sql alchemy set foreign key field to None after commit - sqlalchemy

everyone. I found realy strange behaviour of SQL alchemy. When I'm trying to insert new instance of Device, after commit sql alchemy set type field to None. Here is my models:
class BaseTable(db.Model):
__abstract__ = True
id = db.Column(db.Integer, primary_key=True, unique=True,
index=True, autoincrement=True, nullable=False)
def __init__(self, **kwargs):
super().__init__(**kwargs)
self.date_created = dt.datetime.utcnow()
class Device(BaseTable):
__tablename__ = "device"
node_id = db.Column(db.ForeignKey('node.id', ondelete='CASCADE'), index=True, nullable=False)
name = db.Column(db.Unicode(255), nullable=False)
description = db.Column(db.Unicode, nullable=True)
type = db.Column(db.ForeignKey('device_type.id', ondelete='SET NULL'), index=True, nullable=True)
type_description = db.relationship("DeviceType", lazy='joined')
class DeviceType(BaseTable):
__tablename__ = "device_type"
name = db.Column(db.Unicode(255), unique=True, index=True, nullable=False)
code = db.Column(db.Unicode(255), unique=True, index=True, nullable=False)
In device_type table there is 6 entries with id form 1 to 6.
I do simple post request for creating new device instance. Here is code of controller:
#accepts(schema=post_schema, api=api)
#responds(schema=main_schema, api=api)
def post(self):
strer = ''
result = None
payload = api.payload
try:
entity = post_schema().load(payload)
result = crud_service.post(entity, log, db_session.session)
except Exception as ex:
strer += str(ex)
if result is None:
return response500(log, strer)
else:
return result
In crud_service.post I have:
session.add(entity)
session.commit()
return entity
So, when I make post request in payload I get type of device = 5 (for example). I'm sure that I have this ID in my device_type table. But after sql alchemy do session.commit() type of device is None. After it, if I make put request and rewrite type with
ent = session.query(model)\
.filter(model.id == dict_entity['id'])\
.update(ent_dict)
session.commit()
I got type in my DB as I expected.
So, it seems that sql alchemy by doing session.add() set type field of my device to None,
but session.update() not.
Can enyone explain this behaviour? May be any trouble in my models?

It seems that I've found problem.
I delete type_description = db.relationship("DeviceType", lazy='joined') and now it works fine. I will see another way for make relationship field

Related

SQLAlchemy: Set audit columns prior to insert or update

Each of my mapped class contains created_by and updated_by audit properties that I would like to set automatically upon INSERT and UPDATE of respective objects.
class User(Base):
__tablename__ = 'user'
id = Column(BigInteger, primary_key=True)
name = Column(Text, nullable=False)
...
class Address(Base):
__tablename__ = 'address'
id = Column(BigInteger, primary_key=True)
street = Column(Text, nullable=False)
...
created_by = Column(BigInteger) # references user.id
updated_by = Column(BigInteger) # references user.id
...
Is there a way to handle this centrally in SQLAlchemy? I looked at the events but it appears it needs to be setup for every single mapped class individually (note the SomeClass in the decorator).
#event.listens_for(SomeClass, 'before_insert')
def on_insert(mapper, connection, target):
target.created_by = context["current_user"] # I want to be able to do this not just for 'SomeClass' but for all mapped classes
#event.listens_for(SomeClass, 'before_update')
def on_update(mapper, connection, target):
target.updated_by = context["current_user"] # I want to be able to do this not just for 'SomeClass' but for all mapped classes
One solution here is to use the default parameters in the Column class provided by sqlalchemy. You can actually pass a callable to both default (to execute when first created) and onupdate to execute whenever updated.
def get_current_user():
return context["user"].id
class Address(Base):
__tablename__ = 'address'
...
created_by = Column(default = get_current_user)
updated_by = Column(default = get_current_user, onupdate=get_current_user)
Managed to figure it out, though somewhat concerned about using a dunder method __subclasses__() on declarative_base. If there is a better alternative do suggest.
def on_insert(mapper, connection, target):
target.created_by = context["user"].id
target.updated_at = datetime.utcnow()
def on_update(mapper, connection, target):
target.updated_by = context["user"].id
target.updated_at = datetime.utcnow()
Base.metadata.create_all()
mapped_classes = Base.__subclasses__()
for mapped_class in mapped_classes:
event.listen(mapped_class, 'before_insert', on_insert)
event.listen(mapped_class, 'before_update', on_update)
The context being referred to here is actually starlette-context

