Flask-Sse
A simple Flask extension for HTML5 server-sent events support, powered by Redis
The extension provides 2 things - a blueprint with a single route for streaming events, and a helper function to send messages to subscribers:
from flask import Flask, json
from flask.ext.sse import sse, send_event
app = Flask(__name__)
app.register_blueprint(sse, url_prefix='/stream')
@app.route('/send')
def send_message():
send_event('myevent', json.dumps({"message": "Hello!"}))
You can then subscribe to these events in a supported browser:
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8" />
</head>
<body>
<script>
var source = new EventSource("{{ url_for('sse.stream') }}");
source.addEventListener('myevent', function(e) {
var data = JSON.parse(e.data);
// handle event
}, false);
</script>
</body>
</html>
The source comes with a basic example
Clients can subscribe to different channels by setting 'channel' on the query string, which defaults to 'sse'. These correspond to redis channels.
def send_message():
send_event('myevent', json.dumps({"line": "Something happened"}), channel='logs')
#######
var source = new EventSource("{{ url_for('sse.stream', channel='logs') }}")
Being a blueprint, you can attach a before_request handler to handle things like access control:
@sse.before_request
def check_access():
if request.args.get('channel') == 'firehose' and not g.user.is_admin():
abort(403)
Configuration
Redis connection details are read from the applications config using the following keys (defaults in [])
- SSE_REDIS_HOST [localhost]
- SSE_REDIS_PORT [6379]
- SSE_REDIS_DB [0]
Caveats
Subscribers will connect and block for a long time, so you should seriously consider running under an asynchronous WSGI server, such as gunicorn+gevent (like the example)
I should also say I'm not really maintaining this beyond accepting the odd pull request - it was built as an experiment but I'm not using it in anger on anything production. I wont be publishing it on PyPi myself for the same reasons - if I start using it properly then it will go on PyPi and have some tests put around it.
Credits
Inspired by django-sse