(*

See [1] for the theory behind the analysis


[1] https://fr.m.wiktionary.org/wiki/Annexe:Prononciation/fran%C3%A7ais#Structure_syllabique


   *)



%{
    open Modifiers.Sig

%}
%start<Sounds.t List.t> main

%%

occlusiv:
  | P               { Sounds.p }
  | B               { Sounds.b }

  | T               { Sounds.t }
  | D               { Sounds.d }

  | K               { Sounds.k }
  | G               { Sounds.g }

fricativ:
  | S               { Sounds.s }
  | SZ              { Sounds.sz }
  | Z               { Sounds.z }

  | F               { Sounds.f }
  | V               { Sounds.v }

  | X               { Sounds.ch }
  | J               { Sounds.j  }

obstruent:
  | occlusiv        { $1 }
  | fricativ        { $1 }

liquid:
  | L               { Sounds.l }
  | R               { Sounds.r }

nasal:
  | N               { Sounds.n }
  | M               { Sounds.m }


initial_nasal:
  | nasal           { $1 }
  | G N             { Sounds.gn }

opening_consonant:
  | occlusiv            { $1, None }
  | fricativ            { $1, None }
  | initial_nasal       { $1, None }
  | liquid              { $1, None }

opening_double:
  | obstruent liquid    { $1, Some $2 }
  | occlusiv fricativ   { $1, Some $2 }

(* Each voyel as two associated sounds, depending there is a followng sound or
   not *)
voyels:
  | A               { Sounds.a          }
  | A I             { Sounds.voyel_ai   }
  | E I             { Sounds.voyel_ai   }
  | I               { Sounds.i          }
  | E               { Sounds.schwa } 
  | E_ACUTE E?      { Sounds.e  `Closed }
  | E_AGRAVE        { Sounds.e  `Opened }
  | E U             { Sounds.eu `Opened }
  | O               { Sounds.o          }
  | U               { Sounds.voyel_y    }
  | OU              { Sounds.voyel_u    }
  (* This semi voyel can never be separated *)
  | W voyels        { Sounds.diphtongue Sounds.semi_voyel_w $2 }

voyels_semi:
  | voyels          { $1 }
  | I voyels        { Sounds.diphtongue Sounds.semi_voyel_y $2 }
  | Y voyels        { Sounds.diphtongue Sounds.semi_voyel_y $2 }
  | U voyels        { Sounds.diphtongue Sounds.semi_voyel_u $2 }

ending_consonant: 
  | Nothing         { Some (Sounds.none) }
  | occlusiv        { Some $1 }
  | liquid          { Some $1 }
  | nasal           { Some $1 }


%inline consonant_group(opening)
  : f = fricativ
    o = opening 
    { 
        { ending = None 
        ; opening = f::(fst o)::[]
        ; following = snd o  }
    }
  | o = opening 
    { 
        { ending = None 
        ; opening = [ fst o ]
        ; following = snd o  }
    }
  | e = ending_consonant
    Sep?
    o = opening
    { 
        { ending = Some e
        ; opening = [ fst o ]
        ; following = snd o  }
    }
  | e = ending_consonant
    Sep?
    f = fricativ
    o = opening
    { 
        { ending = Some e
        ; opening = f::[ fst o ]
        ; following = snd o }
    }


(** Exclude a semi voyel in both consonant and voyel position *)
syllable:
  | c = consonant_group(opening_consonant)?
    v = voyels_semi
    Sep?
    { (v, c) }
  | c = consonant_group(opening_double)
    v = voyels
    Sep?
    { (v, Some c) }


syllables:
  | { [] }
  | ss = syllables 
    s = syllable { s::ss }
    

word:
  | Sep? syllables consonant_group(opening_consonant)? 
        { Process.rebuild $3 $2 }

words: 
  | word { $1::[] }
  | ww=words Space w=word  { w:: [Sounds.space $2] ::ww }

main: 
  | words EOL        
        { List.concat (List.rev $1) }