Skip to content

Commit

Permalink
fst-problemer og manglende disambigueringer
Browse files Browse the repository at this point in the history
  • Loading branch information
planggard committed Oct 10, 2023
1 parent 8334944 commit d1a9c76
Show file tree
Hide file tree
Showing 6 changed files with 39 additions and 14 deletions.
21 changes: 17 additions & 4 deletions src/cg3/disambiguator.cg3
Original file line number Diff line number Diff line change
Expand Up @@ -12809,7 +12809,8 @@ SELECT:0515 Par + 3Sg + TRANSVERB IF (*-1 ORATIOOBLVERB BARRIER VFIN LINK *1 N +
REMOVE:0517 Par + 3Sg IF (0 (/"misissor"\ Gram/TV\ HTR\ Der/vv\ Gram/IV\ TUQ/l) LINK -1 Aeq); #Maannali tamanna paasisaqarfiginerulerparput – ilisimatuutut MISISSUISOQ Bonnie Jensen tamatumani qujaffissaavoq.

#Oratio oblique uden eksplicit subjekt
SELECT:0517 Par + 3Sg IF (*-1 ORATIOOBLVERB - (/__UTE\ Der/vn__/l) BARRIER CLB OR V - FossileredeCont LINK 0 V OR POSSESSUM LINK NEGATE *1 CONJ-C BARRIER (*) - KOMMA)(NEGATE 0 TUQ + Sg LINK *1 Par + 3SgO BARRIER V)
SELECT:0517 Par + 3Sg IF (*-1 ORATIOOBLVERB - (/__UTE\ Der/vn__/l) - (/TAQ\ Der/vn\ N/l) BARRIER CLB OR V - FossileredeCont LINK 0 V OR POSSESSUM LINK NEGATE *1 CONJ-C BARRIER (*) - KOMMA)
(NEGATE 0 TUQ + Sg LINK *1 Par + 3SgO BARRIER V)
(NEGATE *1 3SgO LINK *-1 _TARGET_ BARRIER (*) - Gram/Dem LINK -1 Abs + Sg)#
(NEGATE 1 KOMMA LINK 1 Prop LINK *1 3Sg BARRIER (*) - Prop)#Air Greenlandip timmisartunik aqutsisoqarfianeERSOQ, Hans Baadsgård oqaluttuarpoq. Er subjekt pga. appositionen
(NEGATE 0 Abs + &&IV_SUBJ& LINK *1 Par + &&IV_SUBJ& BARRIER V)#ilisimanngilaa PISOQ qanga pisimanersoq
Expand All @@ -12831,7 +12832,7 @@ REMOVE:0519A Par IF (0 TUQ - POSSESSUM)
;

#Den mest generelle regel. NB forudsætter at 0320A først har fjernet 3PlO og at 0519A har fjernet %AmbiguøsPar
REMOVE:0519B Abs + Pl - OTHERTIME IF (*1 TRANSVERB + 3Pl OR 3PlPoss BARRIER CLB)
REMOVE:0519B Abs + Pl - OTHERTIME IF (*1 TRANSVERB + 3Pl OR 3PlPoss - Orth/Arch BARRIER CLB)
(NEGATE *0 3PlO)
(NEGATE *0 INTRANSVERB + 3Pl BARRIER VFIN)
(NEGATE 0 Num LINK -1 OTHERTIME)#Demokraatit pilersaarut tassunga assingusoq ukiut QULIT matuma siorna pinngitsoortippaat.
Expand Down Expand Up @@ -12872,6 +12873,7 @@ REMOVE:0521 Par IF (0 N + $$KASUS + $$NUMERUS LINK *-1 VFIN BARRIER KOMMA OR ORA
REMOVE:0522 Par + INTRANSVERB - TIME IF (0 TUQ + N OR Prop)(*1 V - PASSIV - (/[12][SP][gl]/l) BARRIER KOMMA OR V)(NEGATE *-1 UNA OR UKU)(NEGATE *1 ORATIOOBLVERB + TRANSVERB)(NEGATE *1 ORATIOOBLVERB LINK NOT *1 Par)(NEGATE 0 QQU LINK *-1 ORATIOOBLVERB)(NEGATE 0 CONJ-LU LINK NOT -1 KOMMA OR BOS)(NEGATE *-1 BOS LINK *1 _TARGET_ BARRIER Par)
(NEGATE *-1 Abs + &&IV_SUBJ& LINK *1 _TARGET_ + Par + &&IV_SUBJ& + RIIR OR SSA LINK *1 VFIN + &&IV_SUBJ&)#Kommunimi aningaasaqarnermut ataatsimiititaliaq QNAPP saniatigooralugulu pinialuttartut ATAATSIMEEQAATIGEEREERSULLU isumaqarput qilalukkat qernertartassiissutit 65-t ikippallaartut.
(NEGATE *1 ORATIOOBLVERB OR OQARPOQ LINK *-1 _TARGET_ + Par + &&IV_SUBJ& LINK *-1 Abs + &&IV_SUBJ& BARRIER V)#Partii Naleqqameersoq Jens NapãtôK taama allagarsisarneq TAKORNARTAANNGITSOQ oqarpoq.
(NEGATE *-1 ORATIOOBLVERB LINK *1 _TARGET_ + (/"tassa"\ U/l) LINK *1 Par)#MIPI-p 2008-mut ukiumoortumik nalunaarutaani erserpoq, siulersuisut takusinnaasaat TASSAASOQ Kalaallit Nunaanni meeqqat inuusuttullu pillugit ilisimasaqarnermik siaruarterinermillu inuiaqatigiinni suli pisariaqartitsisoqaqisoq. #PL20231010 Pas på! Skrøbelig regel
;

