Trade data viewer

Technologies

Flask
- click - for additional flask console commands
WebSockets for full duplex realtime communication with server
Plotly for realtime graphics rendering
sqlite for persistent storage of prices fluctuations

No external CDNs are used.

Running via docker

Warning!!! Next operation will try to assign 5000 and 8765 ports on hosts machine.

docker build -t trade_data .
docker run --network=host --rm -it trade-data

Open in browser localhost:5000.

Persistent storage

Mount a docker volume to /opt/data folder to keep all data between container restartings.

Install in dev mode

python3.8 -m venv .venv
source .venv/bin/activate
pip3 install -r requirements.txt
flask init-db

Run

sh init_and_start.sh

Notices for using pudb debugger

Configure pudb to classic shell not bpython.

FLASK_ENV=development flask run --no-debugger --no-reload

Structure

After openning Homepage or selecting a ticker in select element the browser will fetch the last cached data for the selected ticker. The Response headers will contain cache headers for a proxy webserver and for the browser:

result.headers['Cache-Control'] = 'public, smax-age=%d, max-age=%d,' % (ONE_HOUR // 4, ONE_HOUR // 4 - ONE_MINUTE)

After getting precached data for the ticker the browser will subscribe (using WS) to next prices fluctuations using the last received timestamp in precached data.

Then webserver push to the client a portion of all available cached data for the selected ticker.

And then client will receive all new data item in real time.

TODO

Switching to timeseries DB.
Use TimeRange features in Plotly.
Run through gunicorn
Make instruction to run through nginx and configure nginx cache

Potential bugs:

One overlapping value when the WebSocket server push to the client a portion of all available cached data.

gnomeby / trade-data