2.0 KiB
Before diving deeper into the Dependency Injection system, let's upgrade the previous example.
A dict from the previous example
In the previous example, we where returning a dict from our dependency ("dependable"):
{!./tutorial/src/dependencies/tutorial001.py!}
But then we get a dict in the parameter commons of the path operation function.
And we know that dicts can't provide a lot of editor support because they can't know their keys and value types.
Create a Pydantic model
But we are already using Pydantic models in other places and we have already seen all the benefits.
Let's use them here too.
Create a model for the common parameters (and don't pay attention to the rest, for now):
{!./tutorial/src/dependencies/tutorial002.py!}
Return a Pydantic model
Now we can return a Pydantic model from the dependency ("dependable") with the same data as the dict before:
{!./tutorial/src/dependencies/tutorial002.py!}
Declare the Pydantic model
We can now come back to the path operation function and declare the type of the commons parameter to be that Pydantic model:
commons: CommonQueryParams = Depends(common_parameters)
It won't be interpreted as a JSON request Body because we are using Depends:
{!./tutorial/src/dependencies/tutorial002.py!}
!!! info
In the case of dependencies with Depends, the type of the parameter is only to get editor support.
Your dependencies won't be enforced to return a specific type of data.
Use the Pydantic model
And now we can use that model in our code, with all the lovable editor support:
{!./tutorial/src/dependencies/tutorial002.py!}
Recap
By using Pydantic models in your dependencies too you can keep all the editor support that FastAPI is designed to support.