{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6da162ee",
   "metadata": {},
   "source": [
    "Las listas de python y los arrays de numpy son **objetos mutables**. Esto tiene un efecto muy importante que hay que tener en cuenta a la hora de programar, pues nos puede sorprender.\n",
    "\n",
    "En este cuaderno se ilustra con un ejemplo sencillo lo que sucede y se explica cómo resolverlo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f867db8a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.3219883 , 0.89042245, 0.58805226, 0.12659609, 0.14134122])"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Comencemos importando numpy y creando un vector cualquiera\n",
    "\n",
    "import numpy as np\n",
    "import numpy.linalg as la\n",
    "import numpy.random as rnd\n",
    "rnd.seed(2023)\n",
    "\n",
    "n = 5\n",
    "x = rnd.rand(n)\n",
    "\n",
    "x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a7fe4d75",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([42.        ,  0.89042245,  0.58805226,  0.12659609,  0.14134122])"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Ahora creamos un segundo vector $y$, igual a $x$ excepto en una coordenada.\n",
    "y = x\n",
    "y[0] = 42\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c6f2635a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Tal vez nos interesa calcular la distancia entre ambos vectores; esperamos que sea 42\n",
    "la.norm(y - x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "87a1e274",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x = array([42.        ,  0.89042245,  0.58805226,  0.12659609,  0.14134122])\n",
      "y = array([42.        ,  0.89042245,  0.58805226,  0.12659609,  0.14134122])\n"
     ]
    }
   ],
   "source": [
    "# ¿Qué pasó?\n",
    "print(f\"{x = }\")\n",
    "print(f\"{y = }\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d8510b1a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True,  True,  True,  True,  True])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# ¡Los dos vectores son iguales!!!\n",
    "x == y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "069bba98",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# De hecho, son el mismo vector\n",
    "x is y"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d11b063b",
   "metadata": {},
   "source": [
    "Lo que sucede es que las variables en python funcionan como si fueran *punteros*. En este caso hay un único array  (el creado por la función `rnd.rand(n)`) y ambas variables `x` e `y` apuntan a ese mismo array. Al ejecutar `y[0] = 42` se modifica, a través de la variable `y`, la coordenada 0 de dicho array.\n",
    "\n",
    "Esto solo tiene efecto visible en objetos **mutables** como es el caso de los arrays de numpy. Para objetos **inmutables** (como números, cadenas, etc) el hecho que las variables sean punteros no hace ninguna diferencia porque los objetos subyacentes no se pueden modificar.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "25d9cfc4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.        , 0.89042245, 0.58805226, 0.12659609, 0.14134122])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# solución: *copiar* el array antes de modificar la copia\n",
    "z = y.copy()\n",
    "z[0] = 0\n",
    "z"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4fb5b7d6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x = array([42.        ,  0.89042245,  0.58805226,  0.12659609,  0.14134122])\n",
      "y = array([42.        ,  0.89042245,  0.58805226,  0.12659609,  0.14134122])\n",
      "z = array([0.        , 0.89042245, 0.58805226, 0.12659609, 0.14134122])\n"
     ]
    }
   ],
   "source": [
    "print(f\"{x = }\")\n",
    "print(f\"{y = }\")\n",
    "print(f\"{z = }\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6704b0d",
   "metadata": {},
   "source": [
    "En este caso x, y siguen siendo \"punteros\" al mismo array. En cambio z apunta a un nuevo array, inicialmente copia del primer array, pero que puede ser modificado de manera independiente."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "60234fd2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True\n",
      "False\n",
      "False\n"
     ]
    }
   ],
   "source": [
    "#\n",
    "print(x is y)\n",
    "print(x is z)\n",
    "print(y is z)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2849ba61",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(139766962473072, 139766962473072, 139766963050736)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# la función `id` devuelve un identificador del objeto subyacente\n",
    "# aquí se ve que `x` e `y` apuntan al mismo objeto, pero `z` no.\n",
    "id(x), id(y), id(z)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}