Reputation: 1245
I'm using SQLAlchemy's automap to autogenerate the schema from a Postgres database managed by Odoo (direct access of the database is WAAAYYY faster than using the Odoo API). But following examples from the docs are giving me problems.
I'm trying to only parse specific tables per the advice here:
# we can reflect it ourselves from a database, using options
# such as 'only' to limit what tables we look at...
metadata.reflect(engine, only=['user', 'address'])
My code:
import sqlalchemy as sa
from sqlalchemy.ext.automap import automap_base
from app import odb
from sqlalchemy.orm import declarative_base, relationship
from sqlalchemy.orm.relationships import RelationshipProperty
metadata = sa.MetaData()
metadata.reflect(odb, only=['sale_order'])
Base = automap_base(metadata=metadata)
Base.prepare()
Order = Base.classes.sale_order
But I'm erroring out on a mapping for a table that I didn't specify.
[ins] In [1]: from app.omodels import Order
/home/mas/pe/app/omodels/omodels.py:8: SAWarning: Skipped unsupported reflection of expression-based index res_partner_vat_index
metadata.reflect(odb, only=['sale_order'])
---------------------------------------------------------------------------
ArgumentError Traceback (most recent call last)
<ipython-input-1-39a04770d5b4> in <module>
----> 1 from app.omodels import Order
~/pe/app/omodels/__init__.py in <module>
----> 1 from .omodels import *
~/pe/app/omodels/omodels.py in <module>
8 metadata.reflect(odb, only=['sale_order'])
9 Base = automap_base(metadata=metadata)
---> 10 Base.prepare()
11 Order = Base.classes.sale_order
12
<string> in prepare(cls, autoload_with, engine, reflect, schema, classname_for_table, collection_class, name_for_scalar_relationship, name_for_collection_rela
tionship, generate_relationship, reflection_options)
~/pe/.venv/lib/python3.6/site-packages/sqlalchemy/util/deprecations.py in warned(fn, *args, **kwargs)
296 stacklevel=3,
297 )
--> 298 return fn(*args, **kwargs)
299
300 doc = fn.__doc__ is not None and fn.__doc__ or ""
~/pe/.venv/lib/python3.6/site-packages/sqlalchemy/ext/automap.py in prepare(cls, autoload_with, engine, reflect, schema, classname_for_table, collection_class
, name_for_scalar_relationship, name_for_collection_relationship, generate_relationship, reflection_options)
926
927 for map_config in _DeferredMapperConfig.classes_for_base(cls):
--> 928 map_config.map()
929
930 _sa_decl_prepare = True
~/pe/.venv/lib/python3.6/site-packages/sqlalchemy/orm/decl_base.py in map(self, mapper_kw)
1070 def map(self, mapper_kw=util.EMPTY_DICT):
1071 self._configs.pop(self._cls, None)
-> 1072 return super(_DeferredMapperConfig, self).map(mapper_kw)
1073
1074
~/pe/.venv/lib/python3.6/site-packages/sqlalchemy/orm/decl_base.py in map(self, mapper_kw)
990 return self.set_cls_attribute(
991 "__mapper__",
--> 992 mapper_cls(self.cls, self.local_table, **self.mapper_args),
993 )
994
<string> in __init__(self, class_, local_table, properties, primary_key, non_primary, inherits, inherit_condition, inherit_foreign_keys, always_refresh, versi
on_id_col, version_id_generator, polymorphic_on, _polymorphic_map, polymorphic_identity, concrete, with_polymorphic, polymorphic_load, allow_partial_pks, batc
h, column_prefix, include_properties, exclude_properties, passive_updates, passive_deletes, confirm_deleted_rows, eager_defaults, legacy_is_orphan, _compiled_
cache_size)
~/pe/.venv/lib/python3.6/site-packages/sqlalchemy/util/deprecations.py in warned(fn, *args, **kwargs)
296 stacklevel=3,
297 )
--> 298 return fn(*args, **kwargs)
299
300 doc = fn.__doc__ is not None and fn.__doc__ or ""
~/pe/.venv/lib/python3.6/site-packages/sqlalchemy/orm/mapper.py in __init__(self, class_, local_table, properties, primary_key, non_primary, inherits, inherit_condition, inherit_foreign_keys, always_refresh, version_id_col, version_id_generator, polymorphic_on, _polymorphic_map, polymorphic_identity, concrete, with_polymorphic, polymorphic_load, allow_partial_pks, batch, column_prefix, include_properties, exclude_properties, passive_updates, passive_deletes, confirm_deleted_rows, eager_defaults, legacy_is_orphan, _compiled_cache_size)
681 self._configure_inheritance()
682 self._configure_class_instrumentation()
--> 683 self._configure_properties()
684 self._configure_polymorphic_setter()
685 self._configure_pks()
~/pe/.venv/lib/python3.6/site-packages/sqlalchemy/orm/mapper.py in _configure_properties(self)
1439
1440 self._configure_property(
-> 1441 column_key, column, init=False, setparent=True
1442 )
1443
~/pe/.venv/lib/python3.6/site-packages/sqlalchemy/orm/mapper.py in _configure_property(self, key, prop, init, setparent)
1674
1675 if not isinstance(prop, MapperProperty):
-> 1676 prop = self._property_from_column(key, prop)
1677
1678 if isinstance(prop, properties.ColumnProperty):
~/pe/.venv/lib/python3.6/site-packages/sqlalchemy/orm/mapper.py in _property_from_column(self, key, prop)
1866 "use the 'include_properties' or 'exclude_properties' "
1867 "mapper arguments to control specifically which table "
-> 1868 "columns get mapped." % (key, self, column.key, prop)
1869 )
1870
ArgumentError: WARNING: when configuring property 'product_packaging' on mapped class stock_move->stock_move, column 'product_packaging' conflicts with property '<RelationshipProperty at 0x7fe61ba87f48; product_packaging>'. To resolve this, map the column to the class under a different name in the 'properties' dictionary. Or, to remove all awareness of the column entirely (including its availability as a foreign key), use the 'include_properties' or 'exclude_properties' mapper arguments to control specifically which table columns get mapped.
