On Fri, Oct 27, 2017 at 9:54 PM,  <char...@konversion.ca> wrote:
> I'm not querying polymorphically.
>
> Let me restate the problem with more details(my post as done in haste).
>
> With something like the example I gave(completed with what's missing, e.g.
> table names, definition for related table),

thank you, with the stack trace I can construct what is going on.

There is a bug that needs a patch to work, in terms of the descriptor,
this is up at 
https://bitbucket.org/zzzeek/sqlalchemy/issues/4124/abstractconcretebase-transmitting
and the patch you see here:
https://gerrit.sqlalchemy.org/#/c/590/1/lib/sqlalchemy/orm/loading.py
fixes it.

Also, your Concrete1 and Concrete2 classes should specify
__mapper_args__ = {"concrete": True}, without that, you can also get
this problem even without the start/end descriptors, but the fix in
4124 above will also prevent this scenario from occurring.

However this part:

>
> MyBase.start, MyBase.end # that's weird, the Abstract base also has those
> attributes now

that part is normal.   The AbstractConcreteBase indicates you'd like
to query across all subclasses at once, and the way it works out at
the moment is that you'll see every column mapped get associated with
it and you do have start/end columns mapped.   if you don't need to
query in this way (which in general it's inefficient from a SQL
perspective), you should not use AbstractConcreteBase.  Polymorphic
querying across concrete subclasses is a worst case scenario and ACB
is still pretty hacky.





