From 246adbd3a403c7097feab0d6623a4b070637f9bb Mon Sep 17 00:00:00 2001 From: KennyMcFly <35669096+KennyMcFly@users.noreply.github.com> Date: Thu, 31 Aug 2023 10:48:06 +0200 Subject: [PATCH] disam, func, dep og stems --- src/cg3/dep_KalEng.cg3 | 7 ++++--- src/cg3/disambiguator.cg3 | 25 ++++++++++++++++--------- src/fst/stems/nouns.lexc | 4 ++-- src/fst/stems/verbs.lexc | 4 ++-- 4 files changed, 24 insertions(+), 16 deletions(-) diff --git a/src/cg3/dep_KalEng.cg3 b/src/cg3/dep_KalEng.cg3 index fc0ed1a3b..f6bf3238e 100644 --- a/src/cg3/dep_KalEng.cg3 +++ b/src/cg3/dep_KalEng.cg3 @@ -764,9 +764,10 @@ TO (*1A CONT + 3SgO BARRIER KOMMA OR VFIN LINK *1 INTRANSVERB + 3Sg BARRIER KOMM # #MAP:M_uløselighed0004 @POSS> (/"Ukraine"\ Sem/Geo\ iProp\ MIU\ Der/nn\ N\ Rel\ Pl/l) IF (1 (/"timmi"\ Gram/IV\ TAR\ Der/vv\ Gram/IV\ TUQ/l) LINK *0 ("nakkar") OR ("nakkartip"));#Iranimi inuppaaluit UKRAINEMIUT timmisartuannik nakkartitseqataasimasutut tigusarineqarsimapput # -#MAP:M_uløselighed0005 @ADVL> TARGET OTHERTIME + Abs + Pl IF (-1 Abs + Pl - TIME - Gram/Dem LINK *1 VERB-NOTCONT + 3PlO OR VERB-NOTCONT + INTRANSVERB + 3Pl BARRIER CONT + 3PlO) #Aappaatigulli tupaallannarpallaanngilaq suliakka UKIORPASSUIT suliarisarsimagakkit. Der tages forbehold for at at TIME kan være objekt for en CONT og at TIME selv kan have en @Pron>N -#(NEGATE 0 ADVERBIEL OR Der/vn + Abs + Pl LINK *-1 Abs + Pl BARRIER (*) - KOMMA - (Rel Sg) - OBLIQUE - FossileredeCont) #Aningaasat, Novo Nordisk Fondenip coronamut atatillugu suliniutinut atorneqartussanngortitaanit KINGULLIIT Peqqinnissamut Naalakkersuisoqarfimmut tunniunneqarput. -#; +SETPARENT:M_uløselighed0005 @ADVL> + OTHERTIME + Abs + Pl IF (-1 Abs + Pl - TIME - Gram/Dem) +(NEGATE 0 ADVERBIEL OR Der/vn + Abs + Pl LINK *-1 Abs + Pl BARRIER (*) - KOMMA - (Rel Sg) - OBLIQUE - FossileredeCont) #Aningaasat, Novo Nordisk Fondenip coronamut atatillugu suliniutinut atorneqartussanngortitaanit KINGULLIIT Peqqinnissamut Naalakkersuisoqarfimmut tunniunneqarput. +(NEGATE 0 LIK + Abs + $$NUMERUS LINK -1 Abs + $$NUMERUS) #Nunat tamat suliniaqatigiiffiinit Laurent Gbagbo annertuumik akerlilerneqarpoq , nunamit aallaqqusaajunnaartinneqarluni , pigisaalu NALILLIT USA-miittut Tuluit Nunaanniittullu atorneqarsinnaajunnaartinneqarput. +TO (*1A VERB-NOTCONT + 3PlO OR VERB-NOTCONT + INTRANSVERB + 3Pl BARRIER CONT + 