mypy

mypy - Python 用静的型チェッカー

概要

mypy は Python の標準的な静的型チェッカーです。型ヒント（PEP 484）による段階的な型付けと、包括的な型安全性を実現します。実行前に型エラーを検出し、コードのドキュメント性を高め、IDE のサポートを強化しながら、段階的な導入によって Python の動的な性質を維持します。

主な機能：

• 段階的な型付け：既存のコードに型を少しずつ追加できます
• 厳格モード：--strict フラグで最大限の型安全性を実現します
• 型推論：コンテキストから型を自動的に推論します
• プロトコルのサポート：構造的部分型付け（型によるダックタイピング）
• ジェネリック型：TypeVar、Generic、高度な型パターン
• フレームワークとの統合：FastAPI、Django、Pydantic に対応
• プラグインシステム：ライブラリ向けに型チェックを拡張できます
• 差分チェック：大規模なコードベースでも高速に型チェックできます

インストール：

# Basic mypy
pip install mypy

# With common type stubs
pip install mypy types-requests types-PyYAML types-redis

# For FastAPI projects
pip install mypy pydantic

# For Django projects
pip install mypy django-stubs

# Development setup
pip install mypy pre-commit

型アノテーションの基本

1. 変数の型ヒント

# Basic types
name: str = "Alice"
age: int = 30
height: float = 5.9
is_active: bool = True

# Type inference (mypy infers types)
count = 10  # mypy infers: int
message = "Hello"  # mypy infers: str

# Multiple types with Union
from typing import Union

user_id: Union[int, str] = 123  # Can be int OR str
result: Union[int, None] = None  # Nullable int

# Optional (shorthand for Union[T, None])
from typing import Optional

user_email: Optional[str] = None  # Can be str or None

2. 関数の型ヒント

# Basic function typing
def greet(name: str) -> str:
    return f"Hello, {name}"

# Multiple parameters
def add(a: int, b: int) -> int:
    return a + b

# Optional parameters with defaults
def create_user(name: str, age: int = 18) -> dict:
    return {"name": name, "age": age}

# No return value
def log_message(message: str) -> None:
    print(message)

# Functions that never return
from typing import NoReturn

def raise_error() -> NoReturn:
    raise ValueError("Always raises")

3. コレクションの型ヒント

from typing import List, Dict, Set, Tuple

# List with element type
numbers: List[int] = [1, 2, 3, 4]
names: List[str] = ["Alice", "Bob", "Charlie"]

# Dict with key and value types
user_ages: Dict[str, int] = {"Alice": 30, "Bob": 25}
config: Dict[str, Union[str, int]] = {"host": "localhost", "port": 8000}

# Set with element type
unique_ids: Set[int] = {1, 2, 3}

# Tuple with fixed types
coordinate: Tuple[float, float] = (10.5, 20.3)
user_record: Tuple[int, str, bool] = (1, "Alice", True)

# Variable-length tuple
numbers: Tuple[int, ...] = (1, 2, 3, 4, 5)

# Modern syntax (Python 3.9+)
numbers: list[int] = [1, 2, 3]
user_ages: dict[str, int] = {"Alice": 30}

4. クラスの型ヒント

class User:
    # Class attributes
    name: str
    age: int
    email: Optional[str]

    def __init__(self, name: str, age: int, email: Optional[str] = None) -> None:
        self.name = name
        self.age = age
        self.email = email

    def get_info(self) -> Dict[str, Union[str, int]]:
        return {
            "name": self.name,
            "age": self.age,
            "email": self.email or "N/A"
        }

    @classmethod
    def from_dict(cls, data: Dict[str, any]) -> "User":
        return cls(
            name=data["name"],
            age=data["age"],
            email=data.get("email")
        )

高度な型ヒント

1. リテラル型

from typing import Literal

# Restrict to specific values
def set_log_level(level: Literal["debug", "info", "warning", "error"]) -> None:
    print(f"Log level: {level}")

