First-class Pydantic v2+ support #3965

patrick91 · 2025-08-02T09:47:38Z

Summary

Add first-class support for Pydantic v2+ models in Strawberry GraphQL.

This PR introduces a new strawberry.pydantic module that allows you to directly decorate Pydantic BaseModel classes to create GraphQL types, inputs, and interfaces without requiring separate wrapper classes.

Basic Usage

import strawberry
from pydantic import BaseModel

@strawberry.pydantic.type
class User(BaseModel):
    name: str
    age: int

@strawberry.pydantic.input
class CreateUserInput(BaseModel):
    name: str
    age: int

@strawberry.pydantic.interface
class Node(BaseModel):
    id: str

Features Implemented

Core Features

✅ @strawberry.pydantic.type - Convert Pydantic models to GraphQL types
✅ @strawberry.pydantic.input - Convert Pydantic models to GraphQL input types
✅ @strawberry.pydantic.interface - Convert Pydantic models to GraphQL interfaces
✅ Automatic field extraction from Pydantic models
✅ Pydantic field descriptions preserved in GraphQL schema
✅ Pydantic field aliases used as GraphQL field names
✅ Support for strawberry.Private to exclude fields from schema
✅ Support for strawberry.field() with Annotated for directives, permissions, deprecation
✅ Generic Pydantic model support
✅ Nested Pydantic types
✅ strawberry.pydantic.Error type for validation error handling
✅ Computed fields via include_computed=True

Pydantic v2 Features (130 Tests Passing)

✅ Functional Validators - BeforeValidator, AfterValidator, WrapValidator with Annotated types
✅ @model_validator Support - Cross-field validation with mode='before'|'after'|'wrap'
✅ Validation Context - Strawberry Info automatically passed to Pydantic validators
✅ model_config Settings - strict=True, extra='forbid', from_attributes=True all respected
✅ Strict Mode Per-Field - Field(strict=True) for individual field validation
✅ Separate Aliases - validation_alias and serialization_alias supported via by_name=True
✅ Discriminated Unions - Literal type support for type discriminators
✅ TypeAdapter - Use TypeAdapter in resolvers for scalar/list validation
✅ RootModel - Use RootModel for validated list/dict wrappers in resolvers

Migration from Experimental

# Before (experimental)
@strawberry.experimental.pydantic.type(model=UserModel, all_fields=True)
class User:
    pass

# After (first-class)
@strawberry.pydantic.type
class User(BaseModel):
    name: str
    age: int

Validation Context Example

Pydantic validators can access GraphQL context for permission-based validation:

from pydantic import field_validator, ValidationInfo

@strawberry.pydantic.input
class CreatePostInput(BaseModel):
    title: str

    @field_validator('title')
    @classmethod
    def check_permissions(cls, v: str, info: ValidationInfo) -> str:
        strawberry_info = info.context.get('info') if info.context else None
        if strawberry_info:
            user = strawberry_info.context.get('user')
            if user and not user.can_create_posts:
                raise ValueError('User cannot create posts')
        return v

Discriminated Union Example

from typing import Literal

@strawberry.pydantic.type
class Cat(BaseModel):
    pet_type: Literal["cat"]
    meow_volume: int

@strawberry.pydantic.type
class Dog(BaseModel):
    pet_type: Literal["dog"]
    bark_volume: int

@strawberry.type
class Query:
    @strawberry.field
    def pet(self) -> Cat | Dog:
        return Cat(pet_type="cat", meow_volume=10)

Test Results

130 Pydantic-specific tests - All passing ✅
Full test suite (4675+ tests) - All passing ✅

🤖 Generated with Claude Code

sourcery-ai · 2025-08-02T09:47:44Z

Reviewer's Guide

This PR introduces first-class Pydantic support by adding a dedicated strawberry/pydantic module with type/input/interface decorators, overhauls the Pydantic integration documentation and migration guide, cleans up experimental code and unused flags, and restructures and extends the test suite to cover all new features.

