Update data processing pipeline to include device information and align column naming conventions

- Add loading of `devices.csv` to incorporate missing columns (`browser_type` and `os_type`).
- Perform a left join between `events` and `devices` datasets using `device_id`.
- Rename columns (`browser_type` -> `browser_family`, `os_type` -> `os_family`) to align with the rest of the notebook's codebase.
- Improve dataset completeness and maintain consistency in column naming conventions.

Author: divakaivan

bootcamp/materials/3-spark-fundamentals/notebooks/event_data_pyspark.ipynb

@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "id": "81cca085-dba2-42eb-a13b-fa64b6e86583",
    "metadata": {},
    "outputs": [
@@ -46,7 +46,11 @@
     "\n",
     "spark\n",
     "\n",
-    "df = spark.read.option(\"header\", \"true\").csv(\"/home/iceberg/data/events.csv\").withColumn(\"event_date\", expr(\"DATE_TRUNC('day', event_time)\"))\n",
+    "events = spark.read.option(\"header\", \"true\").csv(\"/home/iceberg/data/events.csv\").withColumn(\"event_date\", expr(\"DATE_TRUNC('day', event_time)\"))\n",
+    "devices = spark.read.option(\"header\",\"true\").csv(\"/home/iceberg/data/devices.csv\")\n",
+    "\n",
+    "df = events.join(devices,on=\"device_id\",how=\"left\")\n",
+    "df = df.withColumnsRenamed({'browser_type': 'browser_family', 'os_type': 'os_family'})\n",
     "\n",
     "df.show()"
    ]
@@ -414,17 +418,17 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "outputs": [],
-   "source": [
-    "%%sql \n",
-    "SELECT COUNT(1) FROM bootcamp.matches_bucketed.files"
-   ],
    "metadata": {
     "collapsed": false,
     "pycharm": {
      "name": "#%%\n"
     }
-   }
+   },
+   "outputs": [],
+   "source": [
+    "%%sql \n",
+    "SELECT COUNT(1) FROM bootcamp.matches_bucketed.files"
+   ]
   },
   {
    "cell_type": "code",
@@ -489,4 +493,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 5
-}
+}