REMOVE:0522B N + TUQ OR (".*(s|t)oq"r) IF (0 Par + <fn:Enter> LINK 1 ORATIOOBLVERB) ; #Aasaru maani Kalaallit Nunaanni seqineq tamangajassuarmi pulasoq takuneqarsinnaassaaq. # Til sommer kan man her i Grønland se en næsten fuld solformørkelse.
Expand Down Expand Up @@ -13233,6 +13235,8 @@ SELECT:1690A Abs + Sg - 2SgPoss - Sem/Unit - SUB IF (*1C INTRANSVERB + 3Sg BARRI
(NEGATE 0 ("taama") + IP + Par LINK -1 KOMMA)#Atukkavut aammattaaq assigiiaarsinnaagaluarput, TAAMAATTOQ immikkuutaartuni assigiinngitsutigut naligiinnginneqarsimaqisoq.
(NEGATE 0 ("illit" N) LINK 0 ("illit" Pron))#PL20221222 Una oqaluttuarisara ILLIT kisivit piginagu inoqatinnut oqaluttuariniariuk - immaqa silattuallaataasinnaajumaarpoq.
(NEGATE 0 Prop + Rel + Sg LINK 1 TRANSVERB + 3Sg)#GUUTILLI paasitippaanga inuk kinaasorluunniit illernanngitsumik mingunnartumilluunniit taassanngikkiga
(NEGATE *-1 ORATIOOBLVERB LINK *1 _TARGET_ + (/"tassa"\ U/l) LINK *1 Par)#MIPI-p 2008-mut ukiumoortumik nalunaarutaani erserpoq, siulersuisut takusinnaasaat TASSAASOQ Kalaallit Nunaanni meeqqat inuusuttullu pillugit ilisimasaqarnermik siaruarterinermillu inuiaqatigiinni suli pisariaqartitsisoqaqisoq. #PL20231010 Pas på! Skrøbelig regel

#(NEGATE *1 N + Abs + Sg LINK 1 V + Gram/IV + Sg)
;