# Valid
set_log_level("debug")
set_log_level("error")

# Type error: Argument 1 has incompatible type "verbose"
set_log_level("verbose")

# Multiple literals
Status = Literal["pending", "approved", "rejected"]

def update_status(status: Status) -> None:
    pass

2. 型エイリアス

from typing import Dict, List, Union

# Simple alias
UserId = int
UserName = str

def get_user(user_id: UserId) -> UserName:
    return f"User {user_id}"

# Complex aliases
JSON = Union[Dict[str, "JSON"], List["JSON"], str, int, float, bool, None]
Headers = Dict[str, str]
QueryParams = Dict[str, Union[str, int, List[str]]]

def make_request(
    url: str,
    headers: Headers,
    params: QueryParams
) -> JSON:
    pass

# NewType for distinct types
from typing import NewType

UserId = NewType("UserId", int)
ProductId = NewType("ProductId", int)

def get_user(user_id: UserId) -> str:
    return f"User {user_id}"

user = UserId(123)  # Valid
product = ProductId(456)

get_user(user)  # Valid
get_user(product)  # Type error: ProductId not compatible with UserId

3. ジェネリクスと TypeVar

from typing import TypeVar, Generic, List

# TypeVar for generic functions
T = TypeVar("T")

def first_element(items: List[T]) -> T:
    return items[0]

# Type inference
num = first_element([1, 2, 3])  # mypy infers: int
name = first_element(["Alice", "Bob"])  # mypy infers: str

# Bounded TypeVar
from typing import Union

NumericType = TypeVar("NumericType", int, float)

def add_numbers(a: NumericType, b: NumericType) -> NumericType:
    return a + b

# Generic classes
class Stack(Generic[T]):
    def __init__(self) -> None:
        self._items: List[T] = []

    def push(self, item: T) -> None:
        self._items.append(item)

    def pop(self) -> T:
        return self._items.pop()

    def is_empty(self) -> bool:
        return len(self._items) == 0

# Usage with type inference
int_stack: Stack[int] = Stack()
int_stack.push(42)
int_stack.push("hello")  # Type error: Expected int, got str

str_stack: Stack[str] = Stack()
str_stack.push("hello")  # Valid

4. Protocol（構造的部分型付け）

from typing import Protocol

# Define protocol (interface)
class Drawable(Protocol):
    def draw(self) -> str:
        ...

# Any class with draw() method matches
class Circle:
    def draw(self) -> str:
        return "Drawing circle"

class Square:
    def draw(self) -> str:
        return "Drawing square"

# Function accepts any Drawable
def render(obj: Drawable) -> str:
    return obj.draw()

# Both work (duck typing with types)
circle = Circle()
square = Square()
render(circle)  # Valid
render(square)  # Valid

# Runtime checkable protocols
from typing import runtime_checkable

@runtime_checkable
class Closeable(Protocol):
    def close(self) -> None:
        ...

class File:
    def close(self) -> None:
        pass

# Runtime check
f = File()
isinstance(f, Closeable)  # True

5. Callable 型

from typing import Callable

# Function that takes another function
def apply_twice(func: Callable[[int], int], value: int) -> int:
    return func(func(value))

def double(x: int) -> int:
    return x * 2

result = apply_twice(double, 5)  # Returns 20

# Generic callable
from typing import TypeVar

T = TypeVar("T")
R = TypeVar("R")

def map_values(
    func: Callable[[T], R],
    values: List[T]
) -> List[R]:
    return [func(v) for v in values]

# Callable with multiple arguments
Validator = Callable[[str, int], bool]

def validate_user(name: str, age: int) -> bool:
    return len(name) > 0 and age >= 0

validator: Validator = validate_user

mypy の設定

1. mypy.ini の設定

# mypy.ini
[mypy]
# Python version
python_version = 3.11

# Import discovery
files = src,tests
exclude = build,dist,venv

# Type checking strictness
disallow_untyped_defs = True
disallow_any_unimported = False
no_implicit_optional = True
warn_return_any = True
warn_unused_ignores = True
warn_redundant_casts = True

