Commit Graph

1646 Commits

Author SHA1 Message Date
Anssi Kääriäinen a4e97cf315 Fixed #19707 -- Reset transaction state after requests 2013-02-10 13:55:54 +02:00
Claude Paroz 868e0f3cb4 Removed unneeded imports in db.models init 2013-02-08 17:11:33 +01:00
Claude Paroz e486475a13 Moved import in db.backends.creation 2013-02-08 10:01:56 +01:00
Simon Charette ec469ade2b Fixed #19689 -- Renamed `Model._meta.module_name` to `model_name`. 2013-02-05 04:16:07 -05:00
Claude Paroz 7c5b244826 Fixed #17061 -- Factored out importing object from a dotted path
Thanks Carl Meyer for the report.
2013-02-04 16:38:25 +01:00
Simon Charette fb606e10ac Fixed #19653 -- Removed `Manager.get_empty_query_set`. 2013-02-04 04:02:50 -05:00
Claude Paroz 3c0a81aaf0 Fixed #19709 -- Fixed TimeField introspection on MySQL
Thanks Eugene Grachev for the report.
2013-01-31 20:41:57 +01:00
Tim Graham ee26797cff Fixed typos in docs and comments 2013-01-29 10:55:55 -07:00
Aymeric Augustin f46d7314b5 Fixed #19677 -- Introspection of recursive foreign keys under SQLite.
Thanks Simon Charette.
2013-01-28 10:17:56 +01:00
Aymeric Augustin 6605ac331a Fixed #17158 -- Used a non-ambiguous representation of SQL queries
when a correct representation cannot be obtained.
2013-01-26 17:51:44 +01:00
Marc Tamlyn 389892aae5 Remove dup_select_related method.
This undocumented method was used in an old version of the admin, is
totally untested and hails from 2008. Although it's listed in the
"public methods" section, as it's not documented or used I don't think
it needs a deprecation path.

If we think it's useful I'll write some tests/docs for it instead...
2013-01-23 00:26:20 +00:00
Nick Sandford fe54377dae Fixed #17813 -- Added a .earliest() method to QuerySet
Thanks a lot to everybody participating in developing this feature.
The patch was developed by multiple people, at least Trac aliases
tonnzor, jimmysong, Fandekasp and slurms.

Stylistic changes added by committer.
2013-01-20 06:39:35 +02:00
Jani Tiainen d194f29057 Fixed #19606 -- Adjusted cx_Oracle unicode detection. 2013-01-19 09:27:16 +02:00
Anssi Kääriäinen 56f34f9f22 Removed models/related.py BoundRelatedObject
The class wasn't used anywhere except in RelatedObject.bind(), which
wasn't used anywhere. The class had one method defined as
NotImplemented, yet the class wasn't subclassed anywhere. In short, the
class was dead code.
2013-01-18 16:00:57 +02:00
Anssi Kääriäinen 3647c0a49a Avoided unnecessary recreation of RelatedObjects
Refs #19399. Thanks to Track alias KJ for the patch.
2013-01-18 07:09:39 +02:00
David Cramer abbb88886b Move logic seperation as its not longer repetitive 2013-01-14 13:17:01 -08:00
David Cramer 272de9eb6b Send post_delete signals immediately
In a normal relational construct, if you're listening for an event
that signals a child was deleted, you dont expect that the parent
was deleted already.

This change ensures that post_delete signals are fired immediately
after objects are deleted in the graph.
2013-01-14 13:15:47 -08:00
Claude Paroz 0171ba65db Fixed #17574 -- Implemented missing get_key_columns in PostgreSQL backend 2013-01-12 21:46:08 +01:00
Claude Paroz 223fc8eaf9 Return namedtuple from get_table_description
We don't make use of it currently to not break third-party db
backends.
2013-01-12 21:14:22 +01:00
Alex Gaynor a170c3f755 Removed some now dead code from deletion (thanks to Carl Meyer for noticing it). 2013-01-11 20:59:29 -08:00
Nick Sandford eb6c107624 Fixed #19360 -- Raised an explicit exception for aggregates on date/time fields in sqlite3
Thanks lsaffre for the report and Chris Medrela for the initial
patch.
2013-01-11 18:10:28 +01:00
Simon Charette f58efd07ff Fixed #19576 -- Use `six.with_metaclass` uniformously accross code base. 2013-01-08 20:13:43 +01:00
Anssi Kääriäinen 55da775ce1 Fixed #17541 -- Fixed non-saved/nullable fk querying 2013-01-08 21:02:38 +02:00
Anssi Kääriäinen 23ca3a0194 Fixed #16759 -- Remove use of __deepcopy__ in qs.clone()
The original problem was that queryset cloning was really expensive
when filtering with F() clauses. The __deepcopy__ went too deep copying
_meta attributes of the models used. To fix this the use of
__deepcopy__ in qs cloning was removed.

This commit results in some speed improvements across the djangobench
benchmark suite. Most query_* tests are 20-30% faster, save() is 50%
faster and finally complex filtering situations can see 2x to order
of magnitude improvments.

Thanks to Suor, Alex and lrekucki for valuable feedback.
2013-01-08 19:17:13 +02:00
Claude Paroz c698c55966 Created special PostgreSQL text indexes when unique is True
Refs #19441.
2013-01-07 17:54:30 +01:00
Anssi Kääriäinen a2396a4c8f Fixed #19173 -- Made EmptyQuerySet a marker class only
The guarantee that no queries will be made when accessing results is
done by new EmptyWhere class which is used for query.where and having.

Thanks to Simon Charette for reviewing and valuable suggestions.
2013-01-06 19:18:28 +02:00
Claude Paroz b740da3504 Fixed #19192 -- Allowed running tests with dummy db backend
Thanks Simon Charette for the initial patch, and Jan Bednařík for
his work on the ticket.
2013-01-04 13:55:20 +01:00
mpaolini 6248833d9e Added documentation for the 'db' argument of the post-syncdb signal. 2013-01-03 22:04:55 +01:00
Aymeric Augustin a051a9d929 Fixed PR 478 -- Removed superfluous try/except block. 2013-01-02 22:46:36 +01:00
Aymeric Augustin 07fbc6ae0e Fixed #19547 -- Caching of related instances.
When &'ing or |'ing querysets, wrong values could be cached, and crashes
could happen.

Thanks Marc Tamlyn for figuring out the problem and writing the patch.
2013-01-02 22:21:46 +01:00
Anssi Kääriäinen ce3c71faf1 Minor improvement to proxy model handling
Refs #19385
2012-12-30 11:20:27 +02:00
Anssi Kääriäinen 807eff7439 Made use of PathInfo.direct flag in trim_joins
Refs #19385
2012-12-30 11:20:13 +02:00
Anssi Kääriäinen 68985db482 Added Query.join_parent_model()
This simplifies especially compiler.py a lot, where almost the same
code was repeated multiple times.