Expand Down Expand Up @@ -14607,8 +14611,13 @@ REMOVE:8599A TUQ + Aeq - Der/vn IF (0 TUQ + Aeq + Der/vn)(NEGATE 1 ("ip")); #De
REMOVE:8599B (*) - LEX_FULLFORM_NOUN ; #Jude 20230908
REMOVE:8599C (Sem/Location) (0 (Sem/an) + (/^[12][SP][gl]Poss$/r)) ; #Jude 20230915 qitinnut
REMOVE:8599D ("niaqoq") + LAAQ IF (0 ("niaqulaar")) ; #Parnaarussammi najugaqatikka avatangiiseqannginnertik pissutigalugu, imminnut nakkutingaaramik NIAQULAALILERSUT isigaakka, amiilaarivakkalu.
REMOVE:8599E iN (0 iSem/Hum + QATE) ; #Jude 20231010
REMOVE:8599E iN (0 (/\(iSem/Hum\ iSem/Hprof\ \)?QATE/l)) ; #Jude 20231010
REMOVE:8599F (*) - iSem/Hprof (0 ("sakkutooq")) ; #Jude 20231010
REMOVE:8599G (/"avat"\ ARSUK/l) IF (0 ("avataarsuaq")); #USA-p præsidentiata tullersortaa Mike Pence, USA-p AVATAARSUANUT tunngasunut siunnersuisoqatigiivinut, National Space Councilimut, oqalugiarnermini taama ilaatigut oqarpoq.
REMOVE:8599H (/"ip"\ Gram/[IT]V\ \(Gram/Pass\ \)?TUQ/l) + POSSESSUM IF (0 ("ittu") OR ("ittoq")); #Politiit ITTUAT
REMOVE:8599I ("ittu") IF (0 ("ittoq")) ; #PL20231010 vurderes lidt hyppigere



# Nomen eller pron
REMOVE:8600A (/"sukku"\ N/l) IF (0 (/"suna"\ Gram/Int\ Pron\ Via/l) LINK 1 ("ma"));
Expand Down Expand Up @@ -15825,7 +15834,8 @@ 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") OR ("ittoq") IF (0 ("ip") + TUQ) ;
REMOVE:5959N1 ("ittuk") IF (0 ("ittu") + POSSESSUM OR ("ittoq") + POSSESSUM);
REMOVE:5959N2 ("ittu") - POSSESSUM OR ("ittoq") - POSSESSUM 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")) ;
Expand Down Expand Up @@ -17033,6 +17043,9 @@ REMOVE:9970FU Rel + Sg + MI IF (0 Abs + Pl + MI - HUMAN)(NEGATE *1 3SgPoss BARRI

REMOVE:9970FV Par IF (0 IP + TUQ + $$KASUS + $$NUMERUS LINK NOT *-1 $$KASUS + $$NUMERUS LINK *1 _TARGET_ LINK 1 N + $$KASUS + $$NUMERUS); #Uunnaaviup SANIANIITTOQ igalaaq qiverseriarlugu uunnaavik qamissallugu, eqqarsaammerujussuarpunga.

REMOVE:9970FW (Rel 4SgPoss) IF (0 Lok LINK 1 Lok + CONJ-C); #Pituffimmi SAKKUTOOQARFIMMI Danmarkshavn-imilu aasaanerani nalunaaquttap atuutsinneqarnissaanut tunngasumik aalajangersaasoqarneq ajorpoq.


#Ambiguøse possessum iflg. sandsynlighed

REMOVE:9970_3SgPossA (Abs 3SgPoss) IF (-1 CON_CAU + 1SgO LINK 1 (Abs 1SgPoss) + &&IV_SUBJ& LINK *1 V + &&IV_SUBJ& BARRIER V);#Nujuillisarsinnaagummali INUUNERA qaamanerulissagaluarpoq.
Expand Down
3 changes: 2 additions & 1 deletion src/cg3/kal-pre2.cg3
Original file line number Diff line number Diff line change
Expand Up @@ -81,8 +81,9 @@ ADD:%2smlNP3 (%2smlNP) TARGET <fn:Be> + NIRU + $$KASUS + $$NUMERUS IF (-3 $$KASU
#Parataktiske NP af typen CONT-LU og N
ADD:PartaktiskNP (%ParataktiskNP) TARGET CONT + QAR + 4Pl + CONJ-C IF (1 LIK + N) ;

#PL20231010 Forstår ikke følgende. Forsøgsvist udkommenteret
#Eksplicit tagging af Hprof så <Hprof> er synlig trods evt. transformationer
ADD (<Hprof>) <Hprof> ;
#ADD (<Hprof>) <Hprof> ;

ADD:AappaaAappaa (<aappaq1>) TARGET ("aappaq" 3SgPoss) IF (*1 ("aappaq" 3SgPoss));

