2. Interview Questions- ORM Mapped Class Configuration

Source: https://docs.sqlalchemy.org/en/20/orm/mapper_config.html

---

1. What are the two primary styles of ORM mapper configuration in SQLAlchemy?

Answer:

• Declarative mapping: Defines mapping within the class body using DeclarativeBase.

• Imperative (classical) mapping: Uses mapper() to link a class to a Table object explicitly.

---

2. What is the recommended approach for mapping in modern SQLAlchemy?

Answer: Declarative mapping using DeclarativeBase is recommended for most use cases due to its simplicity and clarity.

---

3. How do you declare a mapped class using declarative mapping?

Answer:

class MyClass(Base):
    __tablename__ = "my_table"
    id: Mapped[int] = mapped_column(primary_key=True)

---

4. How is table metadata associated with a declarative model?

Answer: Declarative models use __tablename__ and mapped_column() definitions to automatically create a Table object via Base.metadata.

---

5. What is imperative mapping?

Answer: It’s a manual approach where you create a Table object and then map it to a class using the mapper() function.

---

6. What function is used for classical (imperative) mapping?

Answer: mapper() from sqlalchemy.orm.

---

7. How do you define a mapped class using imperative mapping?

Answer:

from sqlalchemy.orm import registry
mapper_registry = registry()

my_table = Table(
    "my_table",
    mapper_registry.metadata,
    Column("id", Integer, primary_key=True),
)

class MyClass:
    pass

mapper_registry.map_imperatively(MyClass, my_table)

---

8. What is registry() in SQLAlchemy ORM?

Answer: A factory for generating mappers and managing metadata in both declarative and classical configurations.

---

9. How do you declare mixins in SQLAlchemy ORM?

Answer: By creating a class that is not mapped itself but is inherited by mapped classes to reuse fields or methods.

---

10. What is the role of __mapper_args__?

Answer: It's a dictionary used to pass additional options to the mapper, such as polymorphic_identity, order_by, or eager_defaults.

---

11. What does __table__ signify in a mapped class?

Answer: It directly assigns a Table object to a class, overriding the automatic table construction from __tablename__.

---

12. Can a single class be mapped to multiple tables?

Answer: No, a class can only map to one table directly, but joined-table inheritance can involve multiple tables.

---

13. How do you specify a custom primary key in declarative mapping?

Answer: By using mapped_column(primary_key=True) and not relying on defaults.

---

14. How do mixins interact with __tablename__?

Answer: Mixins should not define __tablename__. The final concrete class must define it.

---

15. Can you define relationships in a mixin class?

Answer: Yes, you can define relationships in mixins using standard ORM syntax.

---

16. What is a declarative registry?

Answer: An object that holds class mappings and metadata, typically instantiated via DeclarativeBase or registry().

---

17. What does __allow_unmapped__ = True do?

Answer: It allows the class to include attributes that SQLAlchemy will ignore (not map) without raising an error.

---

18. How can you override default behavior in class mappings?

Answer: Use __mapper_args__ or custom column definitions using mapped_column() or manually define __table__.

---

19. What’s the difference between mapped_column() and Column()?

Answer:

• mapped_column() is used in declarative classes and includes type hinting.

• Column() is used in classical table definitions.

---

20. How do you reflect a table for use in imperative mapping?

Answer:

my_table = Table("my_table", metadata, autoload_with=engine)
mapper_registry.map_imperatively(MyClass, my_table)

---

21. What does the eager_defaults mapper argument do?

Answer: It enables automatic fetching of default values set by the database (e.g., server-side defaults) after an INSERT.

__mapper_args__ = {"eager_defaults": True}

---

22. What is polymorphic_on used for in __mapper_args__?

Answer: It defines the column used to distinguish between subclasses in a polymorphic inheritance mapping.

---

23. What does polymorphic_identity signify in a mapped class?

Answer: It defines the identifier used in the polymorphic_on column to associate a subclass with a specific value.

---

24. How do you specify ordering for a mapped class using __mapper_args__?

Answer: By setting order_by to a column or list of columns.

__mapper_args__ = {"order_by": id}

---

25. Can you use SQLAlchemy ORM without declaring a Base class?

Answer: Yes, using classical mapping with mapper() and a registry() is possible without a base class.

---

26. What happens if two classes define the same __tablename__?

Answer: SQLAlchemy will raise an error during metadata creation due to table name conflict.

---

27. How do you create an abstract base class that isn’t mapped itself?

Answer: Use __abstract__ = True in the class body.

