Scheduled weekly dependency update for week 01

[pyup-bot]

Updates

Here's a list of all the updates bundled in this pull request. I've added some links to make it easier for you to find all the information you need.

codecov	2.0.9	»	2.0.13	PyPI | Changelog | Repo
django	1.11.7	»	2.0.1	PyPI | Changelog | Homepage
django-parsley	0.6	»	0.7	PyPI | Changelog | Repo
futures	3.1.1	»	3.2.0	PyPI | Changelog | Repo | Docs
html5lib	0.999999999	»	1.0.1	PyPI | Changelog | Repo
Jinja2	2.9.6	»	2.10	PyPI | Homepage
nose2	0.7.2	»	0.7.3	PyPI | Repo
nose2-html-report	0.5.0	»	0.6.0	PyPI | Changelog | Repo
numba	0.35.0	»	0.36.2	PyPI | Changelog | Repo
numpy	1.13.3	»	1.14.0	PyPI | Changelog | Homepage
pandas	0.21.0	»	0.22.0	PyPI | Changelog | Homepage
pillow	4.3.0	»	5.0.0	PyPI | Changelog | Homepage
psutil	5.4.1	»	5.4.3	PyPI | Changelog | Repo
pylint	1.7.4	»	1.8.1	PyPI | Changelog | Repo
pymongo	3.5.1	»	3.6.0	PyPI | Repo
pytest	3.2.3	»	3.3.2	PyPI | Changelog | Repo | Homepage
selenium	3.7.0	»	3.8.1	PyPI | Changelog | Repo
Sphinx	1.6.5	»	1.6.6	PyPI | Changelog | Homepage
sqlalchemy	1.1.15	»	1.2.0	PyPI | Changelog | Homepage

Changelogs

codecov 2.0.9 -> 2.0.13

2.0.13

• fix encoding issues

2.0.12

• revert merge commit fix, back to old way

2.0.11

• fix merge commit when it's a pull request
• remove snapci, business closed
• skip vendor directories for gcov parsing
• run coverage combine not merge
• fix report encoding

2.0.10

• fix uploading when reports contain characters outside of latin-1
• remove reduced_redundancy header from

django 1.11.7 -> 2.0.1

2.0.1

==========================

January 1, 2018

Django 2.0.1 fixes several bugs in 2.0.

Bugfixes

• Fixed a regression in Django 1.11 that added newlines between MultiWidget's
  subwidgets (:ticket:28890).

• Fixed incorrect class-based model index name generation for models with
  quoted db_table (:ticket:28876).

• Fixed incorrect foreign key constraint name for models with quoted
  db_table (:ticket:28876).

• Fixed a regression in caching of a GenericForeignKey when the referenced
  model instance uses more than one level of multi-table inheritance
  (:ticket:28856).

• Reallowed filtering a queryset with GeometryField=None (:ticket:28896).

• Corrected admin check to allow a OneToOneField in
  ModelAdmin.autocomplete_fields (:ticket:28898).

• Fixed a regression on SQLite where DecimalField returned a result with
  trailing zeros in the fractional part truncated (:ticket:28915).

• Fixed crash in the testserver command startup (:ticket:28941).

• Fixed crash when coercing a translatable URL pattern to str
  (:ticket:28947).

• Fixed crash on SQLite when renaming a field in a model referenced by a
  ManyToManyField (:ticket:28884).

• Fixed a crash when chaining values() or values_list() after
  QuerySet.select_for_update(of=(...)) (:ticket:28944).

• Fixed admin changelist crash when using a query expression in the page's
  ordering (:ticket:28958).

========================

2.0

========================

December 2, 2017

Welcome to Django 2.0!

These release notes cover the :ref:new features <whats-new-2.0>, as well as
some :ref:backwards incompatible changes <backwards-incompatible-2.0> you'll
want to be aware of when upgrading from Django 1.11 or earlier. We've
:ref:dropped some features<removed-features-2.0> that have reached the end of
their deprecation cycle, and we've :ref:begun the deprecation process for some features <deprecated-features-2.0>.

This release starts Django's use of a :ref:loose form of semantic versioning <internal-release-cadence>, but there aren't any major backwards incompatible
changes that might be expected of a 2.0 release. Upgrading should be a similar
amount of effort as past feature releases.

See the :doc:/howto/upgrade-version guide if you're updating an existing
project.