Refs #19385
2012-12-30 11:19:35 +02:00
Anssi Kääriäinen 4511aeb6b8 Moved join path generation to Field
Refs #19385
2012-12-30 11:19:19 +02:00
Anssi Kääriäinen 9ef3cab40b Made sure join_field is always available in .join()
Refs #19385
2012-12-30 11:18:25 +02:00
Aymeric Augustin ef017a5f00 Advanced pending deprecation warnings.
Also added stacklevel argument, fixed #18127.
2012-12-29 21:59:07 +01:00
Aymeric Augustin e9c24bef74 Fix #19524 -- Incorrect caching of parents of unsaved model instances.
Thanks qcwxezdas for the report. Refs #13839.
2012-12-28 23:34:54 +01:00
Aymeric Augustin db278c3bf9 Fixed #19525 -- Reverted dcd4383107 and 05d333ba3b.
Refs #9893, #18515.

Thanks Russell for the report.
2012-12-27 09:37:57 +01:00
Anssi Kääriäinen 4007c8f6eb Fixed a regression in distinct_on
Caused by regression fix for #19500.
2012-12-20 22:50:06 +02:00
Anssi Kääriäinen d407164c04 Fixed #18854 -- Join promotion in disjunction cases
The added promotion logic is based on promoting any joins used in only
some of the childs of an OR clause unless the join existed before the
OR clause addition.
2012-12-20 21:45:15 +02:00
Anssi Kääriäinen 3dcd435a0e Fixed #19500 -- Solved a regression in join reuse
The ORM didn't reuse joins for direct foreign key traversals when using
chained filters. For example:
    qs.filter(fk__somefield=1).filter(fk__somefield=2))
produced two joins.

As a bonus, reverse onetoone filters can now reuse joins correctly

The regression was caused by the join() method refactor in commit
68847135bc

Thanks for Simon Charette for spotting some issues with the first draft
of the patch.
2012-12-20 21:27:00 +02:00
Russell Keith-Magee c04c03daa3 Fixed #19401 -- Ensure that swappable model references are case insensitive.
This is necessary because get_model() checks are case insensitive, and if the swapable check isn't, the
swappable logic gets tied up in knots with models that are partially swapped out.

Thanks to chris@cogdon.org for the report and extensive analysis, and Preston for his work on the draft patch.
2012-12-20 16:10:19 +08:00
Claude Paroz 55972ee5c7 Fixed #19441 -- Created PostgreSQL varchar index when unique=True
Thanks Dylan Verheul for the report and Anssi Kääriäinen for the
review.
2012-12-18 09:56:30 +01:00
Anssi Kääriäinen c64b57d166 Replaced '__' with LOOKUP_SEP in sql/query.py
Thanks to Simon Charette for report.
2012-12-17 21:15:01 +02:00
Anssi Kääriäinen 12a96bfa26 Fixed #19197 -- fixed convert_values() for nullable numeric fields
Cleaned up the implementation of base convert_values() a little, and
made sure it accepts None as a value for numeric fields.

There are no tests attached. The reason is that not all of the
convert_values() accept None as a value for numeric fields (for example
sqlite3.convert_values()).

The reason the base convert_values() needs to accept None is that this
situation might arise in custom compilers for 3rd party backends. It
is easy to keep the convert_values() working, so lets do that.
2012-12-16 23:26:16 +02:00
Anssi Kääriäinen 704ee33f50 Fixed #16679 -- Use caching to speed up signal sending 2012-12-16 21:46:02 +02:00
Anssi Kääriäinen 69597e5bcc Fixed #10790 -- Refactored sql.Query.setup_joins()
This is a rather large refactoring. The "lookup traversal" code was
splitted out from the setup_joins. There is now names_to_path() method
which does the lookup traveling, the actual work of setup_joins() is
calling names_to_path() and then adding the joins found into the query.

As a side effect it was possible to remove the "process_extra"
functionality used by genric relations. This never worked for left
joins. Now the extra restriction is appended directly to the join
condition instead of the where clause.

To generate the extra condition we need to have the join field
available in the compiler. This has the side-effect that we need more
ugly code in Query.__getstate__ and __setstate__ as Field objects
aren't pickleable.

The join trimming code got a big change - now we trim all direct joins
and never trim reverse joins. This also fixes the problem in #10790
which was join trimming in null filter cases.
2012-12-16 17:23:26 +02:00
Anssi Kääriäinen f811649710 Fixed #18816 -- Removed "trim" argument from add_filter()
The trim argument was used by split_exclude() only to trim the last
join from the given lookup. It is cleaner to just trim the last part
from the lookup in split_exclude() directly so that there is no need
to burden add_filter() with the logic needed for only split_exclude().
2012-12-16 16:56:28 +02:00
Anssi Kääriäinen d7b49f5b0d Fixed #19469 -- Removed opts.get_ordered_objects() and related code
The code was dead-code since 2006.
2012-12-16 16:22:09 +02:00
Aymeric Augustin a001f3c31e Fixed #2304 -- Documented TRANSACTIONS_MANAGED. 2012-12-14 21:15:48 +01:00
Claude Paroz c91667338a Fixed #19357 -- Allow non-ASCII chars in filesystem paths
Thanks kujiu for the report and Aymeric Augustin for the review.
2012-12-08 11:13:52 +01:00
orblivion 0ff1d4fa49 Typeo in ValuesQuerySet._as_sql docstring 2012-12-01 18:55:26 -08:00
Tai Lee 6ebf115206 Fixed #14694 -- Made ``defer()`` work with reverse relations
Reverse o2o fields are now usable with defer.
2012-11-28 18:17:10 +02:00
Anssi Kääriäinen d37483c533 Removed duplicate opts.pk_index() method 2012-11-27 21:54:46 +02:00
Anssi Kääriäinen 905ea9619b Made sure global settings are changed in test db creation
There was an assumption that changing connection.settings_dict changed
also the settings.DATABASES values. This assumption is now gone.
2012-11-27 19:47:20 +02:00
Anssi Kääriäinen 1893467784 Fixed #19274 -- Made db connection creation overridable in subclasses
Connection creation was done in db backend ._cursor() call. This
included taking a new connection if needed, initializing the session
state for the new connection and finally creating the connection.

To allow easier modifying of these steps in subclasses (for example to
support connection pools) the _cursor() now calls get_new_connection()
and init_connection_state() if there isn't an existing connection. This
was done for all non-gis core backends. In addition the parameters used
for taking a connection are now created by get_connection_params().

