9
       / \
      /   \
     3     7
    / \
   2   4

   N 9 (N 3 (H 2) (H 4)) (H 7)

import Test.QuickCheck

data Arbol a = H a
             | N a (Arbol a) (Arbol a)
  deriving (Show, Eq)

-- Generador de árboles binarios
-- =============================

-- (arbolArbitrario n) es un árbol aleatorio de altura n. Por ejemplo,
--    λ> sample (arbolArbitrario 3 :: Gen (Arbol Int))
--    N 0 (H 0) (H 0)
--    N 1 (N (-2) (H (-1)) (H 1)) (H 2)
--    N 3 (H 1) (H 2)
--    N 6 (N 0 (H 5) (H (-5))) (N (-5) (H (-5)) (H 4))
--    H 7
--    N (-8) (H (-8)) (H 9)
--    H 2
--    N (-1) (H 7) (N 9 (H (-2)) (H (-8)))
--    H (-3)
--    N 0 (N 16 (H (-14)) (H (-18))) (H 7)
--    N (-16) (H 18) (N (-19) (H (-15)) (H (-18)))
arbolArbitrario :: Arbitrary a => Int -> Gen (Arbol a)
arbolArbitrario 0 = H <$> arbitrary
arbolArbitrario n =
  oneof [H <$> arbitrary,
         N <$> arbitrary <*> arbolArbitrario (div n 2) <*> arbolArbitrario (div n 2)]

-- Arbol es subclase de Arbitrary
instance Arbitrary a => Arbitrary (Arbol a) where
  arbitrary = sized arbolArbitrario

        9
       / \
      /   \
     3     7
    / \
   2   4

   N(9, N(3, H(2), H(4)), H(7))

from dataclasses import dataclass
from random import choice, randint
from typing import Generic, TypeVar

A = TypeVar("A")

@dataclass
class Arbol(Generic[A]):
    pass

@dataclass
class H(Arbol[A]):
    x: A

@dataclass
class N(Arbol[A]):
    x: A
    i: Arbol[A]
    d: Arbol[A]

def nHojas(a: Arbol[A]) -> int:
    match a:
        case H(_):
            return 1
        case N(_, i, d):
            return nHojas(i) + nHojas(d)
    assert False

def nNodos(a: Arbol[A]) -> int:
    match a:
        case H(_):
            return 0
        case N(_, i, d):
            return 1 + nNodos(i) + nNodos(d)
    assert False

# Generador de árboles
# ====================

# (arbolArbitrario n) es un árbol aleatorio de orden n. Por ejemplo,
#    >>> arbolArbitrario(4)
#    N(x=2, i=H(x=1), d=H(x=9))
#    >>> arbolArbitrario(4)
#    H(x=10)
#    >>> arbolArbitrario(4)
#    N(x=4, i=N(x=7, i=H(x=4), d=H(x=0)), d=H(x=6))
def arbolArbitrario(n: int) -> Arbol[int]:
    if n <= 1:
        return H(randint(0, 10))
    m = n // 2
    return choice([H(randint(0, 10)),
                   N(randint(0, 10),
                     arbolArbitrario(m),
                     arbolArbitrario(m))])

data Expr = Lit Int
          | Suma Expr Expr
          | Op Expr
          | SiCero Expr Expr Expr
  deriving (Eq, Show)

from dataclasses import dataclass


@dataclass
class Expr:
    pass

@dataclass
class Lit(Expr):
    x: int

@dataclass
class Suma(Expr):
    x: Expr
    y: Expr

@dataclass
class Op(Expr):
    x: Expr

@dataclass
class SiCero(Expr):
    x: Expr
    y: Expr
    z: Expr