Non-tuple exact and in lookups have specialized logic for subqueries that can
be adapted to properly assign select mask if unspecified and ensure the number
of involved members are matching on both side of the operator.
VALUES must be explicitly specified when declaring a sequence of tuples
on SQLite < 3.37 but it's not required on >= 3.37.
See sqlite/sqlite@9289f51 which addressed the last remaining issue with
IN.
This should allow third-party backends to define Tuple.as_vendor()
overrides that are taken into consideration which calling as_sql()
directly prevents.
Now that In.get_source_expression() includes its right-hand-side when it
contains expressions (refs #36025) it no longer requires a specialized
get_refs() method.
In order for Expression.relabeled_clone to work appropriately its
get_source_expressions method must return all resolvable which wasn't the case
for Lookup when its right-hand-side is "direct" (not a compilable).
While refs #22288 added support for non-literals iterable right-hand-side
lookups it predated the subclassing of Lookup(Expression) refs #27021 which
could have been an opportunity to ensure right-hand-sides are always resolvable
(ValueList and ExpressionList).
Addressing all edge case with non-resolvable right-hand-sides would require
a significant refactor and deprecation of some parts of the Lookup interface so
this patch only focuses on FieldGetDbPrepValueIterableMixin (In and Range
lookups) by making sure that a right-hand-side containing resolvables are dealt
with appropriately during the resolving phase.
Thanks Aashay Amballi for the report.
The original queryset._next_is_sticky() call never had the intended effect as
no further filtering was applied internally after the pk__in lookup making it
a noop.
In order to be coherent with how related filters are applied when retrieving
objects from a related manager the effects of what calling _next_is_sticky()
prior to applying annotations and filters to the queryset provided for
prefetching are emulated by allowing the reuse of all pre-existing JOINs.
Thanks David Glenck and Thiago Bellini Ribeiro for the detailed reports and
tests.
Replaced manual caching complexity with cached_property for efficiency.
Enhanced error handling with distinct messages for WKT and PROJ.4.
Thanks to Sarah Boyce for the suggestions.
The get_(local|foreign)_related_value methods of GenericRelation must be
reversed because it defines (from|to)_fields and associated related_fields
in the reversed order as it's effectively a reverse GenericForeignKey
itself.
The related value methods must also account for the fact that referenced
primary key values might be stored as a string on the model defining the
GenericForeignKey but as integer on the model defining the GenericRelation.
This is achieved by calling the to_python method of the involved content type
in get_foreign_related_value just like GenericRelatedObjectManager does.
Lastly reverse many-to-one manager's prefetch_related_querysets should use
set_cached_value instead of direct attribute assignment as direct assignment
might are disallowed on ReverseManyToOneDescriptor descriptors. This is likely
something that was missed in f5233dc (refs #32511) when the is_cached guard
was added.
Thanks 1xinghuan for the report.
When all values of a field with a db_default are DatabaseDefault, which
is the case most of the time, there is no point in specifying explicit
DEFAULT for all INSERT VALUES as that's what the database will do anyway
if not specified.
In the case of PostgreSQL doing so can even be harmful as it prevents
the usage of the UNNEST strategy and in the case of Oracle, which
doesn't support the usage of the DEFAULT keyword, it unnecessarily
requires providing literal db defaults.
Thanks Lily Foote for the review.
Removed obsolete and potentially problematic IDNA 2003 ("punycode")
encoding of international domain names in smart_urlquote() and Urlizer,
which are used (only) by AdminURLFieldWidget and the urlize/urlizetrunc
template filters. Changed to use percent-encoded UTF-8, which defers
IDNA details to the browser (like other URLs rendered by Django).
Remove redundant Case and When.resolve_expression to delegate composite
expression support to BaseExpression.
Thanks Jacob Tyler Walls for the report and test.
Remove redundant Func.resolve_expression and adjust CombinedExpression to
delegate source expression resolving to super() to perform checks against
allows_composite_expressions in a single location.
Rely on ColPairs and TupleIn which support a single column to be specified
to avoid special casing ForwardManyToOne.get_prefetch_querysets().
Thanks Jacob Walls for the report.
The array fields opt-out heuristic failed to account for sized arrays.
Note that we keep relying on db_type as opposed to performing an ArrayField
instance check against the column's field as there could be other
implementations of model fields that use Postgres arrays to store the
optimization must be disabled for all of them.
Refs #35936.
Thanks Claude Paroz for the report and test.
