Commit Graph

382 Commits

Author SHA1 Message Date
Markus Holtermann faad6070ee Fixed #24701 -- Converted model manager names to unicode in migrations
Thanks to Reto Aebersold for reporting the issue and Tim Graham and
Claude Paroz for the review.
2015-04-25 18:17:06 +02:00
Markus Holtermann b93690c465 Fixed #24573 -- Considered new related models for reloading
Thanks tttomekkk for the report.
2015-04-21 11:35:45 +02:00
Marten Kenbeek 1a1f16d67d Fixed #24591 -- Optimized cloning of ModelState objects.
Changed ModelState.clone() to create a shallow copy of self.fields
and self.managers.
2015-04-20 19:53:21 -04:00
Marten Kenbeek 039d7881b4 Refs #24397 -- Sped up model reloading in ProjectState.
Created bulk_update() context manager on StateApps. Sped up unregistering
models in reload_models() by using this context mananger.
2015-04-20 14:42:23 -04:00
Claude Paroz 02260ea3f6 Fixed #24595 -- Prevented loss of null info in MySQL field alteration
Thanks Simon Percivall for the report, and Simon Charette and Tim
Graham for the reviews.
2015-04-17 10:25:15 +02:00
Christopher Luc a7bc00e17b Fixed #24514 -- Made migration writer omit models import if it's unused. 2015-04-10 07:04:26 -04:00
Curtis Maloney 88d7fcebde Use mock.patch in migrations tests
Currently some of the migrations tests rely on the fact 'input' is aliased
because of six, instead of using mock.patch.  Replace this code with proper
use of mock.patch.

Also, replace one case of excessively specific python version check with
testing six.PY3
2015-04-09 11:16:12 +10:00
Patryk Zawadzki 0385dad073 Fixed #24513 -- Made sure a model is only rendered once during reloads
This also prevents state modifications from corrupting previous states.
Previously, when a model defining a relation was unregistered first,
clearing the cache would cause its related models' _meta to be cleared
and would result in the old models losing track of their relations.
2015-04-07 14:49:47 +02:00
Marten Kenbeek e8e4f978dd Fixed #24278 -- Fixed serialization of migration operations.
Fixed MigrationWriter.serialize() to correctly handle migration
operations by utilizing OperationWriter.

Thanks Piotr Maliński for the report.
2015-04-05 20:26:21 +02:00
Marten Kenbeek d597174bd4 Refs #24278 -- Allowed multi-line serializations in OperationWriter.
Changed OperationWriter to support multi-line serialized values with
correct indentation.
2015-04-05 20:19:24 +02:00
Baptiste Mispelon 30a3c2f74c Fixed #24566 -- Added support for serializing timedelta
Thanks to knbk for the report.
2015-04-02 17:21:20 +02:00
Markus Holtermann c5cc332bf2 Fixed #24550 -- Added migration operation description to sqlmigrate output
Thanks Tim Graham for the review.
2015-03-30 16:31:20 +02:00
Marten Kenbeek 75430be86f Refs #24366 -- Fixed recursion depth error in migration graph
Made MigrationGraph forwards_plan() and backwards_plan() fall back to an
iterative approach in case the recursive approach exceeds the recursion
depth limit.
2015-03-29 16:08:07 +02:00
Marten Kenbeek bc83add04c Refs #24366 -- Renamed arguments in MigrationGraph, renamed tests 2015-03-29 16:07:58 +02:00
Markus Holtermann c7ec3c07e7 Fixed #24537 -- Ignored field order in RenameModel detection
Thanks to David Sanders for the report and test and Simon Charette for
the review.
2015-03-28 20:45:15 +01:00
Anssi Kääriäinen 8f30556329 Renamed Field.rel attribute to remote_field
Field.rel is now deprecated. Rel objects have now also remote_field
attribute. This means that self == self.remote_field.remote_field.

