bugoman/jpeg-archive repository overview

⁠jpeg-archive

The bugoman/jpeg-archive Docker image is designed to compress .jpg and .jpeg images using the jpeg-archive tools. This image is versatile and can handle both batch processing of multiple files and single-file processing using the jpeg-recompress tool or other jpeg-archive binaries.

⁠Usage

⁠Running the Image for Single-File Processing

To compress a single image, you can directly call the jpeg-archive tool inside the container. For example:

docker run --rm --user 1000:1000 \
    -v $(pwd)/img:/img \
    bugoman/jpeg-archive \
    jpeg-recompress --quiet --strip --quality medium --method smallfry \
    /img/input.jpg /img/output.jpg

This command will compress the input.jpg file in the /img directory and save it as output.jpg in the same directory.

⁠Batch Processing: Multiple Files in a Folder

If you need to process multiple files located in a single folder, you can use the batch.sh script. This script will automatically find all .jpg and .jpeg files in the /input directory and compress them in parallel.

To run the image for batch processing, use the following command:

docker run --rm --user 1000:1000 \
    -v $(pwd)/input:/input \
    -v $(pwd)/output:/output \
    bugoman/jpeg-archive \
    batch.sh --quiet --strip --quality medium --method smallfry

The batch.sh script will scan the /input directory for .jpg and .jpeg files, compress them using multiple parallel processes, and save the compressed files to the /output directory. By default, the script calculates the number of parallel processes based on the system's available CPU cores to optimize performance.

⁠Controlling the Number of Parallel Processes

If you want to manually control the number of parallel processes used for batch processing, you can set the NUM_PROCESSES environment variable. For example, to use 4 parallel processes:

docker run --rm --user 1000:1000 \
    -v $(pwd)/input:/input \
    -v $(pwd)/output:/output \
    -e NUM_PROCESSES=4 \
    bugoman/jpeg-archive \
    batch.sh --quiet --strip --quality medium --method smallfry