Expand Down
1 change: 1 addition & 0 deletions src/fst/affixes/derivations-inflections.lexc
Original file line number Diff line number Diff line change
Expand Up @@ -8646,6 +8646,7 @@ LEXICON Z2-Zmorf !!= * @CODE@ derivationsmorfemer til tup- paa K. Jf. Z2-qZmorf
+TUQAQ+Der/nn+SUAQ+Der/nn:%>%TRUNCtoqarsuaq Z2aqZ ;
+TUUQ+Der/nn:%>tooq tp ;
+TUUQ+Der/nn+SUAQ+Der/nn:%>toorsuaq Z2aqZ ;
+TUUQ+Der/nn+TAQ+Der/nn:%>toortaq tp ;
+USAAQ+Der/nn:%>%TRUNCusaaq tp ;
+USAQ+Der/nn+ARAQ+Der/nn:%>%TRUNCusaaraq GEMS ;
+USAQ+Der/nn+NNGUAQ+Der/nn:%>%TRUNCusannguaq Z1nnguaqZ ;
Expand Down
11 changes: 5 additions & 6 deletions src/fst/affixes/propernouns.lexc
Original file line number Diff line number Diff line change
Expand Up @@ -4023,7 +4023,6 @@ LEXICON Z1geoSZmorf !Narsaq
LEXICON Z1geoPZmorf !Paamiut og Ivittuut
+Sem/Geo+Prop+LIAR+Der/nv:%>%TRUNCliar IV_r ;
+Sem/Geo+Prop+MIU+Der/nn:%>miu Z1Zmorf ;
!+Sem/Geo+Prop+MIU+Der/nn:%>%TRUNCrmiu Z1VZmorf ;
+Sem/Geo+Prop+MIU+Der/nn+Orth/Alt+N+Abs+Sg:%>%TRUNCrmioq K ;
+Sem/Geo+Prop+MIU+Der/nn+Orth/Alt+N+Abs+Sg:%>mioq K ;
+Sem/[email protected]@+Der/nv:%>%[email protected]@ IV_voq ;
Expand Down Expand Up @@ -4270,13 +4269,13 @@ LEXICON Z1ateqZmorf
+Sem/Hum+KKUT+Der/[email protected]@+Der/nn+Orth/Alt+N+Abs+Sg:%>%[email protected]@ K ;
+Sem/Hum+KULUK+Der/nn+NNGUAQ+Der/nn:%>%TRUNCkulunnguaq Z1nnguaq_ateqZ ; !liv 170329
+Sem/Hum+LIAR+Der/nv:%>%TRUNCliar IV_r ;
+Sem/Hum+NNGUAQ+Der/nn:%>%TRUNCnnguaq Z1nnguaq_ateqZ ;
+Sem/Hum+NNGUAQ+Der/nn:%>%TRUNCnnguaq Z1nnguaq_ateqZ ;
+Sem/Hum+PALAAQ+Der/nn+SUAQ+Der/nn:%>%TRUNCpalaarsuaq Z2suaq_ateqZ ;
+Sem/Hum+SUAQ+Der/nn:%>%TRUNCrsuaq Z2suaq_ateqZ ;
+Sem/Hum+SUAQ+Der/nn+NNGUAQ+Der/nn:%>%TRUNCrsuannguaq Z1nnguaq_ateqZ ;
+Sem/Hum+TUQAQ+Der/nn:%>%TRUNCtoqaq Z1Prop+qaZ ;
+Sem/Hum+Prop+U+Der/nv:%>%TRUNCu IV_voq ;
+Sem/Hum+Prop+U+Der/nv+NIRAR+Der/vv:%>%TRUNCunerar TV ;
+Sem/Hum+SUAQ+Der/nn+NNGUAQ+Der/nn:%>%TRUNCrsuannguaq Z1nnguaq_ateqZ ;
+Sem/Hum+TUQAQ+Der/nn:%>%TRUNCtoqaq Z1Prop+qaZ ;
+Sem/Hum+Prop+U+Der/nv:%>%TRUNCu IV_voq ;
+Sem/Hum+Prop+U+Der/nv+NIRAR+Der/vv:%>%TRUNCunerar TV ;
+UTE+Der/nn+NNGUAQ+Der/nn:%>%TRUNCutinnguaq Z1nnguaq_possessumZ ; !PL20220201 Nyt LEX udelukkende til possessum som Nikkiutinnguarput
Z1ateqZ ;

