forked from GitHub/gf-core
Latvian: finally handles PassV2 clauses in a more or less adequate way
Extends the previously introduced support for specifying the verb valence (verb-dependent subject and object cases), involving functions up to PredVP and RelVP.
In Latvian, the passive voice is not used if the agent (subject) is known; to preserve the information structure (i.e. the word order), a clause like 'A is <done> by B' is linearized in the active voice ('A <does> B') where A has the object case (e.g. Acc), and B - the subject case (e.g. Nom). Thus, the verb valence patterns are swapped on-the-fly.
This is still a rather quick & dirty implementation: parameters and linearization types have to be optimized (VerbLav), the use of PassV2 in AdvVP is problematic as it doesn't apply VPSlashPrep / ComplSlash, etc.
This commit is contained in:
normundsg
@@ -15,6 +15,9 @@ flags
|
||||
lin
|
||||
language_title_Utt = ss "latviešu valoda" ;
|
||||
|
||||
yes_Utt = ss "jā" ;
|
||||
no_Utt = ss "nē" ;
|
||||
|
||||
every_Det = {
|
||||
s = (\\g,c => (mkPronoun_Gend "ikviens").s ! g ! Sg ! c) ;
|
||||
n = Sg ;
|
||||
@@ -23,14 +26,14 @@ lin
|
||||
} ;
|
||||
|
||||
someSg_Det = {
|
||||
s = (\\g,c => (mkPronoun_Gend "kāds").s ! g ! Sg ! c) ; -- lai atļautu arī tukšo, jāliek (\\_,_ => []) | klāt
|
||||
s = (\\g,c => (mkPronoun_Gend "kāds").s ! g ! Sg ! c) ; -- TODO: lai atļautu arī tukšo, jāpieliek alternatīva: (\\_,_ => [])
|
||||
n = Sg ;
|
||||
d = Indef ;
|
||||
isNeg = False
|
||||
} ;
|
||||
|
||||
somePl_Det = {
|
||||
s = (\\g,c => (mkPronoun_Gend "kāds").s ! g ! Pl ! c) ; -- lai atļautu arī tukšo, jāliek (\\_,_ => []) | klāt
|
||||
s = (\\g,c => (mkPronoun_Gend "kāds").s ! g ! Pl ! c) ; -- TODO: lai atļautu arī tukšo, jāpieliek alternatīva: (\\_,_ => [])
|
||||
n = Pl ;
|
||||
d = Indef ;
|
||||
isNeg = False
|
||||
@@ -44,14 +47,14 @@ lin
|
||||
} ;
|
||||
|
||||
many_Det = {
|
||||
s = (\\g,c => (mkPronoun_Gend "daudzs").s ! g ! Pl ! c) ; -- 'daudzs' izlocīsies korekti uz daudzskaitļa 'daudzi'
|
||||
s = (\\g,c => (mkPronoun_Gend "daudzs").s ! g ! Pl ! c) ; -- 'daudzs' izlocīsies korekti uz daudzskaitļa 'daudzi'; tomēr nesmuki...
|
||||
n = Pl ;
|
||||
d = Indef ;
|
||||
isNeg = False
|
||||
} ;
|
||||
|
||||
much_Det = {
|
||||
s = (\\g,c => "daudz") ; -- FIXME - ņem saistību ar ģenitīvu; kā to realizēt?
|
||||
s = (\\g,c => "daudz") ; -- FIXME: piesaista ģenitīvu
|
||||
n = Sg ;
|
||||
d = Indef ;
|
||||
isNeg = False
|
||||
@@ -107,42 +110,36 @@ lin
|
||||
|
||||
it_Pron = mkPronoun_It_Sg Masc ; -- See also: ExtraLav.it8fem_Pron
|
||||
|
||||
-- manuprāt prievārdi tomēr ir valodas-specifiski un nebūtu tieši 1-pret-1 jātulko
|
||||
-- Pronouns; their translation is very ambiguos...
