Field Alias

Alias

Field aliases can be defined explicitly in the model definition in the same way as in regular pydantic.

Terms

Any defined terms are still respected.

SimpleWith Terms

from pydantic_jsonld.compact import BaseModel, Field, Value


class ExampleModel(BaseModel):
    value: Value[int] = Field(..., alias="myValue")


data = {
    "@context": {"@vocab": "http://www.example.org/"},
    "myValue": 1,
}

model = ExampleModel.model_validate(data)
dumped = model.model_dump(by_alias=True, exclude_none=True)

assert dumped == data

from pydantic_jsonld.compact import BaseModel, Field, Value


class ExampleModel(BaseModel):
    value: Value[int] = Field(..., term="ex", alias="myValue")


data = {
    "@context": {"ex": "http://www.example.org/"},
    "ex:myValue": 1,
}

model = ExampleModel.model_validate(data)
dumped = model.model_dump(by_alias=True, exclude_none=True)

assert dumped == data

Tip

Validation and serialization aliases are set to {term}:{alias} if a term is provided.

Validation / Serialization Alias

Similarly, validation as well as serialization aliases can be defined explicitly in the model definition in the same way as in regular pydantic.

Terms

This causes any defined terms to be ignored and may prevent valid JSON-LD data from being parsed correctly.

ValidationValidation with TermsSerializationSerialization with Terms

from pydantic_jsonld.compact import BaseModel, Field, Value


class ExampleModel(BaseModel):
    value: Value[int] = Field(..., validation_alias="myValue")


data = {
    "@context": {"@vocab": "http://www.example.org/"},
    "myValue": 1,
}

ExampleModel.model_validate(data)

from pydantic_jsonld.compact import BaseModel, Field, Value


class ExampleModel(BaseModel):
    value: Value[int] = Field(..., term="ex", validation_alias="myValue")


# Not a valid JSON-LD
data = {
    "@context": {"ex": "http://www.example.org/"},
    "myValue": 1,
}

# The defined term is ignored
ExampleModel.model_validate(data)

from pydantic_jsonld.compact import BaseModel, Field, Value


class ExampleModel(BaseModel):
    value: Value[int] = Field(..., serialization_alias="myValue")


model = ExampleModel(
    context={"@vocab": "http://www.example.org/"},
    value=1,
)
dumped = model.model_dump(by_alias=True, exclude_none=True) 

assert dumped["myValue"] == 1

from pydantic_jsonld.compact import BaseModel, Field, Value


class ExampleModel(BaseModel):
    value: Value[int] = Field(..., term="ex", serialization_alias="myValue")


model = ExampleModel(
    context={"ex": "http://www.example.org/"},
    value=1,
)
# Not a valid JSON-LD
dumped = model.model_dump(by_alias=True, exclude_none=True) 

# The defined term is ignored
assert dumped["myValue"] == 1

Alias Generator

By default BaseModel uses the to_camel alias generator from pydantic. Different alias generators can be defined in the same way as in pydantic.

Context Alias

The only exception to this is the context property which always has alias @context regardless of the chosen alias generator.

No GeneratorPascal GeneratorAdvanced

from pydantic import ConfigDict
from pydantic_jsonld.compact import BaseModel, Value


class ExampleModel(BaseModel):
    model_config = ConfigDict(alias_generator=None)        

    my_value: Value[int]

data = {
    "@context": {"@vocab": "http://www.example.org/"},
    "my_value": 1,
}

model = ExampleModel.model_validate(data)
dumped = model.model_dump(by_alias=True, exclude_none=True)

assert dumped == data

from pydantic import ConfigDict
from pydantic.alias_generators import to_pascal
from pydantic_jsonld.compact import BaseModel, Value


class ExampleModel(BaseModel):
    model_config = ConfigDict(alias_generator=to_pascal)        

    my_value: Value[int]

data = {
    "@context": {"@vocab": "http://www.example.org/"},
    "MyValue": 1,
}

model = ExampleModel.model_validate(data)
dumped = model.model_dump(by_alias=True, exclude_none=True)

assert dumped == data

from pydantic import AliasGenerator, ConfigDict
from pydantic.alias_generators import to_camel, to_pascal
from pydantic_jsonld.compact import BaseModel, Value


class ExampleModel(BaseModel):
    model_config = ConfigDict(
        alias_generator=AliasGenerator(
            validation_alias=to_camel,
            serialization_alias=to_pascal,
        )
    )        

    my_value: Value[int]

data = {
    "@context": {"@vocab": "http://www.example.org/"},
    "myValue": 1,
}

model = ExampleModel.model_validate(data)
dumped = model.model_dump(by_alias=True, exclude_none=True)

assert dumped["MyValue"] == data["myValue"]

Terms

Unlike the case when setting explicit validation and serialization aliases, for the last example, terms are not ignored.