In addition, made the Rel objects a bit more like Field objects. Still,
marked ManyToManyFields as null=True.
2015-03-25 08:16:12 -04:00
David Szotten 80e3444eca Fixed #24483 -- Prevented keepdb from breaking with generator choices.
If Field.choices is provided as an iterator, consume it in __init__ instead
of using itertools.tee (which ends up holding everything in memory
anyway). Fixes a bug where deconstruct() was consuming the iterator but
bypassing the call to `tee`.
2015-03-24 11:42:53 -04:00
Baptiste Mispelon 1aadade373 Fixed #24521 -- Added support for serializing frozensets in migrations. 2015-03-23 15:38:25 +01:00
John Giannelos 8758a63ddb Fixed #24427 -- Stopped writing migration files in dry run mode when merging.
Also added display of migration to stdout when verbosity=3.
2015-03-16 14:04:37 -04:00
Matthew Wilkes ae87ad005f Refs #24354 -- Prevented repointing of relations on superclasses when migrating a subclass's name change
Forwardport of test and release note from stable/1.7.x
2015-03-14 15:37:40 -04:00
Marten Kenbeek e272904ff7 Fixed #23407 -- Extended coverage of makemigrations --noinput option.
Changed --noinput option in makemigrations to suppress all user prompts,
not just when combined with --merge.
2015-03-08 19:44:14 +01:00
Markus Holtermann a9e29fae10 Fixed #24435 -- Prevented m2m field removal and addition in migrations when changing blank
Thanks Mark Tranchant for the report an Tim Graham for the test and
review.
2015-03-04 14:26:49 +01:00
Aymeric Augustin 903d1a57ab Made migrations tests write to a temporary directory.
This is preferrable to writing in the current working directory because
it eliminates the risk to leak unwanted files, which can result in very
weird test failures.

Also this will help if we ever try to run these tests concurrently.
2015-02-23 22:22:58 +01:00
Marten Kenbeek 78d43a5e10 Fixed #24366 -- Optimized traversal of large migration dependency graphs.
Switched from an adjancency list and uncached, iterative depth-first
search to a Node-based design with direct parent/child links and a
cached, recursive depth-first search. With this change, calculating
a migration plan for a large graph takes several seconds instead of
several hours.

Marked test `migrations.test_graph.GraphTests.test_dfs` as an expected
failure due to reaching the maximum recursion depth.
2015-02-23 12:38:53 +01:00
Loic Bistuer bed504d70b Fixed #24351, #24346 -- Changed the signature of allow_migrate().
The new signature enables better support for routing RunPython and
RunSQL operations, especially w.r.t. reusable and third-party apps.

This commit also takes advantage of the deprecation cycle for the old
signature to remove the backward incompatibility introduced in #22583;
RunPython and RunSQL won't call allow_migrate() when when the router
has the old signature.

Thanks Aymeric Augustin and Tim Graham for helping shape up the patch.

Refs 22583.
2015-02-20 21:34:09 +07:00
Shai Berger f7d34fe741 Fixed #24362: Made test run each migration operation in its own schema editor
Thanks Tim Graham for pointing out the fix.
2015-02-18 22:47:15 +02:00
Marten Kenbeek 15dc8d1c9d Fixed #24291 - Fixed migration ModelState generation with unused swappable models
Swapped out models don't have a _default_manager unless they have
explicitly defined managers. ModelState.from_model() now accounts for
this case and uses an empty list for managers if no explicit managers
are defined and a model is swapped out.
2015-02-18 19:05:10 +01:00
Tim Graham 307c0f299a Refs #24324 -- Fixed Python 2 test failures when path to Django source contains non-ASCII characters. 2015-02-17 19:03:03 -05:00
Markus Holtermann d538e37e1b Added tests for get_related_models_recursive 2015-02-16 19:31:52 +01:00
Markus Holtermann 273bc4b667 Refs #24282 -- Added failing test case for assigning models of wrong type to FK
Thanks Jeff Singer for the test case.
2015-02-16 19:31:51 +01:00
Markus Holtermann cc22b009e0 Refs #24264 -- Added failing test case for updating a FK when changing a PK
When the primary key column is altered, foreign keys of referencing
models must be aware of a possible data type change as well and thus
need to be re-rendered.