|
||||
|
||||
above_Prep = mkPrep "virs" Gen Dat ;
|
||||
after_Prep = mkPrep "pēc" Gen Dat ;
|
||||
before_Prep = mkPrep "pirms" Gen Dat ;
|
||||
behind_Prep = mkPrep "aiz" Gen Dat ; -- taču "aiz" nav viennozīmīgi "behind"
|
||||
between_Prep = mkPrep "starp" Acc Dat ;
|
||||
during_Prep = mkPrep Loc ;
|
||||
for_Prep = mkPrep "priekš" Gen Dat ;
|
||||
from_Prep = mkPrep "no" Gen Dat ;
|
||||
on_Prep = mkPrep "uz" Gen Dat ;
|
||||
with_Prep = mkPrep "ar" Acc Dat ; -- ar sievu, ar sievām
|
||||
in_Prep = mkPrep Loc ;
|
||||
to_Prep = mkPrep "uz" Acc Dat ; -- See also: ExtraLav.liidz_Prep, pie_Prep
|
||||
possess_Prep = mkPrep Gen ; -- FIXME: vajadzētu vārdu secību otrādi - pirms paskaidrojamā vārda likt
|
||||
under_Prep = mkPrep "zem" Gen Dat ;
|
||||
without_Prep = mkPrep "bez" Gen Dat ;
|
||||
by8agent_Prep = nom_Prep ; -- TODO: should get rid of this Prep; taču tas parādās Eng2Abs kokos... -- FIXME: nom_Prep vs. dat_Prep/accPrep
|
||||
by8means_Prep = mkPrep "ar" Acc Dat ;
|
||||
during_Prep = mkPrep Loc ;
|
||||
in8front_Prep = mkPrep "priekšā" Dat Dat ;
|
||||
part_Prep = mkPrep Gen ; --FIXME - vajadzētu vārdu secību otrādi - pirms paskaidrojamā vārda likt
|
||||
on_Prep = mkPrep "uz" Gen Dat ;
|
||||
through_Prep = mkPrep "caur" Acc Dat ;
|
||||
except_Prep = mkPrep "izņemot" Acc Acc ;
|
||||
|
||||
to_Prep = mkPrep "uz" Acc Dat ; -- See also: ExtraLav.liidz_Prep, pie_Prep
|
||||
under_Prep = mkPrep "zem" Gen Dat ;
|
||||
with_Prep = mkPrep "ar" Acc Dat ;
|
||||
without_Prep = mkPrep "bez" Gen Dat ;
|
||||
|
||||
by8agent_Prep = nom_Prep ; -- TODO: nom_Prep (default) vs. dat_Prep
|
||||
by8means_Prep = mkPrep "ar" Acc Dat ;
|
||||
except_Prep = mkPrep "izņemot" Acc Acc ; -- Acc - by default?
|
||||
part_Prep = mkPrep Gen ; --FIXME - vajadzētu vārdu secību otrādi - pirms paskaidrojamā vārda likt
|
||||
possess_Prep = mkPrep Gen ; -- FIXME: vajadzētu vārdu secību otrādi - pirms paskaidrojamā vārda likt
|
||||
|
||||
very_AdA = mkAdA "ļoti" ;
|
||||
almost_AdA = mkAdA "gandrīz" ;
|
||||
so_AdA = mkAdA "tik" ;
|
||||
too_AdA = mkAdA "pārāk" ;
|
||||
|
||||
and_Conj = mkConj "un" ;
|
||||
or_Conj = mkConj "vai" Sg ;
|
||||
if_then_Conj = mkConj "ja" "tad" ;
|
||||
|
||||
but_PConj = ss "bet" ;
|
||||
otherwise_PConj = ss "citādi" ;
|
||||
therefore_PConj = ss "tāpēc" ;
|
||||
|
||||
more_CAdv = (mkCAdv [] "nekā" Compar) | (mkCAdv "vairāk" "nekā" Posit) ;
|
||||
less_CAdv = mkCAdv "mazāk" "nekā" Posit ;
|
||||
as_CAdv = mkCAdv "tikpat" "kā" Posit ;
|
||||
@@ -157,10 +154,16 @@ lin
|
