Welcome to Programming and Data Structures

Programs need ways to store information in a way that is accessible to other programs and that will persist after the program ends. The most common way to do this is to write information to a file.

Files come in two main flavors, binary and text based. In reality, all files are binary - they are all just a sequence of 1s and 0s. But in text based files, all of the content is ASCII encoded (or Unicode or some similar encoding) text. Text files are designed to be easy for humans to open and read. All of our source code files in C++ are written as text files. They also are often used for things like configuration files, log files, etc...

In a binary file, there may be some text that is encoded as characters, but most of the content is 1s and 0s that have other meanings - numbers, colors, etc... Binary files are typically more compact and quicker for computers to read and write. But they are much more difficult for humans to work with - you need a special program to edit the raw data and knowledge of how the data has been structured. Binary files are typically used for things like images, movies, complex documents that include formatting (like MS Word documents), and storage of large amounts of data.

In this book we will focus on text files. They are much easier to work with and are perfectly good for learning the basics of reading and writing data. There are some general tricks for working with binary data, but most of the challenges when working with binary files relate to understanding how a particular file is structured rather than how to read/write the data.

We will refer to the files in this book that your programs can work with as datafiles. Here is a datafile named Numbers.txt:

10 20 30
40 50 60

In a program we can refer to "Numbers.txt" to specify that file.

The tools we will use to work with files in C++ are called ofstream and ifstream. The stream part of the name refers to a general idea - a “stream” of data that is produced by one thing and consumed by another.

We have already encountered some streams. cin and cout are streams that send data between the console and a program. cin is an input stream, it bring information from the console to the program. cout on the other hand is an output stream that sends data from the program to the console. Each time we use >> or << we are either removing some data from a stream or inserting some data into it.

As you will see, working with files is not that different than working with the console. That is the power of an abstraction. By thinking of both working with files and the console as examples of “sending information to/from the program”, we can develop and use one set of techniques that lets us solve both problems.