when placing SQL expressions in the query()
list of entities. In particular scalar subqueries
should not "leak" their inner FROM objects out
into the enclosing query.
features keynames which prefer mapped attribute
names over column keys, column keys over
column names, i.e.
Query(Class.foo, Class.bar) will have names
"foo" and "bar" even if those are not the names
of the underlying Column objects. Direct
Column objects such as Query(table.c.col) will
return the "key" attribute of the Column.
- added query_cls keyword argument to sessionmaker(); allows user-defined Query subclasses to be generated by query().
- added @attributes.on_reconstitute decorator, MapperExtension.on_reconstitute, both receieve 'on_load' attribute event allowing
non-__init__ dependent instance initialization routines.
- push memusage to the top to avoid pointless heisenbugs
- renamed '_foostate'/'_fooclass_manager' to '_sa_instance_state'/'_sa_class_manager'
- removed legacy instance ORM state accessors
- query._get() will use _remove_newly_deleted instead of expunge() on ObjectDeleted, so that transaction rollback
restores the previous state
- removed MapperExtension.get(); replaced by a user-defined Query subclass
- removed needless **kwargs from query.get()
- removed Session.get(cls, id); this is redundant against Session.query(cls).get(id)
- removed Query.load() and Session.load(); the use case for this method has never been clear, and the same functionality is available in more explicit ways
- fixed [ticket:1008], count() works with single table inheritance
- changed the relationship of InstrumentedAttribute to class such that each subclass in an inheritance hierarchy gets a unique InstrumentedAttribute per column-oriented attribute, including for the same underlying ColumnProperty. This allows expressions from subclasses to be annotated accurately so that Query can get a hold of the exact entities to be queried when using column-based expressions. This repairs various polymorphic scenarios with both single and joined table inheritance.
- still to be determined is what does something like query(Person.name, Engineer.engineer_info) do; currently it's problematic. Even trickier is query(Person.name, Engineer.engineer_info, Manager.manager_name)
- removed query.compile(); use explicit query.with_labels().statement instead
- moved statement annotation step upwards from query._compile_context() to outliers from_self()/statement. speeds zoomark.step_6_editing by 16%
- pulled out DeclarativeMeta.__init__ into its own function, added instrument_declarative()
which will do the "declarative" thing to any class independent of its lineage (for ctheune)
- added "cls" kwarg to declarative_base() allowing user-defined base class for declarative base [ticket:1042]
- Query.__getitem__ now returns list/scalar in all cases, not generative (#1035)
- added Query.slice_() which provides the simple "limit/offset from a positive range" operation,
we can rename this to range_()/section()/_something_private_because_users_shouldnt_do_this() as needed
- broke up adapter chaining in eagerload, erroneous "wrapping" in row_decorator. column_property() subqueries are now affected only by the ORMAdapter for that mapper. fixes [ticket:1037], and may possibly impact some of [ticket:949]
- @unsupported now only accepts a single target and demands a reason
for not running the test.
- @exclude also demands an exclusion reason
- Greatly expanded @testing.requires.<feature>, eliminating many
decorators in the suite and signficantly easing integration of
multi-driver support.
- New ORM test base class, and a featureful base for mapped tests
- Usage of 'global' for shared setup going away, * imports as well
- this branch changes query.values() to immediately return an iterator, adds a new "aliased" construct which will be the primary method to get at aliased columns when using values()
- tentative ORM versions of _join and _outerjoin are not yet public, would like to integrate with Query better (work continues in the branch)
- lots of fixes to expressions regarding cloning and correlation. Some apparent ORM bug-workarounds removed.
- to fix a recursion issue with anonymous identifiers, bind parameters generated against columns now just use the name of the column instead of the tablename_columnname label (plus the unique integer counter). this way expensive recursive schemes aren't needed for the anon identifier logic. This, as usual, impacted a ton of compiler unit tests which needed a search-n-replace for the new bind names.
- removes everything to do with select_table, which remains as a keyword argument synonymous with
with_polymorphic=('*', select_table).
- all "polymorphic" selectables find their way to Query by way of _set_select_from() now, so that
all joins/aliasing/eager loads/etc. is handled consistently. Mapper has methods for producing
polymorphic selectables so that Query and eagerloaders alike can get to them.
- row aliasing simplified, so that they don't need to nest. they only need the source selectable
and adapt to whatever incoming columns they get.
- Query is more egalitarian about mappers/columns now. Still has a strong sense of "entity zero",
but also introduces new unpublished/experimental _values() method which sets up a columns-only query.
- Query.order_by() and Query.group_by() take *args now (also still take a list, will likely deprecate
in 0.5). May want to do this for select() as well.
- the existing "check for False discriminiator" "fix" was not working completely, added coverage
- orphan detection was broken when the target object was a subclass of the mapper with the orphaned
relation, fixed that too.
loading on an polymorphically-mapped instance mapped
by a select_table mapper.
- added query.with_polymorphic() - specifies a list
of classes which descend from the base class, which will
be added to the FROM clause of the query. Allows subclasses
to be used within filter() criterion as well as eagerly loads
the attributes of those subclasses.
- deprecated Query methods apply_sum(), apply_max(), apply_min(),
apply_avg(). Better methodologies are coming....
- your cries have been heard: removing a pending item
from an attribute or collection with delete-orphan
expunges the item from the session; no FlushError is raised.
Note that if you session.save()'ed the pending item
explicitly, the attribute/collection removal still knocks
it out.
Mike Bayer