Commit Graph

52 Commits

Author SHA1 Message Date
Baptiste Mispelon f1b3ab9c21 Fixed #11629 -- Deprecated callable arguments to queryset methods.
Callable arguments were an untested and undocumented feature.
2013-12-18 18:40:00 +01:00
Anssi Kääriäinen e7b61e5717 Fixed #11320 -- exclude() too aggressive in join promotion 2013-11-07 12:57:02 +02:00
Anssi Kääriäinen 6fe2b001db Fixed #21376 -- New implementation for query join promotion logic
This commit introduced a new class JoinPromoter that can be used to
abstract away join promotion problems for complex filter conditions.
Query._add_q() and Query.combine() now use the new class.

Also, added a lot of comments about why join promotion is done the way
it is.

Thanks to Tim Graham for original report and testing the changes, and
for Loic Bistuer for review.
2013-11-07 12:53:26 +02:00
Jason Myers c3791463a5 Fixing E302 Errors
Signed-off-by: Jason Myers <jason@jasonamyers.com>
2013-11-02 23:48:47 -05:00
Anssi Kääriäinen 2a03a9a9a1 Fixed syntax error in queries/tests.py 2013-11-02 22:42:42 +02:00
Anssi Kääriäinen bec0b2a8c6 Fixed #14511 -- bug in .exclude() query 2013-11-02 22:35:45 +02:00
Anssi Kääriäinen b44d42be6d Fixed #21366 -- regression in join promotion logic
The regression was caused by ecaba36028
and affected OR connected filters.
2013-11-02 20:44:19 +02:00
Alex Gaynor 9d740eb8b1 Fix all violators of E231 2013-10-26 12:15:03 -07:00
Alasdair Nicol c3aa2948c6 Fixed #21298 -- Fixed E301 pep8 warnings 2013-10-23 13:45:03 +01:00
Tim Graham ae48d77ef8 Fixed E225 pep8 warnings. 2013-10-23 06:09:29 -04:00
Tim Graham 1597503a01 Fixed E221 pep8 warnings. 2013-10-22 09:51:39 -04:00
Tim Graham 499cd912ca Fixed E227 pep8 warnings 2013-10-21 08:52:21 -04:00
Alasdair Nicol b289fcf1bf Fixed #21288 -- Fixed E126 pep8 warnings 2013-10-21 08:31:30 -04:00
Tim Graham 96d1d4e292 Removed unused local variables in tests. 2013-10-19 08:31:38 -04:00
Alasdair Nicol bab9123daa Fixed #21268 -- Fixed E303 pep8 warnings 2013-10-18 01:46:24 +01:00
Tim Graham b67ab75e82 Fixed assorted flake8 errors. 2013-10-11 07:25:14 -04:00
Anssi Kääriäinen 93cc6dcdac Fixed #18414 -- qs.exists() for sliced distinct queries 2013-10-05 21:40:36 +03:00
Anssi Kääriäinen 068b1e045f Simplified a test by using CapturedQueriesContext 2013-10-05 21:16:09 +03:00
Anssi Kääriäinen ecaba36028 Improved Query join promotion logic
There were multiple cases where join promotion was a bit too aggressive.
This resulted in using outer joins where not necessary.

Refs #21150.
2013-10-05 14:52:17 +03:00
Anssi Kääriäinen bf13c75c0d Fixed #21203 -- resolve_columns fields misalignment
In queries using .defer() together with .select_related() the values
and fields arguments didn't align properly for resolve_columns().
2013-10-01 20:29:23 +03:00
Simon Charette c4fdd859ec Fixed #15786 -- Added a regression test for o2o excludes using F().
The issue was reported against 1.3.x but has been fixed since.
2013-09-29 16:45:03 -04:00
Michael Manfre b0e06c3662 Fixed #21167 - Improved queries.tests.SubqueryTests
Improve tests to cover slicing scenarios that could be handled in unique
ways by 3rd party database backends.
2013-09-29 17:35:42 +03:00
Simon Charette 4ec6d281cd Fixed #21001 -- Added a regression test for empty string exclusion on Oracle.
The issue was present in 1.4.x but has been reported to be fixed in master.

