Timeseries-based datastore extension

pip install ckanext-timeseries==1.2.0



Timeseries data is streams of data that is indexed by timestamp, such as sensing data. Current default CKAN Datastore plugin only index data based on an auto-incremental integer with no support for timeseries data. Perceived that timeseries data is an important capability when working with sensor network, a new plugin that supports operations based on timestamp natively would be necessary. Thus, the purpose of this repository.

This plugin is fully compatible with current Datastore API <http://docs.ckan.org/en/latest/maintaining/datastore.html#the-datastore-api>, the endpoint is changed from 'datastore_*' to 'datastore_ts_*'. For example, for search query: send request to 'datastore_ts_search' endpoint. Original datastore plugin should still be enabled so that you can view the resources on CKAN web interface.


  • If you are coming from version < v0.1.0, the schema has changed. A command has been created to upgrade the schema. Please run the following command:

` paster --plugin=ckan timeseries -c <path to ini configuration file> upgrade-schema autogen_timestamp _autogen_timestamp `

  • From v1.0.0 the plugin name has changed from "ckanext-datastore_ts" to "ckanext-timeseries", please install ckanext-timeseries as bellow.


Being developed under CKAN 2.6


To install ckanext-timeseries:

  1. Activate your CKAN virtual environment, for example:

    . /usr/lib/ckan/default/bin/activate
  2. Install the ckanext-timeseries Python package into your virtual environment:

    pip install ckanext-timeseries
  3. Add timeseries to the ckan.plugins setting in your CKAN config file (by default the config file is located at /etc/ckan/default/production.ini).

  4. Restart CKAN. For example if you've deployed CKAN with Apache on Ubuntu:

    sudo service apache2 reload

Config Settings

CKAN Timeseries uses configurations from Datastore plugin so make sure those are set. In brief:

sqlalchemy.url = postgresql://ckan_default:pass@localhost/ckan_default

ckan.datastore.write_url = postgresql://ckan_default:password@localhost/datastore_default

ckan.datastore.read_url = postgresql://datastore_default:password@localhost/datastore_default

CKAN Timeseries introduces a new configuration to set the maximum size of a resource table (as we are dealing with real time data). When a resource reaches this limit, it's table will be cleaned, the default 30% of the oldest data will be deleted. This percentage can be customized by user when creating a CKAN Timeseries resource. Please look at the wiki page for more detail.

ckan.timeseries.max_resource_size = 500

The number is observed in MB

Development Installation

To install ckanext-timeseries for development, activate your CKAN virtualenv and do:

git clone https://github.com/namgk/ckan-timeseries.git
cd ckanext-timeseries
python setup.py develop

Running the Tests

To run the tests, do:

nosetests --nologcapture --ckan --with-pylons=test-core.ini ckanext/timeseries/tests/test.. .py

To run the tests and produce a coverage report, first make sure you have coverage installed in your virtualenv (pip install coverage) then run:

nosetests --nologcapture --with-pylons=test.ini --with-coverage --cover-package=ckanext.timeseries --cover-inclusive --cover-erase --cover-tests

Registering ckanext-timeseries on PyPI

ckanext-timeseries should be availabe on PyPI as https://pypi.python.org/pypi/ckanext-timeseries. If that link doesn't work, then you can register the project on PyPI for the first time by following these steps:

  1. (First time only) Create a source distribution of the project:

    python setup.py sdist
  2. (First time only) Register the project:

    python setup.py register
  3. Upload the source distribution to PyPI:

    python setup.py sdist upload
  4. Tag the first release of the project on GitHub with the version number from the setup.py file. For example if the version number in setup.py is 0.0.1 then do:

    git tag 0.0.1
    git push --tags