# Error reporting
show_error_codes = True
show_column_numbers = True
pretty = True

# Incremental type checking
incremental = True
cache_dir = .mypy_cache

# Per-module configuration
[mypy-tests.*]
disallow_untyped_defs = False

[mypy-migrations.*]
ignore_errors = True

# Third-party libraries without stubs
[mypy-redis.*]
ignore_missing_imports = True

[mypy-celery.*]
ignore_missing_imports = True

2. pyproject.toml の設定

# pyproject.toml
[tool.mypy]
python_version = "3.11"
files = ["src", "tests"]
exclude = ["build", "dist", "venv"]

# Strictness
disallow_untyped_defs = true
disallow_any_unimported = false
no_implicit_optional = true
warn_return_any = true
warn_unused_ignores = true
warn_redundant_casts = true
strict_equality = true
strict_concatenate = true

# Error reporting
show_error_codes = true
show_column_numbers = true
pretty = true
color_output = true

# Incremental
incremental = true
cache_dir = ".mypy_cache"

# Per-module overrides
[[tool.mypy.overrides]]
module = "tests.*"
disallow_untyped_defs = false

[[tool.mypy.overrides]]
module = ["redis.*", "celery.*"]
ignore_missing_imports = true

3. 厳格モード

# Enable all strict checks
mypy --strict src/

# Strict mode equivalent flags
mypy \
  --disallow-any-unimported \
  --disallow-any-expr \
  --disallow-any-decorated \
  --disallow-any-explicit \
  --disallow-any-generics \
  --disallow-subclassing-any \
  --disallow-untyped-calls \
  --disallow-untyped-defs \
  --disallow-incomplete-defs \
  --check-untyped-defs \
  --disallow-untyped-decorators \
  --no-implicit-optional \
  --warn-redundant-casts \
  --warn-unused-ignores \
  --warn-return-any \
  --warn-unreachable \
  --strict-equality \
  src/

# mypy.ini strict configuration
[mypy]
strict = True

# Relax specific checks if needed
disallow_any_expr = False  # Too strict for most projects
disallow_any_explicit = False  # Allow explicit Any

段階的な導入方法

1. エントリーポイントから始める

# Start typing from main.py (top-level)
# main.py
from typing import Optional
from app.services import UserService

def main(config_path: Optional[str] = None) -> None:
    """Application entry point."""
    service = UserService()
    service.run()

if __name__ == "__main__":
    main()

# Check only main.py initially
mypy main.py

# Gradually expand scope
mypy main.py app/services.py
mypy src/

2. モジュールごとの厳格モード

# mypy.ini - Gradually enable strict checking
[mypy]
# Lenient global defaults
ignore_missing_imports = True
disallow_untyped_defs = False

# Strict for new modules
[mypy-app.services.user_service]
disallow_untyped_defs = True
warn_return_any = True

[mypy-app.api.*]
disallow_untyped_defs = True
no_implicit_optional = True

# Still lenient for legacy code
[mypy-app.legacy.*]
ignore_errors = True

3. # type: ignore を適切に使う

# Suppress specific errors during migration
import legacy_module  # type: ignore[import]

def process_data(data):  # type: ignore[no-untyped-def]
    # TODO: Add type hints
    return data.transform()

# Ignore specific error codes
user_dict = get_user_dict()
user_id = user_dict["id"]  # type: ignore[index]

# Ignore entire line (use sparingly)
result = external_api.call()  # type: ignore

4. 開発中に型を表示する

from typing import reveal_type

def process_user(user_id: int):
    user = get_user(user_id)
    reveal_type(user)  # mypy will show inferred type

    name = user.name
    reveal_type(name)  # mypy will show: str

FastAPI との統合

1. 型ヒントを使った FastAPI

# main.py
from fastapi import FastAPI, HTTPException
from pydantic import BaseModel
from typing import List, Optional

app = FastAPI()

# Pydantic models (auto-validated)
class User(BaseModel):
    id: int
    name: str
    email: str
    age: Optional[int] = None

class UserCreate(BaseModel):
    name: str
    email: str
    age: Optional[int] = None