We should also do the same for gis backends and encourage 3rd party
backends to use the same pattern. The pattern is not enforced in code,
and as the backends are private API this will not be required by
documentation either.
2012-11-27 19:47:19 +02:00
Aymeric Augustin 2ea80b94d7 Fixed #19362 -- Detected invalid use of @python_2_unicode_compatible.
Thanks m3wolf for the report and akaariai for reproducing the problem.
2012-11-27 09:45:37 +01:00
Anssi Kääriäinen 08b4a22293 Updated stale docstring of setup_joins 2012-11-24 16:26:05 +02:00
Russell Keith-Magee c8985a8a73 Fixed #19806 -- Ensure that content types and permissions aren't created for swapped models.
Thanks to rizumu for the report.
2012-11-24 13:43:20 +08:00
Anssi Kääriäinen 0a0a0d66b3 Fixed #19351 -- SQLite bulk_insert of more than 500 single-field objs 2012-11-24 01:13:12 +02:00
Anssi Kääriäinen a27582484c Fixed SQLite's collapsing of same-valued instances in bulk_create
SQLite used INSERT INTO tbl SELECT %s UNION SELECT %s, the problem
was that there should have been UNION ALL instead of UNION.

Refs #19351
2012-11-24 01:12:17 +02:00
Anssi Kääriäinen 90b86291d0 Fixed #18375 -- Removed dict-ordering dependency for F-expressions
F() expressions reuse joins like any lookup in a .filter() call -
reuse multijoins generated in the same .filter() call else generate
new joins. Also, lookups can now reuse joins generated by F().

This change is backwards incompatible, but it is required to prevent
dict randomization from generating different queries depending on
.filter() kwarg ordering. The new way is also more consistent in how
joins are reused.
2012-11-23 19:53:04 +02:00
Claude Paroz 8c69278764 Fixed #18989 -- Removed unused condition in CursorWrapper
Thanks zimnyx for the report.
2012-11-17 16:34:14 +01:00
Anssi Kääriäinen 1194a96999 Fixed a regression in select_related
The regression was caused by the fix for #13781 (commit
f51e409a5f). Reason was leaving
off some crucial lines when resolving a merge conflict.
2012-11-15 20:17:57 +02:00
Anssi Kääriäinen f51e409a5f Fixed #13781 -- Improved select_related in inheritance situations
The select_related code got confused when it needed to travel a
reverse relation to a model which had different parent than the
originally travelled relation.

Thanks to Trac aliases shauncutts for report and ungenio for original
patch (committed patch is somewhat modified version of that).
2012-11-15 17:15:21 +02:00
Anssi Kääriäinen 92d7f541da Fixed #19058 -- Fixed Oracle GIS crash
The problem is the same as in #10888 which was reintroduced when
bulk_insert was added. Thanks to Jani Tiainen for report, patch and
also testing the final patch on Oracle GIS.
2012-11-15 16:08:06 +02:00
Anssi Kääriäinen ebcf6b36ff Fixed select_related performance regressions
The regression was caused by select_related fix for Oracle, commit
c159d9cec0.
2012-11-13 22:52:58 +02:00
Anssi Kääriäinen ce1af8d702 Removed use of SortedDict for query.alias_refcount
This will have a smallish impact on performance. Refs #19276.
2012-11-13 22:48:25 +02:00
Anssi Kääriäinen cafb266954 Fixed #17144 -- MySQL again groups by PK only
Thanks to Christian Oudard for the report and tests.
2012-11-08 00:56:32 +02:00
Mike Johnson 088f68252d use six.string_types for python3 2012-11-04 14:16:32 -08:00
Mike Johnson fcd3d4c68f model_split: Fixed #19236 - fixed error for abstract models with a split method 2012-11-04 13:43:54 -08:00
Alex Gaynor 4285571c5a Fixed #5805 -- it is now possible to specify multi-column indexes. Thanks to jgelens for the original patch. 2012-11-04 10:16:06 -08:00
Aymeric Augustin fc10418fba Fixed #18963 -- Used a subclass-friendly pattern
for Python 2 object model compatibility methods.
2012-11-03 22:07:35 +01:00
Preston Holmes 965cc0b1ff Deprecated depth kwarg on select_related.
This is the start of a deprecation path for the depth kwarg on
select_related. Removing this will allow us to update select_related so
it chains properly and have an API similar to prefetch_related.

Thanks to Marc Tamlyn for spearheading and initial patch.

refs #16855
2012-11-02 22:03:33 -07:00
Anssi Kääriäinen 68847135bc Removed dupe_avoidance from sql/query and sql/compiler.py
The dupe avoidance logic was removed as it doesn't seem to do anything,
it is complicated, and it has nearly zero documentation.

The removal of dupe_avoidance allowed for refactoring of both the
implementation and signature of Query.join(). This refactoring cascades
again to some other parts. The most significant of them is the changes
in qs.combine(), and compiler.select_related_descent().
2012-10-31 08:19:44 +02:00
Claude Paroz f1cc2be0c5 Fixed #18575 -- Empty DATABASES should default to dummy backend
Thanks delormemarco@gmail.com for the report.
2012-10-28 23:44:03 +01:00
Anssi Kääriäinen 611c4d6f1c Fixed #18823 -- Ensured m2m.clear() works when using through+to_field
There was a potential data-loss issue involved -- when clearing
instance's m2m assignments it was possible some other instance's
m2m data was deleted instead.

This commit also improved None handling for to_field cases.
2012-10-28 17:31:35 +02:00
Anssi Kääriäinen c159d9cec0 Fixed Oracle failure caused by None converted to '' in select_related case 2012-10-27 05:26:53 +03:00
Anssi Kääriäinen 2249bd275c Fixed Oracle failure for "%" in table name 2012-10-27 05:26:42 +03:00
Anssi Kääriäinen 11699ac4b5 Fixed #19190 -- Refactored Query select clause attributes
The Query.select and Query.select_fields were collapsed into one list
because the attributes had to be always in sync. Now that they are in
one attribute it is impossible to edit them out of sync.

Similar collapse was done for Query.related_select_cols and
Query.related_select_fields.
2012-10-27 02:13:02 +03:00
Luke Plant f3a2bcdee9 Fixed #15040 - Boolean fields return 0 and 1 when loaded through select_related
Thanks to homm for the report and ramiro for the patch.
2012-10-26 00:25:59 +01:00
Anssi Kääriäinen bd6d9ea87c Fixed regression caused by #19102 2012-10-25 19:13:10 +03:00
Anssi Kääriäinen 7de439f32d Fixed #19187 -- Raise consistent error from qs.values().delete() 2012-10-25 17:16:56 +03:00
Anssi Kääriäinen 11b8712cc7 Added docstring to DeleteQuery.delete_qs() 2012-10-25 17:16:56 +03:00
Anssi Kääriäinen f64a5ef404 Fixed #19102 -- Fixed fast-path delete for modified SELECT clause cases
There was a bug introduced in #18676 which caused fast-path deletes
implemented as "DELETE WHERE pk IN <subquery>" to fail if the SELECT
clause contained additional stuff (for example extra() and annotate()).