Expand Down
13 changes: 10 additions & 3 deletions src/fst/stems/nouns.lexc
Original file line number Diff line number Diff line change
Expand Up @@ -4326,6 +4326,7 @@ avate+N+Abs+Sg+1PlPoss:avapput Krestr ;
avate+N+Trm+Sg:avammut TrmVb ;
avataaq+Orth/Arch:avataaq Z1Zmorf ;
avataarsersoq Z1Zmorf ;
avataarsuaq Z2aqZmorf ;
avataarsuarsiooq Z1Zmorf ;
avataasaq Z1Zmorf ;
avataasior+Gram/IV+UTE+Der/vn:avataasiut tptmorf ;
Expand Down Expand Up @@ -21539,7 +21540,12 @@ nukeruaq Z1+ssZmorf ;
nukeruartooq Z1Zmorf ;
nukerugaq GEMS ;
nukik Z2-Zmorf ;
nukik Z2kZmorf ; !livm 20170929 nukinga
nukik+N+Abs+Sg+3SgPoss:nukinga K ;
nukik+N+Abs+Sg+3PlPoss:nukingat K ;
nukik+N+Ins+Sg+3SgPoss:nukinganik K ;
nukik+N+Trm+Sg+3SgPoss:nukinganut TrmVb ;
nukik+N+Rel+Sg+3SgPoss:nukingata K ;
!nukik Z2kZmorf ; !livm 20170929 nukinga
nukik+SIUR+Der/nv+Gram/IV+UTE+Der/vn:nukissiuut tptmorf ;
nukik+SSAQ+Der/nn+IAR+Der/nv+Gram/IV+UTE+Der/vn:nukissaajaat tptmorf ;
nukiliorfik Z2-Zmorf ;
Expand Down Expand Up @@ -21572,7 +21578,7 @@ nukik+SIUT+Der/nn+UTE+Der/nn:nukissiuut tptmorf ;
!nukissuaq Z2aqZmorf ; !KW24032023
nukissui Z1eZmorf ;
nukittoorpaluttoq Z1Zmorf ;
nukittoorsuaq Z2aqZmorf ;
!nukittoorsuaq Z2aqZmorf ;
nukittor+Gram/IV+SAR+Der/vv+Gram/IV+UTE+Der/vn:nukittorsaat tptmorf ;
nukittornerusoq Z1Zmorf ;
nukittortoq Z1Zmorf ;
Expand Down Expand Up @@ -29762,7 +29768,8 @@ sakkusersoq Z1Zmorf ;
sakkussamaasivik Z2-Zmorf ;
sakkussamaat tptmorf ;
sakkussiarsuaq Z2aqZmorf ;
sakkutooq+Sem/Hum:sakkutooq Z1Zmorf ;
!sakkutooq+Sem/Hum:sakkutooq Z1Zmorf ;
sakkutooq Z1Zmorf ;
sakkortooq Z1Zmorf ;
sakkutooqaqqusinnginneq Z2Zmorf ;
!sakkutooqarfik Z2-Zmorf ;
Expand Down
4 changes: 4 additions & 0 deletions src/fst/stems/propernouns.lexc
Original file line number Diff line number Diff line change
Expand Up @@ -15887,6 +15887,7 @@ [email protected]@ Z1geoSZmorf ;
Australien Z1geo_oqs-nbestZmorf ;
Avanersuaq Z2geo_aqSZmorf ;
Avannaa ZgeopossessumSZ ;
Avannaarsuaq Z2geo_aqSZmorf ;
Avannaata% Kommunia ZpossessumSZ ;
Avannaata% Qimussersua+Sem/event:Avannaata% Qimussersua ZpossessumSZ ;
Avedøre Z1geopropZ ;
Expand Down Expand Up @@ -19623,6 +19624,9 @@ [email protected]@ Z1geoSZmorf ;
[email protected]@ Z1geoSZmorf ;
Qingdao Z1geopropZ ;
Qinngorput Z2-geo_specielSZmorf ; !til former som Qinngorpummi og Qinngorpumi
Qinngorput+Sem/Geo+Prop+Lok+Sg:Qinngutsinni LokVb ;
Qinngorput+Sem/Geo+Prop+Abl+Sg:Qinngutsinnit AblVb ;
Qinngorput+Sem/Geo+Prop+Trm+Sg:Qinngutsinnut TrmVb ;
Qisavik Z2-geoSZmorf ;
[email protected]@ Z1geoSZmorf ;
[email protected]@ Z1geoSZmorf ;
Expand Down

0 comments on commit d1a9c76

Please sign in to comment.