Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/kennethwussmann/docker-ffmpeg-converter
Docker service designed to simplify and streamline media conversion tasks
https://github.com/kennethwussmann/docker-ffmpeg-converter
converter docker ffmpeg media pipeline
Last synced: 2 days ago
JSON representation
Docker service designed to simplify and streamline media conversion tasks
- Host: GitHub
- URL: https://github.com/kennethwussmann/docker-ffmpeg-converter
- Owner: KennethWussmann
- License: gpl-3.0
- Created: 2023-08-02T18:48:07.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-05-12T21:40:56.000Z (7 months ago)
- Last Synced: 2024-11-02T07:06:04.303Z (about 2 months ago)
- Topics: converter, docker, ffmpeg, media, pipeline
- Language: TypeScript
- Homepage:
- Size: 168 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
🎬 docker-ffmpeg-converter
Dynamic Docker service designed to simplify and streamline media conversion tasks
## 🌟 Overview
docker-ffmpeg-converter automates the process of file conversion by:
1. **Monitoring** an input directory for new files.
2. **Utilizing** [ffmpeg](https://www.ffmpeg.org/) to perform specified conversion operations.
3. **Optionally removing** the source files post conversion.
Whether you are building a small pipeline or conducting background conversions of multiple files, docker-ffmpeg-converter efficiently handles it all. Run multiple instances and make your conversion process more versatile and robust.
## 🚀 Deployment
Deploying docker-ffmpeg-converter is a breeze with Docker Compose. Below is an example YAML configuration file:
```YAML
version: "3.9"
services:
video-converter:
image: ghcr.io/kennethwussmann/docker-ffmpeg-converter:latest
volumes:
# This is where we will add input files and get output files
- ./data:/data
environment:
# See below for all options and their meaning. This is just the required set.
- SOURCE_DIRECTORY_PATH=/data/input
- DESTINATION_DIRECTORY_PATH=/data/output
- GLOB_PATTERNS=*.webm
- FFMPEG_ARGS=-y -fflags +genpts -i %s -r 24 %s.mp4
```
### 🏷️ Tags
This image is built for `arm64` and `amd64`.
- `latest` - Latest stable release
- `x.x.x` - Specific version under Semver ([See all versions](https://github.com/KennethWussmann/docker-ffmpeg-converter/pkgs/container/docker-ffmpeg-converter/versions))
- `develop` - Unstable pre-release development version
### 🔧 Configuration
Configure the container through environment variables. Here's a breakdown of what you can customize:
Variable
Required
Description
SOURCE_DIRECTORY_PATH
Yes
Directory watched for new files
DESTINATION_DIRECTORY_PATH
Yes
Directory for converted files
GLOB_PATTERNS
Yes
Comma-separated list of glob patterns the service will use to filter files in the SOURCE_DIRECTORY_PATH
FFMPEG_ARGS
Yes
Arguments for the ffmpeg binary that specify what to do. See ffmpeg docs. You can use placeholders %s, see below.
REMOVE_SOURCE_AFTER_CONVERT
No (default: false)
true or false. Whether or not to delete source files after successful conversion
SCAN_INTERVAL
No (default: 10)
Interval in seconds when the service will search for new files
FILE_UNCHANGED_INTERVALS
No (default: 3)
How many cycles the service will wait for new files to stay unchanged until conversion starts. See below for detailed info.
## 💼 How the Service Works
### 📂 File Monitoring and Conversion
1. **File Detection**: The service continually polls the `SOURCE_DIRECTORY_PATH`, filtering files using the specified `GLOB_PATTERNS`.
2. **File Verification**: New files are cached and checked for size stability, based on `FILE_UNCHANGED_INTERVALS`.
3. **Conversion**: Stable files are handed to the converter, building an ffmpeg command with `FFMPEG_ARGS`.
Example:
```shell
FFMPEG_ARGS=-y -fflags +genpts -i %s -r 24 %s.mp4
SOURCE_DIRECTORY_PATH=/data/input
DESTINATION_DIRECTORY_PATH=/data/output
```
Resulting command:
```shell
/usr/bin/ffmpeg -y -fflags +genpts -i /data/input/myfile.webm -r 24 /data/input/myfile.mp4
```
> Note: The order of `%s` is vital. The first represents the source file and the second the destination.
4. **Post-Conversion**: Depending on `REMOVE_SOURCE_AFTER_CONVERT`, the source file may be deleted after successful conversion.
### ⏱️ Custom Intervals and Patterns
- **Scan Interval**: Adjust the polling frequency with `SCAN_INTERVAL`, defining the seconds between each search.
- **File Unchanged Intervals**: Customize the number of cycles to wait for file size stability using `FILE_UNCHANGED_INTERVALS`.
## 🔗 Pipeline example
Here is a more complex example on how to use mulitple instances:
```YAML
version: "3.9"
services:
# Same as above
webm-to-mp4:
image: ghcr.io/kennethwussmann/docker-ffmpeg-converter:latest
volumes:
# This is where we will add input files and get output files
- ./data:/data
environment:
- SOURCE_DIRECTORY_PATH=/data/input
- DESTINATION_DIRECTORY_PATH=/data/output
- GLOB_PATTERNS=*.webm
# Convert *.webm files to .mp4
- FFMPEG_ARGS=-y -fflags +genpts -i %s -r 24 %s.mp4
# Another instance to take thumbnails from the *.webm files
webm-thumbnails:
image: ghcr.io/kennethwussmann/docker-ffmpeg-converter:latest
volumes:
- ./data:/data
environment:
- SOURCE_DIRECTORY_PATH=/data/input
- DESTINATION_DIRECTORY_PATH=/data/output
- GLOB_PATTERNS=*.webm
# Take multiple thumbnails from *.webm files
- FFMPEG_ARGS=-y -i %s -vf fps=1/4 %s_%04d.png
```
The output directory will then contain multiple thumbnails and a converted MP4 file of our source material. Because the containers run in parallel they finish quickly as both tasks can run simulatanously.
## 🎉 Conclusion
docker-ffmpeg-converter is a great solution for seamless media conversion tasks, providing robust customization and a simplified deployment process. Experience the ease of automation with this powerful Docker service.
Feel free to explore, contribute, or seek support. Happy converting! 🎬