---

28. What does __abstract__ = True do?

Answer: It tells SQLAlchemy not to map the class even though it inherits from a declarative base.

---

29. What is the purpose of a mixin class in SQLAlchemy?

Answer: To reuse common columns, constraints, relationships, or methods across multiple models.

---

30. Can mixins contain foreign key relationships?

Answer: Yes, mixins can define relationships and foreign key columns, which are inherited by mapped subclasses.

---

31. What is the difference between __tablename__ and __table__ in a model class?

Answer:

• __tablename__ lets SQLAlchemy auto-create the table.

• __table__ directly assigns an existing Table object.

---

32. How can you exclude a column from being mapped in a declarative class?

Answer: Either don’t annotate with Mapped[...] or set __allow_unmapped__ = True to prevent errors for unmapped attributes.

---

33. What is the use of declarative_mixin decorator?

Answer: It marks a class as a mixin, clarifying its purpose and suppressing mapping unless inherited.

from sqlalchemy.orm import declarative_mixin

@declarative_mixin
class TimestampMixin:
    created_at: Mapped[datetime] = mapped_column(...)

---

34. Can mixins override columns from the parent class?

Answer: Yes, but care must be taken to avoid conflicts during mapping.

---

35. How do you define joined-table inheritance in declarative mapping?

Answer:

class Parent(Base):
    __tablename__ = "parent"
    id: Mapped[int] = mapped_column(primary_key=True)
    type: Mapped[str] = mapped_column()
    __mapper_args__ = {"polymorphic_on": type}

class Child(Parent):
    __tablename__ = "child"
    id: Mapped[int] = mapped_column(ForeignKey("parent.id"), primary_key=True)
    __mapper_args__ = {"polymorphic_identity": "child"}

---

36. Can a model define __mapper_args__ without defining __tablename__?

Answer: No, unless __table__ is manually assigned. Otherwise, the table mapping will be incomplete.

---

37. What is the impact of omitting Mapped[...] in a declarative field?

Answer: SQLAlchemy may not include it in the mapping, and a warning or error may be raised in strict configurations.

---

38. How do you ensure compatibility with Python static type checkers in SQLAlchemy?

Answer: Use Mapped[T] for fields and mapped_column() for column definitions.

---

39. What happens when registry().map_imperatively() is called more than once for the same class?

Answer: An error is raised since a class can only be mapped once in a given registry context.

---

40. How does SQLAlchemy handle attribute name clashes between mixins and subclasses?

Answer: The subclass overrides the mixin’s attribute. The last defined attribute takes precedence in the class hierarchy.

---

41. What is the function of __init__() in declarative models?

Answer: SQLAlchemy auto-generates an __init__() method that accepts keyword arguments for mapped fields. You can override it, but it’s typically not necessary unless custom logic is required.

---

42. How does SQLAlchemy determine which attributes to map when using declarative mapping?

Answer: It maps attributes annotated with Mapped[...] and defined using mapped_column(). Unannotated fields are ignored unless explicitly handled.

---

43. How can you control if a class should be mapped at runtime?

Answer: Use __abstract__ = True to skip mapping for base/mixin classes, or delay mapping with registry().map_imperatively() manually.

---

44. What happens if you declare a field with mapped_column() but forget Mapped[...]?

Answer: The field may not be registered correctly in the mapping. Static typing tools may also miss it, and SQLAlchemy may ignore it depending on configuration.

---

45. Why is registry.metadata preferred in imperative mapping over standalone MetaData?

Answer: registry.metadata ensures a consistent metadata registry for all ORM-mapped classes, enabling unified DDL generation and schema management.

---

46. How does SQLAlchemy know a class is "mapped"?

Answer: A class is considered mapped if it’s registered in the declarative base or explicitly mapped via map_imperatively().

---

47. What error will you get if you access .metadata on a model class that isn't properly mapped?

Answer: You may get an AttributeError or mapping exception because the class lacks an associated Table or metadata.

---

48. Can you modify __mapper_args__ dynamically at runtime?

Answer: Yes, though it’s uncommon. You can override it in the class definition based on environment or configuration.

---

49. How does SQLAlchemy differentiate between regular attributes and mapped columns in declarative mapping?

Answer: By type annotations with Mapped[...] and use of mapped_column() function. Attributes without these are treated as regular Python attributes.

---

50. What is the impact of assigning __table_args__ in a declarative class?

Answer: __table_args__ lets you define additional table-level arguments like indexes, constraints, and engine-specific options.

__table_args__ = (UniqueConstraint("email"), {"mysql_engine": "InnoDB"})

---