# Type-safe endpoints
@app.get("/")
def read_root() -> dict[str, str]:
    return {"message": "Hello World"}

@app.get("/users/{user_id}")
def read_user(user_id: int) -> User:
    if user_id == 0:
        raise HTTPException(status_code=404, detail="User not found")
    return User(id=user_id, name=f"User {user_id}", email="[email protected]")

@app.get("/users")
def list_users(skip: int = 0, limit: int = 10) -> List[User]:
    users = [
        User(id=i, name=f"User {i}", email=f"user{i}@example.com")
        for i in range(skip, skip + limit)
    ]
    return users

@app.post("/users")
def create_user(user: UserCreate) -> User:
    # Pydantic ensures type safety
    return User(id=1, name=user.name, email=user.email, age=user.age)

2. 非同期 FastAPI の型チェック

from typing import List, Optional
from fastapi import FastAPI, Depends
from sqlalchemy.ext.asyncio import AsyncSession

app = FastAPI()

# Async dependency
async def get_db() -> AsyncSession:
    async with AsyncSessionLocal() as session:
        yield session

# Async endpoints with types
@app.get("/users/{user_id}")
async def read_user(
    user_id: int,
    db: AsyncSession = Depends(get_db)
) -> User:
    user = await db.get(User, user_id)
    if user is None:
        raise HTTPException(status_code=404, detail="User not found")
    return user

@app.get("/users")
async def list_users(
    skip: int = 0,
    limit: int = 10,
    db: AsyncSession = Depends(get_db)
) -> List[User]:
    result = await db.execute(
        select(User).offset(skip).limit(limit)
    )
    return result.scalars().all()

3. FastAPI の依存性注入の型

from typing import Annotated, Optional
from fastapi import FastAPI, Depends, Header, HTTPException

app = FastAPI()

# Typed dependencies
async def get_current_user(
    authorization: Annotated[Optional[str], Header()] = None
) -> User:
    if authorization is None:
        raise HTTPException(status_code=401, detail="Not authenticated")
    # Verify token and return user
    return User(id=1, name="Current User", email="[email protected]")

# Use dependency with type annotation
@app.get("/me")
async def read_current_user(
    current_user: Annotated[User, Depends(get_current_user)]
) -> User:
    return current_user

# Complex dependency chain
class UserService:
    def __init__(self, db: AsyncSession) -> None:
        self.db = db

    async def get_user(self, user_id: int) -> Optional[User]:
        return await self.db.get(User, user_id)

def get_user_service(
    db: Annotated[AsyncSession, Depends(get_db)]
) -> UserService:
    return UserService(db)

@app.get("/users/{user_id}")
async def get_user_endpoint(
    user_id: int,
    service: Annotated[UserService, Depends(get_user_service)]
) -> User:
    user = await service.get_user(user_id)
    if user is None:
        raise HTTPException(status_code=404, detail="User not found")
    return user

Django との統合

1. django-stubs を使った Django

# Install django-stubs
pip install django-stubs mypy

# Generate mypy configuration
python -m mypy --install-types

# mypy.ini
[mypy]
plugins = mypy_django_plugin.main

[mypy.plugins.django-stubs]
django_settings_module = "myproject.settings"

2. 型ヒントを使った Django モデル

# models.py
from django.db import models
from typing import Optional

class User(models.Model):
    email: models.EmailField = models.EmailField(unique=True)
    name: models.CharField = models.CharField(max_length=100)
    age: models.IntegerField = models.IntegerField(null=True, blank=True)
    is_active: models.BooleanField = models.BooleanField(default=True)
    created_at: models.DateTimeField = models.DateTimeField(auto_now_add=True)

    def get_display_name(self) -> str:
        return self.name or self.email

    @classmethod
    def get_active_users(cls) -> models.QuerySet["User"]:
        return cls.objects.filter(is_active=True)

