django1/docs/releases/3.0.1.txt

==========================
Django 3.0.1 release notes
==========================

*December 18, 2019*

Django 3.0.1 fixes a security issue and several bugs in 3.0.

CVE-2019-19844: Potential account hijack via password reset form
================================================================

By submitting a suitably crafted email address making use of Unicode
characters, that compared equal to an existing user email when lower-cased for
comparison, an attacker could be sent a password reset token for the matched
account.

In order to avoid this vulnerability, password reset requests now compare the
submitted email using the stricter, recommended algorithm for case-insensitive
comparison of two identifiers from `Unicode Technical Report 36, section
2.11.2(B)(2)`__. Upon a match, the email containing the reset token will be
sent to the email address on record rather than the submitted address.

.. __: https://www.unicode.org/reports/tr36/#Recommendations_General

Bugfixes
========

* Fixed a regression in Django 3.0 by restoring the ability to use Django
  inside Jupyter and other environments that force an async context, by adding
  an option to disable :ref:`async-safety` mechanism with
  ``DJANGO_ALLOW_ASYNC_UNSAFE`` environment variable (:ticket:`31056`).

* Fixed a regression in Django 3.0 where ``RegexPattern``, used by
  :func:`~django.urls.re_path`, returned positional arguments to be passed to
  the view when all optional named groups were missing (:ticket:`31061`).

* Reallowed, following a regression in Django 3.0,
  :class:`~django.db.models.expressions.Window` expressions to be used in
  conditions outside of queryset filters, e.g. in
  :class:`~django.db.models.expressions.When` conditions (:ticket:`31060`).

* Fixed a data loss possibility in
  :class:`~django.contrib.postgres.forms.SplitArrayField`. When using with
  ``ArrayField(BooleanField())``, all values after the first ``True`` value
  were marked as checked instead of preserving passed values (:ticket:`31073`).