GitHub/gf-core
1
0
Fork 1
mirror of https://github.com/GrammaticalFramework/gf-core.git synced 2026-05-22 09:32:53 -06:00
Code Issues Packages Projects Releases Wiki Activity

Prasad's sanskrit transliteration ; MiniresourceSan now compiles but is mostly incorrect due to missing paradigms

Browse Source
This commit is contained in:
aarne
2013-05-31 16:25:42 +00:00
parent 54a0642bee
commit f33059ae39
2 changed files with 109 additions and 102 deletions
Download Patch File Download Diff File Expand all files Collapse all files

194
examples/miniresource/MiniresourceSan.gf
View File

@@ -8,99 +8,86 @@ concrete MiniresourceSan of Miniresource = open Prelude in {
S = {s : Str} ; S = {s : Str} ;
Cl = {s : Bool => Str} ; Cl = {s : Bool => Str} ;
NP = NounPhrase ; NP = NounPhrase ;
-- {s : Str} ;
VP = VerbPhrase ; VP = VerbPhrase ;
-- {verb : Verb ; compl : Str} ; AP = Adj ;
AP = {s : Str; monoSyl: Bool} ; CN = Noun ;
CN = Noun ; -- {s : Str; c : Str} ; Det = {s : Gender => Case => Str ; n : Number} ;
Det = {s : Str ; n : Number} ; N = Noun ;
N = Noun ; -- {s : Str; c : Str} ; A = Adj ;
A = Adj ; -- {s : Str; monoSyl: Bool} ; V = Verb;
V = Verb; -- {s : Str ; pp,ds,dp,ep : Str ; neg : Str} V2 = Verb ** {c : Case} ;
V2 = Verb ;
AdA = {s : Str} ; AdA = {s : Str} ;
Pol = {s : Str ; b : Bool} ; Pol = {s : Str ; b : Bool} ;
Tense = {s : Str} ; Tense = {s : Str} ;
Conj = {s : SForm => Str} ; Conj = {s : Str} ;
lin lin
UseCl t p cl = {s = t.s ++ p.s ++ cl.s ! p.b} ; UseCl t p cl = {s = t.s ++ p.s ++ cl.s ! p.b} ;
PredVP np vp = { PredVP np vp = {
s = \\p => np.s ++ neg p ++ vp.verb.s ++ vp.compl s = \\p => np.s ! Nom ++ neg p ++ vp.compl ! np.a ++ vp.verb.s ! VPres np.a.n np.a.p
} ; } ;
ComplV2 v2 np = { ComplV2 v2 np = {
verb = v2 ; verb = v2 ;
compl = np.s compl = \\_ => np.s ! v2.c
} ; } ;
UseV v = { UseV v = {
verb = v ; verb = v ;
compl = [] compl = \\_ => []
} ; } ;
DetCN det cn = case det.n of { DetCN det cn = {
Sg => {s = det.s ++ cn.c ++ cn.s ; n = Sg } ; s = \\c => det.s ! cn.g ! c ++ cn.s ! det.n ! c ;
Pl => {s = det.s ++ "些" ++ cn.s ; n = Pl } a = agr cn.g det.n P3
} ; } ;
ModCN ap cn = case ap.monoSyl of { ModCN ap cn = {s = \\n,c => ap.s ! cn.g ! n ! c ++ cn.s ! n ! c ; g = cn.g} ;
True => {s = ap.s ++ cn.s ; c = cn.c} ;
False => {s = ap.s ++ "的" ++ cn.s ; c = cn.c}
} ;
CompAP ap = { CompAP ap = {
verb = copula ; verb = copula ;
compl = ap.s ++ "的" compl = \\a => ap.s ! a.g ! a.n ! Nom
} ; } ;
AdAP ada ap = { AdAP ada ap = {
s = ada.s ++ ap.s ; s = \\g,n,c => ada.s ++ ap.s ! g ! n ! c
monoSyl = False
} ; } ;
ConjNP co x y = { ConjNP co x y = {
s = x.s ++ co.s ! Phr NPhrase ++ y.s s = \\c => x.s ! c ++ co.s ++ y.s ! c ;
a = y.a ----
} ; } ;
ConjS co x y = {s = x.s ++ co.s ! Sent ++ y.s} ; ConjS co x y = {s = x.s ++ co.s ++ y.s} ;
UseN n = n ; UseN n = n ;
UseA adj = adj ; UseA adj = adj ;
a_Det = mkDet "一" Sg ; a_Det = mkDet "" Sg ;
every_Det = mkDet "每" Sg ; every_Det = mkDet "प्रति" Sg ;
the_Det = mkDet "那" Sg ; the_Det = mkDet "" Sg ;
