Fix support for optional syntax with 3.9

[benedikt-bartscher]

Summary

Add support for future annotations

What is the feature request for?

The core library

The Problem

Currently future annotations are not supported. All custom type annotations are rendered as strings.

The Ideal Solution

Add a flag to enable future annotations

The Current Solution

No response

[Gobot1234]

I don't see much point in enabling this when PEP-649 is just round the corner

[benedikt-bartscher]

using typing.310 and pydantic_dataclasses generated this:

@dataclass(eq=False, repr=False)
class Message(betterproto.Message):
    selection_request: "SelectionRequest" | None = betterproto.message_field(
        1, optional=True, group="message"
    )
    machine_event: "MachineEvent" | None = betterproto.message_field(
        2, optional=True, group="message"
    )

    @model_validator(mode="after")
    def check_oneof(cls, values):
        return cls._validate_field_groups(values)

pyright

/home/***/gen/hello/v1/__init__.py
  /home/***/gen/hello/v1/__init__.py:41:24 - error: Union syntax cannot be used with string operand; use quotes around entire expression (reportGeneralTypeIssues)
  /home/***/gen/hello/v1/__init__.py:44:20 - error: Union syntax cannot be used with string operand; use quotes around entire expression (reportGeneralTypeIssues)

[Gobot1234]

Ah well that's just a bug with the processing. In the meantime, you can generate the imports differently. See the Configuration typing imports section of README