astra/Vidio
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Added audio support for videos

Browse source
This commit is contained in:
Alex Eidt 2022-03-30 20:43:31 -07:00
parent 204ed5b1b5
commit b1fd41b75f
5 changed files with 132 additions and 65 deletions
Download patch file Download diff file Expand all files Collapse all files

2
.gitignore vendored
View file

@ -1,2 +1,4 @@
*.gif
*.mp4
vidio_test.go

26
README.md
View file

@ -1,6 +1,6 @@
# Vidio
A simple Video I/O library written in Go. This library relies on [FFmpeg](https://www.ffmpeg.org/), and [FFProbe](https://www.ffmpeg.org/) which must be downloaded before usage.
A simple Video I/O library written in Go. This library relies on [FFmpeg](https://www.ffmpeg.org/), and [FFProbe](https://www.ffmpeg.org/) which must be downloaded before usage and added to the system path.
All frames are encoded and decoded in 8-bit RGB format.
@ -25,6 +25,7 @@ type Video struct {
duration float64 // Duration in seconds
fps float64 // Frames per second
codec string // Codec used to encode video
audio_codec string // Codec used for audio encoding
pix_fmt string // Pixel format video is stored in
framebuffer []byte // Raw frame data
pipe *io.ReadCloser // Stdout pipe for ffmpeg process
@ -43,7 +44,7 @@ for video.Read() {
## `Camera`
The `Camera` can read from any cameras on the device running Vidio. It takes in the stream index. On most machines the webcam device has index 0.
The `Camera` can read from any cameras on the device running Vidio. It takes in the stream index. On most machines the webcam device has index 0. Note that audio retrieval from the microphone is not yet supported.
```go
type Camera struct {
@ -84,21 +85,25 @@ type Options struct {
fps float64 // Frames per second. Default 25
quality float64 // If bitrate not given, use quality instead. Must be between 0 and 1. 0:best, 1:worst
codec string // Codec for video. Default libx264
audio string // File path for audio for the video. If no audio, audio=nil.
audio_codec string // Codec for audio. Default aac
}
```
```go
type VideoWriter struct {
filename string // Output video filename
filename string // Output filename
audio string // Audio filename.
width int // Frame width
height int // Frame height
bitrate int // Output video bitrate for encoding
bitrate int // Output video bitrate
loop int // Number of times for GIF to loop
delay int // Delay of final frame of GIF
macro int // macro size for determining how to resize frames for codecs
fps float64 // Frames per second for output video
quality float64 // Used if bitrate not given
codec string // Codec to encode video with
delay int // Delay of final frame of GIF. Default -1 (same delay as previous frame)
macro int // Macroblock size for determining how to resize frames for codecs
fps float64 // Frames per second for output video. Default 25
quality float64 // Used if bitrate not given. Default 0.5
codec string // Codec to encode video with. Default libx264
audio_codec string // Codec to encode audio with. Default aac
pipe *io.WriteCloser // Stdout pipe of ffmpeg process
cmd *exec.Cmd // ffmpeg command
}
@ -132,13 +137,14 @@ vidio.Write("output.jpg", w, h, img)
## Examples
Copy `input.mp4` to `output.mp4`.
Copy `input.mp4` to `output.mp4`. Copy the audio from `input.mp4` to `output.mp4` as well.
```go
video := vidio.NewVideo("input.mp4")
options := vidio.Options{
fps: video.fps,
bitrate: video.bitrate
audio: "input.mp4",
}
writer := vidio.NewVideoWriter("output.mp4", video.width, video.height, &options)

7
utils.go
View file

@ -35,7 +35,7 @@ func checkExists(program string) {
}
// Parse ffprobe output to fill in video data.
func parseFFprobe(input []byte, video *Video) {
func parseFFprobe(input []byte) map[string]string {
data := make(map[string]string)
for _, line := range strings.Split(string(input), "|") {
if strings.Contains(line, "=") {
@ -45,7 +45,11 @@ func parseFFprobe(input []byte, video *Video) {
}
}
}
return data
}
// Adds Video data to the video struct from the ffprobe output.
func addVideoData(data map[string]string, video *Video) {
video.width = int(parse(data["width"]))
video.height = int(parse(data["height"]))
video.duration = float64(parse(data["duration"]))
@ -133,7 +137,6 @@ func parseDevices(buffer []byte) []string {
devices = append(devices, pair.name)
}
}
return devices
}

72
video.go
View file

@ -18,6 +18,7 @@ type Video struct {
duration float64 // Duration of video in seconds.
fps float64 // Frames per second.
codec string // Codec used for video encoding.
audio_codec string // Codec used for audio encoding.
pix_fmt string // Pixel format video is stored in.
framebuffer []byte // Raw frame data.
pipe *io.ReadCloser // Stdout pipe for ffmpeg process.
@ -33,40 +34,53 @@ func NewVideo(filename string) *Video {
// Check if ffmpeg and ffprobe are installed on the users machine.
checkExists("ffmpeg")
checkExists("ffprobe")
// Extract video information with ffprobe.
cmd := exec.Command(
"ffprobe",
"-show_streams",
"-select_streams", "v", // Only show video data
"-print_format", "compact",
"-loglevel", "quiet",
filename,
)
pipe, err := cmd.StdoutPipe()
if err != nil {
panic(err)
}
ffprobe := func(stype string) map[string]string {
// "stype" is stream stype. "v" for video, "a" for audio.
// Extract video information with ffprobe.
cmd := exec.Command(
"ffprobe",
"-show_streams",
"-select_streams", stype, // Only show video data
"-print_format", "compact",
"-loglevel", "quiet",
filename,
)
if err := cmd.Start(); err != nil {
panic(err)
}
// Read ffprobe output from Stdout.
buffer := make([]byte, 2<<10)
total := 0
for {
n, err := pipe.Read(buffer[total:])
total += n
if err == io.EOF {
break
pipe, err := cmd.StdoutPipe()
if err != nil {
panic(err)
}
if err := cmd.Start(); err != nil {
panic(err)
}
// Read ffprobe output from Stdout.
buffer := make([]byte, 2<<10)
total := 0
for {
n, err := pipe.Read(buffer[total:])
total += n
if err == io.EOF {
break
}
}
// Wait for ffprobe command to complete.
if err := cmd.Wait(); err != nil {
panic(err)
}
return parseFFprobe(buffer[:total])
}
// Wait for ffprobe command to complete.
if err := cmd.Wait(); err != nil {
panic(err)
}
videoData := ffprobe("v")
audioData := ffprobe("a")
video := &Video{filename: filename, depth: 3}
parseFFprobe(buffer[:total], video)
addVideoData(videoData, video)
video.audio_codec = audioData["codec_name"]
return video
}