SQLAlchemy relationship selection criteria

I have two model classes:
class Programs(db.Model):
__tablename__ = "programs"
id = db.Column(db.Integer, primary_key=True)
title = db.Column(db.String(100), nullable=False)
description = db.Column(db.String(100), nullable=False)
duration = db.Column(db.Integer, nullable=False)
date_created = db.Column(db.DATE, default=datetime.now())
created_by = db.Column(db.String(100))
program_sessions = db.relationship('Program_Session',backref='programs')
class Program_Session(db.Model):
__tablename__ = "program_session"
id = db.Column(db.Integer, primary_key=True)
session_title = db.Column(db.String(100), nullable=False)
session_description = db.Column(db.String(100))
session_year = db.Column(db.Integer)
program_id = db.Column(db.Integer, db.ForeignKey("programs.id"), nullable=False)
students = db.relationship('Student_Registration', backref='program_session')
date_created = db.Column(db.DATE, default=datetime.now())
created_by = db.Column(db.String(100))
I create an object of Programs with:
program = Programs.query.first()
Now I can access all the Program_Sessions from the selected Program:
print(pro.program_sessions)
Is it possible to subquery/query to retrieve only those Program_session in Program whose year is 2021?
Option-1: filter on 'python' (in memory)
Once you get all Program_Sessions (all_sessions = pro.program_sessions), you filter them by sessions_2021 = [item for item in all_sessions if item.session_year == 2021].
Needless to say, this is not efficient at all as lots of data will be loaded from the database to be immediately discarded.
Option2: use Dynamic Relationship Loaders
Define the relationship with lazy="dynamic", which will return a Query and hence you will be able to apply additional criteria to the query:
class Programs(db.Model):
# ...
program_sessions = db.relationship('Program_Session', backref='programs', lazy="dynamic")
program = Programs.query.first()
sessions_2021 = program.program_sessions.filter(Program_Session.year == 2021).all()
Option3: use orm.with_parent [BEST]
sessions_2021 = select(Program_Session).where(with_parent(program, Program_Session.programs)).where(Program_Session.year == 2021)
The answer is yes...
van's answer shows you options for playing with sqlalchemy's query mechanism. But what if you want to write this logic on the Programs class itself? That way anywhere you have a Programs object, you can access the filter.
You can do it in pretty plain python by altering the Programs class like so:
class Programs(db.Model):
__tablename__ = "programs"
id = db.Column(db.Integer, primary_key=True)
title = db.Column(db.String(100), nullable=False)
description = db.Column(db.String(100), nullable=False)
duration = db.Column(db.Integer, nullable=False)
date_created = db.Column(db.DATE, default=datetime.now())
created_by = db.Column(db.String(100))
program_sessions = db.relationship('Program_Session',backref='programs')
'''Here I add a filter that returns only the sessions for a particular year
'''
def program_sessions_by_year(self, year):
return filter(lambda ps: ps.session_year == year, self.program_sessions)
If you care about efficiency, you can get the database to do the filtering for you using a bit more sqlalchemy magic:
from sqlalchemy.orm import object_session
class Programs(db.Model):
__tablename__ = "programs"
id = db.Column(db.Integer, primary_key=True)
title = db.Column(db.String(100), nullable=False)
description = db.Column(db.String(100), nullable=False)
duration = db.Column(db.Integer, nullable=False)
date_created = db.Column(db.DATE, default=datetime.now())
created_by = db.Column(db.String(100))
program_sessions = db.relationship('Program_Session',backref='programs')
'''Improve efficiency by using DB's SQL engine to filter the object.
'''
def program_sessions_by_year(self, year):
return object_session(self)\
.query(Program_Session)\
.filter_by(session_year=year, program_id=self.id)\
.all()
Either way you can then write (where-ever you have a Program object):
# lets say you just want the first program
first_program = Programs.query.first()
# to get the program sessions by year 2021
first_program.program_sessions_by_year(2021)
There's probably a bunch of other ways you could do something like this. SqlAlchemy is a big library. For more background on my answer, have a look at the SQL expressions as Mapped Attributes docs.

Can I specify the join conditions of a relationship when writing a query instead of doing so in my model class?

