mirror of
https://github.com/GrammaticalFramework/gf-core.git
synced 2026-04-09 04:59:31 -06:00
fixes to estonian
This commit is contained in:
inari
@@ -23,8 +23,6 @@ resource HjkEst = open ResEst, Prelude, Predef in {
|
||||
|
||||
oper
|
||||
|
||||
NFS = {s : NForm => Str} ;
|
||||
|
||||
foreign : pattern Str = #("z" | "ž" | "š") ;
|
||||
-- Foreign vowel endings
|
||||
foreign_v : pattern Str = #("ko" | "po" | "to" | "fo" | "ka" | "pa" | "ta" | "fa" | "ku" | "pu" | "tu" | "fu") ;
|
||||
@@ -50,13 +48,14 @@ resource HjkEst = open ResEst, Prelude, Predef in {
|
||||
hjk_type_VI_meeskond,
|
||||
hjk_type_VI_seminar,
|
||||
hjk_type_VII_touge : Str -> NForms ;
|
||||
-- hjk_type_VII_touge : Str -> NFS ;
|
||||
|
||||
-- IVa additionally needs the stem vowel.
|
||||
hjk_type_IVb_audit,
|
||||
hjk_type_IVb_audit1 : Str -> Str -> NForms ; --NFS
|
||||
hjk_type_IVb_audit1 : Str -> Str -> NForms ;
|
||||
|
||||
hjk_type_VI_tukk : Str -> Str -> NForms ;
|
||||
hjk_type_VI_link2 : Str -> Str -> NForms ;
|
||||
|
||||
hjk_type2 : Str -> Str -> NForms ;
|
||||
|
||||
|
||||
-- Definition of the mapping rules.
|
||||
@@ -147,24 +146,19 @@ resource HjkEst = open ResEst, Prelude, Predef in {
|
||||
in
|
||||
nForms6 x (x_n+"i") (x+"i") (x+"i") (x+"ide") (x+"e") ;
|
||||
|
||||
|
||||
--like link but
|
||||
-- gen form given (takes care of vowel and consonant gradation)
|
||||
-- -sid for pl.part (todo: generate short forms depending on vowel?)
|
||||
hjk_type_VI_tukk x x_gen =
|
||||
-- same as hjk_type_VI_link but additionally takes the genitive ending
|
||||
hjk_type_VI_link2 x i =
|
||||
let
|
||||
v_g : Str = last x_gen ;
|
||||
|
||||
{- pl_part : Str =
|
||||
case v_g of {
|
||||
"i" => "e" ;
|
||||
_ => v_g + "sid" } ;
|
||||
-}
|
||||
|
||||
x_n : Str = weaker_noun x ;
|
||||
-- TODO: think about it
|
||||
e : Str = case i of {
|
||||
"a" => "asid" ; -- pikk/pika -> pikkasid
|
||||
_ => "e"
|
||||
}
|
||||
in
|
||||
nForms6 x x_gen (x+v_g) (x+v_g) (x+v_g+"de") (x+v_g+"sid") ;
|
||||
nForms6 x (x_n+i) (x+i) (x+i) (x+i+"de") (x+e) ;
|
||||
|
||||
|
||||
|
||||
hjk_type_VI_imelik x =
|
||||
let
|
||||
x_t : Str = stronger_noun x
|
||||
@@ -312,103 +306,105 @@ resource HjkEst = open ResEst, Prelude, Predef in {
|
||||
-- - comparative and superlative adjective forms (workaround: use mkA instead)
|
||||
-- - type VII (t6uge -> t6uke), as one needs to detect derivation from verb
|
||||
-- - last syllable superlong (rostbiif)
|
||||
hjk_type x =
|
||||
case <(syl_type x), x> of {
|
||||
<S3, _ + "ke">
|
||||
hjk_type x = hjk_type2 x "i" ;
|
||||
|
||||
hjk_type2 x i =
|
||||
case <(syl_type x), x, i> of {
|
||||
<S3, _ + "ke", _>
|
||||
=> hjk_type_Vb_oluline x ;
|
||||
|
||||
<_, _ + "kond">
|
||||
<_, _ + "kond", _>
|
||||
=> hjk_type_VI_meeskond x ;
|
||||
|
||||
-- Some S2 -ik words (voolik), we only cover words with double vowel
|
||||
<_, _ + #vv + ("lik"|"nik"|"stik")>
|
||||
<_, _ + #vv + ("lik"|"nik"|"stik"), _>
|
||||
=> hjk_type_IVb_audit x "u" ;
|
||||
|
||||
-- Other -ik words as in HJKEKS,
|
||||
-- but added 'ndik' which fixes fractions ('kaheksandik')
|
||||
-- and is wrong only for 'kandik'.
|
||||
<_, _ + ("lik"|"nik"|"stik"|"ndik")>
|
||||
<_, _ + ("lik"|"nik"|"stik"|"ndik"), _>
|
||||
=> hjk_type_VI_imelik x ;
|
||||
|
||||
-- Remaining -k words (but need to be S2)
|
||||
-- but not 'konjak'
|
||||
<S2, _ + ("a"|"e"|"i") + ("ng"|"k")>
|
||||
<S2, _ + ("a"|"e"|"i") + ("ng"|"k"), _>
|
||||
=> hjk_type_IVb_audit x "u" ;
|
||||
|
||||
-- Other -ik words (not in HJKEKS)
|
||||
-- including also: alevik, asemik, lobudik, hämarik, sarapik, põletik
|
||||
<_, _ + ("vik"|"mik"|"dik"|"rik"|"pik"|"tik")>
|
||||
<_, _ + ("vik"|"mik"|"dik"|"rik"|"pik"|"tik"), _>
|
||||
=> hjk_type_VI_imelik x ;
|
||||
|
||||
-- kikas
|
||||
<_, ? + #v + #c + #v + "s">
|
||||
<_, ? + #v + #c + #v + "s", _>
|
||||
=> hjk_type_Va_otsene x ;
|
||||
|
||||
<_, _ + ("ngas"|"kas"|"jas"|"nud"|"tud")>
|
||||
<_, _ + ("ngas"|"kas"|"jas"|"nud"|"tud"), _>
|
||||
=> hjk_type_IVb_maakas x ;
|
||||
|
||||
<S1, _ + #v + #v>
|
||||
<S1, _ + #v + #v, _>
|
||||
=> hjk_type_I_koi x ;
|
||||
|
||||
-- 'statiiv' (not like 'karjuv')
|
||||
<S1, _ + #vv + #c>
|
||||
=> hjk_type_VI_link x ;
|
||||
<S1, _ + #vv + #c, i>
|
||||
=> hjk_type_VI_link2 x i ;
|
||||
|
||||
<S3, _ + #c + #v + #lmnr>
|
||||
<S3, _ + #c + #v + #lmnr, _>
|
||||
=> hjk_type_VI_seminar x ;
|
||||
|
||||
<S1, _ + #v + #v + #c>
|
||||
=> hjk_type_VI_link x ;
|
||||
<S1, _ + #v + #v + #c, i>
|
||||
=> hjk_type_VI_link2 x i ;
|
||||
|
||||
<_, _ + ("us"|"is")>
|
||||
<_, _ + ("us"|"is"), _>
|
||||
=> hjk_type_Vb_oluline x ;
|
||||
|
||||
<S3, _ + #v + #v + #c>
|
||||
=> hjk_type_VI_link x ;
|
||||
<S3, _ + #v + #v + #c, i>
|
||||
=> hjk_type_VI_link2 x i ;
|
||||
|
||||
<(S1|S3), _ + #v + #c + #c>
|
||||
=> hjk_type_VI_link x ;
|
||||
<(S1|S3), _ + #v + #c + #c, i>
|
||||
=> hjk_type_VI_link2 x i ;
|
||||
|
||||
<(S1|S3), _ + #v + #c + #c + #c>
|
||||
=> hjk_type_VI_link x ;
|
||||
<(S1|S3), _ + #v + #c + #c + #c, i>
|
||||
=> hjk_type_VI_link2 x i ;
|
||||
|
||||
<_, _ + "nna">
|
||||
<_, _ + "nna", _>
|
||||
=> hjk_type_III_ratsu x ;
|
||||
|
||||
<-(S21|S22), _ + ("nu"|"tu")>
|
||||
<-(S21|S22), _ + ("nu"|"tu"), _>
|
||||
=> hjk_type_IVa_aasta x ;
|
||||
|
||||
-- TODO: improve foreign detection
|
||||
<S2, _ + #foreign + _ + "in">
|
||||
=> hjk_type_IVb_audit x "i" ;
|
||||
<S2, _ + #foreign + _ + "in", i>
|
||||
=> hjk_type_IVb_audit x i ;
|
||||
|
||||
-- TODO: this is not in HJKEKS
|
||||
-- 'absurd' vs 'ebard'
|
||||
<S2, _ + #v + #lmnr + "d">
|
||||
=> hjk_type_IVb_audit x "i" ;
|
||||
<S2, _ + #v + #lmnr + "d", i>
|
||||
=> hjk_type_IVb_audit x i ;
|
||||
|
||||
-- sometimes 'a' (laurits) TODO: this is not in HJKEKS
|
||||
<S2, _ + #v + #kpt + "s">
|
||||
=> hjk_type_IVb_audit x "i" ;
|
||||
<S2, _ + #v + #kpt + "s", i>
|
||||
=> hjk_type_IVb_audit x i ;
|
||||
|
||||
-- TODO: next 3 rules: last syllable must be long
|
||||
-- portfell, TODO: not 'karask'
|
||||
<S2, _ + #v + #c + #c>
|
||||
=> hjk_type_VI_link x ;
|
||||
<S2, _ + #v + #c + #c, i>
|
||||
=> hjk_type_VI_link2 x i ;
|
||||
|
||||
-- rostbiif, not viiul
|
||||
<S2, _ + #c + #v + #v + #c>
|
||||
=> hjk_type_VI_link x ;
|
||||
<S2, _ + #c + #v + #v + #c, i>
|
||||
=> hjk_type_VI_link2 x i ;
|
||||
|
||||
-- impulss
|
||||
<S2, _ + #v + #c + #c + #c>
|
||||
=> hjk_type_VI_link x ;
|
||||
<S2, _ + #v + #c + #c + #c, i>
|
||||
=> hjk_type_VI_link2 x i ;
|
||||
|
||||
-- TODO: sometimes masked by 'maakas'
|
||||
<_, _ + #v + "s">
|
||||
<_, _ + #v + "s", _>
|
||||
=> hjk_type_Va_otsene x ;
|
||||
|
||||
<_, _ + ("v"|"tav")>
|
||||
<_, _ + ("v"|"tav"), _>
|
||||
=> hjk_type_IVb_audit x "a" ;
|
||||
|
||||
-- The choice between Va (pl part: -seid) and Vb (pl part: -si)
|
||||
@@ -416,38 +412,38 @@ resource HjkEst = open ResEst, Prelude, Predef in {
|
||||
-- We just check the ending of the word and require at least 2 letters
|
||||
-- to precede the ending.