Python compatibility

Django 2.0 supports Python 3.4, 3.5, and 3.6. We highly recommend and only
officially support the latest release of each series.

The Django 1.11.x series is the last to support Python 2.7.

Django 2.0 will be the last release series to support Python 3.4. If you plan
a deployment of Python 3.4 beyond the end-of-life for Django 2.0 (April 2019),
stick with Django 1.11 LTS (supported until April 2020) instead. Note, however,
that the end-of-life for Python 3.4 is March 2019.

Third-party library support for older version of Django

Following the release of Django 2.0, we suggest that third-party app authors
drop support for all versions of Django prior to 1.11. At that time, you should
be able to run your package's tests using python -Wd so that deprecation
warnings do appear. After making the deprecation warning fixes, your app should
be compatible with Django 2.0.

.. _whats-new-2.0:

What's new in Django 2.0

Simplified URL routing syntax

The new :func:django.urls.path() function allows a simpler, more readable URL
routing syntax. For example, this example from previous Django releases::

url(r'^articles/(?P<year>[0-9]{4})/$', views.year_archive),

could be written as::

path('articles/<int:year>/', views.year_archive),

The new syntax supports type coercion of URL parameters. In the example, the
view will receive the year keyword argument as an integer rather than as
a string. Also, the URLs that will match are slightly less constrained in the
rewritten example. For example, the year 10000 will now match since the year
integers aren't constrained to be exactly four digits long as they are in the
regular expression.

The django.conf.urls.url() function from previous versions is now available
as :func:django.urls.re_path. The old location remains for backwards
compatibility, without an imminent deprecation. The old
django.conf.urls.include() function is now importable from django.urls
so you can use from django.urls import include, path, re_path in your
URLconfs.

The :doc:/topics/http/urls document is rewritten to feature the new syntax
and provide more details.

Mobile-friendly contrib.admin

The admin is now responsive and supports all major mobile devices. Older
browsers may experience varying levels of graceful degradation.

Window expressions

The new :class:~django.db.models.expressions.Window expression allows
adding an OVER clause to querysets. You can use :ref:window functions <window-functions> and :ref:aggregate functions <aggregation-functions> in
the expression.

Minor features

:mod:django.contrib.admin

• The new :attr:.ModelAdmin.autocomplete_fields attribute and
  :meth:.ModelAdmin.get_autocomplete_fields method allow using an
  Select2 <https://select2.org>_ search widget for ForeignKey and
  ManyToManyField.

:mod:django.contrib.auth

• The default iteration count for the PBKDF2 password hasher is increased from
  36,000 to 100,000.

:mod:django.contrib.gis

• Added MySQL support for the
  :class:~django.contrib.gis.db.models.functions.AsGeoJSON function,
  :class:~django.contrib.gis.db.models.functions.GeoHash function,
  :class:~django.contrib.gis.db.models.functions.IsValid function,
  :lookup:isvalid lookup, and :ref:distance lookups <distance-lookups>.

• Added the :class:~django.contrib.gis.db.models.functions.Azimuth and
  :class:~django.contrib.gis.db.models.functions.LineLocatePoint functions,
  supported on PostGIS and SpatiaLite.

• Any :class:~django.contrib.gis.geos.GEOSGeometry imported from GeoJSON now
  has its SRID set.

• Added the :attr:.OSMWidget.default_zoom attribute to customize the map's
  default zoom level.

• Made metadata readable and editable on rasters through the
  :attr:~django.contrib.gis.gdal.GDALRaster.metadata,
  :attr:~django.contrib.gis.gdal.GDALRaster.info, and
  :attr:~django.contrib.gis.gdal.GDALBand.metadata attributes.

• Allowed passing driver-specific creation options to
  :class:~django.contrib.gis.gdal.GDALRaster objects using papsz_options.

• Allowed creating :class:~django.contrib.gis.gdal.GDALRaster objects in
  GDAL's internal virtual filesystem. Rasters can now be :ref:created from and converted to binary data <gdal-raster-vsimem> in-memory.

• The new :meth:GDALBand.color_interp() <django.contrib.gis.gdal.GDALBand.color_interp> method returns the color
  interpretation for the band.

:mod:django.contrib.postgres

• The new distinct argument for
  :class:~django.contrib.postgres.aggregates.ArrayAgg determines if
  concatenated values will be distinct.

