How to use @st.cache_resource with sqlalchemy.orm.declarative_base?

Using Streamlit
,
1

I’m trying to use streamlit connected to a database, and some ORM classes via SQLAlchemy.

@st.cache_resource
def get_declared_base() -> tuple[Any, Engine]:
    engine = create_engine('sqlite:////content/local_memory.db', echo=True)
    Base = declarative_base()
    Base.metadata.create_all(engine)
    return Base, engine

class Simulation(get_declared_base()[0]):
    __tablename__ = 'tb_simulation'
    id = Column(String, primary_key=True)
    name = Column(String)
    comment = Column(String)
    years = relationship("Year", back_populates="simulation", order_by="Year.year", cascade="all, delete-orphan")

class Year(get_declared_base()[0]):  # type: ignore
    __tablename__ = 'tb_year'
    id = Column(String, primary_key=True)
    is_reference = Column(Integer, default=0)
    year = Column(Integer)
    inflation = Column(Float, default=0)
    simulation_id = Column(String, ForeignKey('tb_simulation.id'))
    simulation = relationship("Simulation", back_populates="years")
    inputs = relationship("Input", back_populates="year", cascade="all, delete-orphan")

This code works as expected, but if I do any changes to the code and rerun, I got the following error:

InvalidRequestError: Table 'tb_simulation' is already defined for this MetaData instance. Specify 'extend_existing=True' to redefine options and columns on an existing Table object.

Any tips on what I’m doing wrong?

2

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.