|
||||
-- We added also -tine and -ldane (which occur with adjectives).
|
||||
<_, _ + ? + ? + ("line"|"lane"|"mine"|"kene"|"tine"|"ldane")>
|
||||
<_, _ + ? + ? + ("line"|"lane"|"mine"|"kene"|"tine"|"ldane"), _>
|
||||
=> hjk_type_Vb_oluline x ;
|
||||
|
||||
-- k6ne
|
||||
<S21, _ + "e">
|
||||
<S21, _ + "e", _>
|
||||
=> hjk_type_III_ratsu x ;
|
||||
|
||||
-- Many adjectives end with "ne" (40% in WordNet)
|
||||
-- We require them to be at least 5 letters long (excluding 'öine'),
|
||||
-- to give a chance to VII_touge (next rule).
|
||||
<_, _ + ? + ? + ? + "ne">
|
||||
<_, _ + ? + ? + ? + "ne", _>
|
||||
=> hjk_type_Va_otsene x ;
|
||||
|
||||
-- Note: this rule does not actually check the derivation from verb.
|
||||
-- verb + e, TODO: masked by S21/e
|
||||
<(S2|S22), _ + "e">
|
||||
<(S2|S22), _ + "e", _>
|
||||
=> hjk_type_VII_touge x ;
|
||||
|
||||
-- ufo, pita, lito
|
||||
<S21, _ + #foreign_v>
|
||||
<S21, _ + #foreign_v, _>
|
||||
=> hjk_type_III_ratsu x ;
|
||||
|
||||
<S21, _ + #v>
|
||||
<S21, _ + #v, _>
|
||||
=> hjk_type_II_ema x ;
|
||||
|
||||
<S22, _ + #v>
|
||||
<S22, _ + #v, _>
|
||||
=> hjk_type_III_ratsu x ;
|
||||
|
||||
<S23, _ + #v>
|
||||
<S23, _ + #v, _>
|
||||
=> hjk_type_IVa_aasta x ;
|
||||
|
||||
<S2, _ + "in">
|
||||
<S2, _ + "in", _>
|
||||
=> hjk_type_IVb_audit x "a" ;
|
||||
|
||||
-- 'e' deletion
|
||||
@@ -456,54 +452,54 @@ resource HjkEst = open ResEst, Prelude, Predef in {
|
||||
-- spikker -> spikri (TODO: not: pokker -> pokkeri)
|
||||
-- Note: pintsel -> pintsli, but not pitser -> pitsri
|
||||
-- Note: 'redel' and 'paber' do not lose the 'e'.
|
||||
<S2, y + kk@("kk"|"pp"|"tt"|"hh") + "e" + l@("l"|"r")>
|
||||
<S2, y + kk@("kk"|"pp"|"tt"|"hh") + "e" + l@("l"|"r"), _>
|
||||
=> hjk_type_IVb_audit1 x (y + (init kk) + l) ;
|
||||
|
||||
-- aaker -> aakri, teater -> teatri
|
||||
<S2, y + vvkpt@(#v + #v + #kpt) + "e" + l@("l"|"r")>
|
||||
<S2, y + vvkpt@(#v + #v + #kpt) + "e" + l@("l"|"r"), _>
|
||||
=> hjk_type_IVb_audit1 x (y+vvkpt+l) ;
|
||||
|
||||
<S2, y + vv@(#vv) + gbd@(#gbd) + "e" + l@("l"|"r")>
|
||||
<S2, y + vv@(#vv) + gbd@(#gbd) + "e" + l@("l"|"r"), _>
|
||||
=> hjk_type_IVb_audit1 x (y+vv+gbd+l) ;
|
||||
|
||||
-- Disabled, 50-50 correctness
|
||||
--<S2, y + vv@(#vv) + lmnr@(#lmnr) + "e" + l@("l"|"r")>
|
||||
--<S2, y + vv@(#vv) + lmnr@(#lmnr) + "e" + l@("l"|"r"), _>
|
||||
-- => hjk_type_IVb_audit1 x (y+vv+lmnr+l) ; -- 50-50
|
||||
|
||||
<S2, y + vv@(#vv) + s@("s"|"v") + "e" + l@("l"|"r")>
|
||||
<S2, y + vv@(#vv) + s@("s"|"v") + "e" + l@("l"|"r"), _>
|
||||
=> hjk_type_IVb_audit1 x (y+vv+s+l) ;
|
||||
|
||||
<S2, y + n@("ht"|"hk"|"hv"|"nts"|"ld"|"lv"|"lb"|"ng"|"nd"|"mb"|"mp"|"nt"|"ps"|"ks"|"sk"|"st") + "e" + l@("l"|"r")>
|
||||
<S2, y + n@("ht"|"hk"|"hv"|"nts"|"ld"|"lv"|"lb"|"ng"|"nd"|"mb"|"mp"|"nt"|"ps"|"ks"|"sk"|"st") + "e" + l@("l"|"r"), _>
|
||||
=> hjk_type_IVb_audit1 x (y+n+l) ;
|
||||
|
||||
<S2, y + "e" + l@("l"|"r")>
|
||||
=> hjk_type_IVb_audit x "i" ;
|
||||
<S2, y + "e" + l@("l"|"r"), i>
|
||||
=> hjk_type_IVb_audit x i ;
|
||||
|
||||
-- TODO: sometimes masked by 'link'
|
||||
<S2, _ + #c>
|
||||
=> hjk_type_IVb_audit x "i" ;
|
||||
<S2, _ + #c, i>
|
||||
=> hjk_type_IVb_audit x i ;
|
||||
|
||||
<S3, _ + #v>
|
||||
<S3, _ + #v, _>
|
||||
=> hjk_type_IVa_aasta x ;
|
||||
|
||||
-- verb + 'e'
|
||||
<_, _ + "e">
|
||||
<_, _ + "e", _>
|
||||
=> hjk_type_VII_touge x ;
|
||||
|
||||
-- catch all that end with consonant
|
||||
<_, _ + #c>
|
||||
=> hjk_type_IVb_audit x "i" ;
|
||||
<_, _ + #c, i>
|
||||
=> hjk_type_IVb_audit x i ;
|
||||
|
||||
-- TODO: not in HJKEKS
|
||||
<_, _ + ("ia"|"ja")> --kündja, not gerilja
|
||||
<_, _ + ("ia"|"ja"), _> --kündja, not gerilja
|
||||
=> hjk_type_IVa_aasta x ;
|
||||
|
||||
--added by Inari 07.10.