3PlO); #Aappaatigulli tupaallannarpallaanngilaq suliakka UKIORPASSUIT suliarisarsimagakkit. Der tages forbehold for at at TIME kan være objekt for en CONT og at TIME selv kan have en @Pron>N # ##Strukturen SUBJ=OBJ=CONT+3SgO=V+3Sg næsten umulig at adskille fra OBJ=@N<=CONT+3SgO=V+3Sg jf. regel M0342 #MAP:M_uløselighed0008 @OBJ> TARGET Abs + Sg + IF (-1 Abs + )(1 CONT + 3SgO);#Ataatani guutipilussiortut palaserigaluaraat milutsitsisuata kristumiunngorsimasup kristumiut upperisaannik ajoqersorsimavaa, arnarlu taanna milutsitsisua UPPERISANI pillugu toqutaasimavoq. diff --git a/src/cg3/disambiguator.cg3 b/src/cg3/disambiguator.cg3 index ea7a87998..522b78234 100644 --- a/src/cg3/disambiguator.cg3 +++ b/src/cg3/disambiguator.cg3 @@ -775,6 +775,7 @@ LIST QQU_HTR_GUNNAIR = QQU\=HTR\=GUNNAIR ; LIST QQU_SSA = QQU\=SSA ; LIST QQU_TAR = QQU\=TAR ; LIST QQUSAAR = QQUSAAR ; +LIST R = R ; LIST RAAR = RAAR ; LIST RAR = RAR ; #LIST RARE = RARE ; @@ -4078,7 +4079,7 @@ SELECT:0019NY ("qanoq") + Adv IF (*1 IP OR ("ip") OR ("ilior") OR ("ili")); REMOVE:0020NX Interj IF (0 Prop)(1 Prop) ; #Jaaku Lyberth SELECT:0020NY Interj - ("tass") - ("tassa") - ("immaqa") IF (-1 BOS)(NEGATE 0 ("ila") + 4SgPoss + Rel LINK 1 POSSESSUM3)(NEGATE 0 KONCESSIV LINK *1 Par BARRIER V)(NEGATE 0 ("taama") OR ("qanoq") LINK *1 V BARRIER (*) - Adv - Conj) (NEGATE 0 Sem/Unit) -(NEGATE 0 ("sooruna") LINK *1 Par BARRIER V) #sooruna sammen med verbum er ikke en interjektion +(NEGATE 0 ("sooruna") LINK *1 Par Or Int BARRIER V) #sooruna sammen med verbum er ikke en interjektion (NEGATE 0 Adv - AA - UNA - UKU); #TASSA holdes uden for reglen og type 'Ilami illuanut iserpoq' ikke interjektion SKAL RAFFINERES YDERLIGERE 20180427. PL: immaqa tilføjet 20180526. Måske skal i stedet immaqa+Interj helt væk?? immaqa + Part og immaqa + Interj udkommenteret fra fst 20180526 REMOVE:0020AdHoc ("immaqa") + Pali ; #fjernet fra fst og vil ikke længere optræde efter update SELECT:7250 Interj IF (-1 BOS OR ("\,"))(1 EOS)(NEGATE 1 ("...")); #Ny 20180407 En isoleret partikel er en interjektion @@ -4851,6 +4852,7 @@ REMOVE:opremsning05 Abs + Pl - Num IF (0 Rel + Pl LINK *1 3PlPoss - ("Kalaallit (NEGATE 0 (/"tassa"\ U/l) + Abs + Pl LINK *-1 (Abs Pl) LINK *1 3PlO BARRIER V LINK *1 _TARGET_ + Abs + Pl LINK NEGATE *1 V - Gram/Exclm BARRIER KOMMA) #Sisimiuni politiit sapaatip akunnerani kingullermi