Thanks to Trac alias pressureman for spotting this regression.
2012-10-25 17:16:44 +03:00
Aymeric Augustin c736a1a9f8 Fixed #19176 -- Typo in copy-pasted error message.
Thanks maxirobaina for the report.
2012-10-24 16:21:30 +02:00
Jan Bednařík 4cceb5cb31 Fixed #19162 -- Wrong indentation. 2012-10-22 19:23:19 +02:00
Jan Bednařík b87e2f46c8 Fixed #19151 -- Added missing methods to EmptyQuerySet.
Added values() and values_list() methods to EmptyQuerySet.
2012-10-21 00:19:38 +02:00
Anssi Kääriäinen b625e8272b Moved F() '&' and '|' to .bitand() and .bitor()
Done for consistency with Q() expressions and QuerySet combining. This
will allow usage of '&' and '|' as boolean logical operators in the
future. Refs #16211.
2012-10-10 01:15:29 +03:00
Anssi Kääriäinen a8b1861fc4 Revert "Fixed #16211 -- Added comparison and negation ops to F() expressions"
This reverts commit 28abf5f0eb.

Conflicts:

	docs/releases/1.5.txt
2012-10-10 01:15:29 +03:00
Michael Manfre c2150d4d2c Fixed #19096 -- Made can_return_id_from_insert more extendable
RETURNING is an extension of the SQL standard, which is not implemented
the same by all databases. Allow DatabaseOperations.return_insert_id to
return a None to allow for other 3rd party backends with a different
implementation.
2012-10-10 01:00:58 +03:00
Aymeric Augustin 0a0fe8f71d Fix exception message from 3190abcd. Refs #18153.
Thanks Preston Holmes.
2012-10-09 22:06:55 +02:00
Aymeric Augustin 3190abcd75 Fixed #18153 -- Reverse OneToOne lookups on unsaved instances.
Thanks David Hatch and Anssi Kääriäinen for their inputs.
2012-10-09 21:36:35 +02:00
Anssi Kääriäinen a62d53c032 Fixed #19087 -- Ensured query's base table is never LOUTER joined
This fixes a regression created by join promotion logic refactoring:
01b9c3d519

Thanks to Ivan Virabyan for the report.
2012-10-08 18:40:09 +03:00
Claude Paroz 34a736b752 Used pkgutil to get list of backend modules
Refs #18827.
2012-10-07 21:59:16 +02:00
Ramiro Morales 35e8dc5169 Removed ad-hoc support for usage of short names of built-in DB backends.
This non-standard naming was deprecated in Django 1.2.
2012-10-06 23:27:08 -03:00
Ramiro Morales 2100da9dcd Ensure we ignore __pycache__ PEP 3174 dirs in a few more places. 2012-10-06 18:40:58 -03:00
Russell Keith-Magee cc337a74f1 Fixed #19069 -- Improved the error message when trying to query a swapped model.
Thanks to Preston Holmes for the suggestion.
2012-10-06 14:21:57 +08:00
Russell Keith-Magee 43530384b7 Fixed #19049 -- Corrected dumb logic negation error from earlier patch. 2012-10-03 09:14:55 +08:00
Russell Keith-Magee 3b6f980bed Fixed #19049 -- Ensure that swapped models aren't included in reverse field caches.
Thanks to Ivan Virabyan for the report.
2012-10-02 22:52:45 +08:00
Anssi Kääriäinen 28abf5f0eb Fixed #16211 -- Added comparison and negation ops to F() expressions
Work done by Walter Doekes and Trac alias knoeb. Reviewed by Simon
Charette.
2012-09-30 17:51:06 +03:00
Anssi Kääriäinen ddd7d1af20 Avoided storing ExpressionNodes in dicts 2012-09-30 17:50:56 +03:00
Anssi Kääriäinen 1cd6e04cd4 Fixed #18676 -- Allow fast-path deletion of objects
Objects can be fast-path deleted if there are no signals, and there are
no further cascades. If fast-path is taken, the objects do not need to
be loaded into memory before deletion.

Thanks to Jeremy Dunck, Simon Charette and Alex Gaynor for reviewing
the patch.
2012-09-28 18:16:08 +03:00
Adrien Lemaire 2c8267bf3d Fixed #17899 -- Rewrote [Ee]-mail to [Ee]mail 2012-09-26 14:14:51 +02:00
Russell Keith-Magee 70a0de37d1 Fixed #3011 -- Added swappable auth.User models.
Thanks to the many people that contributed to the development and review of
this patch, including (but not limited to) Jacob Kaplan-Moss, Anssi
Kääriäinen, Ramiro Morales, Preston Holmes, Josh Ourisman, Thomas Sutton,
and Roger Barnes, as well as the many, many people who have contributed to
the design discussion around this ticket over many years.

Squashed commit of the following:

commit d84749a0f0
Merge: 531e771 7c11b1a
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Wed Sep 26 18:37:04 2012 +0800

    Merge remote-tracking branch 'django/master' into t3011

commit 531e7715da
Merge: 29d1abb 1f84b04
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Wed Sep 26 07:09:23 2012 +0800

    Merged recent trunk changes.

commit 29d1abbe35
Merge: 8a527dd 54c81a1
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Mon Sep 24 07:49:46 2012 +0800

    Merge remote-tracking branch 'django/master' into t3011

commit 8a527dda13
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Mon Sep 24 07:48:05 2012 +0800

    Ensure sequences are reset correctly in the presence of swapped models.

commit e2b6e22f29
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 23 17:53:05 2012 +0800

    Modifications to the handling and docs for auth forms.

commit 98aba856b5
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 23 15:28:57 2012 +0800

    Improved error handling and docs for get_user_model()

commit 0229209c84
Merge: 6494bf9 8599f64
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 23 14:50:11 2012 +0800

    Merged recent Django trunk changes.

commit 6494bf91f2
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Mon Sep 17 21:38:44 2012 +0800

    Improved validation of swappable model settings.

commit 5a04cde342
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Mon Sep 17 07:15:14 2012 +0800

    Removed some unused imports.

commit ffd535e413
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 20:31:28 2012 +0800

    Corrected attribute access on for get_by_natural_key

commit 913e1ac84c
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 20:12:34 2012 +0800

    Added test for proxy model safeguards on swappable models.

commit 280bf19e94
Merge: dbb3900 935a863
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 18:16:49 2012 +0800

    Merge remote-tracking branch 'django/master' into t3011

commit dbb3900775
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 18:09:27 2012 +0800

    Fixes for Python 3 compatibility.

commit dfd72131d8
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 15:54:30 2012 +0800

    Added protection against proxying swapped models.

commit abcb027190
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 15:11:10 2012 +0800

    Cleanup and documentation of AbstractUser base class.

commit a9491a8776
Merge: fd8bb4e 08bcb4a
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 14:46:49 2012 +0800

    Merge commit '08bcb4aec1ed154cefc631b8510ee13e9af0c19d' into t3011

commit fd8bb4e3e4
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 14:20:14 2012 +0800

    Documentation improvements coming from community review.

commit b550a6d06d
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 13:52:47 2012 +0800

    Refactored skipIfCustomUser into the contrib.auth tests.

commit 52a02f1110
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 13:46:10 2012 +0800

    Refactored common 'get' pattern into manager method.

commit b441a6bbc7
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 16 13:41:33 2012 +0800

    Added note about backwards incompatible change to admin login messages.

commit 08bcb4aec1
Author: Anssi Kääriäinen <akaariai@gmail.com>
Date:   Sat Sep 15 18:30:33 2012 +0300

    Splitted User to AbstractUser and User

commit d9f5e5addb
Author: Anssi Kääriäinen <akaariai@gmail.com>
Date:   Sat Sep 15 18:30:02 2012 +0300

    Reworked REQUIRED_FIELDS + create_user() interaction

commit 579f152e4a
Merge: 9184972 93e6733
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sat Sep 15 20:18:37 2012 +0800

    Merge remote-tracking branch 'django/master' into t3011

commit 918497218c
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sat Sep 15 20:18:19 2012 +0800

    Deprecate AUTH_PROFILE_MODULE and get_profile().

commit 334cdfc1bb
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sat Sep 15 20:00:12 2012 +0800

    Added release notes for new swappable User feature.

commit 5d7bb22e8d
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sat Sep 15 19:59:49 2012 +0800

    Ensure swapped models can't be queried.

commit 57ac6e3d32
Merge: f2ec915 abfba3b
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sat Sep 15 14:31:54 2012 +0800

    Merge remote-tracking branch 'django/master' into t3011

commit f2ec915b20
Merge: 1952656 5e99a3d
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 9 08:29:51 2012 +0800

    Merge remote-tracking branch 'django/master' into t3011

commit 19526563b5
Merge: 2c5e833 c4aa26a
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 9 08:22:26 2012 +0800

    Merge recent changes from master.

commit 2c5e833a30
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 9 07:53:46 2012 +0800

    Corrected admin_views tests following removal of the email fallback on admin logins.

commit 20d1892491
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sun Sep 9 01:00:37 2012 +0800

    Added conditional skips for all tests dependent on the default User model

commit 40ea8b8882
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sat Sep 8 23:47:02 2012 +0800

    Added documentation for REQUIRED_FIELDS in custom auth.

commit e6aaf65970
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Sat Sep 8 23:20:02 2012 +0800

    Added first draft of custom User docs.

    Thanks to Greg Turner for the initial text.

commit 75118bd242
Author: Thomas Sutton <me@thomas-sutton.id.au>
Date:   Mon Aug 20 11:17:26 2012 +0800

    Admin app should not allow username discovery

    The admin app login form should not allow users to discover the username
    associated with an email address.

commit d088b3af58
Author: Thomas Sutton <me@thomas-sutton.id.au>
Date:   Mon Aug 20 10:32:13 2012 +0800

    Admin app login form should use swapped user model

commit 7e82e83d67
Merge: e29c010 39aa890
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Fri Sep 7 23:45:03 2012 +0800

    Merged master changes.

commit e29c010beb
Merge: 8e3fd70 30bdf22
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Mon Aug 20 13:12:57 2012 +0800

    Merge remote-tracking branch 'django/master' into t3011

commit 8e3fd703d0
Merge: 507bb50 26e0ba0
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Mon Aug 20 13:09:09 2012 +0800

    Merged recent changes from trunk.

commit 507bb50a92
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Mon Jun 4 20:41:37 2012 +0800

    Modified auth app so that login with alternate auth app is possible.

commit dabe362836
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Mon Jun 4 20:10:51 2012 +0800

    Modified auth management commands to handle custom user definitions.

commit 7cc0baf89d
Author: Russell Keith-Magee <russell@keith-magee.com>
Date:   Mon Jun 4 14:17:28 2012 +0800

    Added model Meta option for swappable models, and made auth.User a swappable model
2012-09-26 18:48:09 +08:00
Claude Paroz 26ff2be787 Imported getLogger directly from logging module
This was a remainder of some 2.4 compatibility code.
2012-09-20 21:03:24 +02:00
Carl Meyer 4e9a74b81d Revert "Fixed #16865 -- Made get_or_create use read database for initial get query."
Thanks to Jeremy Dunck for pointing out the problem with this change. If in a
single transaction, the master deletes a record and then get_or_creates a
similar record, under the new behavior the get_or_create would find the record
in the slave db and fail to re-create it, leaving the record nonexistent, which
violates the contract of get_or_create that the record should always exist
afterwards. We need to do everything against the master here in order to ensure
correctness.

This reverts commit 901af86550.
2012-09-19 11:15:12 -06:00
Carl Meyer 901af86550 Fixed #16865 -- Made get_or_create use read database for initial get query.
Thanks Rick van Hattem for the report and trbs for the patch.
2012-09-19 10:06:53 -06:00
Anssi Kääriäinen f399a804c9 Fixed #17485 regression -- only + select_related interaction
When doing deeper than one level select_related() + only queries(), the
code introduced in b6c356b7bb errored
incorrectly.

Thanks to mrmachine for report & test case.
2012-09-16 22:58:40 +03:00
Claude Paroz 859aa2a6c4 Fixed #18790 -- Encoded database password on Python 2
Thanks thcourbon@gmail.com for the report.
2012-09-12 10:16:49 +02:00
Malcolm Tredinnick c4aa26a983 Internal refactoring; moving LOOKUP_SEP up one level.
In an ideal world, nothing except django.db.models.query should have to
import stuff from django.models.sql.*. A few things were needing to get
hold of sql.constants.LOOKUP_SEP, so this commit moves it up to
django.db.models.constants.LOOKUP_SEP.

There are still a couple of places (admin) poking into sql.* to get
QUERY_TERMS, which is unfortunate, but a slightly different issue and
harder to adjust.
2012-09-08 19:51:36 -04:00
Claude Paroz 09a99714c0 Moved get_primary_key_column implementation to base
Refs #17574.
2012-09-08 10:24:13 +02:00
Alex Gaynor ad50243cd1 [py3k] Fixed the index creation code that I committed a few minutes ago for py3k. 2012-09-07 15:33:02 -04:00
Alex Gaynor 335a9f9cf1 Removed many uses of bare "except:", which were either going to a) silence real issues, or b) were impossible to hit. 2012-09-07 15:08:07 -04:00
Alex Gaynor e4ea536774 Ensued that SQL indexes are alwasy created in the same name.
Previous this used Python's builtin hash() function, which has never been guarnteed to be stable across implementations (CPython/Jython/etc.) or 32/64 bitness. However, this in practice it was stable. However, with the impending release of Python 3.3 hash randomizations is enabled by default, which would mean the index name changed between program invocations.
2012-09-07 14:14:06 -04:00
Alex Gaynor 0e296131bb Cleaned up some small bits of the ORM, including removing an import *. 2012-09-07 10:58:17 -04:00
Claude Paroz 306d34873c Fixed #18212 -- Standardized arguments of GenericIPAddressField
Unlike other model fields, the newly introduced (1.4)
GenericIPAddressField did not accept verbose_name and name as the
first positional arguments. This commit fixes it.
Thanks Dan McGee for the report and the patch.
2012-09-01 18:38:55 +02:00
Claude Paroz a9a773ff38 Made get_table_description also return the size of char fields on SQLite 2012-08-30 19:36:05 +02:00
Claude Paroz 879b245baa Fixed #5725 -- Fixed varchar column size introspection for MySQL
Thanks ferdonline for the initial patch and Karen Tracey for the
related post on django-users.
2012-08-30 19:28:13 +02:00
Claude Paroz ae88e73fa6 Replaced some smart_xxx by force_xxx equivalent
smart_str/smart_text should only be used when a potential lazy
string should be preserved in the result of the function call.
2012-08-30 15:46:16 +02:00
Claude Paroz ebc773ada3 Replaced many smart_bytes by force_bytes
In all those occurrences, we didn't care about preserving the
lazy status of the strings, but we really wanted to obtain a
real bytestring.
2012-08-29 11:20:32 +02:00
Mitar b2ffa2ea90 Code comment fix. 2012-08-26 14:28:44 -07:00
Anssi Kääriäinen 01b9c3d519 Fixed #16715 -- Fixed join promotion logic for nested nullable FKs
The joins for nested nullable foreign keys were often created as INNER
when they should have been OUTER joins. The reason was that only the
first join in the chain was promoted correctly. There were also issues
with select_related etc.

The basic structure for this problem was:
  A -[nullable]-> B -[nonnull]-> C

And the basic problem was that the A->B join was correctly LOUTER,
the B->C join not.

The major change taken in this patch is that now if we promote a join
A->B, we will automatically promote joins B->X for all X in the query.
Also, we now make sure there aren't ever join chains like:
   a LOUTER b INNER c
If the a -> b needs to be LOUTER, then the INNER at the end of the
chain will cancel the LOUTER join and we have a broken query.

Sebastian reported this problem and did also major portions of the
patch.
2012-08-25 14:14:45 +03:00
Alex Gaynor b7c3b044fc Cleaned up the implementation of in_bulk 2012-08-24 16:08:16 -07:00
Aymeric Augustin 5301a9d7b1 [py3] Removed duplicate imports.
Fixed #18837. Refs #18791.
2012-08-23 10:16:16 +02:00
Simon Meers dfe63a52ef Revert "Fixed #18063 -- Avoid unicode in Model.__repr__ in python 2"
This reverts commit 3fce0d2a91.
2012-08-22 11:48:47 +10:00
Anssi Kääriäinen a193372753 Fixed #17886 -- Fixed join promotion in ORed nullable queries
The ORM generated a query with INNER JOIN instead of LEFT OUTER JOIN
in a somewhat complicated case. The main issue was that there was a
chain of nullable FK -> non-nullble FK, and the join promotion logic
didn't see the need to promote the non-nullable FK even if the
previous nullable FK was already promoted to LOUTER JOIN. This resulted
in a query like a LOUTER b INNER c, which incorrectly prunes results.
2012-08-21 21:23:57 +03:00
Claude Paroz bfc380baea [py3] Prepared MySQL backend for Python 3 compatibility 2012-08-20 10:46:21 +02:00
Simon Meers 3fce0d2a91 Fixed #18063 -- Avoid unicode in Model.__repr__ in python 2
Thanks guettli and mrmachine.
2012-08-20 16:47:30 +10:00
Anssi Kääriäinen 4db38cbfe1 [py3] Fixed Oracle specific failures 2012-08-19 23:45:26 +03:00
Aymeric Augustin 547b181046 [py3] Ported django.utils.safestring.
Backwards compatibility aliases were created under Python 2.
2012-08-18 16:04:06 +02:00
Aymeric Augustin f34de7dd6e [py3] Fixed backends tests. 2012-08-18 11:02:38 +02:00
Aymeric Augustin 4c1286cf78 [py3] Added compatibility import of thread/_thread
This commit fixes the auto-reload of the development server.

I should have done that change in ca07fda2.
2012-08-18 10:56:56 +02:00
Dmitry Shevchenko 4412de5d8d [py3] Fixed installed_models filtering. 2012-08-16 01:08:45 -05:00
Claude Paroz de2cb5fede [py3] Fixed F-expression right-hand division
Complementary to commit 62a9ed0ac.
2012-08-15 13:28:47 +02:00
Claude Paroz 187ec5a166 [py3] Pass bytes to md5 in truncate_name 2012-08-15 12:37:08 +02:00
Claude Paroz f8ea12f36b [py3] Removed excessive usage of smart_bytes 2012-08-15 12:13:54 +02:00
Aymeric Augustin e091c18f50 [py3] Removed a remaining use of __metaclass__. 2012-08-14 23:45:12 +02:00
Claude Paroz 9cb80356fc [py3] Fed strftime with unicode on Python 3 2012-08-14 16:19:58 +02:00
Anssi Kääriäinen 62a9ed0ac7 [py3] Fixed F-expression division operators
In Python 3 dividing by int will call obj.__truediv__(). This operator
was missing from F-expressions.
2012-08-14 16:07:32 +03:00
Claude Paroz c2d59e5564 [py3] Fixed admin_views tests
Also changed several occurrences of 'request' to 'response'.
2012-08-14 14:45:28 +02:00
Aymeric Augustin 0c198b85a3 [py3] Replace filter/lambda by list comprehensions
This is more idiomatic and avoids returning a list on Python 2 and
an iterator on Python 3.
2012-08-14 14:31:06 +02:00
Aymeric Augustin 9299dc42ed [py3] Removed unnecessary calls to .keys()
when computing the length of a dictionary. This fails on Python 3.
2012-08-14 14:09:23 +02:00
Anssi Kääriäinen 1930b899bd Refix #13844 -- Made FloatField aggregates work on Python 2.6 + Postgres
Fixed a regression introduced in 59a655988e.
2012-08-13 09:15:20 +03:00
Andrei Antoukh 99321e30ce Fixed #18306 -- Made deferred models issue update_fields on save
Deferred models now automatically update only the fields which are
loaded from the db (with .only() or .defer()). In addition, any field
set manually after the load is updated on save.
2012-08-12 22:39:27 +03:00
Anssi Kääriäinen 59a655988e Fixed #13844 -- Avoid converting unknown db values to float
This patch removes an unconditional float(value) conversion from db
backend default convert_values() method. This can cause problems when
aggregating over character fields for example. In addition, Oracle
and SQLite already return the bare value from their convert_values().

In the long term the converting should be done by fields, and the
fields should then call database backend specific converters when
needed. The current setup is inflexible for 3rd party fields.

Thanks to Merlijn van Deen for the original patch.
2012-08-12 21:52:52 +03:00
Claude Paroz 5513480fe1 [py3] Always convert values from sqlite3 to unicode strings
Thanks Aymeric Augustin for the review.
2012-08-12 20:47:18 +02:00
Claude Paroz dce34dc969 [py3] Made __repr__ return str with Python 3 2012-08-12 20:45:39 +02:00
Anssi Kääriäinen c1684e3dcb Fixed #18731 -- Cleaned up split_exclude's use of can_reuse
The outer query's set of reusable joins (can_reuse) was passed to the
inner query's add_filter call. This was incorrect.
2012-08-12 21:40:22 +03:00
Aymeric Augustin d4a0b27838 [py3] Refactored __unicode__ to __str__.
* Renamed the __unicode__ methods
* Applied the python_2_unicode_compatible decorator
* Removed the StrAndUnicode mix-in that is superseded by
  python_2_unicode_compatible
* Kept the __unicode__ methods in classes that specifically
  test it under Python 2
2012-08-12 14:44:40 +02:00
Aymeric Augustin e7e08fd48b [py3] Replaced some __str__ methods by __unicode__
These methods actually return unicode.
2012-08-12 14:44:40 +02:00
Aymeric Augustin dbb63e56ea [py3] Avoided returning bytes in Model.__str__
on Python 3.
2012-08-12 14:44:40 +02:00
Claude Paroz 15fffcc751 [py3] Fixed reraising of exceptions
reraise with second argument to None is not supported.
2012-08-11 23:46:23 +02:00
Aymeric Augustin 5c09c59bc7 [py3] Renamed `next` to `__next__` in iterators.
See PEP 3114. `next` is retained as an alias for Python 2.
2012-08-09 14:36:05 +02:00
Claude Paroz b8e49d70f2 [py3] Replaced raw_input by input
The six addition has been borrowed from:
https://bitbucket.org/gutworth/six/changeset/733ef740
2012-08-08 19:53:11 +02:00
Alex Gaynor 4c97101b1f remove a bunch of unnescesarry iterkeys() calls 2012-08-08 07:33:15 -07:00
Claude Paroz 576ec12f8e [py3] Replaced __nonzero__ by __bool__
Of course, __nonzero__ alias has been kept for Python 2 compatibility.
2012-08-08 15:02:31 +02:00
Aymeric Augustin 12cda89ffe [py3] Fixed a loop that changed dictionary size. 2012-08-08 15:00:24 +02:00
Aymeric Augustin a8b3ddec5f [py3] Applied minor fixes so the test suite starts 2012-08-07 12:00:22 +02:00
Aymeric Augustin c5ef65bcf3 [py3] Ported django.utils.encoding.
* Renamed smart_unicode to smart_text (but kept the old name under
  Python 2 for backwards compatibility).
* Renamed smart_str to smart_bytes.
* Re-introduced smart_str as an alias for smart_text under Python 3
  and smart_bytes under Python 2 (which is backwards compatible).
  Thus smart_str always returns a str objects.
* Used the new smart_str in a few places where both Python 2 and 3
  want a str.
2012-08-07 12:00:22 +02:00
Aymeric Augustin ee191715ea [py3] Fixed access to dict keys/values/items. 2012-08-07 12:00:22 +02:00
Brendan MacDonell ad237fb72f Fixed #18724 -- Fixed IntegerField validation with value 0 2012-08-06 10:42:21 +02:00
Claude Paroz 9908201d7f Replaced some byte strings by str() calls
This is a useful trick when Python 2 awaits byte strings and
Python 3 Unicode (regular) strings.
2012-08-03 15:18:13 +02:00
Ramiro Morales f758bdab5e Fixed #18271 -- Changed stage at which TransactionTestCase flushes DB tables.
Previously, the flush was done before the test case execution and now
it is performed after it.

Other changes to the testing infrastructure include:

* TransactionTestCase now doesn't reset autoincrement sequences either
  (previous behavior can achieved by using `reset_sequences`.)
  With this, no implicit such reset is performed by any of the provided
  TestCase classes.

* New ordering of test cases: All unittest tes cases are run first and
  doctests are run at the end.

THse changes could be backward-incompatible with test cases that relied
on some kind of state being preserved between tests. Please read the
relevant sections of the release notes and testing documentation for
further details.

Thanks Andreas Pelme for the initial patch. Karen Tracey and Anssi
Kääriäinen for the feedback and Anssi for reviewing.

This also fixes #12408.
2012-07-24 17:24:16 -03:00
Aymeric Augustin ae4125ffce Removed a Python 3-compatibility hack.
Thanks Preston Holmes for the patch.
2012-07-23 13:48:04 +02:00
Aymeric Augustin a84d79f572 [py3] Added Python 3 compatibility for xrange. 2012-07-22 09:29:56 +02:00
Aymeric Augustin ca07fda2ef [py3] Switched to Python 3-compatible imports.
xrange/range will be dealt with in a separate commit due to the huge
number of changes.
2012-07-22 09:29:56 +02:00
Aymeric Augustin bdca5ea345 [py3] Replaced unicode/str by six.text_type/bytes. 2012-07-22 09:29:54 +02:00
Aymeric Augustin 3cb2457f46 [py3] Replaced basestring by six.string_types. 2012-07-22 09:29:54 +02:00
Aymeric Augustin 56dbe924a6 [py3] Removed longs. 2012-07-22 09:29:53 +02:00
Aymeric Augustin d11d45aad9 [py3] Used six.with_metaclass wherever necessary. 2012-07-22 09:29:52 +02:00
Aymeric Augustin 7fa51a24a8 [py3] Exception aren't iterable in Python 3. 2012-07-22 09:29:52 +02:00
Aymeric Augustin d796c94b03 [py3] Used six.reraise wherever necessary. 2012-07-22 09:29:52 +02:00
Alex Gaynor 52df0d50b0 Switched to use a more idiomatic construct. 2012-07-17 07:01:01 -07:00
Anssi Kääriäinen 29132ebdef Fixed #17788 -- Added batch_size argument to qs.bulk_create()
The qs.bulk_create() method did not work with large batches together
with SQLite3. This commit adds a way to split the bulk into smaller
batches. The default batch size is unlimited except for SQLite3 where
the batch size is limited to 999 SQL parameters per batch.