File-Level Changes

Change	Details	Files
Add first-class Pydantic integration decorators	Create strawberry/pydantic/object_type.py with processing logic and decorators Implement _get_pydantic_fields and conversion utilities in strawberry/pydantic/fields.py Export new pydantic module in strawberry/pydantic/init.py Update strawberry/init.py to include pydantic in the public API	`strawberry/pydantic/object_type.py` `strawberry/pydantic/fields.py` `strawberry/pydantic/__init__.py` `strawberry/__init__.py`
Overhaul Pydantic documentation and remove experimental section	Expand docs/integrations/pydantic.md with installation, usage examples, decorator reference, advanced scenarios, and migration guide Remove outdated experimental examples and flags	`docs/integrations/pydantic.md`
Restructure and extend test suite for Pydantic integration	Move experimental tests into a new tests/pydantic directory and split them into focused modules Add comprehensive tests covering basic types, execution (queries/mutations/async), special features, nested types, and validation error handling	`tests/pydantic/test_basic.py` `tests/pydantic/test_execution.py` `tests/pydantic/test_special_features.py` `tests/pydantic/test_queries_mutations.py` `tests/pydantic/test_nested_types.py`
Cleanup unused experimental code and flags	Remove leftover experimental flags and references from strawberry.experimental.pydantic Delete deprecated _strawberry_input_type usage Tidy up compat module imports	`strawberry/experimental/pydantic/_compat.py`
Update CLAUDE.md and PLAN.md with new integration status	Detail completed implementation steps and migration path in PLAN.md Provide guidance and commands for the repository in CLAUDE.md	`PLAN.md` `CLAUDE.md`

Possibly linked issues

Add pytest action #1: The PR implements first-class Pydantic integration, including new decorators and updated documentation, directly addressing Pydantic V2 support and migration.

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

Contact our support team for questions or feedback.
Visit our documentation for detailed guides and information.
Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

codecov · 2025-08-02T09:51:53Z

Codecov Report

❌ Patch coverage is 56.78818% with 1375 lines in your changes missing coverage. Please review.
✅ Project coverage is 91.27%. Comparing base (c03892c) to head (7067816).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3965      +/-   ##
==========================================
- Coverage   94.41%   91.27%   -3.14%     
==========================================
  Files         536      563      +27     
  Lines       35036    38186    +3150     
  Branches     1842     1913      +71     
==========================================
+ Hits        33079    34856    +1777     
- Misses       1659     3023    +1364     
- Partials      298      307       +9

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

codspeed-hq · 2025-08-02T10:01:03Z

CodSpeed Performance Report

Merging #3965 will not alter performance

_{Comparing feature/pydantic-first-class (7067816) with main (c03892c)}

Summary

✅ 28 untouched

patrick91 · 2025-08-02T12:44:11Z

/pre-release

botberry · 2025-08-02T12:44:45Z

Pre-release

👋

Pre-release 0.279.0.dev.1754159379 [70130b4] has been released on PyPi! 🚀
You can try it by doing:

poetry add strawberry-graphql==0.279.0.dev.1754159379

patrick91 · 2025-08-02T17:36:31Z

/pre-release

patrick91 · 2025-08-02T18:28:34Z

/pre-release

patrick91 · 2025-08-07T09:15:27Z

tests/pydantic/test_error.py

+            try:
+                # Validate the input using Pydantic
+                validated = CreateUserModel(name=input.name, age=input.age)
+                # Simulate successful creation
+                return CreateUserSuccess(
+                    user_id=1, message=f"User {validated.name} created successfully"
+                )
+            except pydantic.ValidationError as e:
+                return Error.from_validation_error(e)


Suggested change

try:

# Validate the input using Pydantic

validated = CreateUserModel(name=input.name, age=input.age)

# Simulate successful creation

return CreateUserSuccess(

user_id=1, message=f"User {validated.name} created successfully"

)

except pydantic.ValidationError as e:

return Error.from_validation_error(e)

# Validate the input using Pydantic

validated = CreateUserModel(name=input.name, age=input.age)

return CreateUserSuccess(

user_id=1, message=f"User {validated.name} created successfully"

)

for more information, see https://pre-commit.ci

patrick91 · 2025-08-19T15:56:33Z

strawberry/pydantic/error.py

+    """Represents a single validation error detail."""
+
+    type: str
+    loc: list[str]


this might return something_else instead of somethingElse

patrick91 · 2025-09-08T11:19:42Z

docs/integrations/pydantic.md

+
+
+@strawberry.pydantic.type
+class User(BaseModel):


Suggested change

class User(BaseModel):

class User(Node):

patrick91 · 2025-09-08T11:21:11Z

docs/integrations/pydantic.md

+    name: str = Field(alias="fullName")
+    age: int = Field(alias="yearsOld")


Suggested change

name: str = Field(alias="fullName")

age: int = Field(alias="yearsOld")

name: Annotated[str, Field(alias="fullName")]

age: Annotated[int, Field(alias="yearsOld")]

potentially also allow strawberry.pydantic.field?

patrick91 · 2025-09-08T11:21:34Z

docs/integrations/pydantic.md

+### Optional Fields
+
+Pydantic optional fields are properly handled:
+
+```python
+from typing import Optional
+
+
+@strawberry.pydantic.type
+class User(BaseModel):
+    name: str
+    email: Optional[str] = None
+    age: Optional[int] = None
+```