qamutit aqqanillit normuiarpaat, TASSAANERUSUT taxat, Sisimiut aviisiat Silleq nittartakkamini allappoq. (NEGATE 0 ("IP")) #Issuttuutingaarmanga kavaajara singissartuata eqqaatigut alippoq, ILUANIITTULLU oqorsaatit timmissat meqqui katagalerlutik. (NEGATE *1 OBLIQUE BARRIER V LINK *1 3PlPoss BARRIER V LINK *1 Abs + TAQ BARRIER V) #Assoroorlunga isiginiarsareruttorlugit, malugilerpara soorlu KIINNAT unnummut qitittut akornanni takusakka takoqqillugit. +(NEGATE 0 Abs + Pl LINK 1 KOMMA LINK 1 3PlPoss) #Inuit uku anaanannut naggammik QANIUISUT, inuunerat qanormita oqitsigaa. ; REMOVE:opremsning06 (/"tassa"\ U/l) + Rel + Pl IF (*-1 (Abs Pl) LINK *1 3PlO BARRIER V LINK *1 _TARGET_ + Abs + Pl LINK NEGATE *1 V - Gram/Exclm BARRIER KOMMA); #Sisimiuni politiit sapaatip akunnerani kingullermi qamutit aqqanillit normuiarpaat, TASSAANERUSUT taxat, Sisimiut aviisiat Silleq nittartakkamini allappoq. @@ -13975,7 +13977,7 @@ SELECT:SeHumHum HUMAN - V IF (-*1 HUMVERB) (NEGATE 0 Par + &&IV_SUBJ& LINK *-1 Abs + &&IV_SUBJ& BARRIER (*) - OBLIQUE - Adv LINK *-1 ORATIOOBLVERB BARRIER V)#som foranstående NEGATE men uden paratakse (NEGATE 0 ("angakkoq" Pl) LINK *1 3PlO BARRIER V) (NEGATE 0 Sem/Hum + ("Danmark") LINK 0 Sem/Geo + ("Danmark") ) -(NEGATE 0 ("ittu") LINK 0 (/"ip"\ Gram/IV\ TUQ/l) LINK -1 ("immikkut") OR ("ima") OR ("taama"))#Tamanna qinersinissap eqqarsaatigineqalerneranut aallartitsivoq, tamatumalu kingunerisaanik Inatsisartut immikkut ITTUMIK qinersisoqarnissaa isumaqatigiissutigaat. +(NEGATE 0 ("ittu") OR ("ittoq") LINK 0 (/"ip"\ Gram/IV\ TUQ/l) LINK -1 ("immikkut") OR ("ima") OR ("taama") OR Lok)#Tamanna qinersinissap eqqarsaatigineqalerneranut aallartitsivoq, tamatumalu kingunerisaanik Inatsisartut immikkut ITTUMIK qinersisoqarnissaa isumaqatigiissutigaat. (NEGATE 0 ("inuiaqat") OR (/"inuiak"\ QATE/l)) ; SELECT:Se_illuNot_illoq ("illu") OR ("Illu") ; # "illu" mere sandsynlig end "illoq" @@ -14396,6 +14398,9 @@ REMOVE:5200BM ("ukiaq") + LIR IF (0 ("ukiar") + LIR); REMOVE:5200BN ("tarngup") + TAQ IF (0 ("tarngut")); REMOVE:5200BO NIR + Par + 4Sg IF (0 NIQ + &&POSS34_POSS& LINK -1 Rel + &&POSS34_POSS&);#Aamma atuaruk Canadap inuisa 20.000-t missaat orpippassuit IKUALLANNERINI qimagussorneqartariaqartut REMOVE:5200BP (/"puik"\ \(iSem/f-phys-h\ \)?GUP/l) IF (0 ("puigor")); +REMOVE:5200BQ ("sukaq") + NGA IF (0 ("sukap") + NGA); +REMOVE:5200BR ("megafon") - R + LIR IF (0 ("megafon") + R + LIR); #Angut MEGAFONILERLUNI inuppassuit akorninut pivoq megafonimullu nillialerluni. +REMOVE:5200BS ("attap") + LIR IF (0 ("attat") + LIR); #lav undtagelser når de kommer # Verbum eller verbum? SELECT:Sel0083 (INNAR NNGIT) IF (*1 ("aamma") ) ; #+INNAR=NNGIT aamma Vb fremfor NAR) @@ -14690,7 +14695,8 @@ REMOVE:5490EB ("kissar") IF (0 ("kissap") + Par); #Sukkut tamaana innersuaq alta SELECT:5490EC ("qulip") IF (*0 ("nujaq")); #Kissalaalermat nutsakka QULILLUGIT aallartippakka. REMOVE:5490ED ("maa") + GUNNAR IF (0 ("maajup") + NAR); REMOVE:5490EE ("uppip") + (/^i?Sem/capture/r) IF (0 (/^i?Sem/fall/r)); #Usandsynligt, lav undtagelser når de kommer -REMOVE:5490EF ("najor") + (/^i?Sem/breathe/r) IF (0 (/^i?Sem/lodge/r) LINK NOT *0 ); #Uannut naggammik inuulluaqqusisut NAJORLUGIT tupinnarpoq, sunaaffami inunnguit uannut taamak asannitsigisimapput. +REMOVE:5490EF1 ("najor") + (/^i?Sem/breathe/r) IF (0 (/^i?Sem/lodge/r) LINK NOT *0 ); #Uannut naggammik inuulluaqqusisut NAJORLUGIT tupinnarpoq, sunaaffami inunnguit uannut taamak asannitsigisimapput. +SELECT:5490EF2 ("najor") + (/^i?Sem/breathe/r) IF (*0 ); #Kavaajami iluata kaasarfianiit snapseeraq tiguaa NAJORSILLUNILU. REMOVE:5490EG ("allip") IF (0 ("alli")); #sandsynlighed, lav undtagelser når de kommer REMOVE:5490EH ("isip") IF (0 ("isi") LINK -1 ("qanoq")); #Arnaqaterpassuakka nalusakka aammalumi nalunngisakka, angutinik naapitsinerluttarsimasut paasisinnaaneq ajorpakka, qanoq ISILLUTIK angut iluamik inuttut ilikkarnagulu aapparilersinnaasarneraat. REMOVE:5490EI ("erner")(NEGATE 0 Cont); #bruges kun i kontemporativ @@ -14742,6 +14748,8 @@ REMOVE:5490FV ("nipaa") + (/[12][SP][gl]/r) OR (/"nipi"\ \(iSem/percep-l\ \)?U/l REMOVE:5490FX (/"matu"\ SI\ Der/nv\ Gram/IV/l) IF (0 (/"matu"\ Gram/TV\ HTR\ Der/vv\ Gram/IV/l) LINK *-1 Ins BARRIER V - CONT) ; REMOVE:5490FY (/"uutsisoq"/l) IF (0 (/"uup"\ Gram/TV\ HTR\ Der/vv\ Gram/IV\ TUQ\ Der/vn\ N/l) ) ; REMOVE:5490FZ (/"tunillap"\ Gram/TV\ TIP/l) + Gram/Pass IF (0 (/"tunillap"\ Gram/TV\ TIP/l) + Gram/Refl ); +REMOVE:5490GA ("uffar") + Gram/TV + Gram/Pass IF (0 ("uffar") + Gram/IV); +REMOVE:5490GB ("najor") + SI IF (0 ("najor") + HTR); # Verbum eller partikel? SELECT:5491A ("aarit") IF (-1 BOS); @@ -15393,7 +15401,7 @@ REMOVE:5859J ("seqqoq") IF (0 ("seqqulup"))(*0 ("nipi") OR ("nipaa")) ; REMOVE:5859K ("pujoq") + LIR IF (0 ("pujor") + LIR)(*0 ("ikuma") OR ("kissar") OR ("kissap")) ; REMOVE:5959L ("tun") + OLang/DAN IF (NEGATE *0 ANIMAL OR FOOD) ; REMOVE:5959M ("tuaq") + VIK IF (0 ("tuavior")) ; -REMOVE:5959N ("ittu") IF (0 ("ip") + TUQ) ; +REMOVE:5959N ("ittu") OR ("ittoq") IF (0 ("ip") + TUQ) ; REMOVE:5959O ("aap") + TAQ IF (0 ("aataq") OR ("aata")) ; REMOVE:5959P ("ersik") IF (0 ("ersi")) ; #Ersigamilu REMOVE:5959Q ("eqqoq") + IP IF (0 ("eqqumiip")) ; @@ -17430,7 +17438,10 @@ SELECT:Sem0291 ("ila") + (/^i?Sem/Hprof.*/r) IF (0 4SgPoss OR 4PlPoss OR POSSESS #SELECT:Sem0291 ("ila") + (/^i?Sem/amount.*/r) IF (0 Abs + POSSESSUM LINK -1 TUQ); #Immaqamiaasiit tamakku nappaateqartuusaarlutik nakorsiakulasut ILAAT. #SELECT:Sem0292 ("ila") + (/^i?Sem/amount.*/r) IF (0 ("ila") LINK 1 ("alla")); #Silassami oqaatsit atorneqartartut; natsimaneq, inoriaqqorneq, nakkisaaq, suffarik ILAALU allat amerlasuut isumaat nalusarpaat. -SELECT:Sem0295 ("ilu") + (/^i?Sem/Lh.*/r) IF (-1 ("illu")); #tassagooq illup ILUA sammerujoortarpaa. +REMOVE:Sem0295 ("ilu") + (/^i?Sem/an.*/r) IF (0 3SgPoss LINK -1 Rel + Sg - HUMAN); #Kavaajami ILUATA kaasarfianiit snapseeraq tiguaa najorsillunilu. +SELECT:Sem0296 ("ilu") + (/^i?Sem/Lh.*/r) IF (0 Via + Sg + 1SgPoss LINK 1 ("ippigusup")); #Massakkut allaffigivakkit naarmiut pillugu, tassami Larsilu naapittareernikuulluta ILUKKUT ippigusuttalerama nakorsiarama paasigakku sunaaffa gonorrémik tunillassimagaanga, ilimagilluinnaqqissaarparalu aamma illit nappaateqalersimassasutit. +SELECT:Sem0297 ("ilu") + (/^i?Sem/an.*/r) IF (0 Trm + Sg + 1SgPoss LINK 1 ("mangup")); #ILUNNUT mangutsisinnaasumik. +SELECT:Sem0298 ("ilu") + (/^i?Sem/an.*/r) IF (0 Via + Sg + 1SgPoss); #Mattusuumik, angisuumik takorluuinerma ILUKKUT quinaammerterujussuarpaanga. REMOVE:Sem0300 ("sammi") + (/^i?Sem/do.*/r) IF (0 ("sammi") + (/^i?Sem/work.*/r)); #find eksempel hvor det er "dyrke" @@ -17565,10 +17576,6 @@ REMOVE:Sem0480 ("peer") + (/^i?Sem/activity.*/r) IF (0 (/^i?Sem/remove.*/r)); #l REMOVE:Sem0485 (/"angi"\ Gram/IV\ TUUQ\ Der/vn\ Sem/ac-sign/l)(NEGATE 0 ("angi") LINK *0 ("naqip")); -SELECT:Sem0490 ("ilu") + (/^i?Sem/Lh.