|
||||
<S23, _ + #c + ("la")> --haigla, not gorilla
|
||||
<S23, _ + #c + ("la"), _> --haigla, not gorilla
|
||||
=> hjk_type_IVa_aasta x ;
|
||||
|
||||
-- catch all
|
||||
<_, _>
|
||||
<_, _, _>
|
||||
=> hjk_type_III_ratsu x
|
||||
} ;
|
||||
|
||||
|
||||
@@ -9,27 +9,27 @@ lin
|
||||
alas_Interj = ss "paraku" ;
|
||||
answer_V2S = mkV2 (mkV "vastama" "vastata") callative ;
|
||||
apartment_N = mkN "korter" ;
|
||||
apple_N = mkN "õun" ;
|
||||
apple_N = mkN "õun" "õuna" "õuna";
|
||||
art_N = mkN "kunst" ;
|
||||
ask_V2Q = mkV2 (mkV "küsima") (casePrep ablative) ;
|
||||
baby_N = mkN "beebi" ;
|
||||
bad_A = mkA (mkN "halb" "halva" "halba" "halba" "halbade" "halbu") "halvem" "halvim" ;
|
||||
bank_N = mkN "pank" ;
|
||||
bank_N = mkN "pank" "panga" "panka";
|
||||
beautiful_A = mkA (mkN "kaunis" "kauni" "kaunist" "kaunisse" "kaunite" "kauneid") ;
|
||||
become_VA = mkVA (mkV "saama") ctranslative ; -- tema saab vanaks
|
||||
beer_N = mkN "õlu" ;
|
||||
beg_V2V = mkV2V (mkV "paluma") cpartitive ;
|
||||
big_A = mkA (mkN "suur" "suure" "suurt" "suurde" "suurte" "suuri");
|
||||
bike_N = mkN "ratas" ;
|
||||
bird_N = mkN "lind" ;
|
||||
bike_N = mkN "ratas" "ratta" "ratast" ;
|
||||
bird_N = mkN "lind" "linnu" "lindu" ;
|
||||
black_A = mkA (mkN "must" "musta" "musta") ;
|
||||
blue_A = mkA "sinine" ;
|
||||
boat_N = mkN "paat" ;
|
||||
book_N = mkN "raamat" "raamatu" "raamatut" "raamatusse" "raamatute" "raamatuid" ;
|
||||
boot_N = mkN "saabas" ;
|
||||
boot_N = mkN "saabas" "saapa" "saabast" ;
|
||||
boss_N = mkN "boss" ;
|
||||
boy_N = mkN "poiss" ;
|
||||
bread_N = mkN "leib" ;
|
||||
bread_N = mkN "leib" "leiva" "leiba" ;
|
||||
break_V2 = mkV2 (mkV "katki" (mkV "tegema" "teha")) ;
|
||||
broad_A = mkA (mkN "lai" "laia" "laia" "laia" "laiade" "laiu") "laiem" "TODO" ;
|
||||
brother_N2 = mkN2 (mkN "vend" "venna" "venda" "venda" "vendade" "vendi") ;
|
||||
@@ -39,24 +39,24 @@ lin
|
||||
camera_N = mkN "kaamera" ;
|
||||
cap_N = mkN "müts" ;
|
||||
car_N = mkN "auto" ;
|
||||
carpet_N = mkN "vaip" ;
|
||||
carpet_N = mkN "vaip" "vaiba" "vaipa" ;
|
||||
cat_N = mkN "kass" ;
|
||||
ceiling_N = mkN "lagi" ;
|
||||
ceiling_N = mkN "lagi" "lae" "lage" "lakke" "lagede" "lagesid" ;
|
||||
chair_N = mkN "tool" ;
|
||||
cheese_N = mkN "juust" ;
|
||||
cheese_N = mkN "juust" "juustu" "juustu" ;
|
||||
child_N = mkN "laps" "lapse" "last" "lapsesse" "laste" "lapsi" ;
|
||||
church_N = mkN "kirik" ;
|
||||
city_N = mkN "linn" "linna" "linna" "linna" "linnade" "linnu" ;
|
||||
clean_A = mkA (mkN "puhas" "puhta" "puhast" "puhtasse" "puhaste" "puhtaid");
|
||||
clever_A = mkA "tark" ;
|
||||
clever_A = mkA (mkN "tark" "targa" "tarka") ;
|
||||
close_V2 = mkV2 (mkV "sulgema") ;
|
||||
coat_N = mkN "mantel" ;
|
||||
cold_A = mkA "külm" ;
|
||||
cold_A = mkA (mkN "külm" "külma" "külma") ;
|
||||
come_V = mkV "tulema" ;
|
||||
computer_N = mkN "arvuti" ;
|
||||
country_N = mkN "maa" ;
|
||||
cousin_N = mkN "nõbu" ;
|
||||
cow_N = mkN "lehm" ;
|
||||
cow_N = mkN "lehm" "lehma" "lehma" ;
|
||||
die_V = mkV "surema" ;
|
||||
dirty_A = mkA "räpane" ;
|
||||
-- Using 'terminative' for the destination works better with nouns.
|
||||
@@ -65,71 +65,70 @@ lin
|
||||
-- Unfortunately, we cannot use a similar trick for the source (*Põltsamaast vs Põltsamaalt).
|
||||
distance_N3 = mkN3 (mkN "kaugus") celative (casePrep terminative) ;
|
||||
doctor_N = mkN "arst" ;
|
||||
dog_N = mkN "koer" ;
|
||||
door_N = mkN "uks" ;
|
||||
dog_N = mkN "koer" "koera" "koera" ;
|
||||
door_N = mkN "uks" "ukse" "ust" ;
|
||||
drink_V2 = mkV2 (mkV "jooma") cpartitive ;
|
||||
easy_A2V = mkA2 (mkA (mkN "lihtne")) callative ;
|
||||
easy_A2V = mkA2 (mkA (mkN "lihtne" "lihtsa" "lihtsat")) callative ;
|
||||
eat_V2 = mkV2 (mkV "sööma") cpartitive ;
|
||||
empty_A = mkA "tühi" ;
|
||||
empty_A = mkA (mkN "tühi" "tühja" "tühja" "tühja" "tühjade" "tühje") ;
|
||||
enemy_N = mkN "vaenlane" ;
|
||||
factory_N = mkN "tehas" ;
|
||||
father_N2 = mkN2 (mkN "isa") ;
|
||||
fear_VS = mkVS (mk2V "kartma" "karta") ;
|
||||
find_V2 = mkV2 (mk2V "leidma" "leida") ;
|
||||
fish_N = mkN "kala" ;
|
||||
floor_N = mkN "põrand" ;
|
||||
floor_N = mkN "põrand" "põranda" "põrandat" ;
|
||||
forget_V2 = mkV2 (mkV "unustama") ;
|
||||
fridge_N = mkN "külm" (mkN "kapp") ;
|
||||
friend_N = mkN "sõber" ;
|
||||
fruit_N = mkN "puu" (mkN "vili") ;
|
||||
fun_AV = mkAV (mkA (mkN "lõbus")) ;
|
||||
garden_N = mkN "aed" ;
|
||||
girl_N = mkN "tüdruk" ;
|
||||
glove_N = mkN "kinnas" ;
|
||||
gold_N = mkN "kuld" ;
|
||||
friend_N = mkN "sõber" "sõbra" "sõpra" "sõpra" "sõprade" "sõpru" ;
|
||||
fruit_N = mkN "puu" (mkN "vili" "vilja" "vilja") ;
|
||||
fun_AV = mkAV (mkA (mkN "lõbus" "lõbusa" "lõbusat")) ;
|
||||
garden_N = mkN "aed" "aia" "aeda";
|
||||
girl_N = mkN "tüdruk" "tüdruku" "tüdrukut" ;
|
||||
glove_N = mkN "kinnas" "kinda" "kinnast" ;
|
||||
gold_N = mkN "kuld" "kulla" "kulda" ;
|
||||
good_A = mkA (mkN "hea") "parem" "parim" ;
|
||||
go_V = mkV "minema" "minna" "läheb" "minnakse"
|
||||
"minge" "läks" "läinud" "mindud" ;
|
||||
go_V = vMinema ; -- irregular
|
||||
green_A = mkA "roheline" ;
|
||||
harbour_N = mkN "sadam" ;
|
||||
hate_V2 = mkV2 (mkV "vihkama") cpartitive ;
|
||||
harbour_N = mkN "sadam" "sadama" "sadamat" ;
|
||||
hate_V2 = mkV2 (mkV "vihkama" "vihata") cpartitive ;
|
||||
