Add support for extracting nested object array properties from data models

[jstol]

Currently, using Nested[list[SubObject] to automatically extract properties from data models isn't supported – only using Nested[SubObject] within a TypedDict data model type results in automatic marshalling of the property, even though both object and object[] types can be used for nested properties within a collection. This PR adds support for auto-extracting nested object[] properties in addition to nested object properties via the Nested generic wrapper type.

Example

from typing import TypedDict

import weaviate
import weaviate.classes.config as wc
from weaviate.classes.generics import Nested

client = weaviate.connect_to_local()

class Child(TypedDict):
    age: int
    name: str

class Parent(TypedDict):
    age: int
    name: str
    children: Nested[list[Child]]

collection = client.collections.create(
    name="Parent",
    properties=[
        wc.Property(name="age", data_type=wc.DataType.INT),
        wc.Property(name="name", data_type=wc.DataType.TEXT),
        wc.Property(
            name="children", 
            data_type=wc.DataType.OBJECT_ARRAY,
            nested_properties=[
                wc.Property(name="age", data_type=wc.DataType.INT),
                wc.Property(name="name", data_type=wc.DataType.TEXT),
            ]
        )
    ],
    vectorizer_config=wc.Configure.Vectorizer.none(),
    data_model_properties=Parent,
)
    
collection.data.insert({
    "age": 39,
    "name": "Bob Smith",
    "children": [{
        "age": 8,
        "name": "Joe Smith",
    }, {
        "age": 10,
        "name": "Jane Smith",
    }]
})

res = collection.query.fetch_objects()
print(res.objects[0].properties)

Current State

Prior to this change, running the above snippet resulted in an error. (This was due to the list[SubObject] type being passed to typing.get_type_hints within the extraction logic, rather than the underlying SubObject type.)

>>> TypeError: list[Child] is not a module, class, method, or function.

New State

With this change, the inserted data is now fetched and marshalled back to the TypedDict data model.

{'age': 39, 'name': 'Bob Smith', 'children': [{'age': 8, 'name': 'Joe Smith'}, {'age': 10, 'name': 'Jane Smith'}]}

[orca-security-eu]

Orca Security Scan Summary

Status	Check	Issues by priority
Passed	Infrastructure as Code	0   0   0   0	View in Orca
Passed	SAST	0   0   0   0	View in Orca
Passed	Secrets	0   0   0   0	View in Orca
Passed	Vulnerabilities	0   0   0   0	View in Orca

[weaviate-git-bot]

To avoid any confusion in the future about your contribution to Weaviate, we work with a Contributor License Agreement. If you agree, you can simply add a comment to this PR that you agree with the CLA so that we can merge.

beep boop - the Weaviate bot 👋🤖

PS:
Are you already a member of the Weaviate Slack channel?

[jstol]

To avoid any confusion in the future about your contribution to Weaviate, we work with a Contributor License Agreement. If you agree, you can simply add a comment to this PR that you agree with the CLA so that we can merge.

beep boop - the Weaviate bot 👋🤖

PS:
Are you already a member of the Weaviate Slack channel?

I agree with the CLA.

[dirkkul]

Thank you very much :)

[jstol]

@dirkkul thanks for the speedy approval and merge! 🚀