{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sparkR.session()\n",
    "\n",
    "working_storage <- 'WORKING_STORAGE'\n",
    "output_directory <- 'jupyter/r'\n",
    "protocol_name <- 'PROTOCOL_NAME'\n",
    "\n",
    "storage_path <- function(file_path) {\n",
    "   sprintf('%s://%s/jupyter_dataset/%s', protocol_name, working_storage, file_path)\n",
    "}\n",
    "\n",
    "full_path <- function(file_path) {\n",
    "    sprintf('%s://%s/%s/%s', protocol_name, working_storage, output_directory, file_path)\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Parse and convert Carrier data to parquet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "carriers <- read.df(storage_path(\"carriers.csv\"), \"csv\", header=\"true\", inferSchema=\"true\")\n",
    "write.df(carriers, path=full_path(\"carriers\"), source=\"parquet\", mode=\"overwrite\")\n",
    "createOrReplaceTempView(carriers, \"carriers\")\n",
    "head(carriers, 20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Parse and convert to parquet Airport data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "airports <- read.df(storage_path(\"airports.csv\"), \"csv\", header=\"true\", inferSchema=\"true\")\n",
    "write.df(airports, path=full_path(\"airports\"), source=\"parquet\", mode=\"overwrite\")\n",
    "createOrReplaceTempView(airports, \"airports\")\n",
    "head(airports, 20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Parse and convert Flights data to parquet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "flights_w_na <- read.df(storage_path(\"2008.csv.bz2\"), \"csv\", header=\"true\", inferSchema=\"true\")\n",
    "flights <- fillna(flights_w_na, 0, cols=colnames(flights_w_na)[c(15, 16, 25:29)])\n",
    "write.df(flights, path=full_path(\"flights\"), source=\"parquet\", mode=\"overwrite\")\n",
    "createOrReplaceTempView(flights, \"flights\")\n",
    "colnames(flights)\n",
    "head(flights_w_na, 5)[c(\"ArrDelay\",\"CarrierDelay\",\"WeatherDelay\",\"Distance\")]\n",
    "head(flights, 5)[c(\"ArrDelay\",\"CarrierDelay\",\"WeatherDelay\",\"Distance\")]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Local SparkR (R-3.4.2, Spark-2.1.0)",
   "language": "R",
   "name": "KERNEL_NAME"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "3.4.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}