Skip to main content

PyDataColl is a SCADA-like system, originally developed at GDT.

Project description

PyDataColl
==========

.. image:: https://travis-ci.org/timercrack/pydatacoll.svg
:target: https://travis-ci.org/timercrack/pydatacoll?branch=master

.. image:: https://coveralls.io/repos/timercrack/pydatacoll/badge.svg?branch=master&service=github
:target: https://coveralls.io/github/timercrack/pydatacoll?branch=master

.. image:: https://readthedocs.org/projects/pydatacoll/badge/?version=latest
:target: http://pydatacoll.readthedocs.org/en/latest/?badge=latest
:alt: Documentation Status

.. image:: https://badges.gitter.im/timercrack/pydatacoll.svg
:alt: Join the chat at https://gitter.im/timercrack/pydatacoll
:target: https://gitter.im/timercrack/pydatacoll?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge

PyDataColl is a `SCADA <Ahttps://en.wikipedia.org/wiki/SCADA>`_-like system which use Python as the main language. It
originally inspired by an old program I made as for a core part of a large
`EMS <https://en.wikipedia.org/wiki/Energy_management_system>`_.

PyDataColl can be roughly divided into three parts:

* An APIServer provides `RESTful Services <https://en.wikipedia.org/wiki/Representational_state_transfer>`_
for client to pull/push data from/to devices and perform generic CRUD on devices, terms and items.

* A DeviceManager that manages all devices and terms connected to the system, listens messages send by APIServer
that perform CRUD on devices and terms. It may be combined with some plugins to perform generic operation such as
data checking, database saving and formula calculation.

* Many devices and terms under control of DeviceManager operate with coded data over communication channels(TCP/IP)
so as to provide control of remote equipment(meter or sensor). Each type of Device can communicate with one type of
meter with specify protocol, such as `Modbus TCP <https://en.wikipedia.org/wiki/Modbus>`_,
`IEC 60870-5-104 <https://en.wikipedia.org/wiki/IEC_60870-5#IEC_60870-5-104>`_.

Installation
------------

**Prerequisites**: PyDataColl runs on Python 3.5+. In addition to the requirements showed in requirements.txt,
the following optional packages may be useful:

* `Redis <http://redis.io/>`_ is heavily used by PyDataColl as NoSQL databases and
`IPC <https://en.wikipedia.org/wiki/Inter-process_communication>`_. If you
deploy PyDataColl in local, make sure you have installed and started the Redis server.
* `MySQL <https://www.mysql.com/>`_ is used by DbSaver plugin to store device data in real-time. If you
deploy PyDataColl in local, make sure you have installed and started the MySQL server.
* `ujson <https://pypi.python.org/pypi/ujson>`_ is an ultra fast JSON encoder and decoder written in pure C with
bindings for Python. This is an alternative json library and PyDataColl will use it automatically if possible.
* `hiredis <https://pypi.python.org/pypi/hiredis>`_ is a C library which can provide up to a 10x speed improvement in
parsing responses from the Redis server. This is an alternative redis client and
PyDataColl will use it automatically if possible.

1. Download the latest source code `hosted on github <https://github.com/timercrack/pydatacoll>`_, open a shell
(*CMD* in Windows, *Terminal* in Linux and Mac OS X) and go to the source code directory, install required
package using pip::

pip install -r requirements.txt

2. Then run the following to start PyDataColl server::

python -m pydatacoll.api_server

.. note::
to stop server, press CTRL+C to exit

3. Congratulations! The server is running now. Visit http://localhost:8080 in browser to see the server information, if
success, you will find something like this::

PyDataColl is running, available API:
method: GET URL: http://localhost:8080/
method: GET URL: http://localhost:8080/api/v1/device_protocols
method: GET URL: http://localhost:8080/api/v1/devices
(...more omitted)

4. Now you can send request to server with your favorite http client! check :doc:`restapi` to see the API list.

**Platforms**: PyDataColl should run on any Unix-like platform, although for the best performance and scalability only
Linux (with ``epoll``) and BSD (with ``kqueue``) are recommended for production deployment (even though Mac OS X is
derived from BSD and supports kqueue, its networking performance is generally poor so it is recommended only for
development use). PyDataColl will also run on Windows, although this configuration is not officially supported and is
recommended only for development use.

Quick links
-----------

* `Source (github) <https://github.com/timercrack/pydatacoll>`_
* `Docs <http://pydatacoll.readthedocs.org/>`_

License
-------

``PyDataColl`` is offered under the Apache 2 license.

Project details


Release history Release notifications | RSS feed

This version

0.1

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

PyDataColl-0.1.zip (82.1 kB view details)

Uploaded Source

File details

Details for the file PyDataColl-0.1.zip.

File metadata

  • Download URL: PyDataColl-0.1.zip
  • Upload date:
  • Size: 82.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for PyDataColl-0.1.zip
Algorithm Hash digest
SHA256 1a17af2835ce4e27c1da779a8f6b1eb3ccc33fbab77f9e55473cacac7d3e1f43
MD5 71266f7d37ee0d4701fb5f9bcac775aa
BLAKE2b-256 de569c8d7bc956b2f661a36763377a57be6106ec3d05655eb80f3ae65548bbc3

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page