Thanks Tim Graham for the report.
2015-02-16 19:31:51 +01:00
Claude Paroz 58d0dd9260 Refs #24225 -- Added failing test case for removing a previously added field in migrations
When a related field is deleted, the related model must be updated. As
unchanged models are shared in migration states, the related model must
be re-rendered so that the change applies to a new copy of the related
model.

Thanks Henrik Heimbuerger for the report.
2015-02-16 19:31:51 +01:00
Markus Holtermann f287bec583 Fixed #24184 -- Prevented automatic soft-apply of migrations
Previously Django only checked for the table name in CreateModel
operations in initial migrations and faked the migration automatically.
This led to various errors and unexpected behavior. The newly introduced
--fake-initial flag to the migrate command must be passed to get the
same behavior again. With this change Django will bail out in with a
"duplicate relation / table" error instead.

Thanks Carl Meyer and Tim Graham for the documentation update, report
and review.
2015-02-13 14:29:59 +01:00
Collin Anderson db77915c9f Fixed E265 comment style 2015-02-06 09:30:35 -05:00
Tim Graham 0ed7d15563 Sorted imports with isort; refs #23860. 2015-02-06 08:16:28 -05:00
Tim Graham c0cc8f69e7 Refactored tests that rely on an ImportError for Python 3.5 compatibility
A change in Python test discovery [1] causes the old packages that raised
an error to be discovered; now we use a common directory that's
ignored during discovery. Refs #23763.

[1] http://bugs.python.org/issue7559
2015-02-03 18:30:46 -05:00
Adam Taylor 039465a6a7 Fixed typos in code comments. 2015-01-20 12:18:03 -05:00
Markus Holtermann 7f20041bca Fixed #24155 -- Maintained kwargs and import order in migration writer
Thanks Tomas Dobrovolny for the report and Tim Graham for the review.
2015-01-20 01:24:41 +01:00
Tim Graham 33457cd3b0 Removed IPAddressField per deprecation timeline; refs #20439. 2015-01-19 11:12:57 -05:00
Markus Holtermann 65d55c4093 Cleaned up migration writer tests 2015-01-17 20:29:50 +01:00
Tim Graham 28db4af80a Fixed #24135 -- Made RenameModel rename many-to-many tables.
Thanks Simon and Markus for reviews.
2015-01-15 20:34:33 -05:00
Collin Anderson faf0d66a80 Fixed #23850 -- Fixed a migrations test failure on Mac OS X & Python 3 2015-01-15 19:45:43 -05:00
Markus Holtermann 88786afbff Fixed #24147 -- Prevented managers leaking model during migrations
Thanks Tim Graham for the review.
2015-01-14 16:09:33 +01:00
Markus Holtermann eeb88123e7 Fixed #24129 -- Added indicator that migrations are rendering the initial state
Thanks Tim Graham for the review.
2015-01-12 19:23:46 +01:00
Markus Holtermann bbbed99f62 Fixed #24123 -- Used all available migrations to generate the initial migration state
Thanks Collin Anderson for the input when creating the patch and Tim Graham for the review.
2015-01-12 18:39:18 +01:00
Claude Paroz 68a439a18d Removed supports_binary_field flag as all backends support them
It was mainly for MySQL on Python 3, but now the current
recommended MySQL driver for Python 3 (mysqlclient) does support
binary fields, it is unneeded. Refs #20377.
2015-01-11 23:34:47 +01:00
Markus Holtermann fdc2cc9487 Fixed #24110 -- Rewrote migration unapply to preserve intermediate states 2015-01-10 23:14:15 +01:00
Loic Bistuer 8f4877c89d Fixed #22583 -- Allowed RunPython and RunSQL to provide hints to the db router.
Thanks Markus Holtermann and Tim Graham for the review.
2015-01-10 00:30:48 +07:00
Markus Holtermann c8bac4b556 Fixed #24098 -- Added no-op attributes to RunPython and RunSQL
Thanks Loïc Bistuer and Tim Graham for the discussion and review.
2015-01-09 10:31:32 -05:00
Markus Holtermann 862ea825b5 Fixed #24093 -- Prevented MigrationWriter to write operation kwargs that are not explicitly deconstructed 2015-01-07 17:29:20 -07:00