complete RGL implementation for Mongolian by Nyamsuren Erdenebadrakh

lib/src/mongolian/ParadigmsMonI.gf

@@ -0,0 +1,187 @@
+--# -path=.:../abstract:../prelude:../common
+
+--1 Mongolian Lexical Paradigms
+
+-- This is an API for the user of the resource grammar for adding lexical
+-- items. It gives functions for forming expressions of open categories:
+-- nouns, adjectives, verbs.
+
+-- Closed categories (determiners, pronouns, conjunctions) are accessed
+-- through the resource syntax API, $Structural.gf$.
+
+-- The main difference with $MorphoMon.gf$ is that the result types referred 
+-- to are compiled resource grammar types, i.e. wrapped with $lin Cat$.
+
+interface ParadigmsMonI = open Prelude, MorphoMon, CatMon in {
+
+ flags  coding=utf8 ; optimize=all ; 
+
+oper 
+ regV : Str -> Verb ;
+-- Auxiliary categories not in abstract/Cat.gf
+
+-- Categories $V0, AS, A2S, AV, A2V$ are just $A$, $V0$ is just $V$; 
+-- the second argument is treated as adverb.
+
+ V0, AS, AV, A2S, A2V : Type ;
+
+-- Parameters 
+
+-- To abstract over case names, we define the following case names, following 
+-- the case order standard for Mongolian textbooks. 
+
+ RCase : Type ; 
+ nominative    : RCase ;
+ genitive      : RCase ;
+ dative        : RCase ;
+ accusative    : RCase ; 
+ ablative      : RCase ;
+ instrumental  : RCase ;
+ comitative    : RCase ;
+ directive     : RCase ;
+
+-- To abstract over number names, we define the following.
+
+ Number : Type ;
+ singular : Number ;
+ plural   : Number ;
+
+-- Preposition
+
+ mkPrep : RCase -> Str -> Prep ;
+ noPrep : RCase -> Prep ;
+ 
+-- Conjunctions
+
+ mkConj : Str -> ConjType -> Conj ; 
+                      
+ mkSubj : Str -> Subj ;
+
+-- Nouns
+
+ mkN : overload { mkN : (nNomSg : Str) -> Noun ;
+                  mkN : (nNomSg,nNomPl : Str) -> Noun } ;
+-- Foreign nouns, which are inflected differently.
+
+ mkLN : overload { mkLN : Str -> Noun ;
+                   mkLN : (_,_ : Str) -> Noun } ; 
+				  
+-- Compound nouns constructed with an adjective; only the noun part is inflected.
+  
+ mk2N : (adj : Str) -> Noun -> Noun ;
+
+ mkN2 : N -> N2 ;
+
+ mkN3 : Noun -> Prep -> Prep -> N3 ;
+
+-- Proper names and Noun Phrases.  
+
+ mkPN : Str -> PN ;  
+ mkNP : Str -> Def -> NP ;  
+
+-- Adjectives
+
+-- Mongolian adjectives have only a positive form.
+
+ mkA : (positive : Str) -> A ;
+
+-- Two-place adjectives need a case as an extra argument.
+
+ mkA2 : A -> RCase -> A2 ;  
+
+-- Adverbs
+
+-- Adverbs are not inflected.
+
+ mkAdv : Str -> Adv ;
+
+-- Verbs
+
+-- In our lexicon description there are so far 120 forms: 
+-- 5 Voice * 4 Aspect * 6 indicative tensed Mood 
+
+ Voice: Type; 
+ active      : Voice ;
+ causative   : Voice ;
+ passive     : Voice ;
+ communal    : Voice ;
+ adversative : Voice ;
+
+ Aspect: Type; 
+ simple      : Aspect ;
+ quick       : Aspect ;
+ complete    : Aspect ;
+ collective  : Aspect ;
+
+ Imperative : Type ;
+ intention  : Imperative ;
+ decision   : Imperative ;
+ command    : Imperative ;
+ request    : Imperative ;
+ demand     : Imperative ;
+ admonition : Imperative ;
+ appeal     : Imperative ;
+ permission : Imperative ;
+ hope       : Imperative ;
+ blessing   : Imperative ;
+ 
+ Subordination : Type ;
+ conditional        : Subordination ;
+ concessive         : Subordination ;
+ immediatsucceeding : Subordination ;
+ logicalsucceeding  : Subordination ;
+ intending          : Subordination ;
+ limiting           : Subordination ;
+ progressive        : Subordination ;
+ succeeding         : Subordination ;
+
+ Bool: Type;
+ true        : Bool;
+ false       : Bool;
+
+-- One-place verbs. 
+
+ mkV : Str -> Verb ;
+
+-- Two-place and three-place verbs:
+
+ mkV2 : overload { mkV2 : Verb -> V2 ;
+                   mkV2 : Verb -> Prep -> V2 } ;
+
+ mkV3 : overload { mkV3 : Verb -> V3 ;
+                   mkV3 : Verb -> Prep -> Prep -> V3 } ;
+
+-- Determiners and quantifiers
+
+ mkOrd : Str -> Ord ;
+
+-- Other complement patterns
+
+-- Verbs and adjectives can take complements such as sentences,
+-- questions, verb phrases, and adjectives.
+
+ mkVA : Verb -> VA ;
+ mkV0 : Verb -> V0 ;
+ mkVS : Verb -> VS ;
+ mkVV : {s : VVVerbForm => Str} -> VV ;
+ mkVQ : Verb -> VQ ;
+ mkV2S : Verb -> Prep -> V2S ;
+ mkV2V : Verb -> Prep -> V2V ;
+ mkV2A : Verb -> Prep -> V2A ;
+ mkV2Q : Verb -> Prep -> V2Q ;
+ mkAS : Adjective -> AS ;
+ mkA2S : Adjective -> Prep -> A2S ;
+ mkAV : Adjective -> AV ;
+ mkA2V : Adjective -> Prep -> A2V ;
+
+-- Definitions of parameters and paradigms should not bother the user of the API.
+
+-- We separate the definitions that should be compiled with optimize=all into 
+-- ParadigmsMonO.gf, those which should be compiled with optimize=noexpand into
+-- ParadigmsMon.gf. 
+
+-- Implementation: Why is it much slower if the definitions that should be optimized
+-- are contained here? 
+
+} ;
+