>
> session.expire(results[0])
> results[0].start
>
> Traceback (most recent call last):
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/IPython/core/interactiveshell.py",
> line 3035, in run_code
>     exec(code_obj, self.user_global_ns, self.user_ns)
>   File "<ipython-input-14-25b20252b355>", line 1, in <module>
>     results[0].start
>   File "/home/charles/PycharmProjects/MyProject/my_models.py", line 456, in
> start
>     return self.details.start
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/attributes.py",
> line 237, in __get__
>     return self.impl.get(instance_state(instance), dict_)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/attributes.py",
> line 584, in get
>     value = self.callable_(state, passive)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/strategies.py",
> line 539, in _load_for_state
>     passive
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/strategies.py",
> line 575, in _get_ident_for_use_get
>     for pk in self.mapper.primary_key
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/strategies.py",
> line 575, in <listcomp>
>     for pk in self.mapper.primary_key
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/mapper.py",
> line 2564, in _get_state_attr_by_column
>     return state.manager[prop.key].impl.get(state, dict_, passive=passive)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/attributes.py",
> line 579, in get
>     value = state._load_expired(state, passive)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/state.py",
> line 592, in _load_expired
>     self.manager.deferred_scalar_loader(self, toload)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/loading.py",
> line 713, in load_scalar_attributes
>     only_load_props=attribute_names)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/loading.py",
> line 223, in load_on_ident
>     return q.one()
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/query.py",
> line 2814, in one
>     ret = self.one_or_none()
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/query.py",
> line 2784, in one_or_none
>     ret = list(self)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/loading.py",
> line 90, in instances
>     util.raise_from_cause(err)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/util/compat.py",
> line 203, in raise_from_cause
>     reraise(type(exception), exception, tb=exc_tb, cause=cause)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/util/compat.py",
> line 187, in reraise
>     raise value
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/loading.py",
> line 57, in instances
>     for query_entity in query._entities
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/loading.py",
> line 57, in <listcomp>
>     for query_entity in query._entities
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/query.py",
> line 3700, in row_processor
>     polymorphic_discriminator=self._polymorphic_discriminator
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/loading.py",
> line 299, in _instance_processor
>     mapper._props[k] for k in only_load_props)
>   File
> "/home/charles/PycharmProjects/MyProject/env/lib/python3.6/site-packages/sqlalchemy/orm/loading.py",
> line 299, in <genexpr>
>     mapper._props[k] for k in only_load_props)
> KeyError: 'start'
>
>
> So on refresh, the loading process seems to expect 'start' and 'end' to be
> registered by the mappers(which they would be if they were columns, I
> guess).
>
>
> Le vendredi 27 octobre 2017 13:36:53 UTC-4, Mike Bayer a écrit :
>>
>> On Fri, Oct 27, 2017 at 11:57 AM,  <cha...@konversion.ca> wrote:
>> > My SQLAlchemy version is 1.1.14.
>> >
>> > I'm having difficulty understanding behaviours around a polymorphic
>> > hierarchy.
>> >
>> > I have an AbstractConcreteBase and two subclasses.
>> >
>> > The ACB declares some attributes shared by the two subclasses, and the
>> > subclasses have attributes of their own that are not shared(i.e. they
>> > are
>> > not part of the shared identity represented by the ACB).
>> >
>> > class MyBase(AbstractConcreteBase, Base, Mixin):
>> >     shared_id = Column(Integer,
>> > default=Sequence("database_shared_id_seq",
>> > start=0, increment=1), primary_key=True)
>> >
>> >     @declared_attr
>> >     def name(self):
>> >         return Column(String(100), nullable=False)
>> >
>> >     @declared_attr
>> >     def foreign_key(self):
>> >         return Column(Integer, ForeignKey("MyOtherTable.id"))
>> >
>> >     ...
>> >
>> >
>> > class Concrete1(MyBase):
>> >     start = Column(Date)
>> >     end = Column(Date)
>> >
>> >
>> > class Concrete2(MyBase):
>> >
>> >     @property
>> >     def start(self):
>> >         return self.details.start
>> >
>> >     @property
>> >     def end(self):
>> >         return self.details.end
>> >
>> >     details = relationship("Concrete2Detail", uselist=False,
>> > viewonly=True)
>> >
>> >
>> > As you can see, "start" and "end" are not declared in the ACB and are
>> > implemented differently between the table. I don't want them to be
>> > shared by
>> > the subclasses.
>> >
>> > When I query one of the Concretes, the ACB mapper is updated with the
>> > "start" and "end" attributes, and then the Concretes are expected to
>> > implement the attributes on the mapper level(which is not the case for
>> > Concrete2).
>>
>> This is a natural effect of using an "abstract" base.   When you try
>> to query polymorphically across Concrete1 and Concrete2, you're asking
>> it to query for the Concrete1.start and end columns.   However it
>> should not in any way be modifying the Concrete2 mapper itself.   I'm
>> not sure without running it what "are expected to implement" means.
>> An error is raised?
>>
>>
>> >
>> > How can I prevent this from happening?
>> >
>> > --
>> > SQLAlchemy -
>> > The Python SQL Toolkit and Object Relational Mapper
>> >
>> > http://www.sqlalchemy.org/
>> >
>> > To post example code, please provide an MCVE: Minimal, Complete, and
>> > Verifiable Example. See http://stackoverflow.com/help/mcve for a full
>> > description.
>> > ---
>> > You received this message because you are subscribed to the Google
>> > Groups
>> > "sqlalchemy" group.
>> > To unsubscribe from this group and stop receiving emails from it, send
>> > an
>> > email to sqlalchemy+...@googlegroups.com.
>> > To post to this group, send email to sqlal...@googlegroups.com.
>> > Visit this group at https://groups.google.com/group/sqlalchemy.
>> > For more options, visit https://groups.google.com/d/optout.
>
> --
> SQLAlchemy -
> The Python SQL Toolkit and Object Relational Mapper
>
> http://www.sqlalchemy.org/
>
> To post example code, please provide an MCVE: Minimal, Complete, and
> Verifiable Example. See http://stackoverflow.com/help/mcve for a full
> description.
> ---
> You received this message because you are subscribed to the Google Groups
> "sqlalchemy" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sqlalchemy+unsubscr...@googlegroups.com.
> To post to this group, send email to sqlalchemy@googlegroups.com.
> Visit this group at https://groups.google.com/group/sqlalchemy.
> For more options, visit https://groups.google.com/d/optout.

-- 
SQLAlchemy - 
The Python SQL Toolkit and Object Relational Mapper

http://www.sqlalchemy.org/

To post example code, please provide an MCVE: Minimal, Complete, and Verifiable 
Example.  See  http://stackoverflow.com/help/mcve for a full description.
--- 
You received this message because you are subscribed to the Google Groups 
"sqlalchemy" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sqlalchemy+unsubscr...@googlegroups.com.
To post to this group, send email to sqlalchemy@googlegroups.com.
Visit this group at https://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/d/optout.

Reply via email to