• The new :class:~django.contrib.postgres.functions.RandomUUID database
  function returns a version 4 UUID. It requires use of PostgreSQL's
  pgcrypto extension which can be activated using the new
  :class:~django.contrib.postgres.operations.CryptoExtension migration
  operation.

• :class:django.contrib.postgres.indexes.GinIndex now supports the
  fastupdate and gin_pending_list_limit parameters.

• The new :class:~django.contrib.postgres.indexes.GistIndex class allows
  creating GiST indexes in the database. The new
  :class:~django.contrib.postgres.operations.BtreeGistExtension migration
  operation installs the btree_gist extension to add support for operator
  classes that aren't built-in.

• :djadmin:inspectdb can now introspect JSONField and various
  RangeField\s (django.contrib.postgres must be in INSTALLED_APPS).

:mod:django.contrib.sitemaps

• Added the protocol keyword argument to the
  :class:~django.contrib.sitemaps.GenericSitemap constructor.

Cache

• cache.set_many() now returns a list of keys that failed to be inserted.
  For the built-in backends, failed inserts can only happen on memcached.

File Storage

• :meth:File.open() <django.core.files.File.open> can be used as a context
  manager, e.g. with file.open() as f:.

Forms

• The new date_attrs and time_attrs arguments for
  :class:~django.forms.SplitDateTimeWidget and
  :class:~django.forms.SplitHiddenDateTimeWidget allow specifying different
  HTML attributes for the DateInput and TimeInput (or hidden)
  subwidgets.

• The new :meth:Form.errors.get_json_data() <django.forms.Form.errors.get_json_data> method returns form errors as
  a dictionary suitable for including in a JSON response.

Generic Views

• The new :attr:.ContextMixin.extra_context attribute allows adding context
  in View.as_view().

Management Commands

• :djadmin:inspectdb now translates MySQL's unsigned integer columns to
  PositiveIntegerField or PositiveSmallIntegerField.

• The new :option:makemessages --add-location option controls the comment
  format in PO files.

• :djadmin:loaddata can now :ref:read from stdin <loading-fixtures-stdin>.

• The new :option:diffsettings --output option allows formatting the output
  in a unified diff format.

• On Oracle, :djadmin:inspectdb can now introspect AutoField if the
  column is created as an identity column.

• On MySQL, :djadmin:dbshell now supports client-side TLS certificates.

Migrations

• The new :option:squashmigrations --squashed-name option allows naming the
  squashed migration.

Models

• The new :class:~django.db.models.functions.StrIndex database function
  finds the starting index of a string inside another string.

• On Oracle, AutoField and BigAutoField are now created as identity columns_.

.. _identity columns: https://docs.oracle.com/database/121/DRDAA/migr_tools_feat.htmDRDAA109

• The new chunk_size parameter of :meth:.QuerySet.iterator controls the
  number of rows fetched by the Python database client when streaming results
  from the database. For databases that don't support server-side cursors, it
  controls the number of results Django fetches from the database adapter.

• :meth:.QuerySet.earliest, :meth:.QuerySet.latest, and
  :attr:Meta.get_latest_by <django.db.models.Options.get_latest_by> now
  allow ordering by several fields.

• Added the :class:~django.db.models.functions.ExtractQuarter function to
  extract the quarter from :class:~django.db.models.DateField and
  :class:~django.db.models.DateTimeField, and exposed it through the
  :lookup:quarter lookup.

• Added the :class:~django.db.models.functions.TruncQuarter function to
  truncate :class:~django.db.models.DateField and
  :class:~django.db.models.DateTimeField to the first day of a quarter.

• Added the :attr:~django.db.models.Index.db_tablespace parameter to
  class-based indexes.

• If the database supports a native duration field (Oracle and PostgreSQL),
  :class:~django.db.models.functions.Extract now works with
  :class:~django.db.models.DurationField.

• Added the of argument to :meth:.QuerySet.select_for_update(), supported
  on PostgreSQL and Oracle, to lock only rows from specific tables rather than
  all selected tables. It may be helpful particularly when
  :meth:~.QuerySet.select_for_update() is used in conjunction with
  :meth:~.QuerySet.select_related().

• The new field_name parameter of :meth:.QuerySet.in_bulk allows fetching
  results based on any unique model field.