Thanks to @timgraham for the patch.
2013-09-23 14:18:03 -04:00
Aymeric Augustin b4cd8169de Fixed #11811 -- Data-loss bug in queryset.update.
It's now forbidden to call queryset.update(field=instance) when instance
hasn't been saved to the database ie. instance.pk is None.
2013-09-06 17:15:23 -05:00
Anssi Kääriäinen 8d65b6082c Fixed #20955 -- select_related regression
In cases where the same connection (from model A to model B along the
same field) was needed multiple times in a select_related query, the
join setup code mistakenly reused an existing join.
2013-08-22 10:51:07 +03:00
Anssi Kääriäinen 8dc76c4b91 Fixed test failure caused by different NULL ordering between backends 2013-08-20 11:33:44 +03:00
Anssi Kääriäinen 905409855c Fixed #14056 -- Made sure LEFT JOIN aren't trimmed in ORDER BY
If LEFT JOINs are required for correct results, then trimming the join
can lead to incorrect results. Consider case:

TBL A: ID | TBL B: ID  A_ID
       1           1   1
       2
Now A.order_by('b__a') did use a join to B, and B's a_id column. This
was seen to contain the same value as A's id, and so the join was
trimmed. But this wasn't correct as the join is LEFT JOIN, and for row
A.id = 2 the B.a_id column is NULL.
2013-08-20 10:55:00 +03:00
Anssi Kääriäinen 58c6d0209d Fixed #12807 -- EmptyResultSet ORed condition
The EmptyResultSet wasn't treated correctly so the end results was
incorrect, too. The bug had been already fixed in master so only tests
added.
2013-08-19 16:24:45 +03:00
Anssi Kääriäinen 09a5f5aabe Fixed test failure on MySQL
The fix for #20874 caused a MySQL specific failure.
2013-08-13 15:37:08 +03:00
Anssi Kääriäinen dcdc579d16 Fixed #20874 -- bump_prefix() in nested subqueries
Also made some cleanup to build_filter() code by introducing submethods
solve_lookup_type() and prepare_lookup_value().
2013-08-13 14:11:52 +03:00
Curtis Maloney 07876cf02b Deprecated SortedDict (replaced with collections.OrderedDict)
Thanks Loic Bistuer for the review.
2013-08-04 07:09:39 -04:00
Claude Paroz 5c1143910e Removed most of absolute_import imports
Should be unneeded with Python 2.7 and up.
Added some unicode_literals along the way.
2013-07-29 20:28:13 +02:00
Julien Phalip 47c755327b Fixed a number of minor misspellings. 2013-07-27 18:46:03 -07:00
Anssi Kääriäinen 7cca8d56d2 Fixed related model lookup regression
It has been possible to use models of wrong type in related field
lookups. For example pigs__in=[a_duck] has worked. Changes to
ForeignObject broke that.

It might be a good idea to restrict the model types usable in lookups.
This should be done intentionally, not accidentally and without any
consideration for deprecation path.
2013-07-26 13:13:27 +03:00
Anssi Kääriäinen c00a8525c6 Fixed #20788 -- exclude() generated subquery failure
"Fixed" by adding a test case, the original problem was already fixed
by earlier ORM changes. Thanks to david@judicata.com for the report
and test case.
2013-07-23 13:40:50 +03:00
Tim Graham 2456ffa42c Fixed #20746 -- Removed Python 2.6 specific code/docs 2013-07-14 13:02:55 -04:00
Aymeric Augustin cfcf4b3605 Stopped using django.utils.unittest in the test suite.
Refs #20680.
2013-07-01 14:29:33 +02:00
Florian Hahn 2f35c6f10f Fixed #14930 -- values_list() failure on qs ordered by extra column
Thanks lsaffre for the report and simon29, vicould, and Florian Hahn
for the patch.

