- regPN n = regGenPN n human ; +-- regGenPN n g = nameReg n g ** {g = g ; lock_PN = <>} ; +-- nounPN n = {s = n.s ! singular ; g = n.g ; lock_PN = <>} ; +-- +-- mk2A a b = mkAdjective a a a b ** {lock_A = <>} ; +-- regA a = regADeg a ** {lock_A = <>} ; +-- +-- mkA2 a p = a ** {c2 = p.s ; lock_A2 = <>} ; +-- +-- ADeg = A ; ---- +-- +-- mkADeg a b c d = mkAdjective a b c d ** {lock_A = <>} ; +-- +-- regADeg happy = +-- let +-- happ = init happy ; +-- y = last happy ; +-- happie = case y of { +-- "y" => happ + "ie" ; +-- "e" => happy ; +-- _ => happy + "e" +-- } ; +-- happily : Str = case happy of { +-- _ + "y" => happ + "ily" ; +-- _ + "ll" => happy + "y" ; +-- _ => happy + "ly" +-- } ; +-- in mkADeg happy (happie + "r") (happie + "st") happily ; +-- +-- duplADeg fat = +-- mkADeg fat +-- (fat + last fat + "er") (fat + last fat + "est") (fat + "ly") ; +-- +-- compoundADeg a = +-- let ad = (a.s ! AAdj Posit) +-- in mkADeg ad ("more" ++ ad) ("most" ++ ad) (a.s ! AAdv) ; +-- +-- adegA a = a ; +-- +-- mkAdv x = ss x ** {lock_Adv = <>} ; +-- mkAdV x = ss x ** {lock_AdV = <>} ; +-- mkAdA x = ss x ** {lock_AdA = <>} ; +-- +-- mkPrep p = ss p ** {lock_Prep = <>} ; +-- noPrep = mkPrep [] ; +-- +-- mk5V a b c d e = mkVerb a b c d e ** {s1 = [] ; lock_V = <>} ; +-- +-- regV cry = +-- let +-- cr = init cry ; +-- y = last cry ; +-- cries = (regN cry).s ! Pl ! Nom ; -- ! +-- crie = init cries ; +-- cried = case last crie of { +-- "e" => crie + "d" ; +-- _ => crie + "ed" +-- } ; +-- crying = case y of { +-- "e" => case last cr of { +-- "e" => cry + "ing" ; +-- _ => cr + "ing" +-- } ; +-- _ => cry + "ing" +-- } +-- in mk5V cry cries cried cried crying ; +-- +-- reg2V fit fitted = +-- let fitt = Predef.tk 2 fitted ; +-- in mk5V fit (fit + "s") (fitt + "ed") (fitt + "ed") (fitt + "ing") ; +-- +-- regDuplV fit = +-- case last fit of { +-- ("a" | "e" | "i" | "o" | "u" | "y") => +-- Predef.error (["final duplication makes no sense for"] ++ fit) ; +-- t => +-- let fitt = fit + t in +-- mk5V fit (fit + "s") (fitt + "ed") (fitt + "ed") (fitt + "ing") +-- } ; +-- +-- irregV x y z = let reg = (regV x).s in +-- mk5V x (reg ! VPres) y z (reg ! VPresPart) ** {s1 = [] ; lock_V = <>} ; +-- +-- irreg4V x y z w = let reg = (regV x).s in +-- mk5V x (reg ! VPres) y z w ** {s1 = [] ; lock_V = <>} ; +-- +-- irregDuplV fit y z = +-- let +-- fitting = (regDuplV fit).s ! VPresPart +-- in +-- mk5V fit (fit + "s") y z fitting ; +-- +-- partV v p = verbPart v p ** {lock_V = <>} ; +-- reflV v = {s = v.s ; part = v.part ; lock_V = v.lock_V ; isRefl = True} ; +-- +-- prepV2 v p = v ** {s = v.s ; s1 = v.s1 ; c2 = p.s ; lock_V2 = <>} ; +-- dirV2 v = prepV2 v noPrep ; +-- +-- mkV3 v p q = v ** {s = v.s ; s1 = v.s1 ; c2 = p.s ; c3 = q.s ; lock_V3 = <>} ; +-- dirV3 v p = mkV3 v noPrep p ; +-- dirdirV3 v = dirV3 v noPrep ; +-- +-- mkVS v = v ** {lock_VS = <>} ; +-- mkVV v = { +-- s = table {VVF vf => v.s ! vf ; _ => variants {}} ; +-- isAux = False ; lock_VV = <> +-- } ; +-- mkVQ v = v ** {lock_VQ = <>} ; +-- +-- V0 : Type = V ; +---- V2S, V2V, V2Q : Type = V2 ; +-- AS, A2S, AV : Type = A ; +-- A2V : Type = A2 ; +-- +-- mkV0 v = v ** {lock_V = <>} ; +-- mkV2S v p = prepV2 v p ** {lock_V2S = <>} ; +-- mkV2V v p t = prepV2 v p ** {isAux = False ; lock_V2V = <>} ; +-- mkVA v = v ** {lock_VA = <>} ; +-- mkV2A v p = prepV2 v p ** {lock_V2A = <>} ; +-- mkV2Q v p = prepV2 v p ** {lock_V2Q = <>} ; +-- +-- mkAS v = v ** {lock_A = <>} ; +-- mkA2S v p = mkA2 v p ** {lock_A = <>} ; +-- mkAV v = v ** {lock_A = <>} ; +-- mkA2V v p = mkA2 v p ** {lock_A2 = <>} ; +-- +-- +---- pre-overload API and overload definitions +-- +-- mk4N : (man,men,man's,men's : Str) -> N ; +-- regN : Str -> N ; +-- mk2N : (man,men : Str) -> N ; +-- genderN : Gender -> N -> N ; +-- compoundN : Str -> N -> N ; +-- +-- mkN = overload { +-- mkN : (man,men,man's,men's : Str) -> N = mk4N ; +-- mkN : Str -> N = regN ; +-- mkN : (man,men : Str) -> N = mk2N ; +-- mkN : Gender -> N -> N = genderN ; +-- mkN : Str -> N -> N = compoundN +-- } ; +-- +-- +-- mk2A : (free,freely : Str) -> A ; +-- regA : Str -> A ; +-- +-- mkA = overload { +-- mkA : Str -> A = regA ; +-- mkA : (fat,fatter : Str) -> A = \fat,fatter -> +-- mkAdjective fat fatter (init fatter + "st") (fat + "ly") ** {lock_A = <>} ; +-- mkA : (good,better,best,well : Str) -> A = \a,b,c,d -> +-- mkAdjective a b c d ** {lock_A = <>} +-- } ; +-- +-- compoundA = compoundADeg ; +-- +-- +-- mk5V : (go, goes, went, gone, going : Str) -> V ; +-- regV : (cry : Str) -> V ; +-- reg2V : (stop, stopped : Str) -> V; +-- irregV : (drink, drank, drunk : Str) -> V ; +-- irreg4V : (run, ran, run, running : Str) -> V ; +-- +-- -- Use reg2V instead +-- regDuplV : Str -> V ; +-- -- Use irreg4V instead +-- irregDuplV : (get, got, gotten : Str) -> V ; +-- +-- mkV = overload { +-- mkV : (cry : Str) -> V = regV ; +-- mkV : (stop, stopped : Str) -> V = reg2V ; +-- mkV : (drink, drank, drunk : Str) -> V = irregV ; +-- mkV : (run, ran, run, running : Str) -> V = irreg4V ; +-- mkV : (go, goes, went, gone, going : Str) -> V = mk5V +-- }; +-- +-- prepV2 : V -> Prep -> V2 ; +-- dirV2 : V -> V2 ; +-- +-- mkV2 = overload { +-- mkV2 : V -> Prep -> V2 = prepV2; +-- mkV2 : V -> V2 = dirV2 +-- }; +-- +-- +------ obsolete +-- +---- Comparison adjectives may two more forms. +-- +-- ADeg : Type ; +-- +-- mkADeg : (good,better,best,well : Str) -> ADeg ; +-- +---- The regular pattern recognizes two common variations: +---- "-e" ("rude" - "ruder" - "rudest") and +---- "-y" ("happy - happier - happiest - happily") +-- +-- regADeg : Str -> ADeg ; -- long, longer, longest +-- +---- However, the duplication of the final consonant is nor predicted, +---- but a separate pattern is used: +-- +-- duplADeg : Str -> ADeg ; -- fat, fatter, fattest +-- +---- If comparison is formed by "more", "most", as in general for +---- long adjective, the following pattern is used: +-- +-- compoundADeg : A -> ADeg ; -- -/more/most ridiculous +-- +---- From a given $ADeg$, it is possible to get back to $A$. +-- +-- adegA : ADeg -> A ; +-- +-- +-- regPN : Str -> PN ; +-- regGenPN : Str -> Gender -> PN ; -- John, John's +-- +---- Sometimes you can reuse a common noun as a proper name, e.g. "Bank". +-- +-- nounPN : N -> PN ; +-- +-- +-- +} diff --git a/next-lib/src/hindi/PhraseHin.gf b/next-lib/src/hindi/PhraseHin.gf new file mode 100644 index 000000000..16a492029 --- /dev/null +++ b/next-lib/src/hindi/PhraseHin.gf @@ -0,0 +1,24 @@ +concrete PhraseHin of Phrase = CatHin ** open Prelude, ResHin in { +-- +-- lin +-- PhrUtt pconj utt voc = {s = pconj.s ++ utt.s ++ voc.s} ; +-- +-- UttS s = s ; +-- UttQS qs = {s = qs.s ! QDir} ; +-- UttImpSg pol imp = {s = pol.s ++ imp.s ! contrNeg True pol.p ! ImpF Sg False} ; +-- UttImpPl pol imp = {s = pol.s ++ imp.s ! contrNeg True pol.p ! ImpF Pl False} ; +-- UttImpPol pol imp = {s = pol.s ++ imp.s ! contrNeg True pol.p ! ImpF Sg True} ; +-- +-- UttIP ip = {s = ip.s ! Nom} ; --- Acc also +-- UttIAdv iadv = iadv ; +-- UttNP np = {s = np.s ! Nom} ; +-- UttVP vp = {s = infVP False vp (agrP3 Sg)} ; +-- UttAdv adv = adv ; +-- +-- NoPConj = {s = []} ; +-- PConjConj conj = {s = conj.s2} ; --- +-- +-- NoVoc = {s = []} ; +-- VocNP np = {s = "," ++ np.s ! Nom} ; +-- +} diff --git a/next-lib/src/hindi/QuestionHin.gf b/next-lib/src/hindi/QuestionHin.gf new file mode 100644 index 000000000..e35c2c3d3 --- /dev/null +++ b/next-lib/src/hindi/QuestionHin.gf @@ -0,0 +1,55 @@ +concrete QuestionHin of Question = CatHin ** open ResHin, Prelude in { +-- +-- flags optimize=all_subs ; +-- +-- lin +-- +-- QuestCl cl = { +-- s = \\t,a,p => +-- let cls = cl.s ! t ! a ! p +-- in table { +-- QDir => cls ! OQuest ; +-- QIndir => "if" ++ cls ! ODir +-- } ---- "whether" in ExtHin +-- } ; +-- +-- QuestVP qp vp = +-- let cl = mkClause (qp.s ! Nom) (agrP3 qp.n) vp +-- in {s = \\t,a,b,_ => cl.s ! t ! a ! b ! ODir} ; +-- +-- QuestSlash ip slash = +-- mkQuestion (ss (slash.c2 ++ ip.s ! Acc)) slash ; +-- --- stranding in ExratHin +-- +-- QuestIAdv iadv cl = mkQuestion iadv cl ; +-- +-- QuestIComp icomp np = +-- mkQuestion icomp (mkClause (np.s ! Nom) np.a (predAux auxBe)) ; +-- +-- +-- PrepIP p ip = {s = p.s ++ ip.s ! Acc} ; +-- +-- AdvIP ip adv = { +-- s = \\c => ip.s ! c ++ adv.s ; +-- n = ip.n +-- } ; +-- +-- IdetCN idet cn = { +-- s = \\c => idet.s ++ cn.s ! idet.n ! c ; +-- n = idet.n +-- } ; +-- +-- IdetIP idet = { +-- s = \\c => idet.s ; +-- n = idet.n +-- } ; +-- +-- IdetQuant idet num = { +-- s = idet.s ! num.n ++ num.s ; +-- n = num.n +-- } ; +-- +-- CompIAdv a = a ; +-- CompIP p = ss (p.s ! Nom) ; +-- +} diff --git a/next-lib/src/hindi/RelativeHin.gf b/next-lib/src/hindi/RelativeHin.gf new file mode 100644 index 000000000..c4f3ad5d4 --- /dev/null +++ b/next-lib/src/hindi/RelativeHin.gf @@ -0,0 +1,54 @@ +concrete RelativeHin of Relative = CatHin ** open ResHin in { +-- +-- flags optimize=all_subs ; +-- +-- lin +-- +-- RelCl cl = { +-- s = \\t,a,p,_ => "such" ++ "that" ++ cl.s ! t ! a ! p ! ODir ; +-- c = Nom +-- } ; +-- +-- RelVP rp vp = { +-- s = \\t,ant,b,ag => +-- let +-- agr = case rp.a of { +-- RNoAg => ag ; +-- RAg a => a +-- } ; +-- cl = mkClause (rp.s ! RC (fromAgr agr).g Nom) agr vp +-- in +-- cl.s ! t ! ant ! b ! ODir ; +-- c = Nom +-- } ; +-- +---- Pied piping: "at which we are looking". Stranding and empty +---- relative are defined in $ExtraHin.gf$ ("that we are looking at", +---- "we are looking at"). +-- +-- RelSlash rp slash = { +-- s = \\t,a,p,agr => +-- slash.c2 ++ rp.s ! RPrep (fromAgr agr).g ++ slash.s ! t ! a ! p ! ODir ; +-- c = Acc +-- } ; +-- +-- FunRP p np rp = { +-- s = \\c => np.s ! Acc ++ p.s ++ rp.s ! RPrep (fromAgr np.a).g ; +-- a = RAg np.a +-- } ; +-- +-- IdRP = +-- let varr : Str -> Str = \x -> variants {x ; "that"} --- for bwc +-- in { +-- s = table { +-- RC _ Gen => "whose" ; +-- RC Neutr _ => varr "which" ; +-- RC _ Acc => varr "whom" ; +-- RC _ Nom => varr "who" ; +-- RPrep Neutr => "which" ; +-- RPrep _ => "whom" +-- } ; +-- a = RNoAg +-- } ; +-- +} diff --git a/next-lib/src/hindi/ResHin.gf b/next-lib/src/hindi/ResHin.gf new file mode 100644 index 000000000..17ef91235 --- /dev/null +++ b/next-lib/src/hindi/ResHin.gf @@ -0,0 +1,300 @@ +--# -path=.:../abstract:../common:../../prelude +-- +--1 Hindi auxiliary operations. +-- +-- This module contains operations that are needed to make the +-- resource syntax work. + +resource ResHin = ParamX ** open Prelude in { + + flags optimize=all ; + + param + Case = Dir | Obl | Voc ; + Gender = Masc | Fem ; + + oper + Noun = {s : Number => Case => Str ; g : Gender} ; + + mkNoun : (x1,_,_,_,_,x6 : Str) -> Gender -> Noun = + \sd,so,sv,pd,po,pv,g -> { + s = table Number [table Case [sd;so;sv] ; table Case [pd;po;pv]] ; + g = g + } ; + + reggNoun : Str -> Gender -> Noun = \s,g -> case of { + <-(_ + ("A" | "I")), Fem> => + mkNoun s s s (s + "eM") (s + "oM") (s + "o") Fem ; + _ => regNoun s ** {g = g} + } ; + + regNoun : Str -> Noun = \s -> case s of { + x + "iyA" => + mkNoun s s s (x + "iyAM") (x + "iyoN") (x + "iyo") Fem ; + x + "A" => + mkNoun s (x + "e") (x + "e") (x + "e") (x + "oN") (x + "o") Masc ; + x + "I" => + mkNoun s s s (x + "iyAM") (x + "iyoN") (x + "iyo") Fem ; + _ => + mkNoun s s s s (s + "oN") (s + "o") Masc + } ; + + + Adjective = {s : Gender => Number => Case => Str} ; + + mkAdjective : (x1,x2,x3 : Str) -> Adjective = \smd,sm,f -> { + s = \\g,n,c => case of { + => smd ; + => sm ; + _ => f + } + } ; + + regAdjective : Str -> Adjective = \s -> case s of { + acch + "A" => mkAdjective s (acch + "e") (acch + "I") ; + _ => mkAdjective s s s + } ; + + param + VForm = + VInf + | VStem + | VImpf Gender Number + | VPerf Gender Number + | VSubj Number Person + | VFut Number Person Gender + | VAbs + | VReq + | VImp + | VReqFut + ; + + oper + Verb = {s : VForm => Str} ; + + mkVerb : (x1,_,_,_,_,_,_,_,_,_,_,_,_,_,x15 : Str) -> Verb = + \inf,stem,ims,imp,ifs,ifp,pms,pmp,pfs,pfp,ss1,ss2,sp2,sp3,r -> { + s = + let ga : Number -> Gender -> Str = \n,g -> + (regAdjective "gA").s ! g ! n ! Dir + in table { + VInf => inf ; + VStem => stem ; + VImpf Masc Sg => ims ; + VImpf Masc Pl => imp ; + VImpf Fem Sg => ifs ; + VImpf Fem Pl => ifp ; + VPerf Masc Sg => pms ; + VPerf Masc Pl => pmp ; + VPerf Fem Sg => pfs ; + VPerf Fem Pl => pfp ; + VSubj Sg P1 => ss1 ; + VSubj Sg _ => ss2 ; + VSubj Pl P2 => sp2 ; + VSubj Pl _ => sp3 ; + VFut Sg P1 g => ss1 + ga Sg g ; + VFut Sg _ g => ss2 + ga Sg g ; + VFut Pl P2 g => sp2 + ga Pl g ; + VFut Pl _ g => sp3 + ga Pl g ; + VAbs => stem + "kar" ; --- ke + VReq => r ; + VImp => sp2 ; + VReqFut => stem + "ie-gA" + } + } ; + + regVerb : Str -> Verb = \cal -> + let caly : Str = case cal of { + _ + ("A" | "e") => cal + "y" ; + c + "U" => c + "uy" ; + c + "I" => c + "iy" ; + _ => cal + } + in + mkVerb + (cal + "nA") cal + (cal + "tA") (cal + "te") (cal + "tI") (cal + "tI") + (caly + "A") (caly + "e") (caly + "I") (caly + "IN") + (caly + "UM") (caly + "e") (caly + "o") (caly + "eN") + (caly + "ie-") ; + + param + CTense = CPresent | CPast | CFuture ; + oper + copula : CTense -> Number -> Person -> Gender -> Str = \t,n,p,g -> + case