• :meth:.CursorWrapper.callproc() now takes an optional dictionary of keyword
  parameters, if the backend supports this feature. Of Django's built-in
  backends, only Oracle supports it.

• The new :meth:connection.execute_wrapper() <django.db.backends.base.DatabaseWrapper.execute_wrapper> method allows
  :doc:installing wrappers around execution of database queries </topics/db/instrumentation>.

• The new filter argument for built-in aggregates allows :ref:adding different conditionals <conditional-aggregation> to multiple aggregations
  over the same fields or relations.

• Added support for expressions in :attr:Meta.ordering <django.db.models.Options.ordering>.

• The new named parameter of :meth:.QuerySet.values_list allows fetching
  results as named tuples.

• The new :class:.FilteredRelation class allows adding an ON clause to
  querysets.

Pagination

• Added :meth:Paginator.get_page() <django.core.paginator.Paginator.get_page>
  to provide the documented pattern of handling invalid page numbers.

Requests and Responses

• The :djadmin:runserver Web server supports HTTP 1.1.

Templates

• To increase the usefulness of :meth:.Engine.get_default in third-party
  apps, it now returns the first engine if multiple DjangoTemplates engines
  are configured in TEMPLATES rather than raising ImproperlyConfigured.

• Custom template tags may now accept keyword-only arguments.

Tests

• Added threading support to :class:~django.test.LiveServerTestCase.

• Added settings that allow customizing the test tablespace parameters for
  Oracle: :setting:DATAFILE_SIZE, :setting:DATAFILE_TMP_SIZE,
  :setting:DATAFILE_EXTSIZE, and :setting:DATAFILE_TMP_EXTSIZE.

Validators

• The new :class:.ProhibitNullCharactersValidator disallows the null
  character in the input of the :class:~django.forms.CharField form field
  and its subclasses. Null character input was observed from vulnerability
  scanning tools. Most databases silently discard null characters, but
  psycopg2 2.7+ raises an exception when trying to save a null character to
  a char/text field with PostgreSQL.

.. _backwards-incompatible-2.0:

Backwards incompatible changes in 2.0

Removed support for bytestrings in some places

To support native Python 2 strings, older Django versions had to accept both
bytestrings and unicode strings. Now that Python 2 support is dropped,
bytestrings should only be encountered around input/output boundaries (handling
of binary fields or HTTP streams, for example). You might have to update your
code to limit bytestring usage to a minimum, as Django no longer accepts
bytestrings in certain code paths.

For example, reverse() now uses str() instead of force_text() to
coerce the args and kwargs it receives, prior to their placement in
the URL. For bytestrings, this creates a string with an undesired b prefix
as well as additional quotes (str(b'foo') is "b'foo'"). To adapt, call
decode() on the bytestring before passing it to reverse().

Database backend API

This section describes changes that may be needed in third-party database
backends.

• The DatabaseOperations.datetime_cast_date_sql(),
  datetime_cast_time_sql(), datetime_trunc_sql(),
  datetime_extract_sql(), and date_interval_sql() methods now return
  only the SQL to perform the operation instead of SQL and a list of
  parameters.

• Third-party database backends should add a DatabaseWrapper.display_name
  attribute with the name of the database that your backend works with. Django
  may use it in various messages, such as in system checks.

• The first argument of SchemaEditor._alter_column_type_sql() is now
  model rather than table.

• The first argument of SchemaEditor._create_index_name() is now
  table_name rather than model.

• To enable FOR UPDATE OF support, set
  DatabaseFeatures.has_select_for_update_of = True. If the database
  requires that the arguments to OF be columns rather than tables, set
  DatabaseFeatures.select_for_update_of_column = True.

• To enable support for :class:~django.db.models.expressions.Window
  expressions, set DatabaseFeatures.supports_over_clause to True. You
  may need to customize the DatabaseOperations.window_start_rows_start_end()
  and/or window_start_range_start_end() methods.

• Third-party database backends should add a
  DatabaseOperations.cast_char_field_without_max_length attribute with the
  database data type that will be used in the
  :class:~django.db.models.functions.Cast function for a CharField if the
  max_length argument isn't provided.

• The first argument of DatabaseCreation._clone_test_db() and
  get_test_db_clone_settings() is now suffix rather
  than number (in case you want to rename the signatures in your backend
  for consistency). django.test also now passes those values as strings
  rather than as integers.

