ckanext-twitter

A CKAN extension that enables users to post a tweet every time a dataset is created or updated.


Keywords
CKAN, data, twitter, ckan-extension
License
CNRI-Python-GPL-Compatible
Install
pip install ckanext-twitter==2.2.1

Documentation

Warning

This extension is now deprecated.

Due to the API changes Twitter/X introduced in 2023, this extension no longer works and will not be fixed. There will not be any further releases, new features, or bugfixes.

ckanext-twitter

Tests Coveralls CKAN Python Docs

A CKAN extension that enables users to post a tweet every time a dataset is created or updated.

Overview

This extension connects a CKAN instance to a Twitter account so that when a dataset is updated or created (i.e. the after_update hook is called), the user has the option to post a tweet about the activity.

Installation

Path variables used below:

  • $INSTALL_FOLDER (i.e. where CKAN is installed), e.g. /usr/lib/ckan/default
  • $CONFIG_FILE, e.g. /etc/ckan/default/development.ini

Installing from PyPI

pip install ckanext-twitter

Installing from source

  1. Clone the repository into the src folder:

    cd $INSTALL_FOLDER/src
    git clone https://github.com/NaturalHistoryMuseum/ckanext-twitter.git
  2. Activate the virtual env:

    . $INSTALL_FOLDER/bin/activate
  3. Install via pip:

    pip install $INSTALL_FOLDER/src/ckanext-twitter

Installing in editable mode

Installing from a pyproject.toml in editable mode (i.e. pip install -e) requires setuptools>=64; however, CKAN 2.9 requires setuptools==44.1.0. See our CKAN fork for a version of v2.9 that uses an updated setuptools if this functionality is something you need.

Post-install setup

  1. Add 'twitter' to the list of plugins in your $CONFIG_FILE:

    ckan.plugins = ... twitter
  2. Add a 'tweet' block to read_base.html:

    {% block tweet %}
    {{ super() }}
    {% endblock %}
  3. Optionally, override the styling of the block by creating an ajax_snippets/edit_tweet.html file.

Configuration

These are the options that can be specified in your .ini config file. The only required options are the twitter credentials. Everything else has a sensible default set.

[REQUIRED]

Name Description Options
ckanext.twitter.consumer_key Your Twitter consumer key
ckanext.twitter.consumer_secret Your Twitter consumer secret
ckanext.twitter.token_key Your Twitter token key
ckanext.twitter.token_secret Your Twitter token secret

All of these can be obtained by creating a single-user app at apps.twitter.com. They can be found on the "keys and access tokens" tab when viewing your app.

Tweet Templates

Tweets are generated using Jinja2 and use tokens derived from the package dictionary. See Usage for more detail.

Name Description Default
ckanext.twitter.new Template for tweets about new datasets New dataset: "{{ title }}" by {{ author }} ({%- if records != 0 -%} {{ records }} records {%- else -%} {{ resources }} resource {%- endif -%}).
ckanext.twitter.updated Template for tweets about updated datasets Updated dataset: "{{ title }}" by {{ author }} ({%- if records != 0 -%} {{ records }} records {%- else -%} {{ resources }} resource {%- endif -%}).

If your config is created dynamically using Jinja2, you will have to wrap any custom template in {% raw %}{% endraw %} tags and add a newline after it, e.g.:

ckanext.twitter.new = {% raw %}{{ title }} by {{ author }} ({{ records }} records) has just been published!{% endraw %}

ckanext.twitter.consumer_key = {{ twitter.consumer_key }}
ckanext.twitter...

Other options

Name Description Options Default
ckanext.twitter.debug Is in debug mode; overrides global debug flag if specified True, False False
ckanext.twitter.hours_between_tweets Number of hours between tweets about the same dataset (to prevent spamming) 24
ckanext.twitter.disable_edit If true, users will not be able to edit the tweet about their dataset before it is posted (though they can still decide not to post it) True, False False

Usage

Tweet Templates

Token values for the tweet templates will come from a simplified package dictionary. In these, any collection values (i.e. lists and dictionaries) have been replaced with the number of items, the author list has been significantly shortened, and any long strings will be shortened to fit into the tweet character limit (currently set at 140).

For example, if the package dictionary is:

{
  'author': 'Dippy Diplodocus, Sophie Stegosaurus',
  'author_email': None,
  'dataset_category': ['Citizen Science'],
  'doi': 'DOI_VALUE',
  'license_title': 'Creative Commons Attribution',
  'organization': {'description': '', 'name': 'nhm', 'is_organization': True, 'state': 'active'},
  'resources': [
                  {'mimetype': 'image/jpeg', 'name': 'resource_1.jpg', 'format': 'JPEG'},
                  {'mimetype': 'image/jpeg', 'name': 'resource_2.jpg', 'format': 'JPEG'}
               ],
  'title': 'Dataset Name'
}

Then the tokenised dictionary would be:

{
  'author': 'Diplodocus et al.',  # just the surname of the first author
  # author_email was None so it's excluded
  'dataset_category': 1,  # lists are counted
  'doi': 'DOI_VALUE',  # simple string values stay the same
  'license_title': 'Creative Commons Attribution',
  'organization': 4,  # dicts are also counted
  'resources': 2,
  'title': 'Dataset Name'
}

And if you had defined the tweet template as:

New dataset: "{{ title }}" by {{ author }} ({{ resources }} resources).

Your tweet would then read:

New dataset: "Dataset Name" by Diplodocus et al. (2 resources)

Testing

There is a Docker compose configuration available in this repository to make it easier to run tests. The ckan image uses the Dockerfile in the docker/ folder.

To run the tests against ckan 2.9.x on Python3:

  1. Build the required images:

    docker-compose build
  2. Then run the tests. The root of the repository is mounted into the ckan container as a volume by the Docker compose configuration, so you should only need to rebuild the ckan image if you change the extension's dependencies.

    docker-compose run ckan

Note that the tests shouldn't make any calls to Twitter's API and won't post any tweets.