Integrated Pydantic Support
The why and what
As Argo deals with YAML objects, which are actually a subset of json, Pydantic support is almost built-in to Hera
through Pydantic’s serialization (to/from json) features. Using Pydantic objects (instead of dictionaries) in Script
templates makes them less error-prone, and easier to write! Using Pydantic classes yourself is as simple as inheriting
from Pydantic’s BaseModel. Read more about Pydantic models here.
In Hera
Hera offers some features that explicitly rely on Pydantic. This includes the Hera Runner, which uses Pydantic to
validate the function call. Using Pydantic classes in your function parameters unlocks the powerful serializing and
de-serializing features of Pydantic when running on Argo. Your functions can return objects that are serialized, passed
to another Step as a string argument, and then de-serialized in another function. This flow can be seen in
the callable scripts example.
The new experimental Runner IO feature provides a way to specify composite inputs using the class fields, which become the template’s inputs. Read more in the Script Runner IO guide.