• Third-party database backends should add a
  DatabaseIntrospection.get_sequences() method based on the stub in
  BaseDatabaseIntrospection.

Dropped support for Oracle 11.2

The end of upstream support for Oracle 11.2 is Dec. 2020. Django 1.11 will be
supported until April 2020 which almost reaches this date. Django 2.0
officially supports Oracle 12.1+.

Default MySQL isolation level is read committed

MySQL's default isolation level, repeatable read, may cause data loss in
typical Django usage. To prevent that and for consistency with other databases,
the default isolation level is now read committed. You can use the
:setting:DATABASES setting to :ref:use a different isolation level <mysql-isolation-level>, if needed.

:attr:AbstractUser.last_name <django.contrib.auth.models.User.last_name> max_length increased to 150

A migration for :attr:django.contrib.auth.models.User.last_name is included.
If you have a custom user model inheriting from AbstractUser, you'll need
to generate and apply a database migration for your user model.

If you want to preserve the 30 character limit for last names, use a custom
form::

from django.contrib.auth.forms import UserChangeForm

class MyUserChangeForm(UserChangeForm):
last_name = forms.CharField(max_length=30, required=False)

If you wish to keep this restriction in the admin when editing users, set
UserAdmin.form to use this form::

from django.contrib.auth.admin import UserAdmin
from django.contrib.auth.models import User

class MyUserAdmin(UserAdmin):
form = MyUserChangeForm

admin.site.unregister(User)
admin.site.register(User, MyUserAdmin)

QuerySet.reverse() and last() are prohibited after slicing

Calling QuerySet.reverse() or last() on a sliced queryset leads to
unexpected results due to the slice being applied after reordering. This is
now prohibited, e.g.::

>>> Model.objects.all()[:2].reverse()
Traceback (most recent call last):
...
TypeError: Cannot reverse a query once a slice has been taken.

Form fields no longer accept optional arguments as positional arguments

To help prevent runtime errors due to incorrect ordering of form field
arguments, optional arguments of built-in form fields are no longer accepted
as positional arguments. For example::

forms.IntegerField(25, 10)

raises an exception and should be replaced with::

forms.IntegerField(max_value=25, min_value=10)

call_command() validates the options it receives

call_command() now validates that the argument parser of the command being
called defines all of the options passed to call_command().

For custom management commands that use options not created using
parser.add_argument(), add a stealth_options attribute on the command::

class MyCommand(BaseCommand):
stealth_options = ('option_name', ...)

Indexes no longer accept positional arguments

For example::

models.Index(['headline', '-pub_date'], 'index_name')

raises an exception and should be replaced with::

models.Index(fields=['headline', '-pub_date'], name='index_name')

Foreign key constraints are now enabled on SQLite

This will appear as a backwards-incompatible change (IntegrityError: FOREIGN KEY constraint failed) if attempting to save an existing model
instance that's violating a foreign key constraint.

Foreign keys are now created with DEFERRABLE INITIALLY DEFERRED instead of
DEFERRABLE IMMEDIATE. Thus, tables may need to be rebuilt to recreate
foreign keys with the new definition, particularly if you're using a pattern
like this::

from django.db import transaction

with transaction.atomic():
Book.objects.create(author_id=1)
Author.objects.create(id=1)

If you don't recreate the foreign key as DEFERRED, the first create()
would fail now that foreign key constraints are enforced.

Backup your database first! After upgrading to Django 2.0, you can then
rebuild tables using a script similar to this::

from django.apps import apps
from django.db import connection

for app in apps.get_app_configs():
for model in app.get_models(include_auto_created=True):
if model._meta.managed and not (model._meta.proxy or model._meta.swapped):
for base in model.bases:
if hasattr(base, '_meta'):
base._meta.local_many_to_many = []
model._meta.local_many_to_many = []
with connection.schema_editor() as editor:
editor._remake_table(model)

This script hasn't received extensive testing and needs adaption for various
cases such as multiple databases. Feel free to contribute improvements.

In addition, because of a table alteration limitation of SQLite, it's prohibited
to perform :class:~django.db.migrations.operations.RenameModel and
:class:~django.db.migrations.operations.RenameField operations on models or
fields referenced by other models in a transaction. In order to allow migrations
containing these operations to be applied, you must set the
Migration.atomic attribute to False.

Miscellaneous

• The SessionAuthenticationMiddleware class is removed. It provided no
  functionality since session authentication is unconditionally enabled in
  Django 1.10.