this_Det = mkDet "这" Sg ; this_Det = mkDet "एतद्" Sg ;
these_Det = mkDet "这" Pl ; -- these_Det = mkDet "这" Pl ;
that_Det = mkDet "那" Sg ; that_Det = mkDet "तद्" Sg ;
those_Det = mkDet "那" Pl ; -- those_Det = mkDet "那" Pl ;
i_NP = pronNP "我" ; i_NP = pronNP "" Sg P1 ;
youSg_NP = pronNP "你" ; youSg_NP = pronNP "" Sg P2 ;
he_NP = pronNP "他" ; he_NP = pronNP "" Sg P3 ;
she_NP = pronNP "她" ; she_NP = pronNP "" Sg P3 ;
we_NP = pronNP "我们" ; we_NP = pronNP "" Pl P1 ;
youPl_NP = pronNP "你们" ; youPl_NP = pronNP "" Pl P2 ;
they_NP = pronNP "他们" ; they_NP = pronNP "" Pl P3 ;
very_AdA = ss (word "非常") ; very_AdA = ss "अति" ;
and_Conj = {s = table { and_Conj = {s = " च"} ;
Phr NPhrase => "和" ;
Phr APhrase => "而" ;
Phr VPhrase => "又" ;
Sent => []
}
} ;
or_Conj = {s = table { or_Conj = {s = "अथवा"} ;
Phr _ => "或" ;
Sent => word "还是"
}
} ;
Pos = {s = [] ; b = True} ; Pos = {s = [] ; b = True} ;
Neg = {s = [] ; b = False} ; Neg = {s = [] ; b = False} ;
@@ -110,17 +97,24 @@ concrete MiniresourceSan of Miniresource = open Prelude in {
-- module TestChi -- module TestChi
lin lin
man_N = mkN "男人" ; man_N = mkN "नरः" ;
woman_N = mkN "女人" ; woman_N = mkN "स्त्री" ;
house_N = mkN "房子" ; house_N = mkN "गृहं" ;
tree_N = mkN "树" "棵"; tree_N = mkN "वृक्ष";
big_A = mkA "大" ; big_A = mkA "महाकाय" ;
small_A = mkA "小" ; small_A = mkA "अल्प" ;
green_A = mkA "绿" ; green_A = mkA "हरित" ;
walk_V = mkV "走" ; walk_V = mkV "गम्" ;
arrive_V = mkV "到" ; arrive_V = mkV "अभि-उपा-गम्" ;
love_V2 = mkV2 "爱" ; love_V2 = mkV2 "कम्" ;
please_V2 = mkV2 "麻烦" ; please_V2 = mkV2 "प्री" ;
-- module ResSan -- module ResSan
@@ -132,15 +126,19 @@ param
Gender = Masc | Fem | Neutr ; Gender = Masc | Fem | Neutr ;
Person = P3 | P2 | P1 ; Person = P3 | P2 | P1 ;
Agr = Ag Gender Number Person ;
VForm = VPres Number Person ; VForm = VPres Number Person ;
oper
Agr = {g : Gender ; n : Number ; p : Person} ;
agr : Gender -> Number -> Person -> Agr = \g,n,p -> {g = g ; n = n ; p = p} ;
-- parts of speech -- parts of speech
oper oper
VerbPhrase = {verb : Verb ; compl : Str} ; VerbPhrase = {verb : Verb ; compl : Agr => Str} ;
NounPhrase = {s : Case => Str ; a : Agr} ; NounPhrase = {s : Case => Str ; a : Agr} ;
-- for morphology -- for morphology
@@ -150,7 +148,7 @@ oper
Verb : Type = {s : VForm => Str} ; Verb : Type = {s : VForm => Str} ;
mkNoun : (s1,_,_,_,_,_,_,_, _,_,_, _,_,_,_,_,s17 : Str) -> Gender -> Noun = mkNoun : (s1,_,_,_,_,_,_,_, _,_,_, _,_,_,_,_,s17 : Str) -> Gender -> Noun =
\snon,sacc,sins,sdat,sabl,sgen,sloc,svoc, \snom,sacc,sins,sdat,sabl,sgen,sloc,svoc,
dnomaccvoc,dinsdatabl,dgenloc, dnomaccvoc,dinsdatabl,dgenloc,
pnomvoc,pacc,pins,pdatabl,pgen,ploc, pnomvoc,pacc,pins,pdatabl,pgen,ploc,
gen -> { gen -> {
@@ -159,7 +157,7 @@ oper
Nom => snom ; Acc => sacc ; Ins => sins ; Dat => sdat ; Abl => sabl ; Gen => sgen ; Loc => sloc ; Voc => svoc Nom => snom ; Acc => sacc ; Ins => sins ; Dat => sdat ; Abl => sabl ; Gen => sgen ; Loc => sloc ; Voc => svoc