hat_N = mkN "müts" ;
|
||||
hear_V2 = mkV2 (mkV "kuulma") ;
|
||||
hill_N = mkN "küngas" "künka" "küngast" "künkasse" "küngaste" "künkaid" ;
|
||||
hope_VS = mkVS (mkV "lootma") ;
|
||||
horse_N = mkN "hobune" ;
|
||||
hot_A = mkA "kuum" ;
|
||||
hot_A = mkA (mkN "kuum" "kuuma" "kuuma") ;
|
||||
house_N = mkN "maja" ;
|
||||
important_A = mkA "tähtis" ;
|
||||
important_A = mkA (mkN "tähtis" "tähtsa" "tähtsat") ;
|
||||
industry_N = mkN "tööstus" ;
|
||||
iron_N = mkN "raud" ;
|
||||
iron_N = mkN "raud" "raua" "rauda" ;
|
||||
king_N = mkN "kuningas" ;
|
||||
know_VS = mkVS know_V ;
|
||||
know_VQ = mkVQ know_V ;
|
||||
know_V2 = mkV2 (mkV "tundma") ;
|
||||
lake_N = mkN "järv" ;
|
||||
lake_N = mkN "järv" "järve" "järve" ;
|
||||
lamp_N = mkN "lamp" ;
|
||||
learn_V2 = mkV2 (mkV "õppima") ;
|
||||
leather_N = mkN "nahk" ;
|
||||
leather_N = skin_N ;
|
||||
leave_V2 = mkV2 (mkV "jätma") ;
|
||||
like_V2 = mkV2 (mkV "lugu" (mkV "pidama" "pidada" "peab" "peetakse")) celative ;
|
||||
listen_V2 = mkV2 (mkV "kuulama") partitive ;
|
||||
listen_V2 = mkV2 (mkV "kuulama" "kuulata") partitive ;
|
||||
live_V = mkV "elama" ;
|
||||
long_A = mkA "pikk" ;
|
||||
long_A = mkA (mkN "pikk" "pika" "pikka") ;
|
||||
lose_V2 = mkV2 (mkV "kaotama") ;
|
||||
love_N = mkN "armastus" ;
|
||||
love_V2 = mkV2 (mkV "armastama") partitive ;
|
||||
man_N = mkN "mees" "mehe" "meest" "mehesse" "meeste" "mehi" ;
|
||||
married_A2 = mkA2 (mkA "abielus") (casePrep comitative) ;
|
||||
meat_N = mkN "liha" ;
|
||||
milk_N = mkN "piim" ;
|
||||
milk_N = mkN "piim" "piima" "piima" ;
|
||||
moon_N = mkN "kuu" ;
|
||||
mother_N2 = mkN2 (mkN "ema") ;
|
||||
mountain_N = mkN "mägi" "mäe" "mäge" "mäkke" "mägede" "mägesid" ;
|
||||
music_N = mkN "muusika" ;
|
||||
narrow_A = mkA "kitsas" ;
|
||||
new_A = mkA (mkN "uus" "uue" "uut" "uude" "uute" "uusi") "uuem" "uusim" ;
|
||||
newspaper_N = mkN "aja" (mkN "leht") ;
|
||||
newspaper_N = mkN "aja" (mkN "leht" "lehe" "lehte") ;
|
||||
oil_N = mkN "õli" ;
|
||||
old_A = mkA (mkN "vana" "vana" "vana" "vanasse" "vanade" "vanasid") "vanem" "vanim" ;
|
||||
open_V2 = mkV2 (mkV "avama") ;
|
||||
@@ -149,9 +148,9 @@ lin
|
||||
rain_V0 = mkV0 (mkV "sadama" "sadada" "sajab") ;
|
||||
read_V2 = mkV2 (mkV "lugema" "lugeda" "loeb") ;
|
||||
red_A = mkA "punane" ;
|
||||
religion_N = mkN "usk" ;
|
||||
religion_N = mkN "usk" "usu" "usku" ;
|
||||
restaurant_N = mkN "restoran" ;
|
||||
river_N = mkN "jõgi" ;
|
||||
river_N = mkN "jõgi" "jõe" "jõge" ;
|
||||
rock_N = mkN "kalju" ;
|
||||
roof_N = mkN "katus" ;
|
||||
rubber_N = mkN "kumm" ;
|
||||
@@ -159,55 +158,55 @@ lin
|
||||
say_VS = mkVS (mkV "ütlema") ;
|
||||
school_N = mkN "kool" ;
|
||||
science_N = mkN "teadus" ;
|
||||
sea_N = mkN "meri" ;
|
||||
sea_N = mkN "meri" "mere" "merd" ;
|
||||
seek_V2 = mkV2 (mkV "otsima") cpartitive ;
|
||||
see_V2 = mkV2 (mkV "nägema" "näha") ;
|
||||
sell_V3 = mkV3 (mkV "müüma") accPrep callative ;
|
||||
send_V3 = mkV3 (mkV "saatma") accPrep callative ;
|
||||
sheep_N = mkN "lammas" ;
|
||||
ship_N = mkN "laev" ;
|
||||
sheep_N = mkN "lammas" "lamba" "lammast" ;
|
||||
ship_N = mkN "laev" "laeva" "laeva" ;
|
||||
shirt_N = mkN "särk" ;
|
||||
shoe_N = mkN "king" ;
|
||||
shoe_N = mkN "king" "kinga" "kinga" ;
|
||||
shop_N = mkN "kauplus" ;
|
||||
short_A = mkA "lühike" ;
|
||||
silver_N = mkN "hõbe" ;
|
||||
silver_N = mkN "hõbe" "hõbeda" "hõbedat" ;
|
||||
sister_N = mkN "õde" "õe" "õde" "õesse" "õdede" "õdesid" ;
|
||||
sleep_V = mkV "magama" ;
|
||||
small_A = mkA (mkN "väike" "väikse" "väikest" "väiksesse" "väikeste" "väikseid") ;
|
||||
snake_N = mkN "uss" ;
|
||||
sock_N = mkN "sukk" ;
|
||||
sock_N = mkN "sukk" "suka" "sukka" ;
|
||||
speak_V2 = mkV2 talk_V cpartitive ;
|
||||
star_N = mkN "täht" ;
|
||||
star_N = mkN "täht" "tähe" "tähte" ;
|
||||
steel_N = mkN "teras" ;
|
||||
stone_N = mkN "kivi" ;
|
||||
stove_N = mkN "ahi" ;
|
||||
stove_N = mkN "ahi" "ahju" "ahju" ;
|
||||
student_N = mkN "tudeng" ;
|
||||
stupid_A = mkA "loll" ;
|
||||
sun_N = mkN "päike" ;
|
||||
sun_N = mkN "päike" "päikse" "päikest" ;
|
||||
switch8off_V2 = mkV2 (mkV "välja" (mkV "lülitama")) ;
|
||||
switch8on_V2 = mkV2 (mkV "sisse" (mkV "lülitama")) ;
|
||||
table_N = mkN "laud" ;
|
||||
table_N = mkN "laud" "laua" "lauda" ;
|
||||
talk_V3 = mkV3 talk_V callative celative ;
|
||||
teacher_N = mkN "õpetaja" ;
|
||||
teach_V2 = mkV2 (mkV "õpetama") ;
|
||||
television_N = mkN "televisioon" ;
|
||||
thick_A = mkA "paks" ;
|
||||
thick_A = mkA (mkN "paks" "paksu" "paksu") ;
|
||||
thin_A = mkA "õhuke" ;
|
||||
train_N = mkN "rong" ;
|
||||
travel_V = mkV "reisima" ;
|
||||
tree_N = mkN "puu" ;
|
||||
ugly_A = mkA "kole" ;
|
||||
ugly_A = mkA (mkN "kole" "koleda" "koledat") ;
|
||||
understand_V2 = mkV2 (mkV "aru" (mkV "saama")) celative ;
|
||||
university_N = mkN "ülikool" ;
|
||||
village_N = mkN "küla" ;
|
||||
wait_V2 = mkV2 (mkV "ootama") partitive ;
|
||||
wait_V2 = mkV2 (mkV "ootama" "oodata") partitive ;
|
||||
walk_V = mkV "kõndima" ;
|
||||
warm_A = mkA (mkN "soe" "sooja" "sooja" "sooja" "soojade" "soojasid") "soojem" "soojim" ;
|
||||
war_N = mkN "sõda" ;
|
||||
watch_V2 = mkV2 (mkV "vaatama") cpartitive ;
|
||||
water_N = mkN "vesi" ;
|
||||
war_N = mkN "sõda" "sõja" "sõda" ;
|
||||
watch_V2 = mkV2 (mkV "vaatama" "vaadata") cpartitive ;
|
||||
water_N = mkN "vesi" "vee" "vett" ;