Suggested change

### Optional Fields

Pydantic optional fields are properly handled:

```python

from typing import Optional

@strawberry.pydantic.type

class User(BaseModel):

name: str

email: Optional[str] = None

age: Optional[int] = None

```

patrick91 · 2025-09-08T11:22:36Z

docs/integrations/pydantic.md

+        if user.password:
+            return user
+        return None


this is a silly example, let's change it :D

patrick91 · 2025-09-08T11:23:06Z

docs/integrations/pydantic.md

+### Nested Types
+
+Pydantic models can contain other Pydantic models:
+
+```python
+@strawberry.pydantic.type
+class Address(BaseModel):
+    street: str
+    city: str
+    zipcode: str
+
+
+@strawberry.pydantic.type
+class User(BaseModel):
+    name: str
+    address: Address
+```
+
+### Lists and Collections
+
+Lists of Pydantic models work seamlessly:
+
+```python
+from typing import List
+
+
+@strawberry.pydantic.type
+class User(BaseModel):
+    name: str
+    age: int
+
+
+@strawberry.type
+class Query:
+    @strawberry.field
+    def get_users(self) -> List[User]:
+        return [User(name="John", age=30), User(name="Jane", age=25)]
+```


Suggested change

### Nested Types

Pydantic models can contain other Pydantic models:

```python

@strawberry.pydantic.type

class Address(BaseModel):

street: str

city: str

zipcode: str

@strawberry.pydantic.type

class User(BaseModel):

name: str

address: Address

```

### Lists and Collections

Lists of Pydantic models work seamlessly:

```python

from typing import List

@strawberry.pydantic.type

class User(BaseModel):

name: str

age: int

@strawberry.type

class Query:

@strawberry.field

def get_users(self) -> List[User]:

return [User(name="John", age=30), User(name="Jane", age=25)]

```

patrick91 · 2025-09-08T11:24:17Z

docs/integrations/pydantic.md

+    name: str
+    age: int
+
+    @validator("age")


This is Pydantic v1, we should use field_validator

Let's also maybe use number: Annotated[int, AfterValidator(is_even)]

patrick91 · 2025-09-08T11:25:46Z

docs/integrations/pydantic.md

+## Conversion Methods
+
+Decorated models automatically get conversion methods:
+
+```python
+@strawberry.pydantic.type
+class User(BaseModel):
+    name: str
+    age: int
+
+
+# Create from existing Pydantic instance
+pydantic_user = User(name="John", age=30)
+strawberry_user = User.from_pydantic(pydantic_user)
+
+# Convert back to Pydantic
+converted_back = strawberry_user.to_pydantic()
+```


Suggested change

## Conversion Methods

Decorated models automatically get conversion methods:

```python

@strawberry.pydantic.type

class User(BaseModel):

name: str

age: int

# Create from existing Pydantic instance

pydantic_user = User(name="John", age=30)

strawberry_user = User.from_pydantic(pydantic_user)

# Convert back to Pydantic

converted_back = strawberry_user.to_pydantic()

```

we don't even need this

patrick91 · 2025-09-08T11:26:04Z

docs/integrations/pydantic.md

+converted_back = strawberry_user.to_pydantic()
+```
+
+## Migration from Experimental


remove this

patrick91 · 2025-09-08T11:26:56Z

strawberry/pydantic/__init__.py

+from .error import Error
+from .object_type import input as input_decorator
+from .object_type import interface
+from .object_type import type as type_decorator
+
+# Re-export with proper names
+input = input_decorator
+type = type_decorator


make this good

patrick91 · 2025-09-08T11:28:15Z

strawberry/pydantic/error.py

+    """Represents a single validation error detail."""
+
+    type: str
+    loc: list[str]


Suggested change

loc: list[str]

location: list[str]

patrick91 · 2025-09-08T11:28:37Z

strawberry/pydantic/error.py

+
+    type: str
+    loc: list[str]
+    msg: str


Suggested change

msg: str

message: str

patrick91 · 2025-09-08T11:29:07Z

strawberry/pydantic/error.py

+        """
+        return Error(
+            errors=[
+                ErrorDetail(