*/r) IF (0 Via + Sg + 1SgPoss LINK 1 ("ippigusup")); #Massakkut allaffigivakkit naarmiut pillugu, tassami Larsilu naapittareernikuulluta ILUKKUT ippigusuttalerama nakorsiarama paasigakku sunaaffa gonorrémik tunillassimagaanga, ilimagilluinnaqqissaarparalu aamma illit nappaateqalersimassasutit. -SELECT:Sem0491 ("ilu") + (/^i?Sem/an.*/r) IF (0 Trm + Sg + 1SgPoss LINK 1 ("mangup")); #ILUNNUT mangutsisinnaasumik. -SELECT:Sem0492 ("ilu") + (/^i?Sem/an.*/r) IF (0 Via + Sg + 1SgPoss); #Mattusuumik, angisuumik takorluuinerma ILUKKUT quinaammerterujussuarpaanga. - SELECT:Sem0495 ("iner") + (/^i?Sem/excrete.*/r) IF (*0 ("nipaarsaar") OR ("mangup")); #Eqqarsaatigiinnarlugu tununniit manngersumik, angisuumik ilunnut manguttoqartoq, nipaarsaarlunga INERPUNGA. SELECT:Sem0500 ("qaller") + (/^i?Sem/speak_mnr/r) IF (0 V LINK *-1 DIRTALESTART); #”Qaquguuna eqqartuussaaniartunga?” QALLERPARA. diff --git a/src/fst/stems/nouns.lexc b/src/fst/stems/nouns.lexc index 35e0dd164..d717f5be5 100644 --- a/src/fst/stems/nouns.lexc +++ b/src/fst/stems/nouns.lexc @@ -26423,7 +26423,7 @@ qanittuaralaannguaq Z1nnguaqZmorf ; qanittuararsuaq Z2aqZmorf ; qanittunnguaq Z1nnguaqZmorf ; qaniuartoq Z1Zmorf ; -qaniuisoq Z1Zmorf ; +!qaniuisoq Z1Zmorf ; !KW31082023 qanna+ARAQ+Der/nn:qannaaraq GEMS ; qannerisaannartoq Z1Zmorf ; qannerisoq Z1Zmorf ; @@ -26776,7 +26776,7 @@ qarsaarpaluttoq Z1Zmorf ; qarsisoq Z1Zmorf ; qarsoq Z1Zmorf ; qarsoraluttuinnartoq Z1Zmorf ; -qarsorneq Z2Zmorf ; +!qarsorneq Z2Zmorf ; !KW31082023 qarsorneritsoq Z1Zmorf ; qarsorsaarsersoq Z1Zmorf ; qarsorsaasaq Z1Zmorf ; diff --git a/src/fst/stems/verbs.lexc b/src/fst/stems/verbs.lexc index 847d96d33..8d4b6a3ae 100644 --- a/src/fst/stems/verbs.lexc +++ b/src/fst/stems/verbs.lexc @@ -16833,7 +16833,7 @@ qanimor HTR_r_GAQ_i ; !qanimui XIi_voqXmorf_stem ; !HTR qanimaap XItsXmorf_stem ; !qanimaavip XIiXmorf_stem ; -qaningaar IV_long_stem ; +!qaningaar IV_long_stem ; !KW31082023 qaninngar IV_r_stem ; qaninngar HTR_r_GAQ_i ; qaninngoor HTR_longV_GAQ_i ; @@ -24612,7 +24612,7 @@ ulluerutivip XIiXmorf_stem ; ulluer XIirXmorf_stem ; ulloq+GISSAAR+Der/nv+Orth/Alt:ullugissaar IV_long_stem ; !typisk sportsudtryk "have en god dag" ulloq+GISSAAR+Der/nv+Orth/Alt+GI+Der/vv:ullugissaar XIgujoqX ; !typisk sportsudtryk "have en god dag" -ulluinnarsiutige:ulluinnarsiutig HTR_schwa_SAQ_nnip ; +!ulluinnarsiutige:ulluinnarsiutig HTR_schwa_SAQ_nnip ; !KW31082023 ulluinnar UDEN_HTR_r_GAQ ; !NN 20200104 ulluinnarlugu ulluissaar IV_long_stem ; ullukillisaar IV_long_stem ;