|
||||
white_A = mkA "valge" ;
|
||||
window_N = mkN "aken" ;
|
||||
window_N = mkN "aken" "akna" "akent" ;
|
||||
wine_N = mkN "vein" ;
|
||||
win_V2 = mkV2 (mkV "võitma") ;
|
||||
woman_N = mkN "naine" "naise" "naist" "naisesse" "naiste" "naisi" ;
|
||||
@@ -215,95 +214,95 @@ lin
|
||||
wood_N = mkN "puu" ;
|
||||
write_V2 = mkV2 (mkV "kirjutama") ;
|
||||
yellow_A = mkA "kollane" ;
|
||||
young_A = mkA "noor" ;
|
||||
young_A = mkA (mkN "noor" "noore" "noort") ;
|
||||
|
||||
do_V2 = mkV2 (mkV "tegema" "teha") ;
|
||||
|
||||
now_Adv = mkAdv "nüüd" ;
|
||||
already_Adv = mkAdv "juba" ;
|
||||
song_N = mkN "laul" ;
|
||||
song_N = mkN "laul" "laulu" "laulu" ;
|
||||
add_V3 = mkV3 (mkV "lisama") accPrep callative ;
|
||||
number_N = mkN "number" ;
|
||||
put_V2 = mkV2 (mkV "panema") ;
|
||||
stop_V = mkV "peatuma" ;
|
||||
jump_V = mkV "hüppama" ;
|
||||
jump_V = mkV "hüppama" "hüpata" ;
|
||||
left_Ord = mkOrd1 (mkN "vasak") ;
|
||||
right_Ord = mkOrd1 (mkN "parem") ;
|
||||
right_Ord = mkOrd1 (mkN "parem" "parema" "paremat") ;
|
||||
far_Adv = mkAdv "kaugel" ;
|
||||
correct_A = mkA "õige" ;
|
||||
dry_A = mkA "kuiv" ;
|
||||
correct_A = mkA (mkN "õige" "õige" "õiget" "õigesse" "õigete" "õigeid") ;
|
||||
dry_A = mkA (mkN "kuiv" "kuiva" "kuiva") ;
|
||||
dull_A = mkA "igav" ;
|
||||
full_A = mkA (mkN "täis") "täiem" "täiim" Invariable ; -- 'täis' is one of the non-inflecting adjectives
|
||||
full_A = mkA (mkN "täis" "täie" "täit") "täiem" "täiim" Invariable ; -- 'täis' is one of the non-inflecting adjectives
|
||||
heavy_A = mkA "raske" ;
|
||||
near_A = mkA "lähedane" ;
|
||||
rotten_A = mkA "mäda" ;
|
||||
round_A = mkA "ümmargune" ;
|
||||
sharp_A = mkA "terav" ;
|
||||
smooth_A = mkA "sile" ;
|
||||
smooth_A = mkA (mkN "sile" "sileda" "siledat" "siledasse" "siledate" "siledaid") ;
|
||||
straight_A = mkA "sirge" ;
|
||||
wet_A = mkA "märg" ;
|
||||
wide_A = mkA "lai" ;
|
||||
animal_N = mkN "loom" ;
|
||||
ashes_N = mkN "tuhk" ;
|
||||
back_N = mkN "selg" ;
|
||||
bark_N = mkN "koor" ;
|
||||
belly_N = mkN "kõht" ;
|
||||
blood_N = mkN "veri" ;
|
||||
wet_A = mkA (mkN "märg" "märja" "märga") ;
|
||||
wide_A = mkA (mkN "lai" "laia" "laia") ;
|
||||
animal_N = mkN "loom" "looma" "looma" ;
|
||||
ashes_N = mkN "tuhk" "tuha" "tuhka" ;
|
||||
back_N = mkN "selg" "selja" "selga" ;
|
||||
bark_N = mkN "koor" "koore" "koort" ;
|
||||
belly_N = mkN "kõht" "kõhu" "kõhtu" ;
|
||||
blood_N = mkN "veri" "vere" "verd" "verre" "verede" "veresid" ;
|
||||
bone_N = mkN "luu" ;
|
||||
breast_N = mkN "rind" ;
|
||||
cloud_N = mkN "pilv" ;
|
||||
day_N = mkN "päev" ;
|
||||
dust_N = mkN "tolm" ;
|
||||
ear_N = mkN "kõrv" ;
|
||||
breast_N = mkN "rind" "rinna" "rinda" ;
|
||||
cloud_N = mkN "pilv" "pilve" "pilve" ;
|
||||
day_N = mkN "päev" "päeva" "päeva" ;
|
||||
dust_N = mkN "tolm" "tolmu" "tolmu" ;
|
||||
ear_N = mkN "kõrv" "kõrva" "kõrva" ;
|
||||
earth_N = mkN "maa" ;
|
||||
egg_N = mkN "muna" ;
|
||||
eye_N = mkN "silm" ;
|
||||
fat_N = mkN "rasv" ;
|
||||
feather_N = mkN "sulg" ;
|
||||
fingernail_N = mkN "küüs" ;
|
||||
fire_N = mkN "tuli" ;
|
||||
flower_N = mkN "lill" ;
|
||||
eye_N = mkN "silm" "silma" "silma" ;
|
||||
fat_N = mkN "rasv" "rasva" "rasva" ;
|
||||
feather_N = mkN "sulg" "sule" "sulge" ;
|
||||
fingernail_N = mkN "küüs" "küüne" "küünt" ;
|
||||
fire_N = mkN "tuli" "tule" "tuld" "tulle" "tulede" "tulesid" ;
|
||||
flower_N = mkN "lill" "lille" "lille" ;
|
||||
fog_N = mkN "udu" ;
|
||||
foot_N = mkN "jalg" ;
|
||||
forest_N = mkN "mets" ;
|
||||
grass_N = mkN "rohi" ;
|
||||
foot_N = mkN "jalg" "jala" "jalga" ;
|
||||
forest_N = mkN "mets" "metsa" "metsa" ;
|
||||
grass_N = mkN "rohi" "rohu" "rohtu" ;
|
||||
guts_N = mkN "soolestik" ;
|
||||
hair_N = mkN "juuksed" ; -- TODO: plural
|
||||
hand_N = mkN "käsi" ;
|
||||
hair_N = mkN "juuksed" "juuste" "juukseid" "juustesse" "juuste" "juukseid" ; -- TODO: plural
|
||||
hand_N = mkN "käsi" "käe" "kätt" ;
|
||||
head_N = mkN "pea" ;
|
||||
heart_N = mkN "süda" ;
|
||||
horn_N = mkN "sarv" ;
|
||||
heart_N = mkN "süda" "südame" "südant" ;
|
||||
horn_N = mkN "sarv" "sarve" "sarve" ;
|
||||
husband_N = man_N ;
|
||||
ice_N = mkN "jää" ;
|
||||
knee_N = mkN "põlv" ;
|
||||
leaf_N = mkN "leht" ;
|
||||
leg_N = mkN "jalg" ;
|
||||
liver_N = mkN "maks" ;
|
||||
knee_N = mkN "põlv" "põlve" "põlve" ;
|
||||
leaf_N = mkN "leht" "lehe" "lehte" ;
|
||||
leg_N = mkN "jalg" "jala" "jalga" ;
|
||||
liver_N = mkN "maks" "maksa" "maksa" ;
|
||||
louse_N = mkN "täi" ;
|
||||
mouth_N = mkN "suu" ;
|
||||
name_N = mkN "nimi" ;
|
||||
neck_N = mkN "kael" ;
|
||||
name_N = mkN "nimi" "nime" "nime" ;
|
||||
neck_N = mkN "kael" "kaela" "kaela" ;
|
||||
night_N = mkN "öö" ;
|
||||
nose_N = mkN "nina" ;
|
||||
person_N = mkN "inimene" ;
|
||||
rain_N = mkN "vihm" ;
|
||||
rain_N = mkN "vihm" "vihma" "vihma" ;
|
||||
road_N = mkN "tee" ;
|
||||
root_N = mkN "juur" ;
|
||||
rope_N = mkN "köis" ;
|
||||
salt_N = mkN "sool" ;
|
||||
sand_N = mkN "liiv" ;
|
||||
seed_N = mkN "seeme" ;
|
||||
skin_N = mkN "nahk" ;
|
||||
sky_N = mkN "taevas" ;
|
||||
smoke_N = mkN "suits" ;
|
||||
snow_N = mkN "lumi" ;
|
||||
root_N = mkN "juur" "juure" "juurt" ;
|
||||
rope_N = mkN "köis" "köie" "köit" ;
|
||||
salt_N = mkN "sool" "soola" "soola" ;
|
||||
sand_N = mkN "liiv" "liiva" "liiva" ;
|
||||