Suggested change

ErrorDetail(

ErrorDetail.from_error

patrick91 · 2025-09-08T11:31:17Z

strawberry/pydantic/object_type.py

+def _get_interfaces(cls: builtins.type[Any]) -> list[StrawberryObjectDefinition]:
+    """Extract interfaces from a class's inheritance hierarchy."""
+    interfaces: list[StrawberryObjectDefinition] = []
+
+    for base in cls.__mro__[1:]:  # Exclude current class
+        if hasattr(base, "__strawberry_definition__"):
+            type_definition = base.__strawberry_definition__
+            if type_definition.is_interface:
+                interfaces.append(type_definition)
+
+    return interfaces


reuse the one from strawberry/types/object_type.py

patrick91 · 2025-09-08T11:34:14Z

tests/pydantic/test_basic.py

+        age: int
+        password: Optional[str]
+
+    definition: StrawberryObjectDefinition = User.__strawberry_definition__


let's use the util to get the definition

- Functional validators (BeforeValidator, AfterValidator, WrapValidator) - Model validators with mode='before'|'after'|'wrap' - Validation context passing Strawberry Info to validators - model_config settings (strict, extra, from_attributes) - Per-field strict mode via Field(strict=True) - Separate validation/serialization aliases - Discriminated unions with Literal type support - TypeAdapter and RootModel support in resolvers - Updated documentation with Pydantic v2 examples - Added 9 new test files for comprehensive coverage (130 tests) 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>

for more information, see https://pre-commit.ci

patrick91 added 8 commits July 15, 2025 23:44

Initial version implemented by claude

b6cee1e

Private

54c89f2

Remove unused code and update plan

e925206

Remove unused flags

980942d

Split tests and more tests

13f06ef

Lint

cc75f31

Remove _strawberry_input_type

fa73ce5

Merge branch 'main' into feature/pydantic-first-class

bfd506e

Add empty release notes

410cd5a

Add Pydantic Error type

316fca2

strawberry-graphql deleted a comment from botberry Aug 2, 2025

Support for errors

70130b4

patrick91 commented Aug 7, 2025

View reviewed changes

patrick91 and others added 2 commits August 10, 2025 13:09

Wip errord

21ba5f7

[pre-commit.ci] auto fixes from pre-commit.com hooks

b6216a7

for more information, see https://pre-commit.ci

patrick91 commented Aug 19, 2025

View reviewed changes

patrick91 commented Sep 8, 2025

View reviewed changes

docs/integrations/pydantic.md

converted_back = strawberry_user.to_pydantic()

```

## Migration from Experimental

Copy link

Member Author

patrick91 Sep 8, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

remove this

patrick91 commented Sep 8, 2025

View reviewed changes

strawberry/pydantic/error.py

type: str

loc: list[str]

msg: str

Copy link

Member Author

patrick91 Sep 8, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change

msg: str

message: str

patrick91 commented Sep 8, 2025

View reviewed changes

strawberry/pydantic/error.py

"""

return Error(

errors=[

ErrorDetail(

Copy link

Member Author

patrick91 Sep 8, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change

ErrorDetail(

ErrorDetail.from_error

patrick91 commented Sep 8, 2025

View reviewed changes

patrick91 added 8 commits September 8, 2025 16:53

Remove non useful features

3c57bc5

Restructure tests

f5abc5f

Use annotated syntax

60423be

Add more tests

5d8bec9

Add generic tests

d8253a9

Fix tests

82cdfd7

WIP Annotated + strawberry.field

34bfa44

Merge branch 'main' into feature/pydantic-first-class

389e814

patrick91 changed the title ~~WIP~~ First-class Pydantic v2+ support Nov 25, 2025

patrick91 and others added 2 commits November 25, 2025 17:20

[pre-commit.ci] auto fixes from pre-commit.com hooks

7067816

for more information, see https://pre-commit.ci

		name: str = Field(alias="fullName")
		age: int = Field(alias="yearsOld")

Uh oh!

First-class Pydantic v2+ support #3965

Are you sure you want to change the base?

First-class Pydantic v2+ support #3965

Conversation

patrick91 commented Aug 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Basic Usage

Features Implemented

Core Features

Pydantic v2 Features (130 Tests Passing)

Migration from Experimental

Validation Context Example

Discriminated Union Example

Test Results

Uh oh!

sourcery-ai bot commented Aug 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Reviewer's Guide

File-Level Changes

Possibly linked issues

Interacting with Sourcery

Customizing Your Experience

Getting Help

Uh oh!

codecov bot commented Aug 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

codspeed-hq bot commented Aug 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merging #3965 will not alter performance

Summary

Uh oh!

patrick91 commented Aug 2, 2025

Uh oh!

botberry commented Aug 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Pre-release

Uh oh!

patrick91 commented Aug 2, 2025

Uh oh!

patrick91 commented Aug 2, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

patrick91 commented Aug 2, 2025 •

edited

Loading

sourcery-ai bot commented Aug 2, 2025 •

edited

Loading

codecov bot commented Aug 2, 2025 •

edited

Loading

codspeed-hq bot commented Aug 2, 2025 •

edited

Loading

botberry commented Aug 2, 2025 •

edited

Loading