The OCS makes it easy to coordinate hardware operation and I/O tasks in a distributed system such as an astronomical observatory or test laboratory. OCS relies on the use of a central WAMP router (currently crossbar.io) for coordinating the communication and control of these distributed systems.
The OCS provides Python (and JavaScript) functions and classes to allow "Clients" to talk to "Agents". An Agent is a software program that knows how to do something interesting and useful, such as acquire data from some device or perform cleanup operations on a particular file system. A Control Client could be a web page with control buttons and log windows, or a script written by a user to perform a series of unattended, interlocking data acquisition tasks.
This repository, OCS, contains library code and core system components. Additional code for operating specific hardware can be found in the Simons Observatory Control System (SOCS) repository. Grafana and InfluxDB are supported to provide a near real-time monitoring and historical look back of the housekeeping data.
This code targets Python 3.7+.
- so3g - Required for using the HK Aggregator Agent.
- crossbar (optional) - The supported WAMP router. Controllable via ocsbow. Can be installed with pip or run via Docker.
Install and update with pip:
$ pip3 install -U ocs
If you need to install the optional so3g module you can do so via:
$ pip3 install -U ocs[so3g]
If you are considering contributing to OCS, or would like to use an unreleased feature, you will want to install from source. To do so, clone this repository and install using pip:
$ git clone https://github.com/simonsobs/ocs.git $ cd ocs/ $ pip3 install -r requirements.txt $ pip3 install .
Note: If you want to install locally, not globally, throw the --user flag on the pip3 commands.
Docker images for OCS and each Agent are available on Docker Hub. Official
releases will be tagged with their release version, i.e. v0.1.0
. These are
only built on release, and the latest
tag will point to the latest of these
released tags. These should be considered stable.
Test images will be tagged with the latest released version tag, the number of
commits ahead of that release, the latest commit hash, i.e.
v0.6.0-53-g0e390f6
. These get built on each commit to the main
branch,
and are useful for testing and development, but should be considered unstable.
The OCS documentation can be built using Sphinx. There is a separate
requirements.txt
file in the docs/
directory to install Sphinx and any
additional documentation dependencies:
$ cd docs/ $ pip3 install -r requirements.txt $ make html
You can then open docs/_build/html/index.html
in your preferred web
browser. You can also find a copy hosted on Read the Docs.
The tests for OCS can be run using pytest, and should be run from the
tests/
directory:
$ cd tests/ $ python3 -m pytest
To run the tests within a Docker container (useful if your local environment is missing some dependencies), first make sure you build the latest ocs image, then use docker run:
$ docker build -t ocs . $ docker run --rm -w="/app/ocs/tests/" ocs sh -c "python3 -m pytest -m 'not integtest'"
For more details see tests/README.rst.
A self contained example, demonstrating the operation of a small observatory with a single OCS Agent is contained in example/miniobs/. See the readme in that directory for details.
For guidelines on how to contribute to OCS see CONTRIBUTING.rst.
This project is licensed under the BSD 2-Clause License - see the LICENSE.txt file for details.