||||
somewhere_Adv = mkAdv "kaut kur" ;
|
||||
quite_Adv = mkAdv "diezgan" ;
|
||||
|
||||
both7and_DConj = mkConj "gan" ("," ++ "gan"); --FIXME - komati nav tā kā vajag
|
||||
either7or_DConj = mkConj ("vai" ++ "nu") ("," ++ "vai") Sg ; --FIXME - komati nav tā kā vajag
|
||||
and_Conj = mkConj "un" ;
|
||||
or_Conj = mkConj "vai" Sg ;
|
||||
if_then_Conj = mkConj "ja" "tad" ;
|
||||
|
||||
want_VV = mkVV (mkV "vēlēties" third_conjugation) ;
|
||||
but_PConj = ss "bet" ;
|
||||
otherwise_PConj = ss "citādi" ;
|
||||
therefore_PConj = ss "tāpēc" ;
|
||||
|
||||
both7and_DConj = mkConj "gan" ("," ++ "gan"); -- FIXME: komati nav tā kā vajag
|
||||
either7or_DConj = mkConj ("vai" ++ "nu") ("," ++ "vai") Sg ; -- FIXME: komati nav tā kā vajag
|
||||
|
||||
whoSg_IP = { -- FIXME: Fem
|
||||
s = table {
|
||||
@@ -223,6 +226,11 @@ lin
|
||||
}
|
||||
} ;
|
||||
|
||||
how8many_IDet = {
|
||||
s = table { _ => "cik" } ;
|
||||
n = Pl
|
||||
} ;
|
||||
|
||||
when_Subj = ss "kad" ;
|
||||
although_Subj = ss "kaut arī" ;
|
||||
because_Subj = ss "jo" ;
|
||||
@@ -230,24 +238,15 @@ lin
|
||||
that_Subj = ss "ka" ;
|
||||
|
||||
all_Predet = { s = table { Masc => "visi" ; Fem => "visas" } } ;
|
||||
|
||||
only_Predet = { s = table { _ => "tikai"} } ;
|
||||
most_Predet = { s = table { _ => "vairums"} } ;
|
||||
|
||||
yes_Utt = ss "jā" ;
|
||||
please_Voc = ss "lūdzu" ;
|
||||
|
||||
almost_AdN = mkAdN "gandrīz" ;
|
||||
at_least_AdN = mkAdN "vismaz" ;
|
||||
at_most_AdN = mkAdN "ne vairāk kā" ;
|
||||
|
||||
always_AdV = mkAdV "vienmēr" ;
|
||||
|
||||
how8many_IDet = {
|
||||
s = table { _ => "cik" } ;
|
||||
n = Pl
|
||||
} ;
|
||||
|
||||
somebody_NP = UsePron (mkPronoun_Body "kāds") ;
|
||||
something_NP = UsePron (mkPronoun_Thing "kaut kas") ;
|
||||
everybody_NP = UsePron (mkPronoun_Body "ikviens") ;
|
||||
@@ -258,9 +257,13 @@ lin
|
||||
have_V2 = mkV2 (mkV "būt") nom_Prep Dat ;
|
||||
have_V3 = mkV3 (mkV "būt") nom_Prep dat_Prep Dat ;
|
||||
|
||||
can8know_VV, can_VV = mkVV (mkV "varēt" third_conjugation) ;
|
||||
want_VV = mkVV (mkV "vēlēties" third_conjugation) ;
|
||||
can_VV = mkVV (mkV "varēt" third_conjugation) ;
|
||||
can8know_VV = mkVV (mkV "varēt" third_conjugation) ;
|
||||
must_VV = mkVV (mkV "vajadzēt" third_conjugation) Dat ;
|
||||
|
||||
please_Voc = ss "lūdzu" ;
|
||||
|
||||
oper
|
||||
|
||||
reflPron : Case => Str = table {
|
||||
@@ -288,6 +291,4 @@ oper
|
||||
} ;
|
||||
-}
|
||||
|
||||
no_Utt = ss "nē" ;
|
||||
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user