} ; } ;
Dl => table { Dl => table {
Nom | Voc => dnomaccvoc ; Ins | Dat | Abl => dinsdatabl ; Gen | Loc => dgenloc Nom | Acc | Voc => dnomaccvoc ; Ins | Dat | Abl => dinsdatabl ; Gen | Loc => dgenloc
} ; } ;
Pl => table { Pl => table {
Nom | Voc => pnomvoc ; Acc => pacc ; Ins => pins ; Dat | Abl => pdatabl ; Gen => pgen ; Loc => ploc Nom | Voc => pnomvoc ; Acc => pacc ; Ins => pins ; Dat | Abl => pdatabl ; Gen => pgen ; Loc => ploc
@@ -170,12 +168,12 @@ oper
endingNoun : Str -> (s1,_,_,_,_,_,_,_, _,_,_, _,_,_,_,_,s17 : Str) -> Gender -> Noun = endingNoun : Str -> (s1,_,_,_,_,_,_,_, _,_,_, _,_,_,_,_,s17 : Str) -> Gender -> Noun =
\stem, \stem,
snon,sacc,sins,sdat,sabl,sgen,sloc,svoc, snom,sacc,sins,sdat,sabl,sgen,sloc,svoc,
dnomaccvoc,dinsdatabl,dgenloc, dnomaccvoc,dinsdatabl,dgenloc,
pnomvoc,pacc,pins,pdatabl,pgen,ploc, pnomvoc,pacc,pins,pdatabl,pgen,ploc,
gen -> gen ->
mkNoun mkNoun
(stemm + snon) (stem + sacc) (stem + sins) (stem + sdat) (stem + sabl) (stem + sgen) (stem + sloc) (stem + svoc) (stem + snom) (stem + sacc) (stem + sins) (stem + sdat) (stem + sabl) (stem + sgen) (stem + sloc) (stem + svoc)
(stem + dnomaccvoc) (stem + dinsdatabl) (stem + dgenloc) (stem + dnomaccvoc) (stem + dinsdatabl) (stem + dgenloc)
(stem + pnomvoc) (stem + pacc) (stem + pins) (stem + pdatabl) (stem + pgen) (stem + ploc) (stem + pnomvoc) (stem + pacc) (stem + pins) (stem + pdatabl) (stem + pgen) (stem + ploc)
gen ; gen ;
@@ -184,11 +182,14 @@ oper
ramaNoun : Str -> Noun = \rama -> ramaNoun : Str -> Noun = \rama ->
let ram = init rama in let ram = init rama in
endingNoun ram endingNoun ram
"aH" "amx" "eNe" "a:ya" "a:tx" "asxya" "e" "a" "ः" "म्" "ॆणॆ" "ाय" "ात्" "स्य" "ॆ" ""
"o+" "a:t'xya:mx" "ayo:" "ौ" "ाथ्याम्" "यो"
"a:H" "a:nx" "e+H" "e:t'yaH" "a:Na:ma" "e:Su" "ाः" "ान्" "ैः" "ेथयः" "ाणाम" "ेषु"
Masc ; Masc ;
mkAdj : (m,f,n : Noun) -> Adj = \m,f,n -> {s = table {Masc => m.s ; Fem => f.s ; Neutr => n.s}} ;
mkVerb : (s1,_,_,_,_,_,_,_,s9 : Str) -> Verb = mkVerb : (s1,_,_,_,_,_,_,_,s9 : Str) -> Verb =
\s3,s2,s1,d3,d2,d1,p3,p2,p1 -> { \s3,s2,s1,d3,d2,d1,p3,p2,p1 -> {
s = table { s = table {
@@ -211,58 +212,51 @@ oper
patVerb : Str -> Verb = \pat -> patVerb : Str -> Verb = \pat ->
endingVerb pat endingVerb pat
"ita" "isa" "ima" "ataH" "at'aH" "avaH" "inxta" "at'a" "a:maH" ; "ित" "िस" "िम" "तः" "थः" "ावः" "िनख़त" "थ" "ामः" ;
copula : Verb = {s = \\_ => []} ;
neg : Bool -> Str = \b -> case b of {True => [] ; False => "न"} ;
{-
neg : Bool -> Str = \b -> case b of {True => [] ; False => "不"} ;
-- for structural words -- for structural words
mkDet : Str -> Number -> {s : Str ; n : Number} = \s,n -> { mkDet : Str -> Number -> {s : Gender => Case => Str ; n : Number} = \s,n -> {
s = word s ; s = \\_,_ => s ;
n = n n = n
} ; } ;
pronNP : (s : Str) -> NounPhrase = \s -> { pronNP : (s : Str) -> Number -> Person -> NounPhrase = \s,n,p -> {
s = word s s = \\_ => s ;
a = agr Masc n p
} ; } ;
-- Write the characters that constitute a word separately.
