Skip to content

feat(codegen): use Maybe for nullable input fields #4178

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Draft
patrick91 wants to merge 2 commits into
base: main
Choose a base branch
from
Draft

feat(codegen): use Maybe for nullable input fields #4178

patrick91 wants to merge 2 commits into from
+188 −10

Conversation

@patrick91
Copy link
Member

@patrick91 patrick91 commented Jan 28, 2026 •
edited by sourcery-ai bot
Loading

Summary

  • schema-codegen now generates nullable input fields with strawberry.Maybe[T | None] instead of just T | None
  • This allows generated input types to be instantiated without providing nullable fields

Fixes #4177

Before

@strawberry.input
class HealthResultInput:
    some_number: int | None  # Required - causes TypeError with {}

After

@strawberry.input
class HealthResultInput:
    some_number: strawberry.Maybe[int | None]  # Optional - works with {}

Why Maybe over UNSET?

Per the Maybe documentation, Maybe is the recommended approach because:

  • Better type safety (proper generic type vs Any)
  • Explicit nullability semantics (Maybe[T] vs Maybe[T | None])
  • Clearer distinction between absent and null values

Test plan

  • Added tests for nullable input field generation
  • Verified generated code can be instantiated without nullable fields
  • Verified Some(value) and Some(None) work correctly
  • All existing codegen tests pass

Summary by Sourcery

Generate nullable GraphQL input fields using strawberry.Maybe to allow omitting them when instantiating generated input types.

New Features:

  • Generate nullable input object fields as strawberry.Maybe[...] instead of plain optional types in schema-codegen output.

Tests:

  • Add schema-codegen tests verifying Maybe usage for nullable input fields, mixed required/nullable inputs, and correct handling of Some(value) and Some(None).

@patrick91
feat(codegen): use Maybe for nullable input fields
68eef41 
schema-codegen now generates nullable input fields with
strawberry.Maybe[T | None] instead of just T | None.

This allows generated input types to be instantiated without
providing nullable fields, fixing the issue where empty inputs
like `{}` would cause TypeError.

Before:
```python
@strawberry.input
class HealthResultInput:
    some_number: int | None  # Required - causes TypeError
```

After:
```python
@strawberry.input
class HealthResultInput:
    some_number: strawberry.Maybe[int | None]  # Optional
```

Benefits of using Maybe over UNSET:
- Better type safety (proper generic type)
- Explicit nullability semantics
- Clearer distinction between absent and null values
@sourcery-ai
Copy link
Contributor

sourcery-ai bot commented Jan 28, 2026 •
edited
Loading

Reviewer's Guide

Update schema code generator so nullable GraphQL input fields are emitted as strawberry.Maybe[...] rather than plain T | None, and add tests to verify generation, instantiation, and Maybe semantics for nullable input fields.

File-Level Changes

Change Details Files
Adjust type generation to wrap nullable input fields in strawberry.Maybe[...] while preserving existing behavior for non-null and output types.
  • Introduce _is_nullable helper to detect NonNullTypeNode-wrapped types
  • Extend _get_field_type to accept a wrap_in_maybe flag, propagate through NonNullTypeNode recursion, and build a NullableExpr of T
 None
  • When wrap_in_maybe is True, wrap the nullable type expression in strawberry.Maybe[...] using a cst.Subscript around the T
    		•
    Expand schema-codegen tests to assert Maybe usage for nullable input fields and to validate runtime behavior of generated inputs.
    • Update existing expected input type code to use strawberry.Maybe[...] for nullable list fields instead of plain list[...]
    		 None
  • Add test to assert nullable input fields in generated code use strawberry.Maybe[int
    		•

    Assessment against linked issues

    Issue Objective Addressed Explanation
    #4177 Make nullable fields in generated @strawberry.input classes truly optional at runtime (constructible from {} without raising missing required keyword-only argument errors). ✅
    #4177 Propagate SDL default values for nullable input fields into the generated Python @strawberry.input definitions as Python defaults. ❌ The PR only changes the type annotation for nullable input fields to use strawberry.Maybe[...] and adds related tests. It does not modify how default values are derived or emitted for fields, nor does it add logic to map SDL default values into Python defaults on the generated input classes.

    Possibly linked issues

    Tips and commands

    Interacting with Sourcery

    • Trigger a new review: Comment @sourcery-ai review on the pull request.
    • Continue discussions: Reply directly to Sourcery's review comments.
    • Generate a GitHub issue from a review comment: Ask Sourcery to create an
      issue from a review comment by replying to it. You can also reply to a
      review comment with @sourcery-ai issue to create an issue from it.
    • Generate a pull request title: Write @sourcery-ai anywhere in the pull
      request title to generate a title at any time. You can also comment
      @sourcery-ai title on the pull request to (re-)generate the title at any time.
    • Generate a pull request summary: Write @sourcery-ai summary anywhere in
      the pull request body to generate a PR summary at any time exactly where you
      want it. You can also comment @sourcery-ai summary on the pull request to
      (re-)generate the summary at any time.
    • Generate reviewer's guide: Comment @sourcery-ai guide on the pull
      request to (re-)generate the reviewer's guide at any time.
    • Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
      pull request to resolve all Sourcery comments. Useful if you've already
      addressed all the comments and don't want to see them anymore.
    • Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
      request to dismiss all existing Sourcery reviews. Especially useful if you
      want to start fresh with a new review - don't forget to comment
      @sourcery-ai review to trigger a new review!

    Customizing Your Experience

    Access your dashboard to:

    • Enable or disable review features such as the Sourcery-generated pull request
      summary, the reviewer's guide, and others.
    • Change the review language.
    • Add, remove or edit custom review instructions.
    • Adjust other review settings.

    Getting Help

    @codspeed-hq
    Copy link

    codspeed-hq bot commented Jan 28, 2026 •
    edited
    Loading

    CodSpeed Performance Report

    Merging this PR will not alter performance

    Comparing patrick91/verify-4177-union-alias (13065d3) with main (358a308)

    Summary

    ✅ 31 untouched benchmarks

    @patrick91 patrick91 mentioned this pull request Jan 28, 2026
    Open
    Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

    Reviewers

    No reviews

    Assignees

    No one assigned

    Labels

    None yet

    Projects

    None yet

    Milestone

    No milestone

    2 participants

    @patrick91