Renderer
|
|
|
|
|
|
|
Overview
Definition
A Renderer is a class for converting ExpectationsA verifiable assertion about data., Validation ResultsGenerated when data is Validated against an Expectation or Expectation Suite., etc. into Data DocsHuman readable documentation generated from Great Expectations metadata detailing Expectations, Validation Results, etc. or other output such as email notifications or Slack messages.
Features and promises
Each Expectation has its own Renderer(s) that allows that Expectation to be expressed in Data Docs. From an architectural/contributor standpoint, Expectations are rendered into an intermediate canonical format and then from there turned into docs and other output (think of it more of a rendering pipeline). Likewise, Renderers can be used to send Slack messages and email alerts. Renderers for these purposes are already available in Great Expectations, but you can create custom ones and use them by specifying them in the ActionA Python class with a run method that takes a Validation Result and does something with it that sends the message after ValidationThe act of applying an Expectation Suite to a Batch..
Relationship to other objects
The most common use of Renderers is to render Expectations and Validation Results into output for Data Docs. Renderers can also be used to send Slack messages and email alerts. Great Expectations makes these Renderers available by default. If you create a Custom ExpectationAn extension of the `Expectation` class, developed outside of the Great Expectations library., you may also want to create a custom Renderer for it.
Use cases
Creating Custom Expectations, including their custom Renderers, is a process that falls outside the Great Expectations Validation workflow.
|
|
Create Expectations |
Great Expectations will use a Renderer behind the scenes if you build Data Docs to view your Expectations.
|
|
Validate Data |
If you want to send Slack, email, or other alerts
after a Checkpoint runs you will do so by specifying a
Renderer in an Action in the Checkpoint's
action_list. Renderers will also be used
behind the scenes if you update your Data Docs through
a Checkpoint Action, or if you rebuild your Data Docs
outside the Validation process.
Features
Tailored, human-readable content
Each Expectation comes with a Renderer that is used to create human-readable content for Data Docs. These Renderers are tailored to the Expectation being rendered, ensuring that relevant data and metadata is available in your Data Docs automatically when they are built.
Convenient alerts
The built-in Renderers for email and Slack messages provide a convenient way to alert relevant parties to the result of a Validation run. This can be particularly helpful if your pipeline runs Validations automatically, and you only need to be alerted if an Expectation fails.
API basics
How to access
When looking for Renderers with Expectations, you will
find them specified as part of the Python class
defining the Expectation. With alerts and messaging,
however, you will specify the Renderer to use in an
Action's configuration. To do this, you will
specify a module_name and
class_name indicating the Renderer's
Python class under the renderer key
(found nested under the action key) in an
Action entry in a Checkpoint's
action_list.
Configuration
If you need instructions on how to configure the renderer portion of an Action to send a message or alert, please see the relevant guide: