Commit Graph

22214 Commits

Author SHA1 Message Date
Tim Graham 33db899a21 Fixed #25814 -- Added changing #django topic to release checklist. 2016-03-01 13:02:03 -05:00
Tim Graham 2e895d2870 Added stub release notes for 1.9.4. 2016-03-01 12:39:01 -05:00
Tim Graham 24fc935218 Added CVE-2016-2512/2513 to security release archive. 2016-03-01 12:32:42 -05:00
Florian Apolloner 67b46ba701 Fixed CVE-2016-2513 -- Fixed user enumeration timing attack during login.
This is a security fix.
2016-03-01 11:25:28 -05:00
Mark Striemer c5544d2892 Fixed CVE-2016-2512 -- Prevented spoofing is_safe_url() with basic auth.
This is a security fix.
2016-03-01 11:25:28 -05:00
Tim Graham f43291639b Added stub release notes for security issues. 2016-03-01 11:25:28 -05:00
Michal Petrucha fe8ea3ba3b Fixed #26217 -- Added a warning about format strings to WeekArchiveView docs. 2016-03-01 10:29:52 -05:00
acemaster a1b1688c7d Fixed #26165 -- Added some FAQs about CSRF protection.
Thanks Florian Apolloner and Shai Berger for review.
2016-03-01 08:45:05 -05:00
Taranjeet 11a8207d42 Fixed typos in docs/ref/models/meta.txt. 2016-03-01 08:33:27 -05:00
Alasdair Nicol 65bd053f11 Fixed #26229 -- Improved check for model admin check admin.E124
Refs #22792
2016-03-01 08:20:14 -05:00
Simon Charette 0223e213dd Fixed #26186 -- Documented how app relative relationships of abstract models behave.
This partially reverts commit bc7d201bdb.

Thanks Tim for the review.

Refs #25858.
2016-02-29 22:07:05 -05:00
Jon Dufresne eac1423f9e Removed obsolete test CreatesuperuserManagementCommandTestCase.test_nolocale.
Test was added in 4c934f3921 to verify that
the commend works when locale.getdefaultlocale() doesn't return a locale.
getdefaultlocale() no longer runs at runtime, so the test isn't needed.
2016-02-29 08:46:37 -05:00
Adam Chainz 6a383f773a Removed unused 'Between' lookup.
It was added in 20bab2cf9d and stopped being
used for `Range` in 00aa562884 when
`bilateral` was added to `Transform`.
2016-02-29 08:00:04 -05:00
Shai Berger 72e5778b23 Minor fixes for release-process doc fix
As suggested by Tim Graham
2016-02-28 19:30:18 +02:00
Shai Berger 3dd4e9203a Fixed docs: release-process, Supported Versions section, concrete example
Security & data loss fixes are applied to the two last feature releases,
not just one.

Thanks Loic Bistuer for review
2016-02-28 18:44:47 +02:00
chenesan b84f5ab4ec Fixed #26230 -- Made default_related_name affect related_query_name. 2016-02-27 08:48:32 -05:00
inondle 5fb9756eba Fixed #26275 -- Noted difference between o and Y date format chars. 2016-02-27 08:05:12 -05:00
Attila Tovt 5e2c4d7afb Fixed #26264 -- Fixed prefetch_related() crashes with values_list(flat=True) 2016-02-26 19:26:15 -05:00
Tore Lundqvist 3389c5ea22 Fixed #21608 -- Prevented logged out sessions being resurrected by concurrent requests.
Thanks Simon Charette for the review.
2016-02-26 18:56:56 -05:00
Simon Charette 3938b3ccaa Fixed #26286 -- Prevented content type managers from sharing their cache.
This should prevent managers methods from returning content type instances
registered to foreign apps now that these managers are also attached to models
created during migration phases.

Thanks Tim for the review.