I have three models...
class Customer(db.Model, TimestampMixin):
pk = db.Column(db.Integer, primary_key=True)
name = db.Column(db.Text, nullable=False)
class User(db.Model, TimestampMixin):
pk = db.Column(db.Integer, primary_key=True)
name = db.Column(db.Text, nullable=False)
class CustomerUserStat(db.Model):
__table_args__ = (
UniqueConstraint("customer_pk", "user_pk", name="customer_pk_user_pk"),
)
pk = db.Column(db.Integer, primary_key=True)
customer_pk = db.Column(db.Integer, db.ForeignKey("customers.pk"), nullable=False)
user_pk = db.Column(db.Integer, db.ForeignKey("users.pk"), nullable=False)
data = db.Column(db.Integer, nullable=False, default=0)
I’ve tried...
m.Customer.query.outerjoin(
m.CustomerUserStat,
(m.CustomerUserStat.user_pk == entity.pk)
& (m.CustomerUserStat.customer_pk == m.Customer.pk)
).add_entity(m.CustomerUserStat)
But that gives me a tuple with a Customer object and a CustomerUserStat (or None if one doesn’t exist), close but not quite what I am looking for.
I’ve also tried adding a userstats relationship to the Customer model and ...
m.Customer.query.outerjoin(m.Customer.userstats
).filter(m.CustomerUserStat.user_pk == user.pk
).options(contains_eager(m.Customer.userstats))
But that didn’t produce any results if the user's CustomerUserStat record was missing.
Ultimately, I would like to write a query that will give me a list of Customer objects with an attribute that has an instance of CustomerUserStat for a user that I specify in the query and I need to be able to order by the CustomerUserStat.data field and paginate the results.
It looks to me like you are looking for the AssociationProxy pattern
I came up with this hack to get the query that I needed. In my Flask view function, I defined a new class that inherits from the original and added the relationship to it. Then I use the new class to create the query.
class TempCustomer(m.Customer):
userstats = db.relationship(
m.CustomerUserStat,
primaryjoin=(m.Customer.pk == m.CustomerUserStat.customer_pk) &
(m.CustomerUserStat.user_pk == agent.pk),
uselist=False,
viewonly=True,
lazy="joined",
)
q = TempCustomer.query.outerjoin(m.CustomerUserStat, (m.Customer.pk ==
m.CustomerUserStat.customer_pk) & (m.CustomerUserStat.user_pk == agent.pk),)
This works for me, but doesn't seem like the best solution.
It also results in two LEFT JOINS for the same data which is probably not the most efficient.

Sqlalchemy eager loading of parent all properties in joined table inheritance

I have the following problem:
I have a hierachy of classes with joined table inheritance:
class AdGroupModel(Base, AdwordsRequestMixin):
__tablename__ = 'ad_groups'
db_id = Column(BigInteger, primary_key=True)
created_at = Column(DateTime(timezone=False), nullable=False, default=datetime.datetime.now())
# ----RELATIONS-----
# campaign MANY-to-ONE
campaign_db_id = Column(BigInteger,
ForeignKey('campaigns.db_id', ondelete='CASCADE'),
nullable = True,
)
# # ads ONE-to-MANY
ads = relationship("AdModel",
backref="ad_group",
lazy="subquery",
passive_deletes=True,
single_parent=True,
cascade="all, delete, delete-orphan")
# # # keywords ONE-to-MANY
criteria = relationship("AdGroupCriterionModel",
backref="ad_group",
lazy="subquery",
passive_deletes=True,
single_parent=True,
cascade="all, delete, delete-orphan")
# Joined Table Inheritance
type = Column(Unicode(50))
__mapper_args__ = {
'polymorphic_identity': 'ad_group',
'polymorphic_on': type
}
class AdGroupCriterionModel(Base, AdGroupDependenceMixin):
__tablename__ = 'ad_group_criterion'
db_id = Column(BigInteger, primary_key=True)
destination_url = Column(Unicode, nullable=True)
status = Column(Enum("PAUSED", "ACTIVE", "DELETED",
name='criterion_status'), default="ACTIVE")
# ----RELATIONS---
# ad_group ONE-to-MANY
ad_group_db_id = Column(BigInteger, ForeignKey('ad_groups.db_id',
ondelete='CASCADE'), nullable=True)
# Joined Table Inheritance
criterion_sub_type = Column(Unicode(50))
__mapper_args__ = {
'polymorphic_on': criterion_sub_type
}
class AdGroupKeywordModel(AdGroupCriterionModel):
__tablename__ = 'ad_group_keyword'
__mapper_args__ = {'polymorphic_identity': 'Keyword'}
db_id = Column(Integer, ForeignKey('ad_group_criterion.db_id'), primary_key=True)
text = Column(Unicode, nullable=False)
class AdGroupDependenceMixin(object):
_aggad_id = Column(BigInteger, nullable=True)
_agname = Column(Unicode, nullable=True)
#hybrid_property
def ad_group_GAD_id(self):
if self.ad_group is None:
res = self._aggad_id
else:
res = self.ad_group.GAD_id
return res
#ad_group_GAD_id.setter
def ad_group_GAD_id(self, value):
self._aggad_id = value
if value is not None:
self.ad_group = None
#ad_group_GAD_id.expression
def ad_group_GAD_id(cls):
what = case([( cls._aggad_id != None, cls._aggad_id)], else_=AdGroupModel.GAD_id)
return what.label('adgroupgadid_expression')
#hybrid_property
def ad_group_name(self):
if self.ad_group is None:
return self._agname
else:
return self.ad_group.name
#ad_group_name.setter
def ad_group_name(self, value):
self._agname = value
if value is not None:
self.campaign = None
#ad_group_name.expression
def ad_group_name(cls):
what = case([( cls._agname != None, cls._agname)], else_=AdGroupModel.name)
return what.label('adgroupname_expression')
And I load the Keywords objects from the database with the following query:
all_objects1 = self.database.session.query(AdGroupKeywordModel).join(AdGroupModel)\
.options(subqueryload('ad_group'))\
.filter(AdGroupModel.GAD_id!=None)\
.limit(self.options.limit).all()
which returns obejcts of type AdGroupKeywordModel.
Unfortunately every time I try to access the properties of the AdGroupKeywordModel which are in the parent table (AdGroupCriterionModel) a query of this type is emitted:
sqlalchemy.engine.base.Engine
SELECT ad_group_criterion.destination_url AS ad_group_criterion_destination_url, ad_group_criterion.status AS ad_group_criterion_status, ad_group_criterion.ad_group_db_id AS ad_group_criterion_ad_group_db_id, ad_group_criterion.criterion_sub_type AS ad_group_criterion_criterion_sub_type, ad_group_keyword.text AS ad_group_keyword_text
FROM ad_group_criterion JOIN ad_group_keyword ON ad_group_criterion.db_id = ad_group_keyword.db_id
which is strongly compromising the performace.
What I would like to have is that all the attributes for the class AdGroupKeywordModel which are related to the parent (and other classes defined in the relationship) to be loaded with the initial query and be cached for further use. So that when I access them I do not get any overhead from further sqlstatements.
It seems that eager loading is only defined for relationships but not for hierarchies. Is it possible to have this behaviour in sqlalchemy for hierarchies as well?
Thanks
What I see is: only AdGroupModel has a relationship with a lazy= definition (which is the keyword which defines eager loading for relationships), and the query only has a subqueryload('ad_group').
The only point, in which ad_group or AdGroupModel touch with AdGroupKeywordModel is in AdGroupModel.criteria, which has as backref AdGroupCriterionModel.ad_group. I'm not familiar with the subqueryload syntax, but If I would want to eager-load AdGroupCriterionModel.ad_group, I'd define criteria like this:
criteria = relationship(
"AdGroupCriterionModel", backref=backref("ad_group", lazy="subquery"),
lazy="subquery", passive_deletes=True, single_parent=True,
cascade="all, delete, delete-orphan")
The key would be in defining the right lazy also for the backref.