seed_N = mkN "seeme" "seemne" "seemet" ;
|
||||
skin_N = mkN "nahk" "naha" "nahka" ;
|
||||
sky_N = mkN "taevas" "taeva" "taevast" ;
|
||||
smoke_N = mkN "suits" "suitsu" "suitsu" ;
|
||||
snow_N = mkN "lumi" "lume" "lund" ;
|
||||
stick_N = mkN "kepp" ;
|
||||
tail_N = mkN "saba" ;
|
||||
tongue_N = mkN "keel" ;
|
||||
tooth_N = mkN "hammas" ;
|
||||
tongue_N = mkN "keel" "keele" "keelt" ;
|
||||
tooth_N = mkN "hammas" "hamba" "hammast" ;
|
||||
wife_N = mkN "naine" ;
|
||||
wind_N = mkN "tuul" ;
|
||||
wing_N = mkN "tiib" ;
|
||||
wind_N = mkN "tuul" "tuule" "tuult" ;
|
||||
wing_N = mkN "tiib" "tiiva" "tiiba" ;
|
||||
worm_N = mkN "uss" ;
|
||||
year_N = mkN "aasta" ;
|
||||
bite_V2 = mkV2 (mkV "purema") ;
|
||||
@@ -311,13 +310,13 @@ lin
|
||||
burn_V = mkV "põlema" ;
|
||||
count_V2 = mkV2 (mkV "loendama") ;
|
||||
cut_V2 = mkV2 (mk2V "lõikama" "lõigata") ;
|
||||
dig_V = mkV "kaevama" ;
|
||||
dig_V = mkV "kaevama" "kaevata" ;
|
||||
fall_V = mkV "kukkuma" ;
|
||||
fear_V2 = mkV2 (mkV "kartma") cpartitive ;
|
||||
fight_V2 = mkV2 (mkV "võitlema") (postPrep partitive "vastu") ;
|
||||
float_V = mkV "liuglema" ;
|
||||
flow_V = mkV "voolama" ;
|
||||
fly_V = mkV "lendama" ;
|
||||
flow_V = mkV "voolama" "voolata" ;
|
||||
fly_V = mkV "lendama" "lennata" ;
|
||||
freeze_V = mkV "jäätuma" ;
|
||||
give_V3 = mkV3 (mkV "andma") accPrep callative ;
|
||||
hit_V2 = mkV2 (mkV "lööma") cpartitive ;
|
||||
@@ -327,11 +326,11 @@ lin
|
||||
laugh_V = mkV "naerma" ;
|
||||
lie_V = mkV "lamama" ; -- TODO: maybe: valetama
|
||||
play_V = mkV "mängima" ;
|
||||
pull_V2 = mkV2 (mkV "tõmbama") ;
|
||||
pull_V2 = mkV2 (mkV "tõmbama" "tõmmata") ;
|
||||
push_V2 = mkV2 (mkV "suruma") ;
|
||||
rub_V2 = mkV2 (mkV "hõõruma") cpartitive ;
|
||||
scratch_V2 = mkV2 (mkV "kraapima") cpartitive ;
|
||||
sew_V = mkV "külvama" ;
|
||||
sew_V = mkV "külvama" "külvata" ;
|
||||
sing_V = mkV "laulma" "laulda" "laulab" "lauldakse" ;
|
||||
sit_V = mkV "istuma" ;
|
||||
smell_V = mk2V "haistma" "haista";
|
||||
@@ -344,9 +343,9 @@ lin
|
||||
swell_V = mkV "paistetama" ; -- TODO: üles paistetama
|
||||
swim_V = mkV "ujuma" ;
|
||||
think_V = mkV "mõtlema" "mõtelda" "mõtleb" "mõeldakse";
|
||||
throw_V2 = mkV2 (mkV "viskama") ;
|
||||
throw_V2 = mkV2 (mkV "viskama" "visata") ;
|
||||
tie_V2 = mkV2 (mkV "siduma" "siduda" "seob") ;
|
||||
turn_V = mkV "pöörama" ;
|
||||
turn_V = mkV "pöörama" "pöörata" ;
|
||||
vomit_V = mkV "oksendama" ;
|
||||
wash_V2 = mkV2 (mkV "pesema") ;
|
||||
wipe_V2 = mkV2 (mkV "pühkima") ;
|
||||
@@ -354,7 +353,7 @@ lin
|
||||
breathe_V = mkV "hingama" ;
|
||||
|
||||
grammar_N = mkN "grammatika" ;
|
||||
language_N = mkN "keel" ;
|
||||
language_N = tongue_N ;
|
||||
rule_N = mkN "reegel" ;
|
||||
|
||||
john_PN = mkPN "Juhan" ;
|
||||
@@ -362,7 +361,7 @@ lin
|
||||
ready_A = mkA (mkN "valmis" "valmi" "valmit" "valmisse" "valmite" "valmeid") "valmim" "valmeim" Invariable ; -- 'valmis' is one of the non-inflecting adjectives
|
||||
reason_N = mkN "põhjus" ;
|
||||
today_Adv = mkAdv "täna" ;
|
||||
uncertain_A = mkA "ebakindel" ;
|
||||
uncertain_A = mkA (mkN "ebakindel" "ebakindla" "ebakindlat") ;
|
||||
|
||||
|
||||
oper
|
||||
|
||||
@@ -16,8 +16,12 @@
|
||||
--
|
||||
-- The structure of functions for each word class $C$ is the following:
|
||||
-- there is a polymorphic constructor $mkC$, which takes one or
|
||||
-- a few arguments. In Estonian, one argument is enough in ??? % of
|
||||
-- a few arguments. In Estonian, one argument is enough in 90% of
|
||||
-- cases in average.
|
||||
--
|
||||
-- @author Inari Listenmaa
|
||||
-- @author Kaarel Kaljurand
|
||||
-- @version 2013-10-21
|
||||
|
||||
resource ParadigmsEst = open
|
||||
(Predef=Predef),
|
||||
@@ -59,7 +63,7 @@ oper
|
||||
comitative : Case ; -- e.g. "karbiga"
|
||||
|
||||
infDa : InfForm ; -- e.g. "lugeda"
|
||||
infDes : InfForm ;
|
||||
infDes : InfForm ; -- e.g. "lugedes"
|
||||
infMa : InfForm ; -- e.g. "lugema"
|
||||
infMas : InfForm ; -- e.g. "lugemas"
|
||||
infMaks : InfForm ; -- e.g. "lugemaks"
|
||||
@@ -70,49 +74,30 @@ oper
|
||||
-- of many-place verbs and adjective. A complement can be defined by
|
||||
-- just a case, or a pre/postposition and a case.
|
||||
|
||||
prePrep : Case -> Str -> Prep ; -- preposition, e.g. comitative "koos"
|
||||
prePrep : Case -> Str -> Prep ; -- preposition, e.g. abessive "ilma"
|
||||
postPrep : Case -> Str -> Prep ; -- postposition, e.g. genitive "taga"
|
||||
postGenPrep : Str -> Prep ; -- genitive postposition, e.g. "taga"
|
||||
casePrep : Case -> Prep ; -- just case, e.g. adessive
|
||||
|
||||
-- TODO build the dict
|
||||
NW : Type ; -- Noun from DictEst (WordNet)
|
||||
AW : Type ; -- Adjective from DictEst (WordNet)
|
||||
VW : Type ; -- Verb from DictEst (WordNet)
|
||||
AdvW : Type ; -- Adverb from DictEst (WordNet)
|
||||
|
||||
--2 Nouns
|
||||
|
||||
-- The worst case gives six forms.
|
||||
-- In practice just a couple of forms are needed to define the different
|
||||
-- stems, vowel alternation, and vowel harmony.
|
||||
|
||||
oper
|
||||
|
||||
-- The regular noun heuristic takes just one form (singular
|
||||
-- nominative) and analyses it to pick the correct paradigm.
|
||||
-- It does automatic grade alternation, and is hence not usable
|
||||
-- for words like "auto" (whose genitive would become "audon").
