🎬 PyVideoKit-Libs

Python library for FFmpeg-based video and audio processing. Provides the core functions behind PyVideoKit-CLI and PyVideoKit-GUI, and can be used directly in your own Python projects.

✨ Features

• 📼 VHS effect — retro visual noise, color bleed, and audio degradation
• ✂️ Trim — cut a segment by start/end time with stream copy (no re-encoding)
• 🔗 Concatenate — join multiple videos with stream compatibility validation
• 🎬 Fade — fade-in and/or fade-out on an FFV1 master
• 🔊 Extract audio — dump the audio track to uncompressed WAV (PCM 16-bit)
• 🎞️ Convert to FFV1 — create a lossless MKV master for editing
• 📺 Prepare for YouTube — encode to ProRes 422 HQ MOV, upscaled to 4K

📦 Requirements

• Python ≥ 3.10
• FFmpeg and FFprobe available in PATH
• SoX available in PATH (required by apply_vhs_effect)

No Python package dependencies — the library delegates all heavy lifting to external tools via subprocess.

🔧 Installation

🏗️ Arch Linux (AUR)

yay -S python-pyvideokit-libs

FFmpeg, FFprobe, and SoX are installed automatically as pacman dependencies.

🐍 Other systems (pip)

pip install PyVideoKit-Libs

Make sure FFmpeg, FFprobe, and SoX are available in your PATH.

🖥️ Desktop Integration

If you also have PyVideoKit-GUI installed, you can add it to your application launcher. Arch Linux users get this automatically via the AUR package. For all other systems:

1. Save the icon:

mkdir -p ~/.local/share/icons/hicolor/scalable/apps
cat > ~/.local/share/icons/hicolor/scalable/apps/pvk-gui.svg << 'EOF'
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 48 48">
  <rect x="4" y="14" width="40" height="30" rx="3" fill="#2d3561"/>
  <rect x="10" y="19" width="28" height="20" rx="2" fill="#1a2038"/>
  <polygon points="18,21 18,37 33,29" fill="#ff6b6b"/>
  <rect x="4" y="6" width="40" height="8" rx="2" fill="#ff6b6b"/>
  <rect x="11" y="6" width="6" height="8" fill="#ffffff"/>
  <rect x="23" y="6" width="6" height="8" fill="#ffffff"/>
  <rect x="35" y="6" width="5" height="8" fill="#ffffff"/>
</svg>
EOF

2. Create the .desktop entry:

mkdir -p ~/.local/share/applications
cat > ~/.local/share/applications/pvk-gui.desktop << 'EOF'
[Desktop Entry]
Version=1.0
Type=Application
Name=PyVideoKit
GenericName=Video Processing Tool
Comment=FFmpeg-based video processing GUI
Exec=pvk-gui
Icon=pvk-gui
Categories=AudioVideo;Video;AudioVideoEditing;
Terminal=false
Keywords=video;ffmpeg;trim;fade;vhs;youtube;convert;
EOF

3. Refresh the icon cache (may be required on some desktop environments):

gtk-update-icon-cache ~/.local/share/icons/hicolor/

PyVideoKit will now appear in your application launcher.

🚀 Quick Start

from pathlib import Path
from pyvideokit_libs import convert_to_ffv1, fade_video, prepare_youtube

# 1. Create a lossless master
master = convert_to_ffv1(Path("recording.mp4"))

# 2. Add a 1-second fade in and out
faded = fade_video(master, fade_in=1.0, fade_out=1.0)

# 3. Export for YouTube
result = prepare_youtube(faded)
print(result)

All functions accept an optional on_progress callback (pct: float) -> None for tracking progress.

📖 API Reference

🎞️ Video processing functions

apply_vhs_effect(input_path, output=None, on_progress=None) -> Path

Apply a retro VHS visual and audio effect. The input must have an audio stream. Produces a <name>_vhs.mkv file next to the input.

out = apply_vhs_effect(Path("video.mp4"))

trim_video(input_path, start, end, output=None, on_progress=None) -> Path

Cut a segment from a video using stream copy (no re-encoding). start and end are in seconds.

out = trim_video(Path("video.mkv"), start=10.0, end=90.5)

Use parse_time_to_seconds("00:01:30") to convert a timestamp string to seconds.

join_videos(video_paths, output=None, on_progress=None) -> Path

Concatenate two or more videos using stream copy. Validates stream compatibility (codec, resolution, fps, audio format) before proceeding.

out = join_videos([Path("clip1.mkv"), Path("clip2.mkv"), Path("clip3.mkv")])

fade_video(input_path, fade_in=None, fade_out=None, output=None, on_progress=None) -> Path

Add fade-in and/or fade-out to an FFV1 video. Output is re-encoded as FFV1 with PCM audio.

out = fade_video(Path("master.mkv"), fade_in=1.5, fade_out=2.0)

extract_audio(input_path, output=None, on_progress=None) -> Path

Extract the audio track to an uncompressed WAV file (PCM 16-bit, pcm_s16le).

out = extract_audio(Path("video.mkv"))

convert_to_ffv1(input_path, fps=None, output=None, on_progress=None) -> Path

Convert any video to a lossless FFV1/MKV master. Use this as the first step before applying effects. fps defaults to None (preserve source frame rate); pass an integer to force a specific rate.

master = convert_to_ffv1(Path("recording.mp4"))        # preserve source fps
master = convert_to_ffv1(Path("recording.mp4"), fps=60) # force 60 fps

prepare_youtube(input_path, output=None, on_progress=None) -> Path

Encode an FFV1 master to ProRes 422 HQ MOV for YouTube upload:

• Upscaled to 4K (2160p height), aspect ratio preserved
• ProRes 422 HQ (prores_ks, profile 3), 10-bit 4:2:2 (yuv422p10le)
• Uncompressed 16-bit PCM audio

out = prepare_youtube(Path("master_ffv1.mkv"))

⚙️ Utility functions

🛡️ Error handling

All functions raise FFmpegError when an FFmpeg subprocess exits with a non-zero code.

from pyvideokit_libs import FFmpegError, convert_to_ffv1

try:
    out = convert_to_ffv1(Path("bad_file.mp4"))
except FFmpegError as e:
    print(f"FFmpeg failed (exit {e.returncode}):")
    for line in e.error_lines:
        print(line)

🔄 Typical Workflow

convert_to_ffv1  →  trim_video / fade_video / apply_vhs_effect  →  prepare_youtube

Functions that re-encode (fade_video, apply_vhs_effect, prepare_youtube) expect an FFV1 .mkv as input.

⚖️ License

This project is licensed under the GPLv3 License — see the LICENSE file for details.