QUESTION: How can I exclude this table (or only include the tables that I want) so that SQLA doesn't go needlessly parsing tables I don't need (it also seems to take quite a while to parse this database, which is huge)?
Alternatively, how can I override this specific error so that I can access my Order
object? The Error/Warning says
To resolve this, map the column to the class under a different name in the 'properties' dictionary. Or, to remove all awareness of the column entirely (including its availability as a foreign key), use the 'include_properties' or 'exclude_properties' mapper arguments to control specifically which table columns get mapped.
But I haven't described the schema for this table, and I don't know where to put 'exclude_properties'.
Upvotes: 2
Views: 1498
Reputation: 6474
You should be able to use the automap_base without metadata.reflect
and let Base
handle the reflection. In this way you can add the reflection_options
to prepare
and only the required table is mapped. The issue that you got is because metadata.reflect
is from SQLAlchemy Core and automap_base
is using SQLAlchemy ORM.
from sqlalchemy import create_engine, select
from sqlalchemy.ext.automap import automap_base
from sqlalchemy.orm import Session
DB_URL = ''
Base = automap_base()
engine = create_engine(DB_URL, future=True)
# reflect the tables
Base.prepare(autoload_with=engine, reflection_options={'only': ['res_users']})
# mapped classes are now created with names by default
# matching that of the table name.
OUsers = Base.classes.res_users
for mapped_class in Base.classes:
print(mapped_class)
You can also do it with SQLAlchemy Core in the following way (without automap_base):
meta = MetaData()
meta.reflect(bind=someengine)
OUsers = meta.tables['resusers']
Relevant documentation:
Upvotes: 1
Reputation: 1245
Turns out this is a simple case of naming collision that is addressed in the SQLA docs: Handling Simple Naming Conflicts.
I resolved it with the following method ripped straight from the docs:
def name_for_scalar_relationship(base, local_cls, referred_cls, constraint):
name = referred_cls.__name__.lower()
local_table = local_cls.__table__
if name in local_table.columns:
newname = name + "_"
warnings.warn(
"Already detected name %s present. using %s" %
(name, newname))
return newname
return name
Base.prepare(engine, reflect=True,
name_for_scalar_relationship=name_for_scalar_relationship)
Though I don't know where they're pulling warnings
from, I assume just a stand in for pick-your-own-logger.
It's worth noting I haven't figured out how to get the automap to ignore all tables save the ones I want. This is the state of my code and it still seems to map all the tables:
import sqlalchemy as sa
from sqlalchemy.ext.automap import automap_base
from sqlalchemy.orm import declarative_base, relationship, sessionmaker
from sqlalchemy.orm.relationships import RelationshipProperty
from app import odb
def name_for_scalar_relationship(base, local_cls, referred_cls, constraint):
"""Deals with a naming conflict problem."""
name = referred_cls.__name__.lower()
local_table = local_cls.__table__
if name in local_table.columns:
newname = name + "_"
return newname
return name
Session = sessionmaker(odb)
metadata = sa.MetaData()
metadata.reflect(odb, only=['res_users'])
Base = automap_base(metadata=metadata)
Base.prepare(name_for_scalar_relationship=name_for_scalar_relationship)
OUser = Base.classes.res_users
Order = Base.classes.sale_order
Partner = Base.classes.res_partner
But at least it works.
Upvotes: 1