how to save data in a many to many relationship using turbogears and sqlalchemy

hi i have a many to many relationship between a user and a group.and i will like to add a user with many groups in my database.how do i do that if my database is as follows
user_group_table = Table('tg_user_group', metadata,
Column('user_id', Integer, ForeignKey('tg_user.user_id',
onupdate="CASCADE", ondelete="CASCADE")),
Column('group_id', Integer, ForeignKey('tg_group.group_id',
onupdate="CASCADE", ondelete="CASCADE"))
)
class Group(DeclarativeBase):
"""
Group definition for :mod:`repoze.what`.1
Only the ``group_name`` column is required by :mod:`repoze.what`.
"""
__tablename__ = 'tg_group'
#{ Columns
group_id = Column(Integer, autoincrement=True, primary_key=True)
group_name = Column(Unicode(16), unique=True, nullable=False)
display_name = Column(Unicode(255))
created = Column(DateTime, default=datetime.now)
#{ Relations
users = relation('User', secondary=user_group_table, backref='groups')
#{ Special methods
def __repr__(self):
return '<Group: name=%s>' % self.group_name
def __unicode__(self):
return self.group_name
#}
class User(DeclarativeBase):
"""
User definition.
This is the user definition used by :mod:`repoze.who`, which requires at
least the ``user_name`` column.
"""
__tablename__ = 'tg_user'
#{ Columns
user_id = Column(Integer, autoincrement=True, primary_key=True)
user_name = Column(Unicode(16), unique=True, nullable=False)
email_address = Column(Unicode(255), unique=True, nullable=False,
info={'rum': {'field':'Email'}})
display_name = Column(Unicode(255))
_password = Column('password', Unicode(80),
info={'rum': {'field':'Password'}})
created = Column(DateTime, default=datetime.now)
doing it this way however gives me an error
#expose()
def user_save(self, **kw):
user = User()
user.user_name = kw['user_name']
user.display_name = kw['display_name']
user.email_address = kw['Email']
user._password = kw['password']
user.groups.extend(kw['groups'])
DBSession.add(user)
DBSession.flush()
flash("successfully saved...")
flash(user)
redirect("/user_new")
pls help me solve this.thanks in advance
I believe the answer is in the error message that you havn't posted in the question. user.groups is a list of Group objects, while you assign a list of strings(?) got from form to it. Also I see no explicit DBSession.commit() call. Are you sure TurboGears will do it for you?