A metadata management framework for CKAN.
This extension has been developed and tested with CKAN version 2.8.2.
Extension ckanext-jsonpatch is required for metadata workflows.
RabbitMQ server is required for Elastic search agent integration:
sudo apt install rabbitmq-server
Activate your CKAN virtual environment:
. /usr/lib/ckan/default/bin/activate
Install the latest development version of ckanext-metadata and its dependencies:
cd /usr/lib/ckan/default
pip install -e 'git+https://github.com/SAEONData/ckanext-metadata.git#egg=ckanext-metadata'
pip install -r src/ckanext-metadata/requirements.txt
In a production environment, you'll probably want to pin a specific release version instead, e.g.:
pip install -e 'git+https://github.com/SAEONData/ckanext-metadata.git@v1.0.0#egg=ckanext-metadata'
Create the required database tables:
cd /usr/lib/ckan/default/src/ckanext-metadata
paster metadata_framework initdb -c /etc/ckan/default/development.ini
Add metadata_framework, jsonpatch, metadata_infrastructure_ui (optional, for infrastructure-type groups to
be configurable in the UI), and metadata_elasticsearch (optional, for Elastic search agent integration) to the
list of plugins in your CKAN configuration file (e.g. /etc/ckan/default/production.ini):
ckan.plugins = ... metadata_framework jsonpatch metadata_infrastructure_ui metadata_elasticsearch
Restart your CKAN instance.
The following configuration options are used by plugins as indicated:
| Plugin | Option | Default | Description |
|---|---|---|---|
| metadata_framework | ckan.metadata.admin_org | The name of the admin organization; applicable to the administrator and curator roles. | |
| metadata_framework | ckan.metadata.admin_role | The name of the administrator role. A user with the admin role - within the admin org - can perform functions like configuring metadata schemas and workflow schemas. | |
| metadata_framework | ckan.metadata.curator_role | The name of the curator role. A user with the curator role - either within the admin org or within the org that owns the resources being requested/updated - can perform functions related to metadata collections and metadata workflow. | |
| metadata_framework | ckan.metadata.harvester_role | The name of the harvester role. A harvester is like a contributor, but can also set the workflow state of a metadata record. | |
| metadata_framework | ckan.metadata.contributor_role | The name of the contributor role. A contributor can create and update metadata records owned by the organization in which they have that role. | |
| metadata_framework | ckan.metadata.convert_nested_ids_to_names | True | If True, object IDs are converted to object names in API output dictionaries. Note: this option must be set to True for metadata framework UI forms to work correctly. |
| metadata_framework | ckan.metadata.doi_prefix | The DOI prefix for auto-generation of DOIs (dependent on metadata collection settings). | |
| metadata_elasticsearch | ckan.metadata.elastic.search_agent_url | The URL of the Elastic Search Agent. |
- RABBITMQ_HOST: The host on which the RabbitMQ server is running.
Restart your CKAN instance after any configuration changes.