Add onupdate=“cascade” to existing column in postgres with sqlalchemy

Question

I want to add onupdate="cascade" on the ForeignKey object in an existing table.

I have a many-to-many relationship using an association object:

from sqlalchemy.ext.declarative import declarative_base
from sqlalchemy import Column, Integer, String, ForeignKey, create_engine
from sqlalchemy.orm import relationship

postgres_url = "postgresql:/username:pwd@localhost/myDB"
engine = create_engine(postgres_url)
Base = declarative_base()
Base.metadata.create_all(engine)

class VehicleType(Base):
    __tablename__ = 'vehicleTypes'

    name = Column(String, primary_key=True)

class SearchTermToVehicleType(Base):
    __tablename__ = "searchTermToVehicleType"

    searchTermName = Column(String, ForeignKey('searchTerms.name'), primary_key=True)
    vehicleTypeName = Column(String, ForeignKey('vehicleTypes.name'), primary_key=True)

class SearchTerm(Base):
    __tablename__ = 'searchTerms'

    name = Column(String, primary_key=True)
    vehicleTypes = relationship("VehicleType", secondary="searchTermToVehicleType")

This is already in production, but now I need to allow users to change the search term name. It seems that with an association table sqlalchemy automagically does ondelete="cascade" but not onupdate="cascade" so now I want to change my models to include this on the ForeignKey object on the association table so that I can simply change the name of any searchTerm .

from sqlalchemy.ext.declarative import declarative_base
from sqlalchemy import Column, Integer, String, ForeignKey, create_engine
from sqlalchemy.orm import relationship

postgres_url = "postgresql:/username:pwd@localhost/myDB"
engine = create_engine(postgres_url)
Base = declarative_base()
Base.metadata.create_all(engine)

class VehicleType(Base):
    __tablename__ = 'vehicleTypes'

    name = Column(String, primary_key=True)

class SearchTermToVehicleType(Base):
    __tablename__ = "searchTermToVehicleType"

    searchTermName = Column(
        String, 
        ForeignKey('searchTerms.name', onupdate="cascade"), 
        primary_key=True
    )
    vehicleTypeName = Column(String, ForeignKey('vehicleTypes.name'), primary_key=True)

class SearchTerm(Base):
    __tablename__ = 'searchTerms'

    name = Column(String, primary_key=True)
    vehicleTypes = relationship("VehicleType", secondary="searchTermToVehicleType")

But this will not change the existing underlying postgres tables. I can drop the SearchTerm and SearchTermToVehicleType tables and let sqlalchemy build them again, but I would prefer not to have to repopulate them.

The docs mention that this uses the database's ON UPDATE CASCADE functionality. How can I alter the existing table or column to achieve the result of the above? Or do I have to drop and rebuild the tables?

Answer 1

You need to run DROP CONSTRAINT and ADD CONSTRAINT sql commands on your postgresql database to change your existing constraint to add on update cascade .

ALTER TABLE ...
DROP CONSTRAINT ...,
ADD CONSTRAINT ... FOREIGN KEY ... ON UPDATE CASCADE;

To automate schema management - please check out Alembic project.