|
||||
--
|
||||
-- If the one-argument paradigm does not give the correct result, one can try and give
|
||||
-- two or three forms. Most notably, the two-argument variant is used
|
||||
-- for nouns like "kivi - kiviä", which would otherwise become like
|
||||
-- "rivi - rivejä". Three arguments are used e.g. for
|
||||
-- "auto - auton - autoja", which would otherwise become
|
||||
-- "auto - audon".
|
||||
-- If the 1-argument paradigm does not give the correct result,
|
||||
-- one can try and give 2, 3, 4, or 6 forms.
|
||||
|
||||
mkN : overload {
|
||||
mkN : (kukko : Str) -> N ; -- predictable nouns, covers 82%
|
||||
mkN : (savi,savia : Str) -> N ; -- different pl.part
|
||||
mkN : (vesi,veden,vesia : Str) -> N ; -- also different sg.gen
|
||||
mkN : (ema : Str) -> N ; -- predictable nouns, covers 90%
|
||||
mkN : (tukk,tuku : Str) -> N ; -- sg nom,gen: unpredictable stem vowel
|
||||
mkN : (tukk,tuku,tukku : Str) -> N ; -- sg nom,gen,part
|
||||
mkN : (pank,panga,panka,panku : Str) -> N ; -- sg nom,gen,part, pl.part
|
||||
|
||||
-- mkN : (olo,n,a,na,oon,jen,ja,ina,issa,ihin : Str) -> N ; -- worst case, 10 forms
|
||||
mkN : (oun,ouna,ouna,ounasse,ounte,ounu : Str) -> N ; -- worst case, 6 forms
|
||||
mkN : (oun,ouna,ouna,ounasse,ounte,ounu,ountesse : Str) -> N ; -- worst case, 7 forms
|
||||
mkN : (pika : Str) -> (juna : N) -> N ; -- compound with invariable prefix
|
||||
mkN : (oma : N) -> (tunto : N) -> N ; -- compound with inflecting prefix
|
||||
mkN : NW -> N ; -- noun from DictEst (WordNet)
|
||||
-- mkN : (pika : Str) -> (juna : N) -> N ; -- compound with invariable prefix
|
||||
-- mkN : (oma : N) -> (tunto : N) -> N ; -- compound with inflecting prefix
|
||||
} ;
|
||||
|
||||
-- Nouns used as functions need a case, of which the default is
|
||||
@@ -141,22 +126,20 @@ oper
|
||||
-- The comparative and the superlative
|
||||
-- are always inflected in the same way, so the nominative of them is actually
|
||||
-- enough (TODO: confirm).
|
||||
-- TODO: update these types to include the new boolean non-inflection marker
|
||||
|
||||
mkA : overload {
|
||||
mkA : Str -> A ; -- regular noun made into adjective
|
||||
mkA : N -> A ; -- any noun made into adjective
|
||||
mkA : N -> (infl : Bool) -> A ; -- noun made into adjective, agreement type specified
|
||||
mkA : N -> (infl : Infl) -> A ; -- noun made into adjective, agreement type specified
|
||||
mkA : N -> (parem, parim : Str) -> A ; -- deviating comparison forms
|
||||
mkA : AW -> A ; -- adjective from DictEst (WordNet)
|
||||
} ;
|
||||
|
||||
-- Two-place adjectives need a case for the second argument.
|
||||
|
||||
mkA2 : A -> Prep -> A2 -- e.g. "jaollinen" casePrep adessive
|
||||
mkA2 : A -> Prep -> A2 -- e.g. "vihane" (postGenPrep "peale")
|
||||
= \a,p -> a ** {c2 = p ; lock_A2 = <>};
|
||||
|
||||
genAttrA : Str -> A ; -- genitive attributes ; no agreement to head, no comparison forms.
|
||||
invA : Str -> A ; -- invariable adjectives, such as genitive attributes ; no agreement to head, no comparison forms.
|
||||
|
||||
--2 Verbs
|
||||
--
|
||||
@@ -166,19 +149,18 @@ oper
|
||||
-- The worst case needs eight forms, as shown in the following.
|
||||
|
||||
mkV : overload {
|
||||
mkV : (lugema : Str) -> V ; -- predictable verbs, covers n %
|
||||
mkV : (lugema,lugeda : Str) -> V ; -- deviating past 3sg
|
||||
mkV : (lugema,loeb,lugeda : Str) -> V ; -- also deviating pres. 1sg
|
||||
mkV : (lugema,lugeda,loeb,loetakse : Str) -> V ;
|
||||
mkV : (tegema,teha,teeb,tehakse,tehke,tegi,teinud,tehtud : Str) -> V ; -- worst-case verb
|
||||
mkV : (saama : V) -> (aru : Str) -> V ; -- püsiühendid TODO
|
||||
mkV : VW -> V ; -- verb from DictEst (WordNet)
|
||||
mkV : (lugema : Str) -> V ; -- predictable verbs, covers 90 %
|
||||
mkV : (lugema,lugeda : Str) -> V ; -- ma infinitive, da infinitive
|
||||
mkV : (lugema,lugeda,loeb : Str) -> V ; -- ma, da, present sg 3
|
||||
mkV : (lugema,lugeda,loeb,loetakse : Str) -> V ; --ma, da, pres sg 3, pres passive
|
||||
mkV : (tegema,teha,teeb,tehakse,tehke,tegi,teinud,tehtud : Str) -> V ; -- worst-case verb, 8 forms
|
||||
mkV : (saama : V) -> (aru : Str) -> V ; -- multi-word verbs
|
||||
} ;
|
||||
|
||||
-- All the patterns above have $nominative$ as subject case.
|
||||
-- If another case is wanted, use the following.
|
||||
|
||||
caseV : Case -> V -> V ; -- deviating subj. case, e.g. genitive "täytyä"
|
||||
caseV : Case -> V -> V ; -- deviating subj. case, e.g. allative "meeldima"
|
||||
|
||||
-- The verbs "be" and "go" are special.
|
||||
|
||||
@@ -206,9 +188,9 @@ oper
|
||||
-- Three-place (ditransitive) verbs need two prepositions, of which
|
||||
-- the first one or both can be absent.