of { + => "hUM" ; + => "hE" ; + => "hE" ; + => "hEN" ; + => "ho" ; + => "hEN" ; + => "TA" ; + => "TI" ; + => "Te" ; + => "TIN" ; + => "hUNgA" ; + => "hUNgI" ; + => "hogA" ; + => "hogI" ; + => "hoge" ; + => "hoNge" ; + => "hogi:" ; + => "hoNgi:" + } ; + + param + PronCase = PC Case | PObj | PPoss ; + oper + personalPronoun : Person -> Number -> {s : PronCase => Str} = \p,n -> + case of { + => {s = table PronCase ["mEN" ; "muJ" ; "muJ" ; "muJe" ; "merA"]} ; + => {s = table PronCase ["ham" ; "ham" ; "ham" ; "hameN" ; "hamArA"]} ; + => {s = table PronCase ["tU" ; "tuJ" ; "tuJ" ; "tuJe" ; "terA"]} ; + => {s = table PronCase ["tum" ; "tum" ; "tum" ; "tum" ; "tumhArA"]} ; + => {s = table PronCase ["vah" ; "u-s" ; "u-s" ; "u-se" ; "u-skA"]} ; + => {s = table PronCase ["ve" ; "u-n" ; "u-n" ; "u-nheN" ; "u-nkA"]} + } ; + ---- the third is the vocative - is it really this way? + + -- the Hindi verb phrase + +--- CTense = CPresent | CPast | CFuture ; + + + + param + VPHTense = + VPGenPres -- impf hum nahim "I go" + | VPImpPast -- impf Ta nahim "I went" + | VPContPres -- stem raha hum nahim "I am going" + | VPContPast -- stem raha Ta nahim "I was going" + | VPPerf -- perf na/nahim "I went" + | VPPerfPres -- perf hum na/nahim "I have gone" + | VPPerfPast -- perf Ta na/nahim "I had gone" + | VPSubj -- subj na "I may go" + | VPFut -- fut na/nahim "I shall go" + ; + + VPHForm = + VPTense VPHTense Agr -- 9 * 12 + | VPReq + | VPImp + | VPReqFut + | VPInf + | VPStem + ; + + VType = VIntrans | VTrans | VTransPost ; + + oper + objVType : VType -> NPCase = \vt -> case vt of { + VTrans => NPObj ; + _ => NPC Obl + } ; + + VPH : Type = { + s : Bool => VPHForm => {fin, inf, neg : Str} ; + obj : {s : Str ; a : Agr} ; + subj : VType ; + comp : Agr => Str + } ; + + predV : Verb -> VPH = \verb -> { + s = \\b,vh => + let + na = if_then_Str b [] "na" ; + nahim = if_then_Str b [] "nahIN" ; + in + case vh of { + VPTense VPGenPres (Ag g n p) => + {fin = copula CPresent n p g ; inf = verb.s ! VImpf g n ; neg = nahim} ; + VPTense VPImpPast (Ag g n p) => + {fin = copula CPast n p g ; inf = verb.s ! VImpf g n ; neg = nahim} ; + VPTense VPContPres (Ag g n p) => + {fin = copula CPresent n p g ; + inf = verb.s ! VStem ++ raha g n ; neg = nahim} ; + VPTense VPContPast (Ag g n p) => + {fin = copula CPast n p g ; + inf = verb.s ! VStem ++ raha g n ; neg = nahim} ; + VPTense VPPerf (Ag g n _) => + {fin = verb.s ! VPerf g n ; inf = [] ; neg = nahim} ; + VPTense VPPerfPres (Ag g n p) => + {fin = copula CPresent n p g ; inf = verb.s ! VPerf g n ; neg = nahim} ; + VPTense VPPerfPast (Ag g n p) => + {fin = copula CPast n p g ; inf = verb.s ! VPerf g n ; neg = nahim} ; + VPTense VPSubj (Ag _ n p) => {fin = verb.s ! VSubj n p ; inf = [] ; neg = na} ; + VPTense VPFut (Ag g n p) => {fin = verb.s ! VFut n p g ; inf = [] ; neg = na} ; + VPInf => {fin = verb.s ! VStem ; inf = [] ; neg = na} ; + _ => {fin = verb.s ! VStem ; inf = [] ; neg = na} ---- + } ; + obj = {s = [] ; a = defaultAgr} ; + subj = VIntrans ; + comp = \\_ => [] + } ; + + raha : Gender -> Number -> Str = \g,n -> + (regAdjective "rahA").s ! g ! n ! Dir ; + + VPHSlash = VPH ** {c2 : Compl} ; + + Clause : Type = {s : VPHTense => Bool => Str} ; + + Compl : Type = {s : Str ; c : VType} ; + + insertObject : NP -> VPHSlash -> VPH = \np,vps -> { + s = vps.s ; + obj = {s = vps.obj.s ++ np.s ! objVType vps.c2.c ++ vps.c2.s ; a = np.a} ; + subj = vps.c2.c ; + comp = vps.comp + } ; + + param + Agr = Ag Gender Number Person ; + NPCase = NPC Case | NPObj | NPErg ; + + oper + agrP3 : Gender -> Number -> Agr = \g,n -> Ag g n P3 ; + + defaultAgr : Agr = agrP3 Masc Sg ; + + npcase2case : NPCase -> Case = \npc -> case npc of { + NPC c => c ; + NPObj => Obl ; + NPErg => Obl + } ; + + np2pronCase : NPCase -> PronCase = \np -> case np of { + NPC c => PC c ; + NPObj => PObj ; + NPErg => PC Obl + } ; + + toNP : (Case => Str) -> NPCase -> Str = \pn, npc -> case npc of { + NPC c => pn ! c ; + NPObj => pn ! Obl ; + NPErg => pn ! Obl ++ "ne" + } ; + + NP : Type = {s : NPCase => Str ; a : Agr} ; + + mkClause : NP -> VPH -> Clause = \np,vp -> { + s = \\vt,b => + let + subjagr : NPCase * Agr = case vt of { + VPPerf => case vp.subj of { + VTrans => ; + VTransPost => ; + _ => + } ; + _ => + } ; + subj = subjagr.p1 ; + agr = subjagr.p2 ; + vps = vp.s ! b ! VPTense vt agr ; + in + np.s ! subj ++ vp.obj.s ++ vp.comp ! np.a ++ vps.neg ++ vps.inf ++ vps.fin + } ; + + +} diff --git a/next-lib/src/hindi/SentenceHin.gf b/next-lib/src/hindi/SentenceHin.gf new file mode 100644 index 000000000..d4930c9af --- /dev/null +++ b/next-lib/src/hindi/SentenceHin.gf @@ -0,0 +1,66 @@ +concrete SentenceHin of Sentence = CatHin ** open Prelude, ResHin in { + + flags optimize=all_subs ; + + lin + + PredVP np vp = mkClause np vp ; + +-- PredSCVP sc vp = mkClause sc.s (agrP3 Sg) vp ; +-- +-- ImpVP vp = { +-- s = \\pol,n => +-- let +-- agr = AgP2 (numImp n) ; +-- verb = infVP True vp agr ; +-- dont = case pol of { +-- CNeg True => "don't" ; +-- CNeg False => "do" ++ "not" ; +-- _ => [] +-- } +-- in +-- dont ++ verb +-- } ; +-- +-- SlashVP np vp = +-- mkClause (np.s ! Nom) np.a vp ** {c2 = vp.c2} ; +-- +-- AdvSlash slash adv = { +-- s = \\t,a,b,o => slash.s ! t ! a ! b ! o ++ adv.s ; +-- c2 = slash.c2 +-- } ; +-- +-- SlashPrep cl prep = cl ** {c2 = prep.s} ; +-- +-- SlashVS np vs slash = +-- mkClause (np.s ! Nom) np.a +-- (insertObj (\\_ => conjThat ++ slash.s) (predV vs)) ** +-- {c2 = slash.c2} ; +-- +-- EmbedS s = {s = conjThat ++ s.s} ; +-- EmbedQS qs = {s = qs.s ! QIndir} ; +-- EmbedVP vp = {s = infVP False vp (agrP3 Sg)} ; --- agr +-- +-- UseCl t a p cl = { +-- s = t.s ++ a.s ++ p.s ++ cl.s ! t.t ! a.a ! ctr p.p ! ODir +-- } ; +-- UseQCl t a p cl = { +-- s = \\q => t.s ++ a.s ++ p.s ++ cl.s ! t.t ! a.a ! ctr p.p ! q +-- } ; +-- UseRCl t a p cl = { +-- s = \\r => t.s ++ a.s ++ p.s ++ cl.s ! t.t ! a.a ! ctr p.p ! r ; +-- c = cl.c +-- } ; +-- UseSlash t a p cl = { +-- s = t.s ++ a.s ++ p.s ++ cl.s ! t.t ! a.a ! ctr p.p ! ODir ; +-- c2 = cl.c2 +-- } ; +-- +-- AdvS a s = {s = a.s ++ "," ++ s.s} ; +-- +-- RelS s r = {s = s.s ++ "," ++ r.s ! agrP3 Sg} ; +-- +-- oper +-- ctr = contrNeg True ; -- contracted negations +-- +} diff --git a/next-lib/src/hindi/StructuralHin.gf b/next-lib/src/hindi/StructuralHin.gf new file mode 100644 index 000000000..9ad5d8e53 --- /dev/null +++ b/next-lib/src/hindi/StructuralHin.gf @@ -0,0 +1,132 @@ +concrete StructuralHin of Structural = CatHin ** + open MorphoHin, (P = ParadigmsHin), Prelude in { + + flags optimize=all ; + + lin +-- above_Prep = ss "above" ; +-- after_Prep = ss "after" ; +-- all_Predet = ss "all" ; +-- almost_AdA, almost_AdN = ss "almost" ; +-- although_Subj = ss "although" ; +-- always_AdV = ss "always" ; +-- and_Conj = sd2 [] "and" ** {n = Pl} ; +-----b and_Conj = ss "and" ** {n = Pl} ; +-- because_Subj = ss "because" ; +-- before_Prep = ss "before" ; +-- behind_Prep = ss "behind" ; +-- between_Prep = ss "between" ; +-- both7and_DConj = sd2 "both" "and" ** {n = Pl} ; +-- but_PConj = ss "but" ; +-- by8agent_Prep = ss "by" ; +-- by8means_Prep = ss "by" ; +-- can8know_VV, can_VV = { +-- s = table { +-- VVF VInf => ["be able to"] ; +-- VVF VPres => "can" ; +-- VVF VPPart => ["been able to"] ; +-- VVF VPresPart => ["being able to"] ; +-- VVF VPast => "could" ; --# notpresent +-- VVPastNeg => "couldn't" ; --# notpresent +-- VVPresNeg => "can't" +-- } ; +-- isAux = True +-- } ; +-- during_Prep = ss "during" ; +-- either7or_DConj = sd2 "either" "or" ** {n = Sg} ; +-- everybody_NP = regNP "everybody" Sg ; +-- every_Det = mkDeterminer Sg "every" ; +-- everything_NP = regNP "everything" Sg ; +-- everywhere_Adv = ss "everywhere" ; +-- few_Det = mkDeterminer Pl "few" ; +----- first_Ord = ss "first" ; DEPRECATED +-- for_Prep = ss "for" ; +-- from_Prep = ss "from" ; + he_Pron = personalPronoun P3 Sg ** {a = Ag Masc Sg P3} ; +-- here_Adv = ss "here" ; +-- here7to_Adv = ss ["to here"] ; +-- here7from_Adv = ss ["from here"] ; +-- how_IAdv = ss "how" ; +-- how8many_IDet = mkDeterminer Pl ["how many"] ; +-- if_Subj = ss "if" ; +-- in8front_Prep = ss ["in front of"] ; + i_Pron = personalPronoun P1 Sg ** {a = Ag Masc Sg P1} ; + in_Prep = ss "meN" ; +-- it_Pron = mkNP "it" "it" "its" Sg P3 Neutr ; +-- less_CAdv = ss "less" ; +-- many_Det = mkDeterminer Pl "many" ; +-- more_CAdv = ss "more" ; +-- most_Predet = ss "most" ; +-- much_Det = mkDeterminer Sg "much" ; +-- must_VV = { +-- s = table { +-- VVF VInf => ["have to"] ; +-- VVF VPres => "must" ; +-- VVF VPPart => ["had to"] ; +-- VVF VPresPart => ["having to"] ; +-- VVF VPast => ["had to"] ; --# notpresent +-- VVPastNeg => ["hadn't to"] ; --# notpresent +-- VVPresNeg => "mustn't" +-- } ; +-- isAux = True +-- } ; +-----b no_Phr = ss "no" ; +-- no_Utt = ss "no" ; +-- on_Prep = ss "on" ; +------ one_Quant = mkDeterminer Sg "one" ; -- DEPRECATED +-- only_Predet = ss "only" ; +-- or_Conj = sd2 [] "or" ** {n = Sg} ; +-- otherwise_PConj = ss "otherwise" ; +-- part_Prep = ss "of" ; +-- please_Voc = ss "please" ; +-- possess_Prep = ss "of" ; +-- quite_Adv = ss "quite" ; +-- she_Pron = mkNP "she" "her" "her" Sg P3 Fem ; +-- so_AdA = ss "so" ; +-- somebody_NP = regNP "somebody" Sg ; +-- someSg_Det = mkDeterminer Sg "some" ; +-- somePl_Det = mkDeterminer Pl "some" ; +-- something_NP = regNP "something" Sg ; +-- somewhere_Adv = ss "somewhere" ; +-- that_Quant = mkQuant "that" "those" ; +-- there_Adv = ss "there" ; +-- there7to_Adv = ss "there" ; +-- there7from_Adv = ss ["from there"] ; +-- therefore_PConj = ss "therefore" ; +-- they_Pron = mkNP "they" "them" "their" Pl P3 Masc ; ---- +-- this_Quant = mkQuant "this" "these" ; +-- through_Prep = ss "through" ; +-- too_AdA = ss "too" ; +-- to_Prep = ss "to" ; +-- under_Prep = ss "under" ; +-- very_AdA = ss "very" ; +-- want_VV = P.mkVV (P.regV "want") ; + we_Pron = personalPronoun P1 Pl ** {a = Ag Masc Pl P1} ; + +-- whatPl_IP = mkIP "what" "what" "what's" Sg ; +-- whatSg_IP = mkIP "what" "what" "what's" Sg ; +-- when_IAdv = ss "when" ; +-- when_Subj = ss "when" ; +-- where_IAdv = ss "where" ; +-- which_IQuant = {s = \\_ => "which"} ; +-----b whichPl_IDet = mkDeterminer Pl ["which"] ; +-----b whichSg_IDet = mkDeterminer Sg ["which"] ; +-- whoSg_IP = mkIP "who" "whom" "whose" Sg ; +-- whoPl_IP = mkIP "who" "whom" "whose" Pl ; +-- why_IAdv = ss "why" ; +-- without_Prep = ss "without" ; +-- with_Prep = ss "with" ; +-----b yes_Phr = ss "yes" ; +-- yes_Utt = ss "yes" ; +-- youSg_Pron = mkNP "you" "you" "your" Sg P2 Masc ; +-- youPl_Pron = mkNP "you" "you" "your" Pl P2 Masc ; +-- youPol_Pron = mkNP "you" "you" "your" Sg P2 Masc ; +-- +-- +--oper +-- mkQuant : Str -> Str -> {s : Number => Str} = \x,y -> { +-- s = table Number [x ; y] +-- } ; +-- +} +-- diff --git a/next-lib/src/hindi/VerbHin.gf b/next-lib/src/hindi/VerbHin.gf new file mode 100644 index 000000000..c2abf824d --- /dev/null +++ b/next-lib/src/hindi/VerbHin.gf @@ -0,0 +1,50 @@ +concrete VerbHin of Verb = CatHin ** open ResHin in { + + flags optimize=all_subs ; + + lin + UseV = predV ; + + SlashV2a v = predV v ** {c2 = v.c2} ; + +-- Slash2V3 v np = +-- insertObjc (\\_ => v.c2 ++ np.s ! Acc) (predV v ** {c2 = v.c3}) ; +-- Slash3V3 v np = +-- insertObjc (\\_ => v.c3 ++ np.s ! Acc) (predVc v) ; ---- +-- +-- ComplVV v vp = insertObj (\\a => infVP v.isAux vp a) (predVV v) ; +-- ComplVS v s = insertObj (\\_ => conjThat ++ s.s) (predV v) ; +-- ComplVQ v q = insertObj (\\_ => q.s ! QIndir) (predV v) ; +-- ComplVA v ap = insertObj (ap.s) (predV v) ; +-- +-- SlashV2V v vp = insertObjc (\\a => infVP v.isAux vp a) (predVc v) ; +-- SlashV2S v s = insertObjc (\\_ => conjThat ++ s.s) (predVc v) ; +-- SlashV2Q v q = insertObjc (\\_ => q.s ! QIndir) (predVc v) ; +-- SlashV2A v ap = insertObjc (\\a => ap.s ! a) (predVc v) ; ---- + + ComplSlash vp np = insertObject np vp ; + +-- SlashVV vv vp = +-- insertObj (\\a => infVP vv.isAux vp a) (predVV vv) ** +-- {c2 = vp.c2} ; +-- SlashV2VNP vv np vp = +-- insertObjPre (\\_ => vp.c2 ++ np.s ! Acc) +-- (insertObjc (\\a => infVP vv.isAux vp a) (predVc vv)) ** +-- {c2 = vp.c2} ; +-- +-- UseComp comp = insertComplement comp.s (predAux auxBe) ; +-- +-- AdvVP vp adv = insertObj (\\_ => adv.s) vp ; +-- +-- AdVVP adv vp = insertAdV adv.s vp ; +-- +-- ReflVP v = insertObjPre (\\a => v.c2 ++ reflPron ! a) v ; +-- +-- PassV2 v = insertObj (\\_ => v.s ! VPPart) (predAux auxBe) ; +-- +-- +-- CompAP ap = ap ; +-- CompNP np = {s = \\_ => np.s ! Acc} ; +-- CompAdv a = {s = \\_ => a.s} ; +-- +} diff --git a/next-lib/src/thai/AdjectiveTha.gf b/next-lib/src/thai/AdjectiveTha.gf new file mode 100644 index 000000000..68e8acba9 --- /dev/null +++ b/next-lib/src/thai/AdjectiveTha.gf @@ -0,0 +1,36 @@ +concrete AdjectiveTha of Adjective = CatTha ** open ResTha, Prelude in { + + lin + + PositA a = a ; + +-- ComparA a np = { +-- s = \\_ => a.s ! AAdj Compar ++ "than" ++ np.s ! Nom ; +-- isPre = False +-- } ; +-- +---- $SuperlA$ belongs to determiner syntax in $Noun$. +-- +-- ComplA2 a np = { +-- s = \\_ => a.s ! AAdj Posit ++ a.c2 ++ np.s ! Acc ; +-- isPre = False +-- } ; +-- +-- ReflA2 a = { +-- s = \\ag => a.s ! AAdj Posit ++ a.c2 ++ reflPron ! ag ; +-- isPre = False +-- } ; +-- +-- SentAP ap sc = { +-- s = \\a => ap.s ! a ++ sc.s ; +-- isPre = False +-- } ; +-- +-- AdAP ada ap = { +-- s = \\a => ada.s ++ ap.s ! a ; +-- isPre = ap.isPre +-- } ; +-- +-- UseA2 a = a ; +-- +} diff --git a/next-lib/src/thai/AdverbTha.gf b/next-lib/src/thai/AdverbTha.gf new file mode 100644 index 000000000..876629468 --- /dev/null +++ b/next-lib/src/thai/AdverbTha.gf @@ -0,0 +1,23 @@ +concrete AdverbTha of Adverb = CatTha ** + open ResTha, StringsTha, Prelude in { + + lin + PositAdvAdj a = a ; + +-- ComparAdvAdj cadv a np = { +-- s = cadv.s ++ a.s ! AAdv ++ "than" ++ np.s ! Nom +-- } ; +-- ComparAdvAdjS cadv a s = { +-- s = cadv.s ++ a.s ! AAdv ++ "than" ++ s.s +-- } ; +-- +-- PrepNP prep np = {s = prep.s ++ np.s ! Acc} ; +-- +-- AdAdv = cc2 ; +-- +-- SubjS = cc2 ; +-- AdvSC s = s ; --- this rule give stack overflow in ordinary parsing +-- +-- AdnCAdv cadv = {s = cadv.s ++ "than"} ; +-- +} diff --git a/next-lib/src/thai/AllTha.gf b/next-lib/src/thai/AllTha.gf new file mode 100644 index 000000000..0ca824967 --- /dev/null +++ b/next-lib/src/thai/AllTha.gf @@ -0,0 +1,3 @@ +--# -path=.:../abstract:../common:prelude + +concrete AllTha of AllThaAbs = LangTha ; diff --git a/next-lib/src/thai/AllThaAbs.gf b/next-lib/src/thai/AllThaAbs.gf new file mode 100644 index 000000000..6e1bc90cf --- /dev/null +++ b/next-lib/src/thai/AllThaAbs.gf @@ -0,0 +1,3 @@ +--# -path=.:../abstract:../common:prelude + +abstract AllThaAbs = Lang ; diff --git a/next-lib/src/thai/CatTha.gf b/next-lib/src/thai/CatTha.gf new file mode 100644 index 000000000..d13983f4a --- /dev/null +++ b/next-lib/src/thai/CatTha.gf @@ -0,0 +1,77 @@ +concrete CatTha of Cat = CommonX ** open ResTha, Prelude in { + + flags optimize=all_subs ; + + lincat + +-- Tensed/Untensed + + S = {s : Str} ; + QS = {s : QForm => Str} ; +-- RS = {s : Agr => Str ; c : Case} ; -- c for it clefts +-- +---- Sentence +-- + Cl = {s : Polarity => Str} ; +-- Slash = { +-- s : Tense => Anteriority => CPolarity => Order => Str ; +-- c2 : Str +-- } ; + Imp = {s : Polarity => Str} ; +-- +---- Question +-- + QCl = {s : Polarity => Str} ; +-- IP = {s : Case => Str ; n : Number} ; +-- IComp = {s : Str} ; +-- IDet = {s : Str ; n : Number} ; +-- +---- Relative +-- +-- RCl = {s : Tense => Anteriority => CPolarity => Agr => Str ; c : Case} ; +-- RP = {s : RCase => Str ; a : RAgr} ; +-- +---- Verb +-- + VP = ResTha.VP ; + Comp = ResTha.VP ; +-- +---- Adjective +-- +-- AP = {s : Agr => Str ; isPre : Bool} ; +-- +-- Noun +-- + CN = Noun ; + NP, Pron = SS ; + Det = Determiner ; +-- Predet, Ord = {s : Str} ; + Num, Quant = {s : Str ; hasC : Bool} ; + +-- Numeral + + Numeral = {s : Str} ; + +---- Structural +-- +-- Conj = {s : Str ; n : Number} ; +-- DConj = {s1,s2 : Str ; n : Number} ; +-- Subj = {s : Str} ; +-- Prep = {s : Str} ; +-- +-- Open lexical classes, e.g. Lexicon + + V, VS, VQ, VA = Verb ; + V2, V2A = Verb ** {c2 : Str} ; + V3 = Verb ** {c2, c3 : Str} ; + VV = VVerb ; +-- +-- A = {s : AForm => Str} ; +-- A2 = {s : AForm => Str ; c2 : Str} ; +-- + N = Noun ; +-- N2 = {s : Number => Case => Str} ** {c2 : Str} ; +-- N3 = {s : Number => Case => Str} ** {c2,c3 : Str} ; +-- PN = {s : Case => Str} ; +-- +} diff --git a/next-lib/src/thai/ConjunctionTha.gf b/next-lib/src/thai/ConjunctionTha.gf new file mode 100644 index 000000000..38071f077 --- /dev/null +++ b/next-lib/src/thai/ConjunctionTha.gf @@ -0,0 +1,45 @@ +--concrete ConjunctionTha of Conjunction = +-- CatTha ** open ResTha, Coordination, Prelude in { +-- +-- flags optimize=all_subs ; +-- +-- lin +-- +-- ConjS = conjunctSS ; +-- DConjS = conjunctDistrSS ; +-- +-- ConjAdv = conjunctSS ; +-- DConjAdv = conjunctDistrSS ; +-- +-- ConjNP conj ss = conjunctTable Case conj ss ** { +-- a = {n = conjNumber conj.n ss.a.n ; p = ss.a.p} +-- } ; +-- DConjNP conj ss = conjunctDistrTable Case conj ss ** { +-- a = {n = conjNumber conj.n ss.a.n ; p = ss.a.p} +-- } ; +-- +-- ConjAP conj ss = conjunctTable Agr conj ss ** { +-- isPre = ss.isPre +-- } ; +-- DConjAP conj ss = conjunctDistrTable Agr conj ss ** { +-- isPre = ss.isPre +-- } ; +-- +---- These fun's are generated from the list cat's. +-- +-- BaseS = twoSS ; +-- ConsS = consrSS comma ; +-- BaseAdv = twoSS ; +-- ConsAdv = consrSS comma ; +-- BaseNP x y = twoTable Case x y ** {a = conjAgr x.a y.a} ; +-- ConsNP xs x = consrTable Case comma xs x ** {a = conjAgr xs.a x.a} ; +-- BaseAP x y = twoTable Agr x y ** {isPre = andB x.isPre y.isPre} ; +-- ConsAP xs x = consrTable Agr comma xs x ** {isPre = andB xs.isPre x.isPre} ; +-- +-- lincat +-- [S] = {s1,s2 : Str} ; +-- [Adv] = {s1,s2 : Str} ; +-- [NP] = {s1,s2 : Case => Str ; a : Agr} ; +-- [AP] = {s1,s2 : Agr => Str ; isPre : Bool} ; +-- +--} diff --git a/next-lib/src/thai/ExtraTha.gf b/next-lib/src/thai/ExtraTha.gf new file mode 100644 index 000000000..d987fb7e5 --- /dev/null +++ b/next-lib/src/thai/ExtraTha.gf @@ -0,0 +1,52 @@ +--concrete ExtraTha of ExtraThaAbs = CatTha ** +-- open ResTha, Coordination, Prelude in { +-- +-- lin +-- GenNP np = {s = \\_ => np.s ! Gen} ; +-- ComplBareVS v s = insertObj (\\_ => s.s) (predV v) ; +-- +-- StrandRelSlash rp slash = { +-- s = \\t,a,p,_ => rp.s ! RC Acc ++ slash.s ! t ! a ! p ! ODir ++ slash.c2 ; +-- c = Acc +-- } ; +-- EmptyRelSlash rp slash = { +-- s = \\t,a,p,_ => slash.s ! t ! a ! p ! ODir ++ slash.c2 ; +-- c = Acc +-- } ; +-- +-- StrandQuestSlash ip slash = +-- {s = \\t,a,b,q => +-- (mkQuestion (ss (ip.s ! Acc)) slash).s ! t ! a ! b ! q ++ slash.c2 +-- }; +-- +-- lincat +-- VPI = {s : VPIForm => Agr => Str} ; +-- [VPI] = {s1,s2 : VPIForm => Agr => Str} ; +-- +-- lin +-- BaseVPI = twoTable2 VPIForm Agr ; +-- ConsVPI = consrTable2 VPIForm Agr comma ; +-- +-- MkVPI vp = { +-- s = \\v,a => vp.ad ++ vp.inf ++ vp.s2 ! a +-- } ; +-- ConjVPI = conjunctTable2 VPIForm Agr ; +-- ComplVPIVV vv vpi = +-- insertObj (\\a => (if_then_Str vv.isAux [] "to") ++ vpi.s ! VPIInf ! a) (predVV vv) ; +-- +-- UncNegCl t a cl = {s = t.s ++ a.s ++ cl.s ! t.t ! a.a ! neg ! ODir} ; +-- UncNegQCl t a cl = {s = \\q => t.s ++ a.s ++ cl.s ! t.t ! a.a ! neg !q} ; +-- UncNegRCl t a cl = { +-- s = \\r => t.s ++ a.s ++ cl.s ! t.t ! a.a ! neg ! r ; +-- c = cl.c +-- } ; +-- UncNegImpSg imp = {s = imp.s ! neg ! ImpF Sg False} ; +-- UncNegImpPl imp = {s = imp.s ! neg ! ImpF Pl False} ; +-- +-- CompoundCN a b = {s = \\n,c => a.s ! Sg ! Nom ++ b.s ! n ! c} ; +-- +-- +-- oper +-- neg = CNeg False ; +-- +--} diff --git a/next-lib/src/thai/GrammarTha.gf b/next-lib/src/thai/GrammarTha.gf new file mode 100644 index 000000000..efe05cc45 --- /dev/null +++ b/next-lib/src/thai/GrammarTha.gf @@ -0,0 +1,21 @@ +--# -path=.:../abstract:../common:prelude + +concrete GrammarTha of Grammar = + NounTha, + VerbTha, + AdjectiveTha, + AdverbTha, + NumeralTha, + SentenceTha, + QuestionTha, +-- RelativeTha, +-- ConjunctionTha, + PhraseTha, +-- TextX, + StructuralTha, +-- IdiomTha + ** { + +flags startcat = Phr ; unlexer = text ; lexer = text ; + +} ; diff --git a/next-lib/src/thai/IdiomTha.gf b/next-lib/src/thai/IdiomTha.gf new file mode 100644 index 000000000..40851f3ab --- /dev/null +++ b/next-lib/src/thai/IdiomTha.gf @@ -0,0 +1,30 @@ +--concrete IdiomTha of Idiom = CatTha ** open Prelude, ResTha in { +-- +-- flags optimize=all_subs ; +-- +-- lin +-- ImpersCl vp = mkClause "it" (agrP3 Sg) vp ; +-- GenericCl vp = mkClause "one" (agrP3 Sg) vp ; +-- +-- CleftNP np rs = mkClause "it" (agrP3 Sg) +-- (insertObj (\\_ => rs.s ! np.a) +-- (insertObj (\\_ => np.s ! rs.c) (predAux auxBe))) ; +-- +-- CleftAdv ad s = mkClause "it" (agrP3 Sg) +-- (insertObj (\\_ => conjThat ++ s.s) +-- (insertObj (\\_ => ad.s) (predAux auxBe))) ; +-- +-- ExistNP np = +-- mkClause "there" (agrP3 np.a.n) +-- (insertObj (\\_ => np.s ! Acc) (predAux auxBe)) ; +-- +-- ExistIP ip = +-- mkQuestion (ss (ip.s ! Nom)) +-- (mkClause "there" (agrP3 ip.n) (predAux auxBe)) ; +-- +-- ProgrVP vp = insertObj (\\a => vp.ad ++ vp.prp ++ vp.s2 ! a) (predAux auxBe) ; +-- +-- ImpPl1 vp = {s = "let's" ++ infVP True vp {n = Pl ; p = P1}} ; +-- +--} +-- diff --git a/next-lib/src/thai/IrregTha.gf b/next-lib/src/thai/IrregTha.gf new file mode 100644 index 000000000..27aa23e41 --- /dev/null +++ b/next-lib/src/thai/IrregTha.gf @@ -0,0 +1,181 @@ +----# -path=.:prelude:../abstract:../common +-- +--concrete IrregTha of IrregThaAbs = CatTha ** open ParadigmsTha in { +-- +--flags optimize=values ; +-- +-- lin +-- awake_V = irregV "awake" "awoke" "awoken" ; +-- bear_V = irregV "bear" "bore" "born" ; +-- beat_V = irregV "beat" "beat" "beat" ; +-- become_V = irregV "become" "became" "become" ; +-- begin_V = irregV "begin" "began" "begun" ; +-- bend_V = irregV "bend" "bent" "bent" ; +-- beset_V = irregV "beset" "beset" "beset" ; +-- bet_V = irregDuplV "bet" "bet" "bet" ; +-- bid_V = irregDuplV "bid" (variants {"bid" ; "bade"}) (variants {"bid" ; "bidden"}) ; +-- bind_V = irregV "bind" "bound" "bound" ; +-- bite_V = irregV "bite" "bit" "bitten" ; +-- bleed_V = irregV "bleed" "bled" "bled" ; +-- blow_V = irregV "blow" "blew" "blown" ; +-- break_V = irregV "break" "broke" "broken" ; +-- breed_V = irregV "breed" "bred" "bred" ; +-- bring_V = irregV "bring" "brought" "brought" ; +-- broadcast_V = irregV "broadcast" "broadcast" "broadcast" ; +-- build_V = irregV "build" "built" "built" ; +-- burn_V = irregV "burn" (variants {"burned" ; "burnt"}) (variants {"burned" ; "burnt"}) ; +-- burst_V = irregV "burst" "burst" "burst" ; +-- buy_V = irregV "buy" "bought" "bought" ; +-- cast_V = irregV "cast" "cast" "cast" ; +-- catch_V = irregV "catch" "caught" "caught" ; +-- choose_V = irregV "choose" "chose" "chosen" ; +-- cling_V = irregV "cling" "clung" "clung" ; +-- come_V = irregV "come" "came" "come" ; +-- cost_V = irregV "cost" "cost" "cost" ; +-- creep_V = irregV "creep" "crept" "crept" ; +-- cut_V = irregDuplV "cut" "cut" "cut" ; +-- deal_V = irregV "deal" "dealt" "dealt" ; +-- dig_V = irregDuplV "dig" "dug" "dug" ; +-- dive_V = irregV "dive" (variants {"dived" ; "dove"}) "dived" ; +-- do_V = mkV "do" "does" "did" "done" "doing" ; +-- draw_V = irregV "draw" "drew" "drawn" ; +-- dream_V = irregV "dream" (variants {"dreamed" ; "dreamt"}) (variants {"dreamed" ; "dreamt"}) ; +-- drive_V = irregV "drive" "drove" "driven" ; +-- drink_V = irregV "drink" "drank" "drunk" ; +-- eat_V = irregV "eat" "ate" "eaten" ; +-- fall_V = irregV "fall" "fell" "fallen" ; +-- feed_V = irregV "feed" "fed" "fed" ; +-- feel_V = irregV "feel" "felt" "felt" ; +-- fight_V = irregV "fight" "fought" "fought" ; +-- find_V = irregV "find" "found" "found" ; +-- fit_V = irregDuplV "fit" "fit" "fit" ; +-- flee_V = irregV "flee" "fled" "fled" ; +-- fling_V = irregV "fling" "flung" "flung" ; +-- fly_V = irregV "fly" "flew" "flown" ; +-- forbid_V = irregDuplV "forbid" "forbade" "forbidden" ; +-- forget_V = irregDuplV "forget" "forgot" "forgotten" ; +-- forgive_V = irregV "forgive" "forgave" "forgiven" ; +-- forsake_V = irregV "forsake" "forsook" "forsaken" ; +-- freeze_V = irregV "freeze" "froze" "frozen" ; +-- get_V = irregDuplV "get" "got" "gotten" ; +-- give_V = irregV "give" "gave" "given" ; +-- go_V = mkV "go" "goes" "went" "gone" "going" ; +-- grind_V = irregV "grind" "ground" "ground" ; +-- grow_V = irregV "grow" "grew" "grown" ; +-- hang_V = irregV "hang" "hung" "hung" ; +-- have_V = mkV "have" "has" "had" "had" "having" ; +-- hear_V = irregV "hear" "heard" "heard" ; +-- hide_V = irregV "hide" "hid" "hidden" ; +-- hit_V = irregDuplV "hit" "hit" "hit" ; +-- hold_V = irregV "hold" "held" "held" ; +-- hurt_V = irregV "hurt" "hurt" "hurt" ; +-- keep_V = irregV "keep" "kept" "kept" ; +-- kneel_V = irregV "kneel" "knelt" "knelt" ; +-- knit_V = irregDuplV "knit" "knit" "knit" ; +-- know_V = irregV "know" "knew" "know" ; +-- lay_V = irregV "lay" "laid" "laid" ; +-- lead_V = irregV "lead" "led" "led" ; +-- leap_V = irregV "leap" (variants {"leaped" ; "lept"}) (variants {"leaped" ; "lept"}) ; +-- learn_V = irregV "learn" (variants {"learned" ; "learnt"}) (variants {"learned" ; "learnt"}) ; +-- leave_V = irregV "leave" "left" "left" ; +-- lend_V = irregV "lend" "lent" "lent" ; +-- let_V = irregDuplV "let" "let" "let" ; +-- lie_V = irregV "lie" "lay" "lain" ; +-- light_V = irregV "light" (variants {"lighted" ; "lit"}) "lighted" ; +-- lose_V = irregV "lose" "lost" "lost" ; +-- make_V = irregV "make" "made" "made" ; +-- mean_V = irregV "mean" "meant" "meant" ; +-- meet_V = irregV "meet" "met" "met" ; +-- misspell_V = irregV "misspell" (variants {"misspelled" ; "misspelt"}) (variants {"misspelled" ; "misspelt"}) ; +-- mistake_V = irregV "mistake" "mistook" "mistaken" ; +-- mow_V = irregV "mow" "mowed" (variants {"mowed" ; "mown"}) ; +-- overcome_V = irregV "overcome" "overcame" "overcome" ; +-- overdo_V = mkV "overdo" "overdoes" "overdid" "overdone" "overdoing" ; +-- overtake_V = irregV "overtake" "overtook" "overtaken" ; +-- overthrow_V = irregV "overthrow" "overthrew" "overthrown" ; +-- pay_V = irregV "pay" "paid" "paid" ; +-- plead_V = irregV "plead" "pled" "pled" ; +-- prove_V = irregV "prove" "proved" (variants {"proved" ; "proven"}) ; +-- put_V = irregDuplV "put" "put" "put" ; +-- quit_V = irregDuplV "quit" "quit" "quit" ; +-- read_V = irregV "read" "read" "read" ; +-- rid_V = irregDuplV "rid" "rid" "rid" ; +-- ride_V = irregV "ride" "rode" "ridden" ; +-- ring_V = irregV "ring" "rang" "rung" ; +-- rise_V = irregV "rise" "rose" "risen" ; +-- run_V = irregDuplV "run" "ran" "run" ; +-- saw_V = irregV "saw" "sawed" (variants {"sawed" ; "sawn"}) ; +-- say_V = irregV "say" "said" "said" ; +-- see_V = irregV "see" "saw" "seen" ; +-- seek_V = irregV "seek" "sought" "sought" ; +-- sell_V = irregV "sell" "sold" "sold" ; +-- send_V = irregV "send" "sent" "sent" ; +-- set_V = irregDuplV "set" "set" "set" ; +-- sew_V = irregV "sew" "sewed" (variants {"sewed" ; "sewn"}) ; +-- shake_V = irregV "shake" "shook" "shaken" ; +-- shave_V = irregV "shave" "shaved" (variants {"shaved" ; "shaven"}) ; +-- shear_V = irregV "shear" "shore" "shorn" ; +-- shed_V = irregDuplV "shed" "shed" "shed" ; +-- shine_V = irregV "shine" "shone" "shone" ; +-- shoe_V = irregV "shoe" "shoed" (variants {"shoed" ; "shod"}) ; +-- shoot_V = irregV "shoot" "shot" "shot" ; +-- show_V = irregV "show" "showed" (variants {"showed" ; "shown"}) ; +-- shrink_V = irregV "shrink" "shrank" "shrunk" ; +-- shut_V = irregDuplV "shut" "shut" "shut" ; +-- sing_V = irregV "sing" "sang" "sung" ; +-- sink_V = irregV "sink" "sank" "sunk" ; +-- sit_V = irregDuplV "sit" "sat" "sat" ; +-- sleep_V = irregV "sleep" "slept" "slept" ; +-- slay_V = irregV "slay" "slew" "slain" ; +-- slide_V = irregV "slide" "slid" "slid" ; +-- sling_V = irregV "sling" "slung" "slung" ; +-- slit_V = irregDuplV "slit" "slit" "slit" ; +-- smite_V = irregV "smite" "smote" "smitten" ; +-- sow_V = irregV "sow" "sowed" (variants {"sowed" ; "sown"}) ; +-- speak_V = irregV "speak" "spoke" "spoken" ; +-- speed_V = irregV "speed" "sped" "sped" ; +-- spend_V = irregV "spend" "spent" "spent" ; +-- spill_V = irregV "spill" (variants {"spilled" ; "spilt"}) (variants {"spilled" ; "spilt"}) ; +-- spin_V = irregDuplV "spin" "spun" "spun" ; +-- spit_V = irregDuplV "spit" (variants {"spit" ; "spat"}) "spit" ; +-- split_V = irregDuplV "split" "split" "split" ; +-- spread_V = irregV "spread" "spread" "spread" ; +-- spring_V = irregV "spring" (variants {"sprang" ; "sprung"}) "sprung" ; +-- stand_V = irregV "stand" "stood" "stood" ; +-- steal_V = irregV "steal" "stole" "stolen" ; +-- stick_V = irregV "stick" "stuck" "stuck" ; +-- sting_V = irregV "sting" "stung" "stung" ; +-- stink_V = irregV "stink" "stank" "stunk" ; +-- stride_V = irregV "stride" "strod" "stridden" ; +-- strike_V = irregV "strike" "struck" "struck" ; +-- string_V = irregV "string" "strung" "strung" ; +-- strive_V = irregV "strive" "strove" "striven" ; +-- swear_V = irregV "swear" "swore" "sworn" ; +-- sweep_V = irregV "sweep" "swept" "swept" ; +-- swell_V = irregV "swell" "swelled" (variants {"swelled" ; "swollen"}) ; +-- swim_V = irregDuplV "swim" "swam" "swum" ; +-- swing_V = irregV "swing" "swung" "swung" ; +-- take_V = irregV "take" "took" "taken" ; +-- teach_V = irregV "teach" "taught" "taught" ; +-- tear_V = irregV "tear" "tore" "torn" ; +-- tell_V = irregV "tell" "told" "told" ; +-- think_V = irregV "think" "thought" "thought" ; +-- thrive_V = irregV "thrive" (variants {"thrived" ; "throve"}) "thrived" ; +-- throw_V = irregV "throw" "threw" "thrown" ; +-- thrust_V = irregV "thrust" "thrust" "thrust" ; +-- tread_V = irregV "tread" "trod" "trodden" ; +-- understand_V = irregV "understand" "understood" "understood" ; +-- uphold_V = irregV "uphold" "upheld" "upheld" ; +-- upset_V = irregDuplV "upset" "upset" "upset" ; +-- wake_V = irregV "wake" "woke" "woken" ; +-- wear_V = irregV "wear" "wore" "worn" ; +-- weave_V = irregV "weave" (variants {"weaved" ; "wove"}) (variants {"weaved" ; "woven"}) ; +-- wed_V = irregDuplV "wed" "wed" "wed" ; +-- weep_V = irregV "weep" "wept" "wept" ; +-- wind_V = irregV "wind" "wound" "wound" ; +-- win_V = irregDuplV "win" "won" "won" ; +-- withhold_V = irregV "withhold" "withheld" "withheld" ; +-- withstand_V = irregV "withstand" "withstood" "withstood" ; +-- wring_V = irregV "wring" "wrung" "wrung" ; +-- write_V = irregV "write" "wrote" "written" ; +--} diff --git a/next-lib/src/thai/LangTha.gf b/next-lib/src/thai/LangTha.gf new file mode 100644 index 000000000..77ebd1018 --- /dev/null +++ b/next-lib/src/thai/LangTha.gf @@ -0,0 +1,10 @@ +--# -path=.:../abstract:../common:prelude + +concrete LangTha of Lang = + GrammarTha, + LexiconTha + ** { + +flags startcat = Phr ; unlexer = concat ; lexer = text ; + +} ; diff --git a/next-lib/src/thai/LexiconTha.gf b/next-lib/src/thai/LexiconTha.gf new file mode 100644 index 000000000..c1e047894 --- /dev/null +++ b/next-lib/src/thai/LexiconTha.gf @@ -0,0 +1,368 @@ +--# -path=.:prelude + +concrete LexiconTha of Lexicon = CatTha ** + open StringsTha, ResTha, Prelude in { + +flags + optimize=values ; + +lin +-- airplane_N = regN "airplane" ; +-- answer_V2S = mkV2S (regV "answer") toP ; +-- apartment_N = regN "apartment" ; +-- apple_N = regN "apple" ; +-- art_N = regN "art" ; +-- ask_V2Q = mkV2Q (regV "ask") noPrep ; +-- baby_N = regN "baby" ; +-- bad_A = mkADeg "bad" "badly" "worse" "worst" ; +-- bank_N = regN "bank" ; +-- beautiful_A = regADeg "beautiful" ; +-- become_VA = mkVA (irregV "become" "became" "become") ; + beer_N = mkN biar_s kew_s ; +-- beg_V2V = mkV2V (regDuplV "beg") noPrep toP ; + big_A = ss yay_s ; +-- bike_N = regN "bike" ; +-- bird_N = regN "bird" ; +-- black_A = regADeg "black" ; +-- blue_A = regADeg "blue" ; +-- boat_N = regN "boat" ; + book_N = mkN (nag_s ++ svv_s) lem_s ; +-- boot_N = regN "boot" ; +-- boss_N = regN "boss" ; +-- boy_N = regN "boy" ; +-- bread_N = regN "bread" ; +-- break_V2 = dirV2 (irregV "break" "broke" "broken") ; +-- broad_A = regADeg "broad" ; +-- brother_N2 = regN2 "brother" ; +-- brown_A = regADeg "brown" ; +-- butter_N = regN "butter" ; +-- buy_V2 = dirV2 (irregV "buy" "bought" "bought") ; +-- camera_N = regN "camera" ; +-- cap_N = regN "cap" ; +-- car_N = regN "car" ; +-- carpet_N = regN "carpet" ; +-- cat_N = regN "cat" ; +-- ceiling_N = regN "ceiling" ; +-- chair_N = regN "chair" ; +-- cheese_N = regN "cheese" ; +-- child_N = mk2N "child" "children" ; +-- church_N = regN "church" ; +-- city_N = regN "city" ; +-- clean_A = regADeg "clean" ; +-- clever_A = regADeg "clever" ; +-- close_V2 = dirV2 (regV "close") ; +-- coat_N = regN "coat" ; +-- cold_A = regADeg "cold" ; +-- come_V = (irregV "come" "came" "come") ; +-- computer_N = regN "computer" ; +-- country_N = regN "country" ; +-- cousin_N = regN "cousin" ; +-- cow_N = regN "cow" ; +-- die_V = (regV "die") ; +-- dirty_A = regADeg "dirty" ; +-- distance_N3 = mkN3 (regN "distance") fromP toP ; +-- doctor_N = regN "doctor" ; +-- dog_N = regN "dog" ; +-- door_N = regN "door" ; + drink_V2 = dirV2 (regV dvm_s) ; +-- easy_A2V = mkA2V (regA "easy") forP ; +-- eat_V2 = dirV2 (irregV "eat" "ate" "eaten") ; +-- empty_A = regADeg "empty" ; +-- enemy_N = regN "enemy" ; +-- factory_N = regN "factory" ; +-- father_N2 = regN2 "father" ; +-- fear_VS = mkVS (regV "fear") ; +-- find_V2 = dirV2 (irregV "find" "found" "found") ; +-- fish_N = mk2N "fish" "fish" ; +-- floor_N = regN "floor" ; +-- forget_V2 = dirV2 (irregDuplV "forget" "forgot" "forgotten") ; +-- fridge_N = regN "fridge" ; +-- friend_N = regN "friend" ; +-- fruit_N = regN "fruit" ; +-- fun_AV = mkAV (regA "fun") ; +-- garden_N = regN "garden" ; +-- girl_N = regN "girl" ; +-- glove_N = regN "glove" ; +-- gold_N = regN "gold" ; +-- good_A = mkADeg "good" "well" "better" "best" ; + go_V = regV pay_s ; +-- green_A = regADeg "green" ; +-- harbour_N = regN "harbour" ; +-- hate_V2 = dirV2 (regV "hate") ; +-- hat_N = regN "hat" ; +-- have_V2 = dirV2 (mkV "have" "has" "had" "had" "having") ; +-- hear_V2 = dirV2 (irregV "hear" "heard" "heard") ; +-- hill_N = regN "hill" ; +-- hope_VS = mkVS (regV "hope") ; +-- horse_N = regN "horse" ; +-- hot_A = duplADeg "hot" ; + house_N = mkN baan_s lag_s ; +-- important_A = compoundADeg (regA "important") ; +-- industry_N = regN "industry" ; +-- iron_N = regN "iron" ; +-- king_N = regN "king" ; +-- know_V2 = dirV2 (irregV "know" "knew" "known") ; +-- lake_N = regN "lake" ; +-- lamp_N = regN "lamp" ; +-- learn_V2 = dirV2 (regV "learn") ; +-- leather_N = regN "leather" ; +-- leave_V2 = dirV2 (irregV "leave" "left" "left") ; +-- like_V2 = dirV2 (regV "like") ; +-- listen_V2 = mkV2 (regV "listen") toP ; +-- live_V = (regV "live") ; + john_PN = ss "cn" ; +-- long_A = regADeg "long" ; +-- lose_V2 = dirV2 (irregV "lose" "lost" "lost") ; +-- love_N = regN "love" ; + love_V2 = dirV2 (regV rak_s) ; +-- man_N = mk2N "man" "men" ; +-- married_A2 = mkA2 (regA "married") toP ; +-- meat_N = regN "meat" ; +-- milk_N = regN "milk" ; +-- moon_N = regN "moon" ; +-- mother_N2 = regN2 "mother" ; +-- mountain_N = regN "mountain" ; +-- music_N = regN "music" ; +-- narrow_A = regADeg "narrow" ; +-- new_A = regADeg "new" ; +-- newspaper_N = regN "newspaper" ; +-- oil_N = regN "oil" ; +-- old_A = regADeg "old" ; +-- open_V2 = dirV2 (regV "open") ; +-- paint_V2A = mkV2A (regV "paint") noPrep ; +-- paper_N = regN "paper" ; +-- paris_PN = regPN "Paris" ; +-- peace_N = regN "peace" ; +-- pen_N = regN "pen" ; +-- planet_N = regN "planet" ; +-- plastic_N = regN "plastic" ; +-- play_V2 = dirV2 (regV "play") ; +-- policeman_N = regN "policeman" ; +-- priest_N = regN "priest" ; +-- probable_AS = mkAS (regA "probable") ; +-- queen_N = regN "queen" ; +-- radio_N = regN "radio" ; +-- rain_V0 = mkV0 (regV "rain") ; +-- read_V2 = dirV2 (irregV "read" "read" "read") ; +-- red_A = duplADeg "red" ; +-- religion_N = regN "religion" ; +-- restaurant_N = regN "restaurant" ; +-- river_N = regN "river" ; +-- rock_N = regN "rock" ; +-- roof_N = regN "roof" ; +-- rubber_N = regN "rubber" ; +-- run_V = (irregDuplV "run" "ran" "run") ; +-- say_VS = mkVS (irregV "say" "said" "said") ; +-- school_N = regN "school" ; +-- science_N = regN "science" ; +-- sea_N = regN "sea" ; +-- seek_V2 = dirV2 (irregV "seek" "sought" "sought") ; +-- see_V2 = dirV2 (irregV "see" "saw" "seen") ; +-- sell_V3 = dirV3 (irregV "sell" "sold" "sold") toP ; +-- send_V3 = dirV3 (irregV "send" "sent" "sent") toP ; +-- sheep_N = mk2N "sheep" "sheep" ; +-- ship_N = regN "ship" ; +-- shirt_N = regN "shirt" ; +-- shoe_N = regN "shoe" ; +-- shop_N = regN "shop" ; +-- short_A = regADeg "short" ; +-- silver_N = regN "silver" ; +-- sister_N = regN "sister" ; + sleep_V = resV noon_s lap_s ; +-- small_A = regADeg "small" ; +-- snake_N = regN "snake" ; +-- sock_N = regN "sock" ; +-- speak_V2 = dirV2 (irregV "speak" "spoke" "spoken") ; +-- star_N = regN "star" ; +-- steel_N = regN "steel" ; +-- stone_N = regN "stone" ; +-- stove_N = regN "stove" ; +-- student_N = regN "student" ; +-- stupid_A = regADeg "stupid" ; +-- sun_N = regN "sun" ; +-- switch8off_V2 = dirV2 (partV (regV "switch") "off") ; +-- switch8on_V2 = dirV2 (partV (regV "switch") "on") ; +-- table_N = regN "table" ; +-- talk_V3 = mkV3 (regV "talk") toP aboutP ; +-- teacher_N = regN "teacher" ; +-- teach_V2 = dirV2 (irregV "teach" "taught" "taught") ; +-- television_N = regN "television" ; +-- thick_A = regADeg "thick" ; +-- thin_A = duplADeg "thin" ; +-- train_N = regN "train" ; +-- travel_V = (regDuplV "travel") ; +-- tree_N = regN "tree" ; +-- ---- trousers_N = regN "trousers" ; +-- ugly_A = regADeg "ugly" ; +-- understand_V2 = dirV2 (irregV "understand" "understood" "understood") ; +-- university_N = regN "university" ; +-- village_N = regN "village" ; +-- wait_V2 = mkV2 (regV "wait") forP ; +-- walk_V = (regV "walk") ; +-- warm_A = regADeg "warm" ; +-- war_N = regN "war" ; +-- watch_V2 = dirV2 (regV "watch") ; +-- water_N = regN "water" ; +-- white_A = regADeg "white" ; +-- window_N = regN "window" ; +-- wine_N = regN "wine" ; +-- win_V2 = dirV2 (irregDuplV "win" "won" "won") ; + woman_N = mkN (puu_s ++ yig_s) khon_s ; +-- wonder_VQ = mkVQ (regV "wonder") ; +-- wood_N = regN "wood" ; +-- write_V2 = dirV2 (irregV "write" "wrote" "written") ; +-- yellow_A = regADeg "yellow" ; +-- young_A = regADeg "young" ; +-- +-- do_V2 = dirV2 (mkV "do" "does" "did" "done" "doing") ; +-- now_Adv = mkAdv "now" ; +-- already_Adv = mkAdv "already" ; +-- song_N = regN "song" ; +-- add_V3 = dirV3 (regV "add") toP ; +-- number_N = regN "number" ; +-- put_V2 = mkV2 (irregDuplV "put" "put" "put") noPrep ; +-- stop_V = regDuplV "stop" ; +-- jump_V = regV "jump" ; +-- +-- left_Ord = ss "left" ; +-- right_Ord = ss "right" ; +-- far_Adv = mkAdv "far" ; +-- correct_A = (regA "correct") ; +-- dry_A = regA "dry" ; +-- dull_A = regA "dull" ; +-- full_A = regA "full" ; +-- heavy_A = regA "heavy" ; +-- near_A = regA "near" ; +-- rotten_A = (regA "rotten") ; +-- round_A = regA "round" ; +-- sharp_A = regA "sharp" ; +-- smooth_A = regA "smooth" ; +-- straight_A = regA "straight" ; +-- wet_A = regA "wet" ; ---- +-- wide_A = regA "wide" ; +-- animal_N = regN "animal" ; +-- ashes_N = regN "ash" ; -- FIXME: plural only? +-- back_N = regN "back" ; +-- bark_N = regN "bark" ; +-- belly_N = regN "belly" ; +-- blood_N = regN "blood" ; +-- bone_N = regN "bone" ; +-- breast_N = regN "breast" ; +-- cloud_N = regN "cloud" ; +-- day_N = regN "day" ; +-- dust_N = regN "dust" ; +-- ear_N = regN "ear" ; +-- earth_N = regN "earth" ; +-- egg_N = regN "egg" ; +-- eye_N = regN "eye" ; +-- fat_N = regN "fat" ; +-- feather_N = regN "feather" ; +-- fingernail_N = regN "fingernail" ; +-- fire_N = regN "fire" ; +-- flower_N = regN "flower" ; +-- fog_N = regN "fog" ; +-- foot_N = mk2N "foot" "feet" ; +-- forest_N = regN "forest" ; +-- grass_N = regN "grass" ; +-- guts_N = regN "gut" ; -- FIXME: no singular +-- hair_N = regN "hair" ; +-- hand_N = regN "hand" ; +-- head_N = regN "head" ; +-- heart_N = regN "heart" ; +-- horn_N = regN "horn" ; +-- husband_N = genderN masculine (regN "husband") ; +-- ice_N = regN "ice" ; +-- knee_N = regN "knee" ; +-- leaf_N = mk2N "leaf" "leaves" ; +-- leg_N = regN "leg" ; +-- liver_N = regN "liver" ; +-- louse_N = mk2N "louse" "lice" ; +-- mouth_N = regN "mouth" ; +-- name_N = regN "name" ; +-- neck_N = regN "neck" ; +-- night_N = regN "night" ; +-- nose_N = regN "nose" ; +-- person_N = genderN human (regN "person") ; +-- rain_N = regN "rain" ; +-- road_N = regN "road" ; +-- root_N = regN "root" ; +-- rope_N = regN "rope" ; +-- salt_N = regN "salt" ; +-- sand_N = regN "sand" ; +-- seed_N = regN "seed" ; +-- skin_N = regN "skin" ; +-- sky_N = regN "sky" ; +-- smoke_N = regN "smoke" ; +-- snow_N = regN "snow" ; +-- stick_N = regN "stick" ; +-- tail_N = regN "tail" ; +-- tongue_N = regN "tongue" ; +-- tooth_N = mk2N "tooth" "teeth" ; +-- wife_N = genderN feminine (mk2N "wife" "wives") ; +-- wind_N = regN "wind" ; +-- wing_N = regN "wing" ; +-- worm_N = regN "worm" ; +-- year_N = regN "year" ; +-- blow_V = blow_V ; +-- breathe_V = dirV2 (regV "breathe") ; +-- burn_V = burn_V ; +-- dig_V = dig_V ; +-- fall_V = fall_V ; +-- float_V = regV "float" ; +-- flow_V = regV "flow" ; +-- fly_V = fly_V ; +-- freeze_V = freeze_V ; +-- give_V3 = dirV3 give_V toP ; +-- laugh_V = regV "laugh" ; +-- lie_V = lie_V ; +-- play_V = regV "play" ; +-- sew_V = sew_V ; +-- sing_V = sing_V ; +-- sit_V = sit_V ; +-- smell_V = regV "smell" ; +-- spit_V = spit_V ; +-- stand_V = stand_V ; +-- swell_V = swell_V ; +-- swim_V = swim_V ; +-- think_V = think_V ; +-- turn_V = regV "turn" ; +-- vomit_V = regV "vomit" ; +-- +-- bite_V2 = dirV2 bite_V ; +-- count_V2 = dirV2 (regV "count") ; +-- cut_V2 = dirV2 cut_V ; +-- fear_V2 = dirV2 (regV "fear") ; +-- fight_V2 = dirV2 fight_V ; +-- hit_V2 = dirV2 hit_V ; +-- hold_V2 = dirV2 hold_V ; +-- hunt_V2 = dirV2 (regV "hunt") ; +-- kill_V2 = dirV2 (regV "kill") ; +-- pull_V2 = dirV2 (regV "pull") ; +-- push_V2 = dirV2 (regV "push") ; +-- rub_V2 = dirV2 (regDuplV "rub") ; +-- scratch_V2 = dirV2 (regV "scratch") ; +-- split_V2 = dirV2 split_V ; +-- squeeze_V2 = dirV2 (regV "squeeze") ; +-- stab_V2 = dirV2 (regDuplV "stab") ; +-- suck_V2 = dirV2 (regV "suck") ; +-- throw_V2 = dirV2 throw_V ; +-- tie_V2 = dirV2 (regV "tie") ; +-- wash_V2 = dirV2 (regV "wash") ; +-- wipe_V2 = dirV2 (regV "wipe") ; +-- +---- other_A = regA "other" ; +-- +-- grammar_N = regN "grammar" ; +-- language_N = regN "language" ; +-- rule_N = regN "rule" ; +-- +--oper +-- aboutP = mkPrep "about" ; +-- atP = mkPrep "at" ; +-- forP = mkPrep "for" ; +-- fromP = mkPrep "from" ; +-- inP = mkPrep "in" ; +-- onP = mkPrep "on" ; +-- toP = mkPrep "to" ; +-- +} diff --git a/next-lib/src/thai/MorphoTha.gf b/next-lib/src/thai/MorphoTha.gf new file mode 100644 index 000000000..a703f10ba --- /dev/null +++ b/next-lib/src/thai/MorphoTha.gf @@ -0,0 +1,198 @@ +----# -path=.:../../prelude +-- +----1 A Simple Thai Resource Morphology +---- +---- Aarne Ranta 2002 -- 2005 +---- +---- This resource morphology contains definitions needed in the resource +---- syntax. To build a lexicon, it is better to use $ParadigmsTha$, which +---- gives a higher-level access to this module. +-- +--resource MorphoTha = ResTha ** open Prelude, (Predef=Predef) in { +-- +-- flags optimize=all ; +-- +----2 Phonology +---- +---- To regulate the use of endings for both nouns, adjectives, and verbs: +-- +--oper +-- y2ie : Str -> Str -> Str = \fly,s -> +-- let y = last (init fly) in +-- case y of { +-- "a" => fly + s ; +-- "e" => fly + s ; +-- "o" => fly + s ; +-- "u" => fly + s ; +-- _ => init fly + "ie" + s +-- } ; +-- +-- +----2 Nouns +---- +---- For conciseness and abstraction, we define a worst-case macro for +---- noun inflection. It is used for defining special case that +---- only need one string as argument. +-- +--oper +-- CommonNoun : Type = {s : Number => Case => Str} ; +-- +-- nounGen : Str -> CommonNoun = \dog -> case last dog of { +-- "y" => nounY "dog" ; +-- "s" => nounS (init "dog") ; +-- _ => nounReg "dog" +-- } ; +-- +---- These are auxiliaries to $nounGen$. +-- +-- nounReg : Str -> CommonNoun = \dog -> +-- mkNoun dog (dog + "s") (dog + "'s") (dog + "s'"); +-- nounS : Str -> CommonNoun = \kiss -> +-- mkNoun kiss (kiss + "es") (kiss + "'s") (kiss + "es'") ; +-- nounY : Str -> CommonNoun = \fl -> +-- mkNoun (fl + "y") (fl + "ies") (fl + "y's") (fl + "ies'") ; +-- +-- +----3 Proper names +---- +---- Regular proper names are inflected with "'s" in the genitive. +-- +-- nameReg : Str -> Gender -> {s : Case => Str} = \john,g -> +-- {s = table {Gen => john + "'s" ; _ => john} ; g = g} ; +-- +----2 Determiners +-- +-- mkDeterminer : Number -> Str -> {s : Str ; n : Number} = \n,s -> +-- {s = s ; n = n} ; +-- +----2 Pronouns +---- +---- Here we define personal pronouns. +---- +---- We record the form "mine" and the gender for later use. +-- +-- Pronoun : Type = +-- {s : Case => Str ; a : Agr ; g : Gender} ; +-- +-- mkPronoun : (_,_,_,_ : Str) -> Number -> Person -> Gender -> Pronoun = +-- \I,me,my,mine,n,p,g -> +-- {s = table {Nom => I ; Acc => me ; Gen => my} ; +-- a = {n = n ; p = p} ; +-- g = g +-- } ; +-- +-- human : Gender = Masc ; --- doesn't matter +-- +-- pronI = mkPronoun "I" "me" "my" "mine" Sg P1 human ; +-- pronYouSg = mkPronoun "you" "you" "your" "yours" Sg P2 human ; -- verb agr OK +-- pronHe = mkPronoun "he" "him" "his" "his" Sg P3 Masc ; +-- pronShe = mkPronoun "she" "her" "her" "hers" Sg P3 Fem ; +-- pronIt = mkPronoun "it" "it" "its" "it" Sg P3 Neutr ; +-- +-- pronWe = mkPronoun "we" "us" "our" "ours" Pl P1 human ; +-- pronYouPl = mkPronoun "you" "you" "your" "yours" Pl P2 human ; +-- pronThey = mkPronoun "they" "them" "their" "theirs" Pl P3 human ; --- +-- +-- +----2 Adjectives +---- +---- To form the adjectival and the adverbial forms, two strings are needed +---- in the worst case. (First without degrees.) +-- +-- Adjective = {s : AForm => Str} ; +-- +---- However, most adjectives can be inflected using the final character. +---- N.B. this is not correct for "shy", but $mkAdjective$ has to be used. +-- +-- regAdjective : Str -> Adjective = \free -> +-- let +-- e = last free ; +-- fre = init free ; +-- freely = case e of { +-- "y" => fre + "ily" ; +-- _ => free + "ly" +-- } ; +-- fre = case e of { +-- "e" => fre ; +-- "y" => fre + "i" ; +-- _ => free +-- } +-- in +-- mkAdjective free (fre + "er") (fre + "est") freely ; +-- +---- Many adjectives are 'inflected' by adding a comparison word. +-- +-- adjDegrLong : Str -> Adjective = \ridiculous -> +-- mkAdjective +-- ridiculous +-- ("more" ++ ridiculous) +-- ("most" ++ ridiculous) +-- ((regAdjective ridiculous).s ! AAdv) ; +-- +-- +----3 Verbs +---- +---- The worst case needs five forms. (The verb "be" is treated separately.) +-- +-- mkVerb4 : (_,_,_,_: Str) -> Verb = \go,goes,went,gone -> +-- let going = case last go of { +-- "e" => init go + "ing" ; +-- _ => go + "ing" +-- } +-- in +-- mkVerb go goes went gone going ; +-- +---- This is what we use to derive the irregular forms in almost all cases +-- +-- mkVerbIrreg : (_,_,_ : Str) -> Verb = \bite,bit,bitten -> +-- let bites = case last bite of { +-- "y" => y2ie bite "s" ; +-- "s" => init bite + "es" ; +-- _ => bite + "s" +-- } +-- in mkVerb4 bite bites bit bitten ; +-- +---- This is used to derive regular forms. +-- +-- mkVerbReg : Str -> Verb = \soak -> +-- let +-- soaks = case last soak of { +-- "y" => y2ie soak "s" ; +-- "s" => init soak + "es" ; +-- _ => soak + "s" +-- } ; +-- soaked = case last soak of { +-- "e" => init soak + "s" ; +-- _ => soak + "ed" +-- } +-- in +-- mkVerb4 soak soaks soaked soaked ; +-- +-- verbGen : Str -> Verb = \kill -> case last kill of { +-- "y" => verbP3y (init kill) ; +-- "e" => verbP3e (init kill) ; +-- "s" => verbP3s (init kill) ; +-- _ => regVerbP3 kill +-- } ; +-- +---- These are just auxiliary to $verbGen$. +-- +-- regVerbP3 : Str -> Verb = \walk -> +-- mkVerbIrreg walk (walk + "ed") (walk + "ed") ; +-- verbP3s : Str -> Verb = \kiss -> +-- mkVerb4 kiss (kiss + "es") (kiss + "ed") (kiss + "ed") ; +-- verbP3e : Str -> Verb = \love -> +-- mkVerb4 love (love + "s") (love + "d") (love + "d") ; +-- verbP3y : Str -> Verb = \cr -> +-- mkVerb4 (cr + "y") (cr + "ies") (cr + "ied") (cr + "ied") ; +-- +----- The particle always appears right after the verb. +-- +-- verbPart : Verb -> Str -> Verb = \v,p -> +-- {s = \\f => v.s ! f ++ p ; isRefl = v.isRefl} ; +-- +-- verbNoPart : Verb -> Verb = \v -> verbPart v [] ; +-- +-- +--} ; +-- diff --git a/next-lib/src/thai/NounTha.gf b/next-lib/src/thai/NounTha.gf new file mode 100644 index 000000000..3bc9ab578 --- /dev/null +++ b/next-lib/src/thai/NounTha.gf @@ -0,0 +1,76 @@ +concrete NounTha of Noun = CatTha ** open StringsTha, ResTha, Prelude in { + + flags optimize=all_subs ; + + lin + DetCN det cn = + let cnc = if_then_Str det.hasC cn.c [] + in ss (cn.s ++ det.s1 ++ cnc ++ det.s2) ; + UsePN pn = pn ; + UsePron p = p ; +-- +-- PredetNP pred np = { +-- s = \\c => pred.s ++ np.s ! c ; +-- a = np.a +-- } ; +-- +-- PPartNP np v2 = { +-- s = \\c => np.s ! c ++ v2.s ! VPPart ; +-- a = np.a +-- } ; +-- +-- AdvNP np adv = { +-- s = \\c => np.s ! c ++ adv.s ; +-- a = np.a +-- } ; + + DetQuant quant num = { + s1 = [] ; + s2 = quant.s ++ num.s ; + hasC = quant.hasC ; + } ; + DetQuantOrd quant num ord = { + s1 = num.s ; + s2 = quant.s ++ ord.s ; + hasC = orB num.hasC quant.hasC ; + } ; + + PossPron p = { + s = khoog_s ++ p.s ; + hasC = False + } ; + + NumSg, NumPl = {s = [] ; hasC = False} ; + + NumCard n = n ** {hasC = True} ; +-- OrdInt n = {s = n.s ++ "th"} ; --- +-- + NumNumeral numeral = numeral ** {hasC = True} ; + OrdNumeral numeral = {s = thii_s ++ numeral.s} ; +-- +-- AdNum adn num = {s = adn.s ++ num.s} ; +-- +-- OrdSuperl a = {s = a.s ! AAdj Superl} ; +-- + DefArt = {s = [] ; hasC = False} ; + IndefArt = {s = [] ; hasC = False} ; + + MassNP cn = cn ; + + UseN n = n ; +-- UseN2 n = n ; +-- UseN3 n = n ; +-- +-- ComplN2 f x = {s = \\n,c => f.s ! n ! Nom ++ f.c2 ++ x.s ! c} ; +-- ComplN3 f x = {s = \\n,c => f.s ! n ! Nom ++ f.c2 ++ x.s ! c ; c2 = f.c3} ; + + AdjCN ap cn = {s = cn.s ++ ap.s ; c = cn.c} ; + +-- RelCN cn rs = {s = \\n,c => cn.s ! n ! c ++ rs.s ! {n = n ; p = P3}} ; +-- AdvCN cn ad = {s = \\n,c => cn.s ! n ! c ++ ad.s} ; +-- +-- SentCN cn sc = {s = \\n,c => cn.s ! n ! c ++ sc.s} ; +-- +-- ApposCN cn np = {s = \\n,c => cn.s ! n ! Nom ++ np.s ! c} ; +-- +} diff --git a/next-lib/src/thai/NumeralTha.gf b/next-lib/src/thai/NumeralTha.gf new file mode 100644 index 000000000..80c1e3788 --- /dev/null +++ b/next-lib/src/thai/NumeralTha.gf @@ -0,0 +1,66 @@ +concrete NumeralTha of Numeral = CatTha ** open ResTha, StringsTha in { + +lincat +-- Numeral = {s : Str} ; + Digit = {s : DForm => Str} ; + Sub10 = {s : DForm => Str} ; + Sub100 = {s : NForm => Str} ; + Sub1000 = {s : NForm => Str} ; + Sub1000000 = {s : Str} ; + +lin + num x = x ; + + pot01 = mkNum nvg_s nvg_s et_s ; + + n2 = mkNum soog_s yii_s soog_s ; + n3 = regNum saam_s ; + n4 = regNum sii_s ; + n5 = regNum haa_s ; + n6 = regNum hok_s ; + n7 = regNum cet_s ; + n8 = regNum peet_s ; + n9 = regNum kaaw_s ; + + + pot0 d = d ; + + pot110 = {s = sip} ; + pot111 = {s = table { + Unit => sip_s ++ et_s ; + Thousand => nvg_s ++ mvvn_s ++ nvg_s ++ phan_s + } + } ; + pot1to19 d = {s = table { + Unit => sip_s ++ d.s ! After ; + Thousand => nvg_s ++ mvvn_s ++ d.s ! Indep ++ phan_s + } + } ; + pot0as1 d = {s = \\n => d.s ! Indep ++ phan ! n} ; + pot1 d = {s = \\n => d.s ! ModTen ++ sip ! n} ; + pot1plus d e = { + s = \\n => d.s ! ModTen ++ sip ! n ++ e.s ! After ++ phan ! n + } ; + pot1as2 n = n ; + pot2 d = {s = \\n => d.s ! Indep ++ roy ! n} ; + pot2plus d e = {s = \\n => d.s ! Indep ++ roy ! n ++ e.s ! n} ; + pot2as3 n = {s = n.s ! Unit} ; + pot3 n = {s = n.s ! Thousand} ; + pot3plus n m = {s = n.s ! Thousand ++ m.s ! Unit} ; + +param + DForm = Indep | ModTen | After ; + NForm = Unit | Thousand ; + +oper + mkNum : Str -> Str -> Str -> {s : DForm => Str} = \x,y,z -> + {s = table {Indep => x ; ModTen => y ; After => z}} ; + regNum : Str -> {s : DForm => Str} = \x -> + mkNum x x x ; + + + sip = table {Unit => sip_s ; Thousand => mvvn_s} ; + roy = table {Unit => rooy_s ; Thousand => seen_s} ; + phan = table {Unit => [] ; Thousand => phan_s} ; + +} diff --git a/next-lib/src/thai/OverloadTha.gf