stacktrace_translator

Web-based frontend for translating stack traces of SpringRTS engine crashes.

The frontend is a lightweight Dash (Python) app that talks to the Python XML-RPC backend. It replaces the previous PHP/Apache frontend, giving a smaller footprint (a single Python stack instead of PHP + Apache + Python).

Requirements:

• Docker
• docker-compose

Installation:

1. Clone the repository:

   git clone https://github.com/beyond-all-reason/stacktrace_translator.git

2. Navigate to the directory:

   cd stacktrace_translator

3. Build Docker image:

   docker-compose build

   Or completely rebuild it via (also needs a complete rebuild if git repo has changes!):

   docker-compose build --no-cache

4. Deploy Docker image:

   docker-compose up

5. Deployed:

   http://127.0.0.1:80
   

Usage

Use a URL or manually copy and paste into the textbox the infolog.txt of a crashed Spring.

Keep in mind:

• Only Windows versions can be translated.
• At max 100.000 bytes are downloaded.
• The app takes a couple of minutes to process the data.

Architecture

A single container (built from the root Dockerfile) runs both services, started by entrypoint.sh:

• XML-RPC backend — core translation logic on port 8000. Published on host port 8000 so external services can keep calling translate_stacktrace directly.
• Dash frontend — Dash web UI served by gunicorn on port 8050 (published on host port 80). It forwards infologs to the backend on 127.0.0.1:8000 and renders the translated frames with GitHub source links. The backend URL is configurable via the TRANSLATOR_URL environment variable.

docker-compose orchestrates this single container, and Docker's restart: unless-stopped policy recovers it if either process exits.

Local changes

The image copies the local python-server/ and dash-frontend/ sources at build time, so any local edits are picked up on the next build.

Build and run with

docker-compose build && docker-compose up

The frontend is published on host port 80, so open http://127.0.0.1:80 after docker-compose up. The XML-RPC backend is available on http://127.0.0.1:8000.

Running without Docker

# backend (needs binutils/addr2line + p7zip installed on the host)
python3 python-server/stacktrace_translator.py   # serves XML-RPC on :8000

# frontend (in another shell)
cd dash-frontend
pip install -r requirements.txt
TRANSLATOR_URL=http://127.0.0.1:8000 python app.py   # serves on :8050

Then open http://127.0.0.1:8050.