|
||||
|
||||
mkV3 : V -> Prep -> Prep -> V3 ; -- e.g. puhua, allative, elative
|
||||
dirV3 : V -> Case -> V3 ; -- siirtää, (accusative), illative
|
||||
dirdirV3 : V -> V3 ; -- antaa, (accusative), (allative)
|
||||
mkV3 : V -> Prep -> Prep -> V3 ; -- e.g. rääkima, allative, elative
|
||||
dirV3 : V -> Case -> V3 ; -- liigutama, (accusative), illative
|
||||
dirdirV3 : V -> V3 ; -- andma, (accusative), (allative)
|
||||
|
||||
|
||||
--3 Other complement patterns
|
||||
@@ -218,15 +200,15 @@ oper
|
||||
|
||||
mkV0 : V -> V0 ; --%
|
||||
mkVS : V -> VS ;
|
||||
mkV2S : V -> Prep -> V2S ; -- e.g. "sanoa" allative
|
||||
mkVV : V -> VV ; -- e.g. "alkaa"
|
||||
mkV2S : V -> Prep -> V2S ; -- e.g. "ütlema" allative
|
||||
mkVV : V -> VV ; -- e.g. "hakkama"
|
||||
mkVVf : V -> InfForm -> VV ; -- e.g. "hakkama" infMa
|
||||
mkV2V : V -> Prep -> V2V ; -- e.g. "käskeä" genitive
|
||||
mkV2Vf : V -> Prep -> InfForm -> V2V ; -- e.g. "kieltää" partitive infMast
|
||||
mkVA : V -> Prep -> VA ; -- e.g. "maistua" ablative
|
||||
mkV2A : V -> Prep -> Prep -> V2A ; -- e.g. "maalata" accusative translative
|
||||
mkV2V : V -> Prep -> V2V ; -- e.g. "käskima" adessive
|
||||
mkV2Vf : V -> Prep -> InfForm -> V2V ; -- e.g. "keelama" partitive infMast
|
||||
mkVA : V -> Prep -> VA ; -- e.g. "muutuma" translative
|
||||
mkV2A : V -> Prep -> Prep -> V2A ; -- e.g. "värvima" genitive translative
|
||||
mkVQ : V -> VQ ;
|
||||
mkV2Q : V -> Prep -> V2Q ; -- e.g. "kysyä" ablative
|
||||
mkV2Q : V -> Prep -> V2Q ; -- e.g. "küsima" ablative
|
||||
|
||||
mkAS : A -> AS ; --%
|
||||
mkA2S : A -> Prep -> A2S ; --%
|
||||
@@ -279,22 +261,16 @@ oper
|
||||
\c -> {c = NPCase c ; s = [] ; isPre = True ; lock_Prep = <>} ;
|
||||
accPrep = {c = NPAcc ; s = [] ; isPre = True ; lock_Prep = <>} ;
|
||||
|
||||
NW = {s : NForms ; lock_NW : {}} ;
|
||||
AW = {s : NForms ; lock_AW : {}} ;
|
||||
VW = {s : VForms ; lock_VW : {}} ;
|
||||
AdvW = {s : Str ; lock_AdvW : {}} ;
|
||||
|
||||
|
||||
mkN = overload {
|
||||
mkN : (nisu : Str) -> N = mk1N ;
|
||||
mkN : (link,lingi : Str) -> N = mk2N ;
|
||||
mkN : (tukk,tuku,tukku : Str) -> N = mk3N ;
|
||||
mkN : (paat,paadi,paati,paatide : Str) -> N = mk4N ;
|
||||
mkN : (paat,paadi,paati,paate : Str) -> N = mk4N ;
|
||||
mkN : (oun,ouna,ouna,ounasse,ounte,ounu : Str) -> N = mk6N ;
|
||||
|
||||
mkN : (sora : Str) -> (tie : N) -> N = mkStrN ;
|
||||
mkN : (oma,tunto : N) -> N = mkNN ;
|
||||
mkN : (sana : NW) -> N = \w -> nForms2N w.s ;
|
||||
} ;
|
||||
|
||||
-- Adjective forms (incl. comp and sup) are derived from noun forms
|
||||
@@ -309,6 +285,8 @@ oper
|
||||
|
||||
mk1N : (link : Str) -> N = \s -> nForms2N (hjk_type s) ** {lock_N = <> } ;
|
||||
|
||||
-- mk2N, mk3N, mk4N make sure that the user specified forms end up in the paradigm,
|
||||
-- even though the rest is wrong
|
||||
mk2N : (link,lingi : Str) -> N = \link,lingi ->
|
||||
let nfs : NForms = (nForms2 link lingi) ;
|
||||
nfs_fixed : NForms = table {
|
||||
@@ -334,24 +312,16 @@ oper
|
||||
} ;
|
||||
in nForms2N nfs_fixed ** {lock_N = <> } ;
|
||||
|
||||
{- mk1N : (link : Str) -> N = \s -> nForms2N (hjk_type s) ** {lock_N = <> } ;
|
||||
mk2N : (link,lingi : Str) -> N = \s,t -> nForms2N (nForms2 s t) ** {lock_N = <>} ;
|
||||
mk3N : (tukk,tuku,tukku : Str) -> N = \s,t,u -> nForms2N (nForms3 s t u) ** {lock_N = <>} ;
|
||||
--regular mk4N
|
||||
mk4N : (paat,paadi,paati,paate : Str) -> N = \s,t,u,v -> nForms2N (nForms4 s t u v) ** {lock_N = <>} ;
|
||||
-}
|
||||
|
||||
--experimental: make sure that the user specified forms end up in the paradigm, even though the rest is wrong
|
||||
--this is using pl part
|
||||
mk4N : (paat,paadi,paati,paatide : Str) -> N = \paat,paadi,paati,paate ->
|
||||
let nfs : NForms = (nForms4 paat paadi paati paate) ;
|
||||
mk4N : (paat,paadi,paati,paate : Str) -> N = \paat,paadi,paati,paate ->
|
||||
let nfs : NForms = (nForms4 paat paadi paati paate) ;
|
||||
nfs_fixed : NForms = table {
|
||||
0 => paat ;
|
||||
1 => paadi ;
|
||||
2 => paati ;
|
||||
3 => nfs ! 3 ;
|
||||
4 => nfs ! 4 ;
|
||||
5 => paate
|
||||
5 => paate
|
||||
} ;
|
||||
in nForms2N nfs_fixed ** {lock_N = <> } ;
|
||||
|
||||
@@ -403,7 +373,7 @@ oper
|
||||
--heuristics to catch palk:palga but not maakas:maaka (for longer words, same with more ?s)
|
||||
--didn't work, don't try this
|
||||
--<? + ? + #c, ? + ? + #c + #v> => hjk_type_IVb_audit link i ;
|
||||
_ => hjk_type link
|
||||
_ => hjk_type2 link i
|
||||
} ;
|
||||
|
||||
nForms3 : (_,_,_ : Str) -> NForms = \tukk,tuku,tukku ->
|
||||
@@ -422,13 +392,12 @@ oper
|
||||
|
||||
<_ + "ik", _ + "iku", _ + "ikku"> => hjk_type_VI_imelik tukk ; --imelik:_:imelikku caught here
|
||||
|
||||
<_ + #c, _ + #v, _ + #v> => hjk_type_VI_tukk tukk tuku ;
|
||||
<_ + "ud", _ + "u", _ + "ut"> => nForms2 tukk tuku ; -- -nud/-tud participles are not like 'voolik'
|
||||
<_ + #c, _ + #v, _ + #v + "t"> => hjk_type_IVb_audit tukk u ; --voolik:_:voolikut caught here
|
||||
_ => nForms2 tukk tuku
|
||||
} ;
|
||||
|
||||
nForms4 : (_,_,_,_ : Str) -> NForms = \paat,paadi,paati,paate ->
|
||||
nForms4 : (_,_,_,_ : Str) -> NForms = \paat,paadi,paati,paate ->
|
||||
case <paat,paadi,paati,paate> of {
|
||||
-- distinguish between joonis and segadus
|
||||
<_ +("ne"|"s"), _+"se", _+"st", _+"seid"> => hjk_type_Va_otsene paat ;
|
||||
@@ -442,6 +411,7 @@ oper
|
||||
|
||||
_ => nForms3 paat paadi paati
|
||||
} ;
|
||||
|
||||
{-
|
||||
--Version that uses pl gen instead of pl part
|
||||
nForms4 : (_,_,_,_ : Str) -> NForms = \paat,paadi,paati,paatide ->
|
||||
@@ -468,7 +438,7 @@ oper
|
||||
|
||||
mmkN2 : N -> Prep -> N2 = \n,c -> n ** {c2 = c ; isPre = mkIsPre c ; lock_N2 = <>} ;
|
||||
mkN3 = \n,c,e -> n ** {c2 = c ; c3 = e ;
|
||||
isPre = mkIsPre c ; -- matka Lontoosta Pariisiin
|
||||
isPre = mkIsPre c ; -- matka Londonist Pariisi
|
||||
isPre2 = mkIsPre e ; -- Suomen voitto Ruotsista
|
||||
lock_N3 = <>
|
||||
} ;
|
||||
@@ -495,10 +465,9 @@ oper
|
||||
-- TODO: temporary usage of regAdjective1
|
||||
mkA : N -> (valmim,valmeim : Str) -> (infl : Infl) -> A =
|
||||
\n,c,s,infl -> (regAdjective1 n c s) ** {infl = infl ; lock_A = <>} ;
|
||||
mkA : (sana : AW) -> A = \w -> noun2adjDeg (nForms2N w.s) ** {infl = Regular} ;
|
||||
} ;
|
||||
|
||||
genAttrA balti = {s = \\_,_ => balti ; infl = Invariable ; lock_A = <>} ;
|
||||
invA balti = {s = \\_,_ => balti ; infl = Invariable ; lock_A = <>} ;
|
||||
|
||||
mkA_1 : Str -> A = \x -> noun2adjDeg (mk1N x) ** {infl = Regular ; lock_A = <>} ;
|
||||
|
||||
@@ -550,7 +519,6 @@ oper
|
||||
mkV : (lugema,lugeda,loeb,loetakse : Str) -> V = mk4V ;
|
||||
mkV : (tegema,teha,teeb,tehakse,tehke,tegi,teinud,tehtud : Str) -> V = mk8V ;
|
||||
mkV : (aru : Str) -> (saama : V) -> V = mkPV ; -- particle verbs
|
||||
mkV : (sana : VW) -> V = \w -> vforms2V w.s ** {sc = NPCase Nom ; lock_V = <>} ;
|
||||
} ;
|
||||
|
||||
mk1V : Str -> V = \s ->
|
||||
@@ -742,7 +710,6 @@ oper
|
||||
|
||||
mkAdv = overload {
|
||||
mkAdv : Str -> Adv = \s -> {s = s ; lock_Adv = <>} ;
|
||||
mkAdv : AdvW -> Adv = \s -> {s = s.s ; lock_Adv = <>} ;
|
||||
} ;
|
||||
|
||||
mkV2 = overload {
|
||||
|
||||
Reference in New Issue
Block a user