feat: enable processing of large video files over 2GB

Update audio extraction to use file-based streaming
instead of in-memory buffers to support processing
video files larger than 2GB.

**Changes:**
- Modified `IAudioExtractor.ExtractMp3FromMp4Async()`
  to accept an output audio stream parameter instead
  of creating a MemoryStream internally
- Updated `AudioExtractor` to write directly to the
  provided audio stream, eliminating the need to load
  entire video into memory
- Enhanced `FFMpegService` to detect FileStream inputs
  and use direct file-to-file processing when possible,
  falling back to pipe-based streaming for other stream types
- Refactored `DefaultCommand` to create output directory
  and audio file stream before extraction, enabling direct
  file-based audio extraction
- Updated all unit and integration tests to accommodate the
  new audio stream parameter

This change prevents `OutOfMemoryException` errors when processing
large video files by streaming data directly to disk rather than
buffering in memory. Additionally, it removes the possibility of
overflowing the 2GB size limit that `MemoryStream` has, which
would result in `IOException` errors when trying to write or
copy to it.
This commit is contained in:
Stevan Freeborn
2025-10-10 16:23:02 -05:00
parent 6dcffa6b75
commit e611522c7b
6 changed files with 57 additions and 38 deletions
@@ -25,7 +25,7 @@ public sealed class AudioExtractor : IAudioExtractor
_videoService = videoService ?? throw new ArgumentNullException(nameof(videoService), $"{nameof(videoService)} cannot be null");
}
public async Task<Stream> ExtractMp3FromMp4Async(Stream video)
public async Task<Stream> ExtractMp3FromMp4Async(Stream video, Stream audio)
{
if (video is null)
{
@@ -42,25 +42,29 @@ public sealed class AudioExtractor : IAudioExtractor
throw new ArgumentException("Video stream must be seekable", nameof(video));
}
if (audio is null)
{
throw new ArgumentNullException(nameof(audio), "Audio stream cannot be null");
}
if (audio.CanWrite is false)
{
throw new ArgumentException("Audio stream must be writable", nameof(audio));
}
var originalPosition = video.Position;
try
{
var mp3Stream = new MemoryStream();
using var mp4Stream = new MemoryStream();
await video.CopyToAsync(mp4Stream).ConfigureAwait(false);
mp4Stream.Position = 0;
var wasExtracted = await _videoService.ExtractAudioFromVideoAsync(mp4Stream, mp3Stream).ConfigureAwait(false);
var wasExtracted = await _videoService.ExtractAudioFromVideoAsync(video, audio).ConfigureAwait(false);
if (wasExtracted is false)
{
throw new FailedAudioExtractionException("Failed to extract audio from the video stream.");
}
mp3Stream.Position = 0;
return mp3Stream;
audio.Position = 0;
return audio;
}
catch (Exception e) when (e is not FailedAudioExtractionException)
{
@@ -9,11 +9,14 @@ public interface IAudioExtractor
/// Extracts MP3 audio from an MP4 video stream.
/// </summary>
/// <param name="video">The input video stream.</param>
/// <param name="audio">The output audio stream where the extracted MP3 will be written.</param>
/// <returns>A stream containing the extracted MP3 audio.</returns>
/// <exception cref="ArgumentNullException">Thrown when the video stream is null.</exception>
/// <exception cref="ArgumentException">Thrown when the video stream is not readable.</exception>
/// <exception cref="ArgumentException">Thrown when the video stream is not seekable.</exception>
/// <exception cref="ArgumentNullException">Thrown when the audio stream is null.</exception>
/// <exception cref="ArgumentException">Thrown when the audio stream is not writable.</exception>
/// <exception cref="FailedAudioExtractionException">Thrown when the audio extraction fails.</exception>
/// <remarks>The method will preserve the passed video stream's data and position.</remarks>
Task<Stream> ExtractMp3FromMp4Async(Stream video);
Task<Stream> ExtractMp3FromMp4Async(Stream video, Stream audio);
}
@@ -12,14 +12,17 @@ internal sealed class FFMpegService : IVideoService
{
public async Task<bool> ExtractAudioFromVideoAsync(Stream video, Stream audio)
{
return await FFMpegArguments
.FromPipeInput(new StreamPipeSource(video))
.OutputToPipe(
new StreamPipeSink(audio),
static o => o.DisableChannel(Channel.Video).ForceFormat("mp3")
)
.ProcessAsynchronously()
.ConfigureAwait(false);
var inputArguments = video is FileStream videoFileStream
? FFMpegArguments.FromFileInput(videoFileStream.Name)
: FFMpegArguments.FromPipeInput(new StreamPipeSource(video));
Action<FFMpegArgumentOptions> arguments = static o => o.DisableChannel(Channel.Video).ForceFormat("mp3");
var processor = audio is FileStream audioFileStream
? inputArguments.OutputToFile(audioFileStream.Name, addArguments: arguments)
: inputArguments.OutputToPipe(new StreamPipeSink(audio), addArguments: arguments);
return await processor.ProcessAsynchronously().ConfigureAwait(false);
}
public async Task CreateClipFromVideoAsync(