PySarra

A Python application that can read from Sarracenia and relay the messages however you'd like. It is designed to be significantly simpler than the official client, and is meant for people who don't need all the bells and whistles. It is only tested on Linux, and does not seem to work on macOS.

To create a processor, you must define process() in your processor that accepts one argument which is a dictionary with keys of topic and message. message is the full URL of the file that is now available under the Environment Canada datamart. A Google Cloud relay processor is included as an example.

Notes

• To use the Google Cloud processor, you must:

  • have a working Google Cloud Platform account

  • change the project name to your project name as created in GCP

  • pass in your Google Cloud credentials as an environment variable as described right here unless you are directly running this from a Compute Engine instance that already has access to the topic you are relaying the messages to.

Usage

• Clone this repository.

• Install all requirements by running pip3 install -r requirements.txt

• Edit config.json as necessary. Tasks is a list of dictionaries, where each dictionary has entries of topic, regex, and processor.

  • processor is the filename of your processor, without the .py.

  • You can subscribe to any topics by using the . delimeter - for example to be notified of any changes under http://dd.weather.gc.ca/alerts/cap/20180725 can be tracked by using a topic value of alerts.cap.20180725.

  • regex is optional. If it is defined, it is tested with each incoming message and the processor is only called if it is a match. Please note that due to Python's Global Interpreter Lock, it is recommended to keep regexes simple.

• Run main.py!