Skip to content

opendatadiscovery/odd-great-expectations

Repository files navigation

OpenDataDiscovery Action for handling Great Expectations tests results.

PyPI version

image

What is it?

odd_great_expectation.action.ODDAction Is a class derived from GX ValidationAction which will be run by GreatExpectations at runtime with ValidationResults for checkpoint.

How to use it?:

Install odd-great-expectations package

pip install odd-great-expectations

Add ODDAction action to some checkpoint's action list:

name: <CHECKPOINT_NAME>
config_version: 1.0
...
action_list:
  # other actions
  - name: store_metadata_to_odd
    action:
      module_name: odd_great_expectations.action
      class_name: ODDAction
      platform_host: <PLATFORM_HOST>
      platform_token: <PLATFORM_TOKEN>  # collector token
      data_source_name: <DATA_SOURCE_NAME>

Parameters:

platform_host - Location of OpenDataDiscovery platform, i.e. http://localhost:8080

platform_token - OpenDataDiscovery token, how to get it - https://docs.opendatadiscovery.org/configuration-and-deployment/trylocally#create-collector-entity

data_source_name - Unique name for data source, i.e. local_qa_test

Both platform_host and platform_token can be set using ODD_PLATFORM_HOST and ODD_PLATFORM_TOKEN env variables accordingly.

Run checkpoint

great_expectations checkpoint run <CHECKPOINT_NAME>

Check results on PLATFORM_HOST UI.

Supporting features

Feature Supporting
V3 API + +
SqlAlchemyEngine +
PandasEngine +
Great Expectations V2 API - -
Cloud Solution -

About

Integration for collecting metadata from Great Expectations

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published