CS 3005: Programming in C++

WAV File Output

This is a continuation of the previous assignment. Please read the instructions there and complete the required programming tasks before starting this assignment.

Introduction

A WAV file is a file format that can be used to store sound data. It is actually a specific use of the RIFF file format. The WAV file consists of single RIFF chuck, that contains the “RIFF” header, a “fmt ” sub-chunk and a “data” sub-chunk. The “RIFF” header describes the file format, and the “fmt ” sub-chunk contains information about the audio format, such as the number of channels, and the “data” sub-chunk contains the actual audio data.

RIFF Header

The “RIFF” header has three pieces of information.

• The ID, consisting of the 4 characters “RIFF”
• The Size, a 32-bit integer, which is the number of bytes in the file after the size. Or, in other words, the number of bytes in the file minus 8.
• The Format, consisting of the 4 characters “WAVE”.

“fmt ” Sub-chunk

The “fmt ” sub-chunk contains information about the audio format, such as the number of channels, and the sample rate. Each of the properties has a specific size and order. Below we describe each of them.

We want ByteRate and BlockAlign to be an integers, so we will only allow BitsPerSample to be multiples of 8.

“data” Sub-chunk

This block contains the actual sound data. There is a small header at the beginning of the sub-chunk.

This header is followed by the actual sound data. The first sample is written for all channels. Followed by the second sample for all channels, etc.

The maximum integer value depends on the bits per sample. If there are 8 bits, then the maximum integer value is 127. This is the value achieved by setting all but the most significant bit in the integer to 1. Likewise, if there are 16 bits, then the maximum integer value is 32767. For 24 and 32 bits per sample the maximum values are 8388607 and 2147483647.

Assignment

In this assignment, you will finish creating a class, WAVFile that supports the writing of WAV files.

Programming Requirements

Depends on the AudioTrack class. Depends on the endian_io module. Depends on the WAVFile class started in the previous assignment.

Update library-audiofiles/WAVFile.{h,cpp}

WAVFile Class

Data Members:

The WAVFile class should contain data members to track the following information. These data members should be protected or private. They are not allowed to be public.

• unsigned int data subchunk position; The position in the file of the data subchunk. Computed while writing a file, and used to write the subchunk size in the correct location, after it has been determined.

public Methods:

• void writeFile(const std::string& filename, const std::vector<AudioTrack>& tracks); Opens an output file stream using the open method, passes the output file stream to writeFile, and uses the close method to close the output file stream.
• void writeFile(std::ostream& output_stream, const std::vector<AudioTrack>& tracks); Sets the data subchunk position to 0 and writes the RIFF header, the FMT subchunk, then writes the data subchunk header. It then writes the track data, and updates the sizes in the headers. Use the other methods of the class to do this work.

protected Methods:

• void open(const std::string& filename, std::ofstream& output_stream); Open the file named filename with the output_stream, in binary mode.
• void writeRIFFHeader(std::ostream& output_stream); Write the RIFF header. Put 0 as filler data in the chunk size field.
• void writeFMTSubchunk(std::ostream& output_stream); Write the “fmt ” subchunk.
• void writeDataSubchunkHeader(std::ostream& output_stream); Write the header to the data subchunk. Put 0 as filler data in the chunk size field. Use tellp to record the position at the start of the subchunk in the data subchunk position member.
• void writeOneTrackData(std::ostream& output_stream, const double track_data, int maximum_amplitude, int bytes_per_sample); Computes the integer value to write to the stream from the product of track_data and maximum_amplitude. Writes the value in little endian order to output_stream.
• void writeTracks(std::ostream& output_stream, const std::vector<AudioTrack>& tracks); Write the track data in little endian order. If there are not exactly 2 tracks, or if the two tracks are not the same size, do nothing. Note that the bytes per sample is the bits per sample divided by 8. Also, the maximum amplitude depends on the bits per sample. The track data passed to writeOneTrackData is the value pulled directly from the AudioTrack objects.
• void writeSizes(std::ostream& output_stream); Now that the sizes of chunks and file are known, go back to the headers and write the correct values in the size fields. Assumes that the output_stream is now positioned at the end of the file. This will allow us to compute the total file size. Use tellp() to find the current stream position. Use seekp() to move to the location where the size should be written.
• void close(std::ofstream& output_stream); Close output_stream, if it is open.

Additional Documentation

• WAVE PCM soundfile format
• std::ostream
• std::ofstream
• namespace
• types
• Member initializer list
• if
• comparison operators
• logical operators
• std::stringstream
• arithmetic operators
• std::vector
• std::string
• #include
• const method
• references
• public/protected/private
• size_t
• classes
• class declaration
• class implementation file
• static_cast
• auto
• for (counted)
• for (range)
• address-of

Grading Instructions

To receive credit for this assignment:

• your code must be pushed to your repository for this class on GitHub
• all unit tests must pass
• all acceptance tests must pass
• all programs must build, run, and execute as described in the assignment descriptions.

Extra Challenges (Not Required)

• None at this time

Last Updated 02/04/2025