-- This enables straightforward tokenization.
bword : Str -> Str -> Str = \x,y -> x ++ y ;
-- change to x + y to treat words as single tok ens
word : Str -> Str = \s -> case s of { -- module ParadigmsSan
x@? + y@? + z@? + u@? => bword x (bword y (bword z u)) ;
x@? + y@? + z@? => bword x (bword y z) ;
x@? + y@? => bword x y ;
_ => s
} ;
-- module ParadigmsChi
oper oper
mkN = overload { mkN = overload {
mkN : (man : Str) -> N mkN : (man : Str) -> N
= \n -> lin N (mkNoun n "个") ; = \s -> lin N (ramaNoun s) ;
mkN : (man : Str) -> Str -> N
= \n,c -> lin N (mkNoun n c)
} ; } ;
mkA : (small : Str) -> A mkA : (small : Str) -> A
= \a -> lin A (mkAdj a) ; = \s -> let n = ramaNoun s in lin A (mkAdj n n n) ;
mkV : (walk : Str) -> V mkV : (walk : Str) -> V
= \s -> lin V (mkVerb s) ; = \s -> lin V (patVerb s) ;
mkV2 = overload { mkV2 = overload {
mkV2 : (love : Str) -> V2 mkV2 : (love : Str) -> V2
= \love -> lin V2 (mkVerb love) ; = \love -> lin V2 (mkV love ** {c = Acc}) ;
mkV2 : (love : V) -> V2 --- mkV2 : (love : V) -> V2
= \love -> lin V2 love ; --- = \love -> lin V2 love ;
} ; } ;
-}
} }

13
src/compiler/GF/Text/Transliterations.hs
View File

@@ -43,6 +43,7 @@ allTransliterations = Map.fromAscList [
("greek", transGreek), ("greek", transGreek),
("hebrew", transHebrew), ("hebrew", transHebrew),
("persian", transPersian), ("persian", transPersian),
("sanskrit", transSanskrit),
("sindhi", transSindhi), ("sindhi", transSindhi),
("nepali", transNepali), ("nepali", transNepali),
("telugu", transTelugu), ("telugu", transTelugu),
@@ -291,3 +292,15 @@ allTrans = words $
" p. p- p' p( p) p p? p*" " p. p- p' p( p) p p? p*"
allCodes = [0x1200..0x1357] allCodes = [0x1200..0x1357]
-- by Prasad 31/5/2013
transSanskrit :: Transliteration
transSanskrit = (mkTransliteration "Sanskrit" allTrans allCodes) {invisible_chars = ["a"]} where
allTrans = words $
"- n~ m. h. - A A: I I: U U: R. L. - - E: " ++
"E+ - O O: O+ k k' g g' n- c c' j j' n* T " ++
"T' D D' N t t' d d' n - p p' b b' m y " ++
"r - l L - v s* S s h - - - v- a: i " ++
"i: u u: r. r.: - e e: e+ - o o: o+ a_ - - " ++
"o~ - - - - - - - q x G z R R' f - " ++
"R.: L.: l. l.: p, p. N0 N1 N2 N3 N4 N5 N6 N7 N8 N9 "
allCodes = [0x0900 .. 0x097f]