3. 型ヒントを使った Django ビュー

# views.py
from django.http import HttpRequest, HttpResponse, JsonResponse
from django.shortcuts import get_object_or_404
from typing import Dict, Any
from .models import User

def user_detail(request: HttpRequest, user_id: int) -> JsonResponse:
    user: User = get_object_or_404(User, pk=user_id)
    data: Dict[str, Any] = {
        "id": user.id,
        "name": user.name,
        "email": user.email,
    }
    return JsonResponse(data)

def user_list(request: HttpRequest) -> JsonResponse:
    users = User.get_active_users()
    data = {
        "users": list(users.values("id", "name", "email"))
    }
    return JsonResponse(data)

型スタブとサードパーティライブラリ

1. 型スタブのインストール

# Install stubs for popular libraries
pip install types-requests
pip install types-PyYAML
pip install types-redis
pip install types-boto3

# Search for available stubs
pip search types-

# Auto-install missing stubs
mypy --install-types

2. カスタムスタブの作成

# stubs/external_lib.pyi
from typing import Optional, List

class Client:
    def __init__(self, api_key: str) -> None: ...

    def get_user(self, user_id: int) -> Optional[dict]: ...

    def list_users(self, limit: int = 10) -> List[dict]: ...

def connect(host: str, port: int) -> Client: ...

# mypy.ini
[mypy]
mypy_path = stubs

3. インポート不足を無視する

# mypy.ini
[mypy-external_lib.*]
ignore_missing_imports = True

# For multiple libraries
[mypy-celery.*,redis.*,boto3.*]
ignore_missing_imports = True

CI/CD との統合

1. GitHub Actions

# .github/workflows/type-check.yml
name: Type Check

on: [push, pull_request]

jobs:
  mypy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3

      - name: Set up Python
        uses: actions/setup-python@v4
        with:
          python-version: '3.11'

      - name: Install dependencies
        run: |
          pip install mypy
          pip install -r requirements.txt
          pip install types-requests types-PyYAML

      - name: Run mypy
        run: mypy src/

      - name: Run mypy strict on new code
        run: mypy --strict src/api/

2. pre-commit フック

# .pre-commit-config.yaml
repos:
  - repo: https://github.com/pre-commit/mirrors-mypy
    rev: v1.8.0
    hooks:
      - id: mypy
        args: [--strict, --ignore-missing-imports]
        additional_dependencies:
          - types-requests
          - types-PyYAML

# Install pre-commit
pip install pre-commit
pre-commit install

# Run manually
pre-commit run mypy --all-files

3. Make ターゲット

# Makefile
.PHONY: typecheck
typecheck:
	mypy src/

.PHONY: typecheck-strict
typecheck-strict:
	mypy --strict src/

.PHONY: typecheck-report
typecheck-report:
	mypy src/ --html-report mypy-report
	@echo "Report: mypy-report/index.html"

.PHONY: ci
ci: typecheck test lint

よく使われるパターンと慣用表現

1. 複数のシグネチャに対するオーバーロード

from typing import overload, Union

@overload
def process(data: str) -> str: ...

@overload
def process(data: int) -> int: ...

@overload
def process(data: list) -> list: ...

def process(data: Union[str, int, list]) -> Union[str, int, list]:
    """Process different data types."""
    if isinstance(data, str):
        return data.upper()
    elif isinstance(data, int):
        return data * 2
    else:
        return [x * 2 for x in data]

# mypy knows return types
result1: str = process("hello")  # Valid
result2: int = process(42)  # Valid
result3: str = process(42)  # Type error

2. 構造化された辞書に対する TypedDict

from typing import TypedDict, Optional

class UserDict(TypedDict):
    id: int
    name: str
    email: str
    age: Optional[int]

# Type-safe dict usage
def create_user(data: UserDict) -> UserDict:
    return {
        "id": 1,
        "name": data["name"],
        "email": data["email"],
        "age": data.get("age"),
    }

user: UserDict = {
    "id": 1,
    "name": "Alice",
    "email": "[email protected]",
    "age": 30
}