88
videowriter.go
View file

@ -11,29 +11,33 @@ import (
)
type VideoWriter struct {
filename string // Output filename.
width int // Frame width.
height int // Frame height.
bitrate int // Output video bitrate.
loop int // Number of times for GIF to loop.
delay int // Delay of final frame of GIF. Default -1 (same delay as previous frame).
macro int // Macroblock size for determining how to resize frames for codecs.
fps float64 // Frames per second for output video. Default 25.
quality float64 // Used if bitrate not given. Default 0.5.
codec string // Codec to encode video with. Default libx264.
pipe *io.WriteCloser // Stdout pipe of ffmpeg process.
cmd *exec.Cmd // ffmpeg command.
filename string // Output filename.
audio string // Audio filename.
width int // Frame width.
height int // Frame height.
bitrate int // Output video bitrate.
loop int // Number of times for GIF to loop.
delay int // Delay of final frame of GIF. Default -1 (same delay as previous frame).
macro int // Macroblock size for determining how to resize frames for codecs.
fps float64 // Frames per second for output video. Default 25.
quality float64 // Used if bitrate not given. Default 0.5.
codec string // Codec to encode video with. Default libx264.
audio_codec string // Codec to encode audio with. Default aac.
pipe *io.WriteCloser // Stdout pipe of ffmpeg process.
cmd *exec.Cmd // ffmpeg command.
}
// Optional parameters for VideoWriter.
type Options struct {
bitrate int // Bitrate.
loop int // For GIFs only. -1=no loop, 0=infinite loop, >0=number of loops.
delay int // Delay for final frame of GIFs.
macro int // Macroblock size for determining how to resize frames for codecs.
fps float64 // Frames per second for output video.
quality float64 // If bitrate not given, use quality instead. Must be between 0 and 1. 0:best, 1:worst.
codec string // Codec for video.
bitrate int // Bitrate.
loop int // For GIFs only. -1=no loop, 0=infinite loop, >0=number of loops.
delay int // Delay for final frame of GIFs.
macro int // Macroblock size for determining how to resize frames for codecs.
fps float64 // Frames per second for output video.
quality float64 // If bitrate not given, use quality instead. Must be between 0 and 1. 0:best, 1:worst.
codec string // Codec for video.
audio string // File path for audio. If no audio, audio=nil.
audio_codec string // Codec for audio.
}
// Creates a new VideoWriter struct with default values from the Options struct.
@ -86,6 +90,20 @@ func NewVideoWriter(filename string, width, height int, options *Options) *Video
} else {
writer.codec = options.codec
}
if options.audio != "" {
if !exists(options.audio) {
panic("Audio file does not exist.")
}
writer.audio = options.audio
if options.audio_codec == "" {
writer.audio_codec = "aac"
} else {
writer.audio_codec = options.audio_codec
}
}
return &writer
}
@ -104,11 +122,25 @@ func initVideoWriter(writer *VideoWriter) {
"-pix_fmt", "rgb24",
"-r", fmt.Sprintf("%.02f", writer.fps), // frames per second.
"-i", "-", // The input comes from stdin.
"-an", // Tells ffmpeg not to expect any audio.
"-vcodec", writer.codec,
"-pix_fmt", "yuv420p", // Output is 8-bit RGB, no alpha.
}
gif := strings.HasSuffix(strings.ToLower(writer.filename), ".gif")
if writer.audio == "" || gif {
command = append(command, "-an") // No audio.
} else {
command = append(
command,
"-i", writer.audio,
)
}
command = append(
command,
"-vcodec", writer.codec,
"-pix_fmt", "yuv420p", // Output is 8-but RGB, no alpha.
)
// Code from the imageio-ffmpeg project.
// https://github.com/imageio/imageio-ffmpeg/blob/master/imageio_ffmpeg/_io.py#L399.
// If bitrate not given, use a default.
@ -124,7 +156,7 @@ func initVideoWriter(writer *VideoWriter) {
command = append(command, "-b:v", fmt.Sprintf("%d", writer.bitrate))
}
// For GIFs, add looping and delay parameters.
if strings.HasSuffix(strings.ToLower(writer.filename), ".gif") {
if gif {
command = append(
command,
"-loop", fmt.Sprintf("%d", writer.loop),
@ -152,6 +184,16 @@ func initVideoWriter(writer *VideoWriter) {
}
}
// If audio was included, then specify video and audio channels.
if writer.audio != "" {
command = append(
command,
"-acodec", writer.audio_codec,
"-map", "0:v:0",
"-map", "1:a:0",
)
}
command = append(command, writer.filename)
cmd := exec.Command("ffmpeg", command...)
writer.cmd = cmd