Serializing SQLAlchemy with Marshmallow

Question

I'm following a tutorial and using the below code. I'm also using Postman to view the status of the server for http://localhost:5000/planets , but I'm getting 500 INTERNAL SERVER ERROR, when I should see my JSON data of the planets I created. In the command line I also see: AttributeError: 'list' object has no attribute 'data'

I feel it might have to do with the line that has: return jsonify(result.data) but I'm not sure.

from flask import Flask, jsonify, request
from flask_sqlalchemy import SQLAlchemy
from sqlalchemy import Column, Integer, String, Float
import os
from flask_marshmallow import Marshmallow
from marshmallow import Schema

app = Flask(__name__)

basedir = os.path.abspath(os.path.dirname(__file__))
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///C:/Users/krist/Py3/flask2/planets.db'

db = SQLAlchemy(app)
ma = Marshmallow(app)

@app.cli.command('db_create')
def db_create():
    db.create_all()
    print("DB Created")

@app.cli.command('db_seed')
def deb_seed():
    mercury = Planet(planet_name='Mercury',
                     planet_type='Class D',
                     home_star='Sol',
                     mass=3.25e23,
                     radius=1516,
                     distance=35.98e6)

    venus = Planet(planet_name='Venus',
                     planet_type='Class K',
                     home_star='Sol',
                     mass=8.95e24,
                     radius=3516,
                     distance=67.98e6)

    earth = Planet(planet_name='Earth',
                     planet_type='Class M',
                     home_star='Sol',
                     mass=5.97e24,
                     radius=3916,
                     distance=92.96e6)

    db.session.add(mercury)
    db.session.add(venus)
    db.session.add(earth)

    test_user = User(first_name='William',
                     last_name='Hershel',
                     email='test@test.com',
                     password='p@ssw0rd')

    db.session.add(test_user)
    db.session.commit()
    print("DB Seeded")

@app.route('/planets', methods=['GET'])
def planets():
    planets_list = Planet.query.all()
    result = planets_schema.dump(planets_list)
    return jsonify(result.data)


class User(db.Model):
    __tablename__ = 'users'
    id = Column(Integer, primary_key=True)
    first_name = Column(String)
    last_name = Column(String)
    email = Column(String, unique=True)
    password = Column(String)

class Planet(db.Model):
    __tablename__ = 'planets'
    planet_id = Column(Integer, primary_key=True)
    planet_name = Column(String)
    planet_type = Column(String)
    home_star = Column(String)
    mass = Column(Float)
    radius = Column(Float)
    distance = Column(Float)


class UserSchema(ma.Schema):
    class Meta:
        fields = ('id', 'first_name', 'last_name', 'email', 'password')

class PlanetSchema(ma.Schema):
    class Meta:
        fields = ('planet_id', 'planet_name', 'planet_type', 'home_star', 'mass', 'radius', 'distance')

user_schema = UserSchema()
users_schema = UserSchema(many=True)

planet_schema = PlanetSchema()
planets_schema = PlanetSchema(many=True)


if __name__ == '__main__':
    app.run()

Answer 1

Instead of

result = planets_schema.dump(planets_list)
return jsonify(result.data)

Try

result = planets_schema.dump(planets_list)
return jsonify(result)

Why this works:

Here you are querying the Planet Mapper to return a list of Planet ORM objects

planets_list = Planet.query.all()

Then the Marshmallow schema is used to marshal, or transform the ORM object into a python dictionary object. This is the basic principle of marshaling - transforming data from one format into another when the data is about to be transmitted or stored. So in this case you transform you data from a list of SQLAlchemy ORM objects into a list of Python dictionary objects.

result = planets_schema.dump(planets_list)

Now you have result (which could more aptly be names results that contains a list of dictionary objects. Then you are attempting to access the data variable on this list object. However Python lists have no data variable, so you get an error.

return jsonify(result.data)

The jsonify method from flask accepts a list of dictionaries as input, so simply modifying this line to the below should work:

return jsonify(result)