SQLAlchemy IntegrityError duplicate key violates unique-constraint

Question

This problem has been doing my head in for far too long and I cannot find any solution in other questions.

In SQLAlchemy I am creating a schema (2 tables) which are linked by keys. When I INSERT data into the tables, I get a UNIQUE-CONSTRAINT error:

    (IntegrityError) ERROR:  duplicate key violates UNIQUE-CONSTRAINT e_data_pkey
    DETAIL:  KEY (id)=(2) already exists.
    'INSERT INTO e_data (id, iso3, year, value) VALUES (%(id)s, %(iso3)s, %(year)s, %(value)s)' 
  ({'iso3': 'ABW', 'id': 1, 'value': 5.5, 'year': 2009}, 
   {'iso3': 'ZZZ', **'id': 2,** 'value': 9.9, 'year': 1977}, 
   {'iso3': 'ZZY', **'id': 2**, 'value': 9.876, 'year': 1971})

Obviously, the class is allocating the same ID to each instance in the list.

Shouldn't the IDs be auto-generated by SQLAlchemy and why is this error occurring? I have adapted this example from other sources, so it should be easily implemented.

Does anyone see the problem? Here's the code I am using:

from sqlalchemy import *
from sqlalchemy.ext.declarative import declarative_base
from sqlalchemy.orm import relation, sessionmaker

Base = declarative_base()

class E_data(Base):
    __tablename__ = 'e_data'

    id = Column(Integer, ForeignKey('e_indicator.id'), primary_key=True) 
    iso3 = Column(String(3), nullable=False)
    year = Column(Integer)
    value = Column(FLOAT)

    eind = relation("E_indicator", backref='e_data', lazy=False)

    def __init__(self, iso3=None, year=None, value=None):
        self.iso3 = iso3
        self.year = year
        self.value = value
    def __repr__(self):
        return "E_Data (%r, %r, %r, %r)" % (self.iso3, self.year, self.value, self.eind)

class E_indicator(Base):
    __tablename__ = 'e_indicator'

    id = Column(Integer, primary_key=True)
    name = Column(String(50), nullable=False, unique=True)

    def __init__(self, name=None):
        self.name = name

    def __repr__(self):
        return "E_indicator (%r)" % (self.name)

engine = create_engine('postgresql://postgres:password@localhost/world')
Base.metadata.create_all(engine)

Session = sessionmaker(bind=engine)
session = Session()

#INSERT INTO DB
Session = sessionmaker(bind=engine)
session = Session()

m1 = E_data("ABW", 2009, 5.5)
m1.eind = E_indicator("GDP_TEST")

d2 = E_indicator("GDP THE OTHER WAY ROUND")
d2.e_data = [E_data("ZZZ", 1977, 9.9), E_data("ZZY", 1971, 9.876)]


#try:
session.add(m1)
session.add(d2)
session.commit()
#except:
    #session.rollback()

Thanks for any advice.

Answer 1

and sorry but both you suggestions did not solve my problem. The only way I could make it work to specify a primaryjoin condition like this:

eind = relation("E_indicator", backref='e_data', primaryjoin="E_indicator.id==E_data.id")

This solved the issue. I hope this will help someone.

Cheers, A

Answer 2

My problem was that I inserted 4 initial records when the app runs for the first time. Somehow Postgres has restarted the counter, I still do not know why, and after my 5th try and every future try it works.