PyLingual Docker

This repository provides Docker images for PyLingual, a powerful Python bytecode decompiler.

We provide four image variants to balance between image size and pre-installed compatibility.

Variant	Base Python	Pre-installed Pythons	Purpose
lite-cli	3.14	3.14	Thin wrapper around PyLingual's CLI.
full-cli	3.14	3.6 - 3.14	CLI tool with all supported Python versions pre-installed.
lite-server	3.14	3.14	Lightweight REST API. (Requires internet for first model download)
full-server	3.14	3.6 - 3.14	REST API with all supported Python versions pre-installed and models pre-loaded for offline use.

Building the Images

All images are built from the same Dockerfile using different targets.

Note: The full images download several gigabytes of models during the build process.

# Lite CLI
docker build --target lite-cli -t pylingual:lite-cli .

# Full CLI (Offline capable)
docker build --target full-cli -t pylingual:full-cli .

# Lite Server
docker build --target lite-server -t pylingual:lite-server .

# Full Server (Offline capable)
docker build --target full-server -t pylingual:full-server .

Build Requirements

The build process uses docker buildx features (Mounts). Ensure you have a recent version of Docker. Intermediate build stages are used to keep the final image size optimized.

Using the CLI

Using the CLI

Run the CLI image and mount the directory containing your .pyc files.

docker run --rm -v $(pwd):/data pylingual:lite-cli /data/example.pyc -o /data

Using the Server

Run the server image and map port 8000.

docker run -d -p 8000:8000 pylingual:lite-server

API Endpoint: /decompile

Decompile a .pyc file by sending a POST request to the /decompile endpoint.

Example with curl:

curl -X POST http://localhost:8000/decompile \
  -F "file=@example.pyc"

Response Format:

{
  "decompiled_source": "<decompiled source code>",
  "equivalence_report": ["..."],
  "version": "3.<minor version>"
}