In previous lessons, you learned how to efficiently manage datasets stored in compressed formats using Scala. Building on that foundation, today's lesson will introduce you to reading and processing data in batches from multiple CSV files using Scala. Processing data in smaller chunks, or batches, is essential for enhancing code efficiency and speed, especially when dealing with large datasets.

Our focus in this lesson will be on a practical scenario where car information is spread across multiple CSV files. You'll learn how to read, process, and analyze this data to extract valuable insights, such as identifying the car with the lowest price.

In this lesson, we will work with a set of CSV files that hold car data. Here is an example of the CSV format:

csv
Copy to clipboard
1model,price,transmission,year,distance_traveled_km,color
2Chevrolet Silverado,43725.23,Manual,2013,55504,Silver
3BMW X5,5643.78,Semi-Automatic,2014,11902,Red
4Honda Accord,42850.79,Manual,2010,102223,Black
5BMW Series 3,53359.81,Automatic,2009,237231,Gray
6...

A typical record might look like this:

• Model: Chevrolet Silverado
• Price: 43725.23
• Transmission: Manual
• Year: 2013
• Distance Traveled (km): 55504
• Color: Silver

Understanding the structure of these files is crucial as you learn to read and process them efficiently, especially when dealing with multiple files in batches.

To effectively read and process CSV files in batches, we need to set up our environment by defining the necessary classes and data structures using Scala's syntax. We will specify the filenames for our CSV data files.

First, we'll define a Car case class to map each row of the CSV file into a Scala object. This class includes fields that correspond to the columns in the CSV file.

Scala
Copy to clipboard
1// Case class to represent a car
2case class Car(model: String, price: Double, transmission: String, year: Int, distanceTraveled: Int, color: String)

Next, we declare an array filenames that will hold the names of the CSV files we want to read. Additionally, we create a list carData to store all the car data read from these files, enabling us to process and analyze this data collectively.

Scala
Copy to clipboard
1// Filenames to read
2val filenames = Array("data_part1.csv", "data_part2.csv", "data_part3.csv")
3
4// List to store all car data
5var carData = List.empty[Car]

By setting up these structures, we prepare to efficiently read and store data from multiple CSV files, allowing us to handle large datasets by processing information in manageable batches.

Now, we'll loop through each filename, read the file contents, and map CSV records to Car objects using Scala’s capabilities, such as os-lib for file operations and string manipulation to process data.

Scala
Copy to clipboard
1// Read and process each CSV file
2filenames.foreach { filename =>
3  val filePath = os.pwd / filename
4
5  // Read all lines from the CSV file
6  val lines = os.read.lines.stream(filePath).toList
7
8  // Skip the header line
9  val dataLines = lines.drop(1)
10
11  // Process each line of data
12  dataLines.foreach { line =>
13    // Split the line into its components
14    val Array(model, priceStr, transmission, yearStr, distanceStr, color) = line.split(",")
15
16    // Convert the strings to appropriate types
17    val price = priceStr.toDouble
18    val year = yearStr.toInt
19    val distanceTraveled = distanceStr.toInt
20
21    // Create a Car object with all the attributes
22    val car = Car(model, price, transmission, year, distanceTraveled, color)
23
24    // Prepend it to the carData list
25    carData ::= car
26  }
27}

Here, we read the file and convert each CSV line into a Car object. This is stored in a list for further processing, using Scala’s collections and os-lib for efficient file reading.

With all data combined in carData, the next step is identifying the car with the lowest price using Scala's higher-order functions, like minBy.

Scala
Copy to clipboard
1// Check if car data is not empty before finding the lowest cost car
2if carData.nonEmpty then
3  // Find the car with the lowest price
4  val lowestCostCar = carData.minBy(_.price)
5
6  // Output the model and price of the lowest cost car
7  println(s"Model: ${lowestCostCar.model}")
8  println(f"Price: $$${lowestCostCar.price}%.2f")
9else
10  // If no valid car data is available, output an appropriate message
11  println("No valid car data available.")

Using minBy, we can succinctly find the car with the lowest price in carData, demonstrating the power and simplicity of Scala’s functional programming paradigms.

In this lesson, you have learned how to:

• Read data in batches from multiple CSV files using Scala and the os-lib.
• Process the data efficiently by mapping CSV records to Scala objects using case classes.
• Identify insights, such as the car with the lowest price, using Scala's concise syntax and higher-order functions.

These techniques prepare you to handle similar datasets efficiently using Scala. Practice these skills with exercises designed to reinforce your understanding, focusing on efficient data-handling techniques with modern Scala libraries.