• The default HTTP error handlers (handler404, etc.) are now callables
  instead of dotted Python path strings. Django favors callable references
  since they provide better performance and debugging experience.

• :class:~django.views.generic.base.RedirectView no longer silences
  NoReverseMatch if the pattern_name doesn't exist.

• When :setting:USE_L10N is off, :class:~django.forms.FloatField and
  :class:~django.forms.DecimalField now respect :setting:DECIMAL_SEPARATOR
  and :setting:THOUSAND_SEPARATOR during validation. For example, with the
  settings::

USE_L10N = False
USE_THOUSAND_SEPARATOR = True
DECIMAL_SEPARATOR = ','
THOUSAND_SEPARATOR = '.'

an input of "1.345" is now converted to 1345 instead of 1.345.

• Subclasses of :class:~django.contrib.auth.models.AbstractBaseUser are no
  longer required to implement get_short_name() and get_full_name().
  (The base implementations that raise NotImplementedError are removed.)
  django.contrib.admin uses these methods if implemented but doesn't
  require them. Third-party apps that use these methods may want to adopt a
  similar approach.

• The FIRST_DAY_OF_WEEK and NUMBER_GROUPING format settings are now
  kept as integers in JavaScript and JSON i18n view outputs.

• :meth:~django.test.TransactionTestCase.assertNumQueries now ignores
  connection configuration queries. Previously, if a test opened a new database
  connection, those queries could be included as part of the
  assertNumQueries() count.

• The default size of the Oracle test tablespace is increased from 20M to 50M
  and the default autoextend size is increased from 10M to 25M.

• To improve performance when streaming large result sets from the database,
  :meth:.QuerySet.iterator now fetches 2000 rows at a time instead of 100.
  The old behavior can be restored using the chunk_size parameter. For
  example::

 Book.objects.iterator(chunk_size=100)

• Providing unknown package names in the packages argument of the
  :class:~django.views.i18n.JavaScriptCatalog view now raises ValueError
  instead of passing silently.

• A model instance's primary key now appears in the default Model.__str__()
  method, e.g. Question object (1).

• makemigrations now detects changes to the model field limit_choices_to
  option. Add this to your existing migrations or accept an auto-generated
  migration for fields that use it.

• Performing queries that require :ref:automatic spatial transformations <automatic-spatial-transformations> now raises NotImplementedError
  on MySQL instead of silently using non-transformed geometries.

• django.core.exceptions.DjangoRuntimeWarning is removed. It was only used
  in the cache backend as an intermediate class in CacheKeyWarning's
  inheritance of RuntimeWarning.

• Renamed BaseExpression._output_field to output_field. You may need
  to update custom expressions.

• In older versions, forms and formsets combine their Media with widget
  Media by concatenating the two. The combining now tries to :ref:preserve the relative order of elements in each list <form-media-asset-order>.
  MediaOrderConflictWarning is issued if the order can't be preserved.

• django.contrib.gis.gdal.OGRException is removed. It's been an alias for
  GDALException since Django 1.8.

• Support for GEOS 3.3.x is dropped.

• The way data is selected for GeometryField is changed to improve
  performance, and in raw SQL queries, those fields must now be wrapped in
  connection.ops.select. See the :ref:Raw queries note<gis-raw-sql> in
  the GIS tutorial for an example.

.. _deprecated-features-2.0:

Features deprecated in 2.0

context argument of Field.from_db_value() and Expression.convert_value()

The context argument of Field.from_db_value() and
Expression.convert_value() is unused as it's always an empty dictionary.
The signature of both methods is now::

(self, value, expression, connection)

instead of::

(self, value, expression, connection, context)

Support for the old signature in custom fields and expressions remains until
Django 3.0.

Miscellaneous

• The django.db.backends.postgresql_psycopg2 module is deprecated in favor
  of django.db.backends.postgresql. It's been an alias since Django 1.9.
  This only affects code that imports from the module directly. The
  DATABASES setting can still use
  'django.db.backends.postgresql_psycopg2', though you can simplify that by
  using the 'django.db.backends.postgresql' name added in Django 1.9.

• django.shortcuts.render_to_response() is deprecated in favor of
  :func:django.shortcuts.render. render() takes the same arguments
  except that it also requires a request.

• The DEFAULT_CONTENT_TYPE setting is deprecated. It doesn't interact well
  well with third-party apps and is obsolete since HTML5 has mostly superseded
  XHTML.

• HttpRequest.xreadlines() is deprecated in favor of iterating over the
  request.

• The field_name keyword argument to :meth:.QuerySet.earliest and
  :meth:.QuerySet.latest is deprecated in favor of passing the field
  names as arguments. Write .earliest('pub_date') instead of
  .earliest(field_name='pub_date').

.. _removed-features-2.0:

Features removed in 2.0

These features have reached the end of their deprecation cycle and are removed
in Django 2.0.

See :ref:deprecated-features-1.9 for details on these changes, including how
to remove usage of these features.

• The weak argument to django.dispatch.signals.Signal.disconnect() is
  removed.

• django.db.backends.base.BaseDatabaseOperations.check_aggregate_support()
  is removed.

• The django.forms.extras package is removed.

• The assignment_tag helper is removed.

• The host argument to SimpleTestCase.assertsRedirects() is removed.
  The compatibility layer which allows absolute URLs to be considered equal to
  relative ones when the path is identical is also removed.

• Field.rel and Field.remote_field.to are removed.

• The on_delete argument for ForeignKey and OneToOneField is now
  required in models and migrations. Consider squashing migrations so that you
  have fewer of them to update.

• django.db.models.fields.add_lazy_relation() is removed.

• When time zone support is enabled, database backends that don't support time
  zones no longer convert aware datetimes to naive values in UTC anymore when
  such values are passed as parameters to SQL queries executed outside of the
  ORM, e.g. with cursor.execute().

• django.contrib.auth.tests.utils.skipIfCustomUser() is removed.

• The GeoManager and GeoQuerySet classes are removed.

• The django.contrib.gis.geoip module is removed.

• The supports_recursion check for template loaders is removed from:

• django.template.engine.Engine.find_template()
• django.template.loader_tags.ExtendsNode.find_template()
• django.template.loaders.base.Loader.supports_recursion()
• django.template.loaders.cached.Loader.supports_recursion()

• The load_template and load_template_sources template loader methods
  are removed.

• The template_dirs argument for template loaders is removed:

• django.template.loaders.base.Loader.get_template()
• django.template.loaders.cached.Loader.cache_key()
• django.template.loaders.cached.Loader.get_template()
• django.template.loaders.cached.Loader.get_template_sources()
• django.template.loaders.filesystem.Loader.get_template_sources()

• django.template.loaders.base.Loader.__call__() is removed.

• Support for custom error views that don't accept an exception parameter
  is removed.

• The mime_type attribute of django.utils.feedgenerator.Atom1Feed and
  django.utils.feedgenerator.RssFeed is removed.

• The app_name argument to include() is removed.

• Support for passing a 3-tuple (including admin.site.urls) as the first
  argument to include() is removed.

• Support for setting a URL instance namespace without an application namespace
  is removed.

• Field._get_val_from_obj() is removed.

• django.template.loaders.eggs.Loader is removed.

• The current_app parameter to the contrib.auth function-based views is
  removed.

• The callable_obj keyword argument to
  SimpleTestCase.assertRaisesMessage() is removed.

• Support for the allow_tags attribute on ModelAdmin methods is
  removed.

• The enclosure keyword argument to SyndicationFeed.add_item() is
  removed.

• The django.template.loader.LoaderOrigin and
  django.template.base.StringOrigin aliases for
  django.template.base.Origin are removed.

See :ref:deprecated-features-1.10 for details on these changes.

• The makemigrations --exit option is removed.

• Support for direct assignment to a reverse foreign key or many-to-many
  relation is removed.

• The get_srid() and set_srid() methods of
  django.contrib.gis.geos.GEOSGeometry are removed.

• The get_x(), set_x(), get_y(), set_y(), get_z(), and
  set_z() methods of django.contrib.gis.geos.Point are removed.

• The get_coords() and set_coords() methods of
  django.contrib.gis.geos.Point are removed.

• The cascaded_union property of django.contrib.gis.geos.MultiPolygon
  is removed.

• django.utils.functional.allow_lazy() is removed.

• The shell --plain option is removed.

• The django.core.urlresolvers module is removed in favor of its new
  location, django.urls.

• CommaSeparatedIntegerField is removed, except for support in historical
  migrations.