Some changes done by committer.
2013-06-18 23:56:51 +03:00
Anssi Kääriäinen 89bf7a4525 Fixed #20528 -- regression in select_related join promotion
The join used by select_related was incorrectly INNER when the query
had an ORed filter for nullable join that was trimmed away. Fixed this
by forcing the join type to LOUTER even when a join was trimmed away
in ORed queries.
2013-06-14 16:29:06 +03:00
Tim Graham fffb55860e Changed assertEquals -> assertEqual. 2013-06-01 06:31:25 -04:00
Anssi Kääriäinen 84909377f2 Fixed #18785 -- Added Test join trimming regression
The regression was caused by patch to ticket #15316 and was fixed by a
patch to #10790.
2013-06-01 00:28:44 +03:00
Tim Graham 7426e72302 Fixed #20250 - Added a regression test for negated Q + annotate
Thanks nott.
2013-05-29 08:49:42 -04:00
Anssi Kääriäinen d467e11785 Fixed #20507 -- SubqueryConstraint alias relabeling
The SubqueryConstraint defined relabeled_clone(), but that was never
called. Instead there is now clone() and relabel_aliases() methods for
SubqueryConstraint.

A related problem was that SubqueryConstraint didn't correctly use
quote_name_unless_alias() of the outer query. This resulted in failures
when running under PostgreSQL.
2013-05-27 12:25:29 +03:00
Anssi Kääriäinen 904084611d Fixed #19895 -- Second iterator over invalid qs raises errors
The fix was already in the patch for #18702, so only test added for
ticket #19895.
2013-05-22 08:30:41 +03:00
Anssi Kääriäinen 70679243d1 Fixed #18702 -- Removed chunked reads from QuerySet iteration 2013-05-21 19:51:59 +03:00
Anssi Kääriäinen f53059b411 Fixed qs.values() regression when used in subquery 2013-05-20 15:38:47 +03:00
Anssi Kääriäinen e17fa9e877 Fixed #20091 -- Oracle null promotion for empty strings 2013-03-26 14:19:54 +02:00
Anssi Kääriäinen 80e68ee2ff Added tests for already fixed #20101
The ticket dealt with a case where one query had .exclude() that
produced a subquery, the other query had a join to the same model that
was subqueried in the first query. This was already fixed in master, so
only test added.
2013-03-21 00:43:26 +02:00
Anssi Kääriäinen d3f00bd570 Refactored qs.add_q() and utils/tree.py
The sql/query.py add_q method did a lot of where/having tree hacking to
get complex queries to work correctly. The logic was refactored so that
it should be simpler to understand. The new logic should also produce
leaner WHERE conditions.

The changes cascade somewhat, as some other parts of Django (like
add_filter() and WhereNode) expect boolean trees in certain format or
they fail to work. So to fix the add_q() one must fix utils/tree.py,
some things in add_filter(), WhereNode and so on.

This commit also fixed add_filter to see negate clauses up the path.
A query like .exclude(Q(reversefk__in=a_list)) didn't work similarly to
.filter(~Q(reversefk__in=a_list)). The reason for this is that only
the immediate parent negate clauses were seen by add_filter, and thus a
tree like AND: (NOT AND: (AND: condition)) will not be handled
correctly, as there is one intermediary AND node in the tree. The
example tree is generated by .exclude(~Q(reversefk__in=a_list)).

Still, aggregation lost connectors in OR cases, and F() objects and
aggregates in same filter clause caused GROUP BY problems on some
databases.

Fixed #17600, fixed #13198, fixed #17025, fixed #17000, fixed #11293.
2013-03-13 10:44:49 +02:00
Anssi Kääriäinen d744c550d5 Fixed #19964 -- Removed relabel_aliases from some structs
Before there was need to have both .relabel_aliases() and .clone() for
many structs. Now there is only relabeled_clone() for those structs
where alias is the only mutable attribute.
2013-03-12 21:33:47 +02:00