{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f9c836a8-aa45-4b3f-8392-660ffdf0daf9",
   "metadata": {},
   "outputs": [],
   "source": [
    "from functools import cache\n",
    "import pandas as pd\n",
    "\n",
    "pd.set_option(\"display.max_columns\", None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "faf0690b-1462-4d93-b66d-3b3e017fcc88",
   "metadata": {},
   "outputs": [],
   "source": [
    "@cache\n",
    "def get_dataset_for(year):\n",
    "    return pd.read_csv(f\"datasets/sinac{year}DatosAbiertos.csv\", dtype=object)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "4436b11a-5e07-4b5f-9aa8-9f60b9328e70",
   "metadata": {},
   "outputs": [],
   "source": [
    "columns_selected = [\n",
    "    \"edo_captura\",\n",
    "    \"edo_nac_madre\",\n",
    "    \"fecha_nac_madre\",\n",
    "    \"edad_madre\",\n",
    "    \"estado_conyugal\",\n",
    "    \"entidad_residencia_madre\",\n",
    "    \"numero_embarazos\",\n",
    "    \"hijos_nacidos_muertos\",\n",
    "    \"hijos_nacidos_vivos\",\n",
    "    \"hijos_sobrevivientes\",\n",
    "    \"el_hijo_anterior_nacio\",\n",
    "    \"vive_aun_hijo_anterior\",\n",
    "    \"orden_nacimiento\",\n",
    "    \"recibio_atencion_prenatal\",\n",
    "    \"trimestre_recibio_primera_consulta\",\n",
    "    \"total_consultas_recibidas\",\n",
    "    \"madre_sobrevivio_al_parto\",\n",
    "    \"escolaridad_madre\",\n",
    "    \"ocupacion_habitual_madre\",\n",
    "    \"trabaja_actualmente\",\n",
    "    \"fecha_nacimiento_nac_vivo\",\n",
    "    \"hora_nacimiento_nac_vivo\",\n",
    "    \"sexo_nac_vivo\",\n",
    "    \"semanas_gestacion_nac_vivo\",\n",
    "    \"talla_nac_vivo\",\n",
    "    \"peso_nac_vivo\",\n",
    "    \"valoracion_apgar_nac_vivo\",\n",
    "    \"valoracion_silverman_nac_vivo\",\n",
    "    \"producto_de_un_embarazo\",\n",
    "    \"codigo_anomalia\",\n",
    "    \"anomalia_congenita_nac_vivo\",\n",
    "    \"lugar_de_nacimiento\",\n",
    "    \"entidad_certifico\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "71959de6-9879-4471-a331-99a054b7f5af",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat([get_dataset_for(year)[columns_selected] for year in range(2010, 2017)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9147e139-6a91-4660-81e2-db49ec151ff0",
   "metadata": {},
   "outputs": [],
   "source": [
    "def _ano_nacimiento_vivo_func(str_date):\n",
    "    try:\n",
    "        return str_date.split(\"/\")[-1]\n",
    "    except:\n",
    "        return \"\"\n",
    "\n",
    "\n",
    "df[\"año_de_nacimiento_vivo\"] = df[\"fecha_nacimiento_nac_vivo\"].apply(\n",
    "    _ano_nacimiento_vivo_func\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f71063d3-5c78-426f-a484-80fcf543051f",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "columnas_finales = [\n",
    "    \"edo_captura\",\n",
    "    \"edo_nac_madre\",\n",
    "    \"fecha_nac_madre\",\n",
    "    \"edad_madre\",\n",
    "    \"estado_conyugal\",\n",
    "    \"entidad_residencia_madre\",\n",
    "    \"numero_embarazos\",\n",
    "    \"hijos_nacidos_muertos\",\n",
    "    \"hijos_nacidos_vivos\",\n",
    "    \"hijos_sobrevivientes\",\n",
    "    \"el_hijo_anterior_nacio\",\n",
    "    \"vive_aun_hijo_anterior\",\n",
    "    \"orden_nacimiento\",\n",
    "    \"recibio_atencion_prenatal\",\n",
    "    \"trimestre_recibio_primera_consulta\",\n",
    "    \"total_consultas_recibidas\",\n",
    "    \"madre_sobrevivio_al_parto\",\n",
    "    \"escolaridad_madre\",\n",
    "    \"ocupacion_habitual_madre\",\n",
    "    \"trabaja_actualmente\",\n",
    "    \"fecha_nacimiento_nac_vivo\",\n",
    "    \"hora_nacimiento_nac_vivo\",\n",
    "    \"sexo_nac_vivo\",\n",
    "    \"semanas_gestacion_nac_vivo\",\n",
    "    \"talla_nac_vivo\",\n",
    "    \"peso_nac_vivo\",\n",
    "    \"valoracion_apgar_nac_vivo\",\n",
    "    \"valoracion_silverman_nac_vivo\",\n",
    "    \"producto_de_un_embarazo\",\n",
    "    \"codigo_anomalia\",\n",
    "    \"entidad_certifico\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "937d0447-e6e6-477d-a43b-471302229fca",
   "metadata": {},
   "outputs": [],
   "source": [
    "df[columnas_finales].to_csv(\"2010-2016.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dfe420d2-b0f3-48ce-b30d-06f8d901dbd8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "69940ce2-da7a-4752-9d0d-c6046eb03992",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20918084"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "5873377 + 15044707"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c3db8468-11e0-4758-a9b1-f037e633b729",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}