Thanks to everybody who participated in the discussions at Trac.
2012-07-17 15:24:41 +03:00
Anssi Kääriäinen fcad6c48f0 Fixed #17497 -- Corrected FieldError message in add_fields()
The erroneous message was user visible in values_list() calls.

Thanks to ojii for report and review, and to antoviaque for the patch.
2012-07-17 12:49:46 +03:00
Nuno Maltez bebbbb7af0 Fixed #18056 - Cleared aggregations on DateQuery.add_date_select
Cleared aggregations on add_date_select method so only distinct dates
are returned when dealing with a QuerySet that contained aggregations.
That would cause the query set to return repeated dates because it
would look for distinct (date kind, aggregation) pairs.
2012-07-16 19:52:31 +03:00
Alex Gaynor 9877d25dc6 Switched from usign a method that was about to be deprecated to normal code. 2012-07-14 16:02:30 -07:00
Alex Gaynor fb8ec2855b Remove some code that has been dead since newforms-admin was merged, many moons ago. 2012-07-12 09:27:08 -07:00
Claude Paroz 828f7b62e8 Fixed #18602 -- Improved error message when database NAME is missing
Thanks Kristian Glass for the report.
2012-07-10 13:22:55 +02:00
Claude Paroz 146aff3bac Fixed #18590 - Reverted Python 2.4 workaround for Model pickling
Revert of 08d521efa0. Refs #10547, #12121.
Thanks Michal Petrucha for the report.
2012-07-08 20:08:28 +02:00
Anssi Kääriäinen d3c2eb103f Fixed #18330 - Made cache culling 3rd party db backend friendly
This is Ian Kelly's patch from #15580 with minor modifications.
2012-07-05 17:20:48 +03:00
Andrei Antoukh 0f49b2bce2 Fixed #18362 - Made model.save() update_fields accept attnames 2012-07-05 16:43:28 +03:00
Anssi Kääriäinen ab7f071058 Fixed comment wording in sql/where.py
Thanks to Simon Charette for noticing this.
2012-07-03 10:31:06 +03:00
Anssi Kääriäinen e74787391e Fixed a regression introduced in where.as_sql() refactor
At least Oracle needs parentheses in negated where conditions, even if
there is only single condition negated. Fixed this by reverting to old
logic in that part of as_sql() and adding a comment about this.

I did not investigate why the parentheses are needed. The original
offending commit was bd283aa844.
2012-07-01 22:52:35 +03:00
Anssi Kääriäinen f572ee0c65 Fixed #16047 -- Restore autocommit state correctly on psycopg2
When the postgresql_psycopg2 backend was used with DB-level autocommit
mode enabled, after entering transaction management and then leaving
it, the isolation level was never set back to autocommit mode.

Thanks brodie for report and working on this issue.
2012-07-01 19:36:43 +03:00
Anssi Kääriäinen da573fbb41 Fixed some locations to work with autocommit=True
- backends: supports_transactions()
 - select_for_update tests
2012-07-01 19:27:36 +03:00
Anssi Kääriäinen bd283aa844 Refactored the empty/full result logic in WhereNode.as_sql()
Made sure the WhereNode.as_sql() handles various EmptyResultSet and
FullResultSet conditions correctly. Also, got rid of the FullResultSet
exception class. It is now represented by '', [] return value in the
as_sql() methods.
2012-07-01 17:21:34 +03:00
Anssi Kääriäinen 2b9fb2e644 Fixed #18251 -- Removed a deadlock possibility in apploading
Thanks to harm for the report and comments.
2012-07-01 16:14:27 +03:00
Josh Smeaton fa182e8ae8 Fixed #18465 -- Set date formats correctly on Oracle
Correctly configure NLS_SESSION_PARAMETERS to format Date and DateTime
on Oracle backend.

Thanks to Josh Smeaton for report & patch.
2012-06-26 19:39:14 +03:00
Claude Paroz 05d333ba3b Fixed #18515 -- Conditionally regenerated filename in FileField validation
When a FileField value has been saved, a new validation should not
regenerate a new filename when checking the length. Refs #9893.
2012-06-26 18:18:44 +02:00
Anssi Kääriäinen b6c356b7bb Fixed #17485 -- Made defer work with select_related
This commit tackles a couple of issues. First, in certain cases there
were some mixups if field.attname or field.name should be deferred.
Field.attname is now always used.

Another issue tackled is a case where field is both deferred by
.only(), and selected by select_related. This case is now an error.

A lot of thanks to koniiiik (Michal Petrucha) for the patch, and
to Andrei Antoukh for review.
2012-06-26 18:08:42 +03:00
Alex Gaynor e0fce8706d Switch to using context managers for acquiring and releasing locks. 2012-06-23 08:11:15 -07:00
Luke Plant a54a8bab0c Fixed #17776 - DoesNotExist is not picklable
Thanks to ambv for the report
2012-06-22 13:28:15 +01:00
Anssi Kääriäinen 86c20e39eb Fixed connection.queries encoding handling on Oracle
In addition, removed a possibly problematic .filter() call from
backends.test_query_encoding test. It is possible the .filter could
cause collation problems on MySQL, and as it wasn't absolutely needed
for the test it seemed better to get rid of the call.

Refs #18461.
2012-06-13 14:16:34 +03:00
Claude Paroz e9ef9776d1 Fixed #18461 -- Ensured that last_executed_query returns Unicode
Thanks Anssi Kääriäinen for the review.
2012-06-13 12:04:46 +02:00
Claude Paroz 3dd5d726d1 Fixed #18463 -- Forced type() argument to be a byte string 2012-06-11 22:14:06 +02:00
Claude Paroz aa423575e7 Fixed #17760 -- Implemented callable database features as cached properties
This does remove the requirement to call features.confirm() method
before checking the properties.
Thanks cdestiger and Ramiro Morales for their work on the patch.
2012-06-09 15:59:52 +02:00
Jann Kleen 1a412dda62 Fixed #18432 -- Prevented the ForeignKey field from creating an invalid query when chained. Thanks, Jann Kleen. 2012-06-07 18:49:19 +02:00
Claude Paroz 4a103086d5 Fixed #18269 -- Applied unicode_literals for Python 3 compatibility.
Thanks Vinay Sajip for the support of his django3 branch and
Jannis Leidel for the review.
2012-06-07 18:08:47 +02:00