Refs #23822.
2016-02-26 16:18:16 -05:00
Tim Graham b9519b2730 Added 'prefetches to docs/spelling_wordlist. 2016-02-26 16:06:34 -05:00
Adam Chainz ef33bc2d4d Fixed #25279 -- Made prefetch_related_objects() public. 2016-02-26 14:55:01 -05:00
Yoong Kang Lim d5f89ff6e8 Fixed #24974 -- Fixed inheritance of formfield_callback for modelform_factory forms. 2016-02-26 12:27:27 -05:00
Simon Charette 766afc22a1 Fixed #24793 -- Unified temporal difference support. 2016-02-26 12:25:12 -05:00
Simon Charette 31098e3288 Used setUpTestData for the timedelta expression tests. 2016-02-26 12:25:12 -05:00
zshimanchik 65aa94200b Fixed #24653 -- Fixed MySQL database introspection when using read_default_file. 2016-02-26 12:02:13 -05:00
Simon Charette 62ea86448e Cleaned up session backends tests.
Made SessionTestsMixin backend agnostic and removed code obsoleted by the test
discovery refactor.
2016-02-26 11:22:33 -05:00
Ivan Tsouvarev 8890c533e0 Fixed #26280 -- Fixed cached template loader crash when loading nonexistent template. 2016-02-26 08:02:10 -05:00
Edwar Baron eb44172760 Fixed #25811 -- Added a helpful error when making _in queries across different databases. 2016-02-26 07:31:56 -05:00
Tim Graham 7fec264e46 Removed try/fail antipattern from model_options tests. 2016-02-25 20:04:51 -05:00
Sjoerd Job Postmus bbe136e1a2 Fixed #26231 -- Used .get_username in admin login template. 2016-02-25 19:29:53 -05:00
Nick Malakhov ee69789f45 Fixed #26269 -- Prohibited spaces in is_valid_ipv6_address(). 2016-02-25 18:52:50 -05:00
Tim Graham 22d2a5b00a Corrected a run on sentence in doc/topics/db/models.txt. 2016-02-25 14:22:41 -05:00
Yoong Kang Lim 4b1529e2cb Fixed #26151 -- Refactored MigrationWriter.serialize()
Thanks Markus Holtermann for review.
2016-02-25 14:01:06 -05:00
Scott Sexton fc584f0685 Fixed #26117 -- Consulted database routers in initial migration detection.
Thanks Simon Charette for help.
2016-02-25 09:56:00 -05:00
Tim Graham 1f8cfcf3b4 Fixed #26278 -- Clarified apps.ready docs. 2016-02-25 08:55:10 -05:00
Tim Graham 7a7e403325 Refs #26270 -- Reorganized TestCase docs. 2016-02-25 07:58:22 -05:00
Olivier Le Thanh Duong 10781b4c6f Fixed #12233 -- Allowed redirecting authenticated users away from the login view.
contrib.auth.views.login() has a new parameter `redirect_authenticated_user`
to automatically redirect authenticated users visiting the login page.

Thanks to dmathieu and Alex Buchanan for the original code and to Carl Meyer
for the help and review.
2016-02-25 07:18:33 -05:00
Claude Paroz 4c18a8a378 Fixed #14098 -- Prevented crash for introspection errors in inspectdb
Thanks Tim Graham for the review.
2016-02-25 08:43:56 +01:00
Tim Graham 441c537b66 Fixed a function signature in docs/topics/auth/default.txt. 2016-02-24 16:24:33 -05:00
Tim Graham 8ad7b8118c Used addCleanup() to call recorder.flush() in migration loader tests. 2016-02-24 11:22:09 -05:00
Claude Paroz c5517b9e74 Fixed #26266 -- Output the primary key in the GeoJSON serializer properties
Thanks Tim Graham for the review.
2016-02-24 16:10:46 +01:00
Tim Graham 6637cd0ef2 Removed docs of deprecated SimpleTestCase warnings behavior.
Removed in Django 1.7 (4f6be9a0c4).
2016-02-24 09:57:39 -05:00
Jon Dufresne b412681359 Fixed #26267 -- Fixed BoundField to reallow slices of subwidgets. 2016-02-24 07:02:51 -05:00
James Aylett 1ff6e37de4 Fixed #23832 -- Added timezone aware Storage API.
New Storage.get_{accessed,created,modified}_time() methods convert the
naive time from now-deprecated {accessed,created_modified}_time()
methods into aware objects in UTC if USE_TZ=True.
2016-02-23 18:51:43 -05:00
Claude Paroz eda306f1ce Fixed #26232 -- Fixed Popen mocking environment in i18n tests
Refs #25925. Thanks Jeroen Pulles for the report.
2016-02-23 20:06:18 +01:00
Simon Charette c30086159d Used setupTestData in prefetch_related tests. 2016-02-23 13:53:58 -05:00
Aymeric Augustin 7f6fbc906a Prevented static file corruption when URL fragment contains '..'.
When running collectstatic with a hashing static file storage backend,
URLs referencing other files were normalized with posixpath.normpath.
This could corrupt URLs: for example 'a.css#b/../c' became just 'c'.

Normalization seems to be an artifact of the historical implementation.
It contained a home-grown implementation of posixpath.join which relied
on counting occurrences of .. and /, so multiple / had to be collapsed.

The new implementation introduced in the previous commit doesn't suffer
from this issue. So it seems safe to remove the normalization.

There was a test for this normalization behavior but I don't think it's
a good test. Django shouldn't modify CSS that way. If a developer has
rendundant /s, it's mostly an aesthetic issue and it isn't Django's job
to fix it. Conversely, if the user wants a series of /s, perhaps in the
URL fragment, Django shouldn't destroy it.

Refs #26249.
2016-02-23 19:35:16 +01:00
Aymeric Augustin 706b33fef8 Fixed #26249 -- Fixed collectstatic crash for files in STATIC_ROOT referenced by absolute URL.
collectstatic crashed when:

* a hashing static file storage backend was used
* a static file referenced another static file located directly in
  STATIC_ROOT (not a subdirectory) with an absolute URL (which must
  start with STATIC_URL, which cannot be empty)

It seems to me that the current code reimplements relative path joining
and doesn't handle edge cases correctly. I suspect it assumes that
STATIC_URL is of the form r'/[^/]+/'.

Throwing out that code in favor of the posixpath module makes the logic
easier to follow. Handling absolute paths correctly also becomes easier.
2016-02-23 19:34:21 +01:00
Tim Graham c62807968d Fixed a stray __unicode__() method in auth_tests. 2016-02-23 13:20:50 -05:00