# Type error: Missing required key "email"
invalid_user: UserDict = {
    "id": 1,
    "name": "Bob",
}

3. Final と定数値

from typing import Final

# Constants that should never change
API_VERSION: Final = "v1"
MAX_RETRIES: Final[int] = 3

# Type error: Cannot assign to final name
API_VERSION = "v2"

# Final class (cannot be subclassed)
from typing import final

@final
class BaseConfig:
    pass

# Type error: Cannot inherit from final class
class AppConfig(BaseConfig):  # Error!
    pass

4. メソッドチェーンのための Self 型

from typing import Self  # Python 3.11+

class Builder:
    def __init__(self) -> None:
        self._value = 0

    def add(self, value: int) -> Self:
        self._value += value
        return self

    def multiply(self, value: int) -> Self:
        self._value *= value
        return self

    def build(self) -> int:
        return self._value

# Type-safe method chaining
result = Builder().add(5).multiply(2).add(3).build()

mypy と pyright の比較

機能の比較

機能	mypy	pyright
型チェッカー	Python の公式型チェッカー	Microsoft の型チェッカー
速度	大規模なコードベースでは低速	高速、差分チェックに対応
厳格さ	設定可能な厳格モード	デフォルトで非常に厳格
IDE との統合	良好（LSP をサポート）	優秀（VS Code の Pylance）
プラグインシステム	あり（mypy プラグイン）	限定的
エラーメッセージ	明確で詳細	非常に詳細で役立つ
コミュニティ	大規模、公式	拡大中、Microsoft の支援あり
型推論	良好	優秀
設定	mypy.ini、pyproject.toml	pyrightconfig.json、pyproject.toml

mypy を使う場合

# Use mypy for:
# - Official Python type checking standard
# - Plugin ecosystem (Django, SQLAlchemy, Pydantic)
# - Gradual typing with fine-grained control
# - Compatibility with existing mypy configurations
# - CI/CD pipelines (industry standard)

pyright を使う場合

# Use pyright for:
# - VS Code development (Pylance)
# - Faster type checking on large codebases
# - Stricter type checking by default
# - Better type inference
# - Real-time IDE feedback

両方を使う場合

# pyproject.toml - Configure both
[tool.mypy]
strict = true
files = ["src"]

[tool.pyright]
include = ["src"]
strict = ["src/api"]
reportMissingTypeStubs = false

# Run both in CI
mypy src/
pyright src/

ローカル mypy プロファイル（ご自身のリポジトリ）

Python プロジェクトでよく使われるパターン：

• 厳格なデフォルト設定（edgar、kuzu-memory、mcp-browser）： disallow_untyped_defs = true、check_untyped_defs = true、no_implicit_optional = true、warn_return_any = true、strict_equality = true。
• 緩和プロファイル（mcp-ticketer）：パッチリリース向けに、disable_error_code リストを使って厳格なフラグを一時的に無効にします。
• 段階的な導入（mcp-vector-search）：型を安定させる間、ignore_errors = true を設定します。
• インポートの不足：mcp-memory と mcp-ticketer では ignore_missing_imports = true を使用します。

参考：edgar、kuzu-memory、mcp-vector-search、mcp-ticketer の pyproject.toml をご覧ください。

ベストプラクティス

1. 重要なモジュールから始める

# ✅ GOOD: Type critical business logic first
# services/user_service.py
from typing import Optional

class UserService:
    def get_user(self, user_id: int) -> Optional[User]:
        """Retrieve user by ID."""
        return self.db.query(User).get(user_id)

    def create_user(self, data: UserCreate) -> User:
        """Create new user."""
        user = User(**data.dict())
        self.db.add(user)
        self.db.commit()
        return user

2. 読みやすさのために型エイリアスを使う

# ✅ GOOD: Clear, reusable type aliases
from typing import Dict, List, Union

JSON = Union[Dict[str, "JSON"], List["JSON"], str, int, float, bool, None]
Headers = Dict[str, str]
UserId = int

def parse_response(data: JSON, headers: Headers) -> UserId:
    pass

# ❌ BAD: Complex inline types
def parse_response(
    data: Union[Dict[str, Union[...]], List[...], str, int, float, bool, None],
    headers: Dict[str, str]
) -> int:
    pass

