SQLAlchemy update metadata.tables with schema

9 hours ago 2

ARTICLE AD BOX

I must be misunderstanding the usage of Metadata in SQLAlchemy... I'm trying to define a data model in one repository and then instantiate that model in a different repository using Alembic. The following is a minimal example:

# test.py from uuid import uuid4, UUID from sqlalchemy import MetaData from sqlalchemy.orm import DeclarativeBase, Mapped, mapped_column class Base(DeclarativeBase): metadata = MetaData() class Test(Base): __tablename__ = "test" id: Mapped[UUID] = mapped_column(primary_key=True, default=uuid4)

And then:

import test as t print(t.Base.metadata.schema) # expected to be `None` print(t.Base.metadata.tables) # Dict of tables with key = 'test' t.Base.metadata.schema = 'new_schema' print(t.Base.metadata.schema) # correctly set as 'new_schema' print(t.Base.metadata.tables) # Keys not updated to be 'new_schema.test'

In my real example, I'm setting the Metadata schema in Alembic's env.py, but Alembic checks both Metadata.tables and Metadata.sorted_tables, neither of which auto-update with the schema. Thanks in advance for your help!