b/next-lib/src/thai/OverloadTha.gf new file mode 100644 index 000000000..6074cf7ad --- /dev/null +++ b/next-lib/src/thai/OverloadTha.gf @@ -0,0 +1 @@ +--resource OverloadTha = Overload with (Grammar = GrammarTha) ; diff --git a/next-lib/src/thai/ParadigmsTha.gf b/next-lib/src/thai/ParadigmsTha.gf new file mode 100644 index 000000000..a0dc0d4d3 --- /dev/null +++ b/next-lib/src/thai/ParadigmsTha.gf @@ -0,0 +1,496 @@ +----# -path=.:../abstract:../../prelude:../common +-- +----1 Thai Lexical Paradigms +---- +---- Aarne Ranta 2003--2005 +---- +---- This is an API for the user of the resource grammar +---- for adding lexical items. It gives functions for forming +---- expressions of open categories: nouns, adjectives, verbs. +---- +---- Closed categories (determiners, pronouns, conjunctions) are +---- accessed through the resource syntax API, $Structural.gf$. +---- +---- The main difference with $MorphoTha.gf$ is that the types +---- referred to are compiled resource grammar types. We have moreover +---- had the design principle of always having existing forms, rather +---- than stems, as string arguments of the paradigms. +---- +---- The structure of functions for each word class $C$ is the following: +---- first we give a handful of patterns that aim to cover all +---- regular cases. Then we give a worst-case function $mkC$, which serves as an +---- escape to construct the most irregular words of type $C$. +---- However, this function should only seldom be needed: we have a +---- separate module [``IrregTha`` ../../english/IrregTha.gf], +---- which covers irregular verbss. +-- +--resource ParadigmsTha = open +-- (Predef=Predef), +-- Prelude, +-- MorphoTha, +-- CatTha +-- in { +----2 Parameters +---- +---- To abstract over gender names, we define the following identifiers. +-- +--oper +-- Gender : Type ; +-- +-- human : Gender ; +-- nonhuman : Gender ; +-- masculine : Gender ; +-- +---- To abstract over number names, we define the following. +-- +-- Number : Type ; +-- +-- singular : Number ; +-- plural : Number ; +-- +---- To abstract over case names, we define the following. +-- +-- Case : Type ; +-- +-- nominative : Case ; +-- genitive : Case ; +-- +---- Prepositions are used in many-argument functions for rection. +---- The resource category $Prep$ is used. +-- +-- +-- +----2 Nouns +-- +---- Nouns are constructed by the function $mkN$, which takes a varying +---- number of arguments. +-- +-- mkN : overload { +-- +---- Worst case: give all four forms. +-- +-- mkN : (man,men,man's,men's : Str) -> N ; +-- +---- The regular function captures the variants for nouns ending with +---- "s","sh","x","z" or "y": "kiss - kisses", "flash - flashes"; +---- "fly - flies" (but "toy - toys"), +-- +-- mkN : (flash : Str) -> N ; +-- +---- In practice the worst case is just: give singular and plural nominative. +-- +-- mkN : (man,men : Str) -> N ; +-- +---- All nouns created by the previous functions are marked as +---- $nonhuman$. If you want a $human$ noun, wrap it with the following +---- function: +-- +-- mkN : Gender -> N -> N ; +-- +----3 Compound nouns +---- +---- A compound noun is an uninflected string attached to an inflected noun, +---- such as "baby boom", "chief executive officer". +-- +-- mkN : Str -> N -> N +-- } ; +-- +-- +----3 Relational nouns +---- +---- Relational nouns ("daughter of x") need a preposition. +-- +-- mkN2 : N -> Prep -> N2 ; +-- +---- The most common preposition is "of", and the following is a +---- shortcut for regular relational nouns with "of". +-- +-- regN2 : Str -> N2 ; +-- +---- Use the function $mkPrep$ or see the section on prepositions below to +---- form other prepositions. +---- +---- Three-place relational nouns ("the connection from x to y") need two prepositions. +-- +-- mkN3 : N -> Prep -> Prep -> N3 ; +-- +-- +----3 Relational common noun phrases +---- +---- In some cases, you may want to make a complex $CN$ into a +---- relational noun (e.g. "the old town hall of"). +-- +-- cnN2 : CN -> Prep -> N2 ; +-- cnN3 : CN -> Prep -> Prep -> N3 ; +-- +---- +----3 Proper names and noun phrases +---- +---- Proper names, with a regular genitive, are formed as follows +-- +-- regPN : Str -> PN ; +-- regGenPN : Str -> Gender -> PN ; -- John, John's +-- +---- Sometimes you can reuse a common noun as a proper name, e.g. "Bank". +-- +-- nounPN : N -> PN ; +-- +---- To form a noun phrase that can also be plural and have an irregular +---- genitive, you can use the worst-case function. +-- +-- mkNP : Str -> Str -> Number -> Gender -> NP ; +-- +----2 Adjectives +-- +---- Non-comparison one-place adjectives need two forms: one for +---- the adjectival and one for the adverbial form ("free - freely") +-- +-- mkA : (free,freely : Str) -> A ; +-- +---- For regular adjectives, the adverbial form is derived. This holds +---- even for cases with the variation "happy - happily". +-- +-- regA : Str -> A ; +-- +----3 Two-place adjectives +---- +---- Two-place adjectives need a preposition for their second argument. +-- +-- mkA2 : A -> Prep -> A2 ; +-- +---- Comparison adjectives may two more forms. +-- +-- ADeg : Type ; +-- +-- mkADeg : (good,better,best,well : Str) -> ADeg ; +-- +---- The regular pattern recognizes two common variations: +---- "-e" ("rude" - "ruder" - "rudest") and +---- "-y" ("happy - happier - happiest - happily") +-- +-- regADeg : Str -> ADeg ; -- long, longer, longest +-- +---- However, the duplication of the final consonant is nor predicted, +---- but a separate pattern is used: +-- +-- duplADeg : Str -> ADeg ; -- fat, fatter, fattest +-- +---- If comparison is formed by "more", "most", as in general for +---- long adjective, the following pattern is used: +-- +-- compoundADeg : A -> ADeg ; -- -/more/most ridiculous +-- +---- From a given $ADeg$, it is possible to get back to $A$. +-- +-- adegA : ADeg -> A ; +-- +-- +----2 Adverbs +-- +---- Adverbs are not inflected. Most lexical ones have position +---- after the verb. Some can be preverbal (e.g. "always"). +-- +-- mkAdv : Str -> Adv ; +-- mkAdV : Str -> AdV ; +-- +---- Adverbs modifying adjectives and sentences can also be formed. +-- +-- mkAdA : Str -> AdA ; +-- +----2 Prepositions +---- +---- A preposition as used for rection in the lexicon, as well as to +---- build $PP$s in the resource API, just requires a string. +-- +-- mkPrep : Str -> Prep ; +-- noPrep : Prep ; +-- +---- (These two functions are synonyms.) +-- +----2 Verbs +---- +---- Except for "be", the worst case needs five forms: the infinitive and +---- the third person singular present, the past indicative, and the +---- past and present participles. +-- +-- mkV : (go, goes, went, gone, going : Str) -> V ; +-- +---- The regular verb function recognizes the special cases where the last +---- character is "y" ("cry - cries" but "buy - buys") or "s", "sh", "x", "z" +---- ("fix - fixes", etc). +-- +-- regV : Str -> V ; +-- +---- The following variant duplicates the last letter in the forms like +---- "rip - ripped - ripping". +-- +-- regDuplV : Str -> V ; +-- +---- There is an extensive list of irregular verbs in the module $IrregularTha$. +---- In practice, it is enough to give three forms, +---- e.g. "drink - drank - drunk", with a variant indicating consonant +---- duplication in the present participle. +-- +-- irregV : (drink, drank, drunk : Str) -> V ; +-- irregDuplV : (get, got, gotten : Str) -> V ; +-- +-- +----3 Verbs with a particle. +---- +---- The particle, such as in "switch on", is given as a string. +-- +-- partV : V -> Str -> V ; +-- +----3 Reflexive verbs +---- +---- By default, verbs are not reflexive; this function makes them that. +-- +-- reflV : V -> V ; +-- +----3 Two-place verbs +---- +---- Two-place verbs need a preposition, except the special case with direct object. +---- (transitive verbs). Notice that a particle comes from the $V$. +-- +-- mkV2 : V -> Prep -> V2 ; +-- +-- dirV2 : V -> V2 ; +-- +----3 Three-place verbs +---- +---- Three-place (ditransitive) verbs need two prepositions, of which +---- the first one or both can be absent. +-- +-- mkV3 : V -> Prep -> Prep -> V3 ; -- speak, with, about +-- dirV3 : V -> Prep -> V3 ; -- give,_,to +-- dirdirV3 : V -> V3 ; -- give,_,_ +-- +----3 Other complement patterns +---- +---- Verbs and adjectives can take complements such as sentences, +---- questions, verb phrases, and adjectives. +-- +-- mkV0 : V -> V0 ; +-- mkVS : V -> VS ; +-- mkV2S : V -> Prep -> V2S ; +-- mkVV : V -> VV ; +-- mkV2V : V -> Prep -> Prep -> V2V ; +-- mkVA : V -> VA ; +-- mkV2A : V -> Prep -> V2A ; +-- mkVQ : V -> VQ ; +-- mkV2Q : V -> Prep -> V2Q ; +-- +-- mkAS : A -> AS ; +-- mkA2S : A -> Prep -> A2S ; +-- mkAV : A -> AV ; +-- mkA2V : A -> Prep -> A2V ; +-- +---- Notice: categories $V2S, V2V, V2A, V2Q$ are in v 1.0 treated +---- just as synonyms of $V2$, and the second argument is given +---- as an adverb. Likewise $AS, A2S, AV, A2V$ are just $A$. +---- $V0$ is just $V$. +-- +-- V0, V2S, V2V, V2A, V2Q : Type ; +-- AS, A2S, AV, A2V : Type ; +-- +----. +----2 Definitions of paradigms +---- +---- The definitions should not bother the user of the API. So they are +---- hidden from the document. +-- +-- Gender = MorphoTha.Gender ; +-- Number = MorphoTha.Number ; +-- Case = MorphoTha.Case ; +-- human = Masc ; +-- nonhuman = Neutr ; +-- masculine = Masc ; +-- feminine = Fem ; +-- singular = Sg ; +-- plural = Pl ; +-- nominative = Nom ; +-- genitive = Gen ; +-- +-- Preposition : Type = Str ; -- obsolete +-- +-- regN = \ray -> +-- let +-- ra = Predef.tk 1 ray ; +-- y = Predef.dp 1 ray ; +-- r = Predef.tk 2 ray ; +-- ay = Predef.dp 2 ray ; +-- rays = +-- case y of { +-- "y" => y2ie ray "s" ; +-- "s" => ray + "es" ; +-- "z" => ray + "es" ; +-- "x" => ray + "es" ; +-- _ => case ay of { +-- "sh" => ray + "es" ; +-- "ch" => ray + "es" ; +-- _ => ray + "s" +-- } +-- } +-- in +-- mk2N ray rays ; +-- +-- mk2N = \man,men -> +-- let mens = case last men of { +-- "s" => men + "'" ; +-- _ => men + "'s" +-- } +-- in +-- mk4N man men (man + "'s") mens ; +-- +-- mk4N = \man,men,man's,men's -> +-- mkNoun man man's men men's ** {g = Neutr ; lock_N = <>} ; +-- +-- genderN g man = {s = man.s ; g = g ; lock_N = <>} ; +-- +-- compoundN s n = {s = \\x,y => s ++ n.s ! x ! y ; g=n.g ; lock_N = <>} ; +-- +-- mkN2 = \n,p -> n ** {lock_N2 = <> ; c2 = p.s} ; +-- regN2 n = mkN2 (regN n) (mkPrep "of") ; +-- mkN3 = \n,p,q -> n ** {lock_N3 = <> ; c2 = p.s ; c3 = q.s} ; +-- cnN2 = \n,p -> n ** {lock_N2 = <> ; c2 = p.s} ; +-- cnN3 = \n,p,q -> n ** {lock_N3 = <> ; c2 = p.s ; c3 = q.s} ; +-- +-- regPN n = regGenPN n human ; +-- regGenPN n g = nameReg n g ** {lock_PN = <>} ; +-- nounPN n = {s = n.s ! singular ; g = n.g ; lock_PN = <>} ; +-- mkNP x y n g = {s = table {Gen => x ; _ => y} ; a = agrP3 n ; +-- lock_NP = <>} ; +-- +-- mkA a b = mkAdjective a a a b ** {lock_A = <>} ; +-- regA a = regAdjective a ** {lock_A = <>} ; +-- +-- mkA2 a p = a ** {c2 = p.s ; lock_A2 = <>} ; +-- +-- ADeg = A ; ---- +-- +-- mkADeg a b c d = mkAdjective a b c d ** {lock_A = <>} ; +-- +-- regADeg happy = +-- let +-- happ = init happy ; +-- y = last happy ; +-- happie = case y of { +-- "y" => happ + "ie" ; +-- "e" => happy ; +-- _ => happy + "e" +-- } ; +-- happily = case y of { +-- "y" => happ + "ily" ; +-- _ => happy + "ly" +-- } ; +-- in mkADeg happy (happie + "r") (happie + "st") happily ; +-- +-- duplADeg fat = +-- mkADeg fat +-- (fat + last fat + "er") (fat + last fat + "est") (fat + "ly") ; +-- +-- compoundADeg a = +-- let ad = (a.s ! AAdj Posit) +-- in mkADeg ad ("more" ++ ad) ("most" ++ ad) (a.s ! AAdv) ; +-- +-- adegA a = a ; +-- +-- mkAdv x = ss x ** {lock_Adv = <>} ; +-- mkAdV x = ss x ** {lock_AdV = <>} ; +-- mkAdA x = ss x ** {lock_AdA = <>} ; +-- +-- mkPrep p = ss p ** {lock_Prep = <>} ; +-- noPrep = mkPrep [] ; +-- +-- mkV a b c d e = mkVerb a b c d e ** {s1 = [] ; lock_V = <>} ; +-- +-- regV cry = +-- let +-- cr = init cry ; +-- y = last cry ; +-- cries = (regN cry).s ! Pl ! Nom ; -- ! +-- crie = init cries ; +-- cried = case last crie of { +-- "e" => crie + "d" ; +-- _ => crie + "ed" +-- } ; +-- crying = case y of { +-- "e" => case last cr of { +-- "e" => cry + "ing" ; +-- _ => cr + "ing" +-- } ; +-- _ => cry + "ing" +-- } +-- in mkV cry cries cried cried crying ; +-- +-- regDuplV fit = +-- case last fit of { +-- ("a" | "e" | "i" | "o" | "u" | "y") => +-- Predef.error (["final duplication makes no sense for"] ++ fit) ; +-- t => +-- let fitt = fit + t in +-- mkV fit (fit + "s") (fitt + "ed") (fitt + "ed") (fitt + "ing") +-- } ; +-- +-- irregV x y z = let reg = (regV x).s in +-- mkV x (reg ! VPres) y z (reg ! VPresPart) ** {s1 = [] ; lock_V = <>} ; +-- +-- irregDuplV fit y z = +-- let +-- fitting = (regDuplV fit).s ! VPresPart +-- in +-- mkV fit (fit + "s") y z fitting ; +-- +-- partV v p = verbPart v p ** {lock_V = <>} ; +-- reflV v = {s = v.s ; part = v.part ; lock_V = v.lock_V ; isRefl = True} ; +-- +-- mkV2 v p = v ** {s = v.s ; s1 = v.s1 ; c2 = p.s ; lock_V2 = <>} ; +-- dirV2 v = mkV2 v noPrep ; +-- +-- mkV3 v p q = v ** {s = v.s ; s1 = v.s1 ; c2 = p.s ; c3 = q.s ; lock_V3 = <>} ; +-- dirV3 v p = mkV3 v noPrep p ; +-- dirdirV3 v = dirV3 v noPrep ; +-- +-- mkVS v = v ** {lock_VS = <>} ; +-- mkVV v = { +-- s = table {VVF vf => v.s ! vf ; _ => variants {}} ; +-- isAux = False ; lock_VV = <> +-- } ; +-- mkVQ v = v ** {lock_VQ = <>} ; +-- +-- V0 : Type = V ; +-- V2S, V2V, V2Q, V2A : Type = V2 ; +-- AS, A2S, AV : Type = A ; +-- A2V : Type = A2 ; +-- +-- mkV0 v = v ** {lock_V = <>} ; +-- mkV2S v p = mkV2 v p ** {lock_V2 = <>} ; +-- mkV2V v p t = mkV2 v p ** {s4 = t ; lock_V2 = <>} ; +-- mkVA v = v ** {lock_VA = <>} ; +-- mkV2A v p = mkV2 v p ** {lock_V2A = <>} ; +-- mkV2Q v p = mkV2 v p ** {lock_V2 = <>} ; +-- +-- mkAS v = v ** {lock_A = <>} ; +-- mkA2S v p = mkA2 v p ** {lock_A = <>} ; +-- mkAV v = v ** {lock_A = <>} ; +-- mkA2V v p = mkA2 v p ** {lock_A2 = <>} ; +-- +-- +---- pre-overload API and overload definitions +-- +-- mk4N : (man,men,man's,men's : Str) -> N ; +-- regN : Str -> N ; +-- mk2N : (man,men : Str) -> N ; +-- genderN : Gender -> N -> N ; +-- compoundN : Str -> N -> N ; +-- +-- mkN = overload { +-- mkN : (man,men,man's,men's : Str) -> N = mk4N ; +-- mkN : Str -> N = regN ; +-- mkN : (man,men : Str) -> N = mk2N ; +-- mkN : Gender -> N -> N = genderN ; +-- mkN : Str -> N -> N = compoundN +-- } ; +-- +-- +--} ; diff --git a/next-lib/src/thai/PhraseTha.gf b/next-lib/src/thai/PhraseTha.gf new file mode 100644 index 000000000..c51d85a20 --- /dev/null +++ b/next-lib/src/thai/PhraseTha.gf @@ -0,0 +1,24 @@ +concrete PhraseTha of Phrase = CatTha ** open Prelude, ResTha in { + + lin + PhrUtt pconj utt voc = {s = pconj.s ++ utt.s ++ voc.s} ; + + UttS s = s ; + UttQS qs = {s = qs.s ! QDir} ; + UttImpSg pol imp = {s = pol.s ++ imp.s ! pol.p} ; + UttImpPl pol imp = {s = pol.s ++ imp.s ! pol.p} ; + UttImpPol pol imp = {s = pol.s ++ imp.s ! pol.p} ; --- add politeness here? + + UttIP ip = ip ; + UttIAdv iadv = iadv ; + UttNP np = np ; + UttVP vp = {s = vp.s ! Pos} ; + UttAdv adv = adv ; + + NoPConj = {s = []} ; + PConjConj conj = conj ; + + NoVoc = {s = []} ; + VocNP np = {s = np.s} ; ---- ?? + +} diff --git a/next-lib/src/thai/QuestionTha.gf b/next-lib/src/thai/QuestionTha.gf new file mode 100644 index 000000000..033539bdf --- /dev/null +++ b/next-lib/src/thai/QuestionTha.gf @@ -0,0 +1,41 @@ +concrete QuestionTha of Question = CatTha ** + open ResTha, StringsTha, Prelude in { + + flags optimize=all_subs ; + + lin + +-- pos. may, neg. chay may - not always the proper forms --- + + QuestCl cl = {s = \\p => cl.s ! Pos ++ polStr chay_s p ++ m'ay_s} ; + +-- +-- QuestVP qp vp = +-- let cl = mkClause (qp.s ! Nom) {n = qp.n ; p = P3} vp +-- in {s = \\t,a,b,_ => cl.s ! t ! a ! b ! ODir} ; +-- +-- QuestSlash ip slash = +-- mkQuestion (ss (slash.c2 ++ ip.s ! Acc)) slash ; +-- --- stranding in ExratTha +-- +-- QuestIAdv iadv cl = mkQuestion iadv cl ; +-- +-- QuestIComp icomp np = +-- mkQuestion icomp (mkClause (np.s ! Nom) np.a (predAux auxBe)) ; +-- +-- +-- PrepIP p ip = {s = p.s ++ ip.s ! Nom} ; +-- +-- AdvIP ip adv = { +-- s = \\c => ip.s ! c ++ adv.s ; +-- n = ip.n +-- } ; +-- +-- IDetCN idet num ord cn = { +-- s = \\c => idet.s ++ num.s ++ ord.s ++ cn.s ! idet.n ! c ; +-- n = idet.n +-- } ; +-- +-- CompIAdv a = a ; +-- +} diff --git a/next-lib/src/thai/RelativeTha.gf b/next-lib/src/thai/RelativeTha.gf new file mode 100644 index 000000000..1dadd0ccf --- /dev/null +++ b/next-lib/src/thai/RelativeTha.gf @@ -0,0 +1,48 @@ +--concrete RelativeTha of Relative = CatTha ** open ResTha in { +-- +-- flags optimize=all_subs ; +-- +-- lin +-- +-- RelCl cl = { +-- s = \\t,a,p,_ => "such" ++ "that" ++ cl.s ! t ! a ! p ! ODir ; +-- c = Nom +-- } ; +-- +-- RelVP rp vp = { +-- s = \\t,ant,b,ag => +-- let +-- agr = case rp.a of { +-- RNoAg => ag ; +-- RAg a => a +-- } ; +-- cl = mkClause (rp.s ! RC Nom) agr vp +-- in +-- cl.s ! t ! ant ! b ! ODir ; +-- c = Nom +-- } ; +-- +---- Pied piping: "at which we are looking". Stranding and empty +---- relative are defined in $ExtraTha.gf$ ("that we are looking at", +---- "we are looking at"). +-- +-- RelSlash rp slash = { +-- s = \\t,a,p,_ => slash.c2 ++ rp.s ! RPrep ++ slash.s ! t ! a ! p ! ODir ; +-- c = Acc +-- } ; +-- +-- FunRP p np rp = { +-- s = \\c => np.s ! Acc ++ p.s ++ rp.s ! RPrep ; +-- a = RAg np.a +-- } ; +-- +-- IdRP = { +-- s = table { +-- RC Gen => "whose" ; +-- RC _ => "that" ; +-- RPrep => "which" +-- } ; +-- a = RNoAg +-- } ; +-- +--} diff --git a/next-lib/src/thai/ResTha.gf b/next-lib/src/thai/ResTha.gf new file mode 100644 index 000000000..6c1657151 --- /dev/null +++ b/next-lib/src/thai/ResTha.gf @@ -0,0 +1,484 @@ +--# -path=.:../abstract:../common:../../prelude + +--1 Thai auxiliary operations. +-- +---- This module contains operations that are needed to make the +---- resource syntax work. To define everything that is needed to +---- implement $Test$, it moreover contains regular lexical +---- patterns needed for $Lex$. +-- +resource ResTha = ParamX ** open StringsTha, Prelude in { + + oper + +-- noun and classifier + + Noun = {s,c : Str} ; + + mkN : Str -> Str -> Noun = \s,c -> {s = s ; c = c} ; + +-- before and after classifier; whether classifier needed (default) + + Determiner = {s1, s2 : Str ; hasC : Bool} ; + + mkDet : Str -> Str -> Determiner = + \s,c -> {s1 = s ; s2 = c ; hasC = True} ; + +-- Part before and after negation (mai_s) + + Verb = {s1,s2 : Str} ; + + resV : Str -> Str -> Verb = \s,c -> {s1 = s ; s2 = c} ; + + regV : Str -> Verb = \s -> resV [] s ; + + dirV2 : Verb -> Verb ** {c2 : Str} = \v -> v ** {c2 = []} ; + +-- Auxiliary verbs, according to order and negation. +-- The three types are $VV may VP | may VV VP | VP may VV$ + + param VVTyp = VVPre | VVMid | VVPost ; + + oper VVerb = {s : Str ; typ : VVTyp} ; + +-- Verb phrases: form negation and question, too. + + VP = { + s : Polarity => Str + } ; + + mkVP : Verb -> VP = \v -> { + s = \\p => v.s1 ++ polStr may_s p ++ v.s2 + } ; + + insertObject : Str -> VP -> VP = \np,vp -> { + s = \\p => vp.s ! p ++ np + } ; + + polStr : Str -> Polarity -> Str = \m,p -> case p of { + Pos => [] ; + Neg => m + } ; + +-- flags optimize=all ; +-- +-- +---- Some parameters, such as $Number$, are inherited from $ParamX$. +-- +----2 For $Noun$ +-- +---- This is the worst-case $Case$ needed for pronouns. +-- +-- param +-- Case = Nom | Acc | Gen ; +-- +---- Agreement of $NP$ is a record. We'll add $Gender$ later. +-- +-- oper +-- Agr = {n : Number ; p : Person} ; +-- +-- param +-- Gender = Neutr | Masc | Fem ; +-- +----2 For $Verb$ +-- +---- Only these five forms are needed for open-lexicon verbs. +-- +-- param +-- VForm = +-- VInf +-- | VPres +-- | VPPart +-- | VPresPart +-- | VPast --# notpresent +-- ; +-- +---- Auxiliary verbs have special negative forms. +-- +-- VVForm = +-- VVF VForm +-- | VVPresNeg +-- | VVPastNeg --# notpresent +-- ; +-- +---- The order of sentence is needed already in $VP$. +-- +-- Order = ODir | OQuest ; +-- +-- +----2 For $Adjective$ +-- +-- AForm = AAdj Degree | AAdv ; +-- +----2 For $Relative$ +-- +-- RAgr = RNoAg | RAg {n : Number ; p : Person} ; +-- RCase = RPrep | RC Case ; +-- +----2 For $Numeral$ +-- +-- CardOrd = NCard | NOrd ; +-- DForm = unit | teen | ten ; +-- +----2 Transformations between parameter types +-- +-- oper +-- agrP3 : Number -> Agr = \n -> +-- {n = n ; p = P3} ; +-- +-- conjAgr : Agr -> Agr -> Agr = \a,b -> { +-- n = conjNumber a.n b.n ; +-- p = conjPerson a.p b.p +-- } ; +-- +---- For $Lex$. +-- +---- For each lexical category, here are the worst-case constructors. +-- +-- mkNoun : (_,_,_,_ : Str) -> {s : Number => Case => Str} = +-- \man,mans,men,mens -> { +-- s = table { +-- Sg => table { +-- Gen => mans ; +-- _ => man +-- } ; +-- Pl => table { +-- Gen => mens ; +-- _ => men +-- } +-- } +-- } ; +-- +-- mkAdjective : (_,_,_,_ : Str) -> {s : AForm => Str} = +-- \good,better,best,well -> { +-- s = table { +-- AAdj Posit => good ; +-- AAdj Compar => better ; +-- AAdj Superl => best ; +-- AAdv => well +-- } +-- } ; +-- +-- mkVerb : (_,_,_,_,_ : Str) -> Verb = +-- \go,goes,went,gone,going -> { +-- s = table { +-- VInf => go ; +-- VPres => goes ; +-- VPast => went ; --# notpresent +-- VPPart => gone ; +-- VPresPart => going +-- } ; +-- isRefl = False +-- } ; +-- +-- mkIP : (i,me,my : Str) -> Number -> {s : Case => Str ; n : Number} = +-- \i,me,my,n -> let who = mkNP i me my n P3 in {s = who.s ; n = n} ; +-- +-- mkNP : (i,me,my : Str) -> Number -> Person -> {s : Case => Str ; a : Agr} = +-- \i,me,my,n,p -> { +-- s = table { +-- Nom => i ; +-- Acc => me ; +-- Gen => my +-- } ; +-- a = { +-- n = n ; +-- p = p +-- } +-- } ; +-- +---- These functions cover many cases; full coverage inflectional patterns are +---- in $MorphoTha$. +-- +-- regN : Str -> {s : Number => Case => Str} = \car -> +-- mkNoun car (car + "'s") (car + "s") (car + "s'") ; +-- +-- regA : Str -> {s : AForm => Str} = \warm -> +-- mkAdjective warm (warm + "er") (warm + "est") (warm + "ly") ; +-- +-- regV : Str -> Verb = \walk -> +-- mkVerb walk (walk + "s") (walk + "ed") (walk + "ed") (walk + "ing") ; +-- +-- regNP : Str -> Number -> {s : Case => Str ; a : Agr} = \that,n -> +-- mkNP that that (that + "'s") n P3 ; +-- +---- We have just a heuristic definition of the indefinite article. +---- There are lots of exceptions: consonantic "e" ("euphemism"), consonantic +---- "o" ("one-sided"), vocalic "u" ("umbrella"). +-- +-- artIndef = pre { +-- "a" ; +-- "an" / strs {"a" ; "e" ; "i" ; "o" ; "A" ; "E" ; "I" ; "O" } +-- } ; +-- +-- artDef = "the" ; +-- +---- For $Verb$. +-- +-- Verb : Type = { +-- s : VForm => Str ; +-- isRefl : Bool +-- } ; +-- +-- param +-- CPolarity = +-- CPos +-- | CNeg Bool ; -- contracted or not +-- +-- oper +-- contrNeg : Bool -> Polarity -> CPolarity = \b,p -> case p of { +-- Pos => CPos ; +-- Neg => CNeg b +-- } ; +-- +-- VerbForms : Type = +-- Tense => Anteriority => CPolarity => Order => Agr => {fin, inf : Str} ; +-- +-- VP : Type = { +-- s : VerbForms ; +-- prp : Str ; -- present participle +-- inf : Str ; -- the infinitive form ; VerbForms would be the logical place +-- ad : Str ; -- sentential adverb +-- s2 : Agr => Str -- complement +-- } ; +-- +-- +-- predV : Verb -> VP = \verb -> { +-- s = \\t,ant,b,ord,agr => +-- let +-- inf = verb.s ! VInf ; +-- fin = presVerb verb agr ; +-- part = verb.s ! VPPart ; +-- in +-- case of { +-- => vf fin [] ; +-- => vf (does agr) inf ; +-- => vf (have agr) part ; --# notpresent +-- => vfn c (have agr) (havent agr) part ; --# notpresent +-- => vf (verb.s ! VPast) [] ; --# notpresent +-- => vf "did" inf ; --# notpresent +-- => vfn c "did" "didn't" inf ; --# notpresent +-- => vf "had" part ; --# notpresent +-- => vfn c "had" "hadn't" part ; --# notpresent +-- => vf "will" inf ; --# notpresent +-- => vfn c "will" "won't" inf ; --# notpresent +-- => vf "will" ("have" ++ part) ; --# notpresent +-- => vfn c "will" "won't"("have" ++ part) ; --# notpresent +-- => vf "would" inf ; --# notpresent +-- => vfn c "would" "wouldn't" inf ; --# notpresent +-- => vf "would" ("have" ++ part) ; --# notpresent +-- => vfn c "would" "wouldn't" ("have" ++ part) ; --# notpresent +-- => vfn c (does agr) (doesnt agr) inf +-- } ; +-- prp = verb.s ! VPresPart ; +-- inf = verb.s ! VInf ; +-- ad = [] ; +-- s2 = \\a => if_then_Str verb.isRefl (reflPron ! a) [] +-- } ; +-- +-- predAux : Aux -> VP = \verb -> { +-- s = \\t,ant,cb,ord,agr => +-- let +-- b = case cb of { +-- CPos => Pos ; +-- _ => Neg +-- } ; +-- inf = verb.inf ; +-- fin = verb.pres ! b ! agr ; +-- finp = verb.pres ! Pos ! agr ; +-- part = verb.ppart ; +-- in +-- case of { +-- => vf (have agr) part ; --# notpresent +-- => vfn c (have agr) (havent agr) part ; --# notpresent +-- => vf (verb.past ! b ! agr) [] ; --# notpresent +-- => vfn c (verb.past!Pos!agr)(verb.past!Neg!agr) [] ; --# notpresent +-- => vf "had" part ; --# notpresent +-- => vfn c "had" "hadn't" part ; --# notpresent +-- => vf "will" inf ; --# notpresent +-- => vfn c "will" "won't" inf ; --# notpresent +-- => vf "will" ("have" ++ part) ; --# notpresent +-- => vfn c "will" "won't"("have" ++ part) ; --# notpresent +-- => vf "would" inf ; --# notpresent +-- => vfn c "would" "wouldn't" inf ; --# notpresent +-- => vf "would" ("have" ++ part) ; --# notpresent +-- => vfn c "would" "wouldn't" ("have" ++ part) ; --# notpresent +-- => vf fin [] ; +-- => vfn c finp fin [] +-- } ; +-- prp = verb.prpart ; +-- inf = verb.inf ; +-- ad = [] ; +-- s2 = \\_ => [] +-- } ; +-- +-- vf : Str -> Str -> {fin, inf : Str} = \x,y -> vfn True x x y ; +-- +-- vfn : Bool -> Str -> Str -> Str -> {fin, inf : Str} = \contr,x,y,z -> +-- case contr of { +-- True => {fin = y ; inf = z} ; +-- False => {fin = x ; inf = "not" ++ z} +-- } ; +-- +-- insertObj : (Agr => Str) -> VP -> VP = \obj,vp -> { +-- s = vp.s ; +-- prp = vp.prp ; +-- inf = vp.inf ; +-- ad = vp.ad ; +-- s2 = \\a => vp.s2 ! a ++ obj ! a +-- } ; +-- +----- The adverb should be before the finite verb. +-- +-- insertAdV : Str -> VP -> VP = \adv,vp -> { +-- s = vp.s ; +-- prp = vp.prp ; +-- inf = vp.inf ; +-- ad = vp.ad ++ adv ; +-- s2 = \\a => vp.s2 ! a +-- } ; +-- +---- +-- +-- predVV : {s : VVForm => Str ; isAux : Bool} -> VP = \verb -> +-- let verbs = verb.s +-- in +-- case verb.isAux of { +-- True => predAux { +-- pres = table { +-- Pos => \\_ => verbs ! VVF VPres ; +-- Neg => \\_ => verbs ! VVPresNeg +-- } ; +-- past = table { --# notpresent +-- Pos => \\_ => verbs ! VVF VPast ; --# notpresent +-- Neg => \\_ => verbs ! VVPastNeg --# notpresent +-- } ; --# notpresent +-- inf = verbs ! VVF VInf ; +-- ppart = verbs ! VVF VPPart ; +-- prpart = verbs ! VVF VPresPart ; +-- } ; +-- _ => predV {s = \\vf => verbs ! VVF vf ; isRefl = False} +-- } ; +-- +-- presVerb : {s : VForm => Str} -> Agr -> Str = \verb -> +-- agrVerb (verb.s ! VPres) (verb.s ! VInf) ; +-- +-- infVP : Bool -> VP -> Agr -> Str = \isAux,vp,a -> +-- vp.ad ++ if_then_Str isAux [] "to" ++ +-- vp.inf ++ vp.s2 ! a ; +-- +-- agrVerb : Str -> Str -> Agr -> Str = \has,have,agr -> +-- case agr of { +-- {n = Sg ; p = P3} => has ; +-- _ => have +-- } ; +-- +-- have = agrVerb "has" "have" ; +-- havent = agrVerb "hasn't" "haven't" ; +-- does = agrVerb "does" "do" ; +-- doesnt = agrVerb "doesn't" "don't" ; +-- +-- Aux = { +-- pres : Polarity => Agr => Str ; +-- past : Polarity => Agr => Str ; --# notpresent +-- inf,ppart,prpart : Str +-- } ; +-- +-- auxBe : Aux = { +-- pres = \\b,a => case of { +-- => "am" ; +-- => ["am not"] ; --- am not I +-- _ => agrVerb (posneg b "is") (posneg b "are") a +-- } ; +-- past = \\b,a => case a of { --# notpresent +-- {n = Sg ; p = P1|P3} => (posneg b "was") ; --# notpresent +-- _ => (posneg b "were") --# notpresent +-- } ; --# notpresent +-- inf = "be" ; +-- ppart = "been" ; +-- prpart = "being" +-- } ; +-- +-- posneg : Polarity -> Str -> Str = \p,s -> case p of { +-- Pos => s ; +-- Neg => s + "n't" +-- } ; +-- +-- conjThat : Str = "that" ; +-- +-- reflPron : Agr => Str = table { +-- {n = Sg ; p = P1} => "myself" ; +-- {n = Sg ; p = P2} => "yourself" ; +-- {n = Sg ; p = P3} => "itself" ; ---- +-- {n = Pl ; p = P1} => "ourselves" ; +-- {n = Pl ; p = P2} => "yourselves" ; +-- {n = Pl ; p = P3} => "themselves" +-- } ; +-- +---- For $Sentence$. +-- +-- Clause : Type = { +-- s : Tense => Anteriority => CPolarity => Order => Str +-- } ; +-- +-- mkClause : Str -> Agr -> VP -> Clause = +-- \subj,agr,vp -> { +-- s = \\t,a,b,o => +-- let +-- verb = vp.s ! t ! a ! b ! o ! agr ; +-- compl = vp.s2 ! agr +-- in +-- case o of { +-- ODir => subj ++ verb.fin ++ vp.ad ++ verb.inf ++ compl ; +-- OQuest => verb.fin ++ subj ++ vp.ad ++ verb.inf ++ compl +-- } +-- } ; +-- +-- +---- For $Numeral$. +-- +-- mkNum : Str -> Str -> Str -> Str -> {s : DForm => CardOrd => Str} = +-- \two, twelve, twenty, second -> +-- {s = table { +-- unit => table {NCard => two ; NOrd => second} ; +-- teen => \\c => mkCard c twelve ; +-- ten => \\c => mkCard c twenty +-- } +-- } ; +-- +-- regNum : Str -> {s : DForm => CardOrd => Str} = +-- \six -> mkNum six (six + "teen") (six + "ty") (regOrd six) ; +-- +-- regCardOrd : Str -> {s : CardOrd => Str} = \ten -> +-- {s = table {NCard => ten ; NOrd => regOrd ten}} ; +-- +-- mkCard : CardOrd -> Str -> Str = \c,ten -> +-- (regCardOrd ten).s ! c ; +-- +-- regOrd : Str -> Str = \ten -> +-- case last ten of { +-- "y" => init ten + "ieth" ; +-- _ => ten + "th" +-- } ; +-- +-- mkQuestion : +-- {s : Str} -> Clause -> +-- {s : Tense => Anteriority => CPolarity => QForm => Str} = \wh,cl -> +-- { +-- s = \\t,a,p => +-- let +-- cls = cl.s ! t ! a ! p ; +-- why = wh.s +-- in table { +-- QDir => why ++ cls ! OQuest ; +-- QIndir => why ++ cls ! ODir +-- } +-- } ; +-- +---- for VP conjunction +-- +-- param +-- VPIForm = VPIInf | VPIPPart ; +-- +-- +} diff --git a/next-lib/src/thai/SentenceTha.gf b/next-lib/src/thai/SentenceTha.gf new file mode 100644 index 000000000..3d46348c8 --- /dev/null +++ b/next-lib/src/thai/SentenceTha.gf @@ -0,0 +1,61 @@ +concrete SentenceTha of Sentence = CatTha ** + open Prelude, StringsTha, ResTha in { + + flags optimize=all_subs ; + + lin + + PredVP np vp = {s = \\p => np.s ++ vp.s ! p} ; + +-- PredSCVP sc vp = mkClause sc.s (agrP3 Sg) vp ; + + ImpVP vp = { + s = table { + Pos => vp.s ! Pos ++ si_s ; + Neg => yaa_s ++ vp.s ! Pos + } + } ; +-- SlashV2 np v2 = +-- mkClause (np.s ! Nom) np.a (predV v2) ** {c2 = v2.c2} ; +-- +-- SlashVVV2 np vv v2 = +-- mkClause (np.s ! Nom) np.a +-- (insertObj (\\a => infVP vv.isAux (predV v2) a) (predVV vv)) ** +-- {c2 = v2.c2} ; +-- +-- AdvSlash slash adv = { +-- s = \\t,a,b,o => slash.s ! t ! a ! b ! o ++ adv.s ; +-- c2 = slash.c2 +-- } ; +-- +-- SlashPrep cl prep = cl ** {c2 = prep.s} ; +-- +-- EmbedS s = {s = conjThat ++ s.s} ; +-- EmbedQS qs = {s = qs.s ! QIndir} ; +-- EmbedVP vp = {s = infVP False vp (agrP3 Sg)} ; --- agr +-- + UseCl t p cl = {s = t.s ++ p.s ++ cl.s ! p.p} ; + UseQCl t p cl = { + s = \\q => t.s ++ p.s ++ + case q of {QIndir => waa_s ; _ => []} ++ + cl.s ! p.p + } ; +-- UseRCl t a p cl = { +-- s = \\r => t.s ++ a.s ++ p.s ++ cl.s ! t.t ! a.a ! ctr p.p ! r ; +-- c = cl.c +-- } ; +-- +-- AdvS a s = {s = a.s ++ "," ++ s.s} ; +-- +-- oper +-- ctr = contrNeg True ; -- contracted negations +--} +-- +--{- +----- todo: tense of embedded Slash +-- +-- SlashVSS np vs s = +-- mkClause (np.s ! Nom) np.a +-- (insertObj (\\_ => conjThat ++ s.s) (predV vs)) ** +-- {c2 = s.c2} ; +} diff --git a/next-lib/src/thai/StringsTha.gf b/next-lib/src/thai/StringsTha.gf new file mode 100644 index 000000000..1f7f932f6 --- /dev/null +++ b/next-lib/src/thai/StringsTha.gf @@ -0,0 +1,87 @@ +-- The only place where literal Thai strings are defined +-- (except for Lexicon and Structural). +-- Convert this into StringsThai by 'make strings' + +resource StringsTha = { + +oper + +aphai_s = "Op3ay" ; -- excuse2 +baan_s = "bT2a:n" ; -- house +biar_s = "ebi:OrK" ; -- beer +ca_s = "ca." ; -- Modal +cet_s = "ecSd" ; -- seven +chan_s = "c1an" ; -- I +chay_s = "a%c2T1" ; -- yes +cheut_s = "ec2v:T2t" ; -- shirt2 +chuay_s = "c2T1wy" ; -- help1 +di_s = "di" ; -- I (fem)1 +dii_s = "di:" ; -- hello2 +duay_s = "dT2wy" ; -- help2 +dvm_s = "dvm" ; -- drink +et_s = "eOSd" ; -- one' +haa_s = "hT2a:" ; -- five +hay_s = "a%hT2" ; -- give +hoog_s = "hT2Og" ; -- room +hok_s = "ho?k" ; -- six +jai_s = "a%j" ; -- understand2 +kaaw_s = "eka:" ; -- nine +kew_s = "e'kT2w" ; -- glass (drink Classif) +khaw_s = "ek1a:" ; -- he +khon_s = "k2n" ; -- people Classif +khoo_s = "k1O" ; -- please +khoog_s = "k1Og" ; -- Possessive +khoop_s = "k1Ob" ; -- thank +khow_s = "ek1T2w" ; -- understand1 +khun_s = "k2un'" ; -- you +koon_s = "kT1On" ; -- bye2 +laa_s = "la:" ; -- bye1 +lag_s = "hlag" ; -- houses Classif +lap_s = "hlab" ; -- sleep2 +lem_s = "elT1m" ; -- books Classif +may_s = "a&mT1" ; -- not +m'ay_s = "a&hm" ; -- Question +mvvn_s = "hmv:T1n" ; -- ten thousand +nag_s = "hna.g" ; -- book1 +nai_s = "a&hn" ; -- where2 +nam_s = "na+" ; -- water +nan_s = "naT2n" ; -- that +nii_s = "ni:T2" ; -- this +nit_s = "nid" ; -- little1 +noon_s = "nOn" ; -- sleep1 +noi_s = "hnT1Oy" ; -- little2 +nvg_s = "hnvg" ; -- one +pay_s = "a&p" ; -- go +peet_s = "e'pd" ; -- eight +pen_s = "epSn" ; -- be, can-know +phan_s = "p2an" ; -- thousand +phom_s = "p1m" ; -- I (masc) +puu_s = "p1u:T2" ; -- woman1 +rai_s = "a&r" ; -- how-much2 +rak_s = "rak" ; -- love +raw_s = "era:" ; -- we +rooy_s = "rT2Oy" ; -- hundred +saam_s = "sa:m" ; -- three +sawat_s = "swas" ; -- hello1 +seen_s = "e'sn" ; -- hundred thousand +seua_s = "esv:T2O" ; -- shirt1 +si_s = "s'i" ; -- Imperative +sii_s = "si:T1" ; -- four +sip_s = "sib" ; -- ten +soog_s = "sOg" ; -- two +svv_s = "sv:O" ; -- book2 +thii_s = "t5i:T1" ; -- Ord +thoot_s = "o:t5r'" ; -- sorry2 +thao_s = "et5T1a:" ; -- how-much1 +tog_s = "tT2Og" ; -- must +waa_s = "wT1a:" ; -- that Conj +way_s = "a&hw" ; -- can-potent +yaa_s = "OyT1a:" ; -- Neg Imper +yaak_s = "Oya:k" ; -- want +yay_s = "a%hy'" ; -- big +yig_s = "hy'ig" ; -- woman2 +yii_s = "yi:T1" ; -- two' +yin_s = "yin" ; -- you're-welcome1 +yuu_s = "yu:" ; -- where0(?) + +} diff --git a/next-lib/src/thai/StructuralTha.gf b/next-lib/src/thai/StructuralTha.gf new file mode 100644 index 000000000..1767c4652 --- /dev/null +++ b/next-lib/src/thai/StructuralTha.gf @@ -0,0 +1,110 @@ +concrete StructuralTha of Structural = CatTha ** + open StringsTha, ResTha, Prelude in { + + flags optimize=all ; + + lin +-- above_Prep = ss "above" ; +-- after_Prep = ss "after" ; +-- all_Predet = ss "all" ; +-- almost_AdA, almost_AdN = ss "almost" ; +-- although_Subj = ss "although" ; +-- always_AdV = ss "always" ; +-- and_Conj = ss "and" ** {n = Pl} ; +-- because_Subj = ss "because" ; +-- before_Prep = ss "before" ; +-- behind_Prep = ss "behind" ; +-- between_Prep = ss "between" ; +-- both7and_DConj = sd2 "both" "and" ** {n = Pl} ; +-- but_PConj = ss "but" ; +-- by8agent_Prep = ss "by" ; +-- by8means_Prep = ss "by" ; + can8know_VV = {s = pen_s ; typ = VVPost} ; + can_VV = {s = way_s ; typ = VVPost} ; +-- during_Prep = ss "during" ; +-- either7or_DConj = sd2 "either" "or" ** {n = Sg} ; +-- everybody_NP = regNP "everybody" Sg ; +-- every_Det = mkDeterminer Sg "every" ; +-- everything_NP = regNP "everything" Sg ; +-- everywhere_Adv = ss "everywhere" ; +-- few_Det = mkDeterminer Pl "few" ; +-- first_Ord = ss "first" ; +-- for_Prep = ss "for" ; +-- from_Prep = ss "from" ; + he_Pron = ss khaw_s ; +-- here_Adv = ss "here" ; +-- here7to_Adv = ss ["to here"] ; +-- here7from_Adv = ss ["from here"] ; +-- how_IAdv = ss "how" ; +-- how8many_IDet = mkDeterminer Pl ["how many"] ; +-- if_Subj = ss "if" ; +-- in8front_Prep = ss ["in front of"] ; + i_Pron = ss chan_s ; +-- in_Prep = ss "in" ; +-- it_Pron = mkNP "it" "it" "its" Sg P3 ; +-- less_CAdv = ss "less" ; +-- many_Det = mkDeterminer Pl "many" ; +-- more_CAdv = ss "more" ; +-- most_Predet = ss "most" ; +-- much_Det = mkDeterminer Sg "much" ; + must_VV = {s = tog_s ; typ = VVPre} ; +-- no_Phr = ss "no" ; +-- on_Prep = ss "on" ; +-- one_Quant = mkDeterminer Sg "one" ; +-- only_Predet = ss "only" ; +-- or_Conj = ss "or" ** {n = Sg} ; +-- otherwise_PConj = ss "otherwise" ; +-- part_Prep = ss "of" ; +-- please_Voc = ss "please" ; +-- possess_Prep = ss "of" ; +-- quite_Adv = ss "quite" ; + she_Pron = ss khaw_s ; +-- so_AdA = ss "so" ; +-- somebody_NP = regNP "somebody" Sg ; +-- someSg_Det = mkDeterminer Sg "some" ; +-- somePl_Det = mkDeterminer Pl "some" ; +-- something_NP = regNP "something" Sg ; +-- somewhere_Adv = ss "somewhere" ; + that_Quant = ss nan_s ** {hasC = True} ; +-- that_NP = regNP "that" Sg ; +-- there_Adv = ss "there" ; +-- there7to_Adv = ss "there" ; +-- there7from_Adv = ss ["from there"] ; +-- therefore_PConj = ss "therefore" ; +-- these_NP = regNP "these" Pl ; +-- they_Pron = mkNP "they" "them" "their" Pl P3 ; +-- this_Quant = mkQuant "this" "these" ; +-- this_NP = regNP "this" Sg ; +-- those_NP = regNP "those" Pl ; +-- through_Prep = ss "through" ; +-- too_AdA = ss "too" ; +-- to_Prep = ss "to" ; +-- under_Prep = ss "under" ; +-- very_AdA = ss "very" ; + want_VV = {s = yaak_s ; typ = VVMid} ; + we_Pron = ss raw_s ; +-- whatPl_IP = mkIP "what" "what" "what's" Sg ; +-- whatSg_IP = mkIP "what" "what" "what's" Sg ; +-- when_IAdv = ss "when" ; +-- when_Subj = ss "when" ; +-- where_IAdv = ss "where" ; +-- whichPl_IDet = mkDeterminer Pl ["which"] ; +-- whichSg_IDet = mkDeterminer Sg ["which"] ; +-- whoSg_IP = mkIP "who" "whom" "whose" Sg ; +-- whoPl_IP = mkIP "who" "whom" "whose" Pl ; +-- why_IAdv = ss "why" ; +-- without_Prep = ss "without" ; +-- with_Prep = ss "with" ; +-- yes_Phr = ss "yes" ; + youSg_Pron = ss khun_s ; + youPl_Pron = ss khun_s ; + youPol_Pron = ss khun_s ; +-- +-- +--oper +-- mkQuant : Str -> Str -> {s : Number => Str} = \x,y -> { +-- s = table Number [x ; y] +-- } ; +-- +} + diff --git a/next-lib/src/thai/VerbTha.gf b/next-lib/src/thai/VerbTha.gf new file mode 100644 index 000000000..b894a04f9 --- /dev/null +++ b/next-lib/src/thai/VerbTha.gf @@ -0,0 +1,51 @@ +concrete VerbTha of Verb = CatTha ** open ResTha, StringsTha, Prelude in { + + flags optimize=all_subs ; + + lin + UseV = mkVP ; +-- ComplV2 v np = insertObject (v.c2 ++ np.s) (mkVP v) ; +-- ComplV3 v np np2 = insertObject (v.c2 ++ np.s ++ v.c3 ++ np2.s) (mkVP v) ; + + ComplVV vv vp = { + s = \\p => + let + neg = polStr may_s p ; + v = vp.s ! Pos + in + case vv.typ of { + VVPre => vv.s ++ neg ++ v ; + VVMid => neg ++ vv.s ++ v ; + VVPost => v ++ neg ++ vv.s + } + } ; + +-- +-- ComplVS v s = insertObj (\\_ => conjThat ++ s.s) (predV v) ; +-- ComplVQ v q = insertObj (\\_ => q.s ! QIndir) (predV v) ; +-- +-- ComplVA v ap = insertObj (ap.s) (predV v) ; +-- ComplV2A v np ap = +-- insertObj (\\_ => v.c2 ++ np.s ! Acc ++ ap.s ! np.a) (predV v) ; +-- + UseComp comp = comp ; +-- +-- AdvVP vp adv = insertObj (\\_ => adv.s) vp ; +-- +-- AdVVP adv vp = insertAdV adv.s vp ; +-- +-- ReflV2 v = insertObj (\\a => v.c2 ++ reflPron ! a) (predV v) ; +-- +-- PassV2 v = insertObj (\\_ => v.s ! VPPart) (predAux auxBe) ; +-- +-- UseVS, UseVQ = \vv -> {s = vv.s ; c2 = [] ; isRefl = vv.isRefl} ; + + CompAP ap = {s = \\p => polStr may_s p ++ ap.s} ; + CompNP np = {s = table { + Pos => pen_s ++ np.s ; + Neg => may_s ++ chay_s ++ np.s + } + } ; + CompAdv a = {s = \\p => polStr may_s p ++ a.s} ; --- ?? + +}