3. 暗黙的な指定より明示的な指定を優先する

# ✅ GOOD: Explicit types for public APIs
def get_user(user_id: int) -> Optional[User]:
    return db.query(User).get(user_id)

# ❌ ACCEPTABLE: Type inference for internal helpers
def _format_name(first, last):  # mypy infers str -> str
    return f"{first} {last}"

4. デバッグには reveal_type を使う

# During development, check inferred types
from typing import reveal_type

def process_data(data):
    result = transform(data)
    reveal_type(result)  # mypy: Revealed type is "int"
    return result * 2

5. 型チェックの無視を文書化する

# ✅ GOOD: Document why type checking is disabled
import legacy_module  # type: ignore[import]  # TODO: Add type stubs

# ❌ BAD: No explanation
import legacy_module  # type: ignore

よくある落とし穴

❌ アンチパターン 1：あらゆる場所で Any を使う

# WRONG: Defeats purpose of type checking
from typing import Any

def process(data: Any) -> Any:
    return data.transform()

正しい例：

from typing import Union, Protocol

class Transformable(Protocol):
    def transform(self) -> dict: ...

def process(data: Transformable) -> dict:
    return data.transform()

❌ アンチパターン 2：型エラーを全体的に無視する

# WRONG: Disables type checking
[mypy]
ignore_errors = True

正しい例：

# Ignore specific modules only
[mypy-legacy.*]
ignore_errors = True

[mypy]
strict = True

❌ アンチパターン 3：Optional を使わない

# WRONG: Nullable without Optional
def get_user(user_id: int) -> User:
    user = db.get(user_id)  # Can be None!
    return user  # Runtime error if None

正しい例：

from typing import Optional

def get_user(user_id: int) -> Optional[User]:
    return db.get(user_id)

# Handle None explicitly
user = get_user(123)
if user is not None:
    print(user.name)

クイックリファレンス

よく使うコマンド

# Basic type checking
mypy main.py
mypy src/

# Strict mode
mypy --strict src/

# Install missing type stubs
mypy --install-types

# Generate HTML report
mypy src/ --html-report mypy-report

# Check specific error codes
mypy --show-error-codes src/

# Ignore missing imports
mypy --ignore-missing-imports src/

# Follow imports
mypy --follow-imports=silent src/

# Incremental mode (faster)
mypy --incremental src/

# Verbose output
mypy --verbose src/

エラーコードのリファレンス

# Common error codes
[attr-defined]      # Attribute not defined
[arg-type]          # Argument type mismatch
[return-value]      # Return type mismatch
[assignment]        # Assignment type mismatch
[call-overload]     # No matching overload
[index]             # Invalid index operation
[operator]          # Unsupported operand type
[import]            # Cannot find import
[misc]              # Miscellaneous type error
[no-untyped-def]    # Function missing type annotation
[var-annotated]     # Variable needs type annotation

リソース

• 公式ドキュメント：https://mypy.readthedocs.io/
• 型ヒントの PEP：https://peps.python.org/pep-0484/
• typing モジュール：https://docs.python.org/3/library/typing.html
• mypy の GitHub：https://github.com/python/mypy
• 型スタブ：https://github.com/python/typeshed
• django-stubs：https://github.com/typeddjango/django-stubs
• FastAPI + mypy：https://fastapi.tiangolo.com/tutorial/type-hints/

関連スキル

mypy を使用する際は、次の補完的なスキルもご検討ください（スキルライブラリから利用できます）。

• pytest：mypy による型安全なテスト — テストスイートに型チェックを組み込み、包括的な型カバレッジを実現します
• fastapi-local-dev：完全な型安全性を備えた FastAPI — FastAPI の実行時検証と mypy の静的チェックを組み合わせます
• pydantic：mypy に対応した実行時の型検証 — mypy がコンパイル時に検証する一方で、実行時にデータを検証します

---

**mypy のバージョン互換性：**このスキルは mypy 1.8 以降を対象としており、2025 年時点の Python 型チェックのベストプラクティスを反映しています。