• The template Context.has_key() method is removed.

• Support for the django.core.files.storage.Storage.accessed_time(),
  created_time(), and modified_time() methods is removed.

• Support for query lookups using the model name when
  Meta.default_related_name is set is removed.

• The MySQL __search lookup is removed.

• The shim for supporting custom related manager classes without a
  _apply_rel_filters() method is removed.

• Using User.is_authenticated() and User.is_anonymous() as methods
  rather than properties is no longer supported.

• The Model._meta.virtual_fields attribute is removed.

• The keyword arguments virtual_only in Field.contribute_to_class() and
  virtual in Model._meta.add_field() are removed.

• The javascript_catalog() and json_catalog() views are removed.

• django.contrib.gis.utils.precision_wkt() is removed.

• In multi-table inheritance, implicit promotion of a OneToOneField to a
  parent_link is removed.

• Support for Widget._format_value() is removed.

• FileField methods get_directory_name() and get_filename() are
  removed.

• The mark_for_escaping() function and the classes it uses: EscapeData,
  EscapeBytes, EscapeText, EscapeString, and EscapeUnicode are
  removed.

• The escape filter now uses django.utils.html.conditional_escape().

• Manager.use_for_related_fields is removed.

• Model Manager inheritance follows MRO inheritance rules. The requirement
  to use Meta.manager_inheritance_from_future to opt-in to the behavior is
  removed.

• Support for old-style middleware using settings.MIDDLEWARE_CLASSES is
  removed.

===========================

1.11.9

===========================

January 1, 2018

Django 1.11.9 fixes several bugs in 1.11.8.

Bugfixes

• Fixed a regression in Django 1.11 that added newlines between MultiWidget's
  subwidgets (:ticket:28890).

• Fixed incorrect class-based model index name generation for models with
  quoted db_table (:ticket:28876).

• Fixed incorrect foreign key constraint name for models with quoted
  db_table (:ticket:28876).

• Fixed a regression in caching of a GenericForeignKey when the referenced
  model instance uses more than one level of multi-table inheritance
  (:ticket:28856).

===========================

1.11.8

===========================

December 2, 2017

Django 1.11.8 fixes several bugs in 1.11.7.

Bugfixes

• Reallowed, following a regression in Django 1.10, AuthenticationForm to
  raise the inactive user error when using ModelBackend (:ticket:28645).

• Added support for QuerySet.values() and values_list() for
  union(), difference(), and intersection() queries
  (:ticket:28781).

• Fixed incorrect index name truncation when using a namespaced db_table
  (:ticket:28792).

• Made QuerySet.iterator() use server-side cursors on PostgreSQL after
  values() and values_list() (:ticket:28817).

• Fixed crash on SQLite and MySQL when ordering by a filtered subquery that
  uses nulls_first or nulls_last (:ticket:28848).

• Made query lookups for CICharField, CIEmailField, and CITextField
  use a citext cast (:ticket:28702).

• Fixed a regression in caching of a GenericForeignKey when the referenced
  model instance uses multi-table inheritance (:ticket:28856).

• Fixed "Cannot change column 'x': used in a foreign key constraint" crash on
  MySQL with a sequence of AlterField and/or RenameField operations in
  a migration (:ticket:28305).

===========================

django-parsley 0.6 -> 0.7

0.7

• Added support for django 1.10 and 1.11
• Dropped support for django 1.3 - 1.7
• Fixed issue where django version would be forced to 1.8 when installing django-parsley. More details <a href="pip install django-parsley replaces Django 1.8.3 with 1.8 agiliq/Django-parsley#72" target="_blank">here</a>.
• Added parsleyfy template tag

futures 3.1.1 -> 3.2.0

3.2.0

=====

• The ThreadPoolExecutor class constructor now accepts an optional thread_name_prefix
  argument to make it possible to customize the names of the threads created by the pool.
  Upstream contribution by Gregory P. Smith in https://bugs.python.org/issue27664.
• Backported fixes from upstream (thanks Lisandro Dalcin):

• python/cpython1560
• python/cpython3270
• python/cpython3830

html5lib 0.999999999 -> 1.0.1

1.0.1

Released on December 7, 2017

Breaking changes:

• Drop support for Python 2.6. (330) (Thank you, Hugo, Will Kahn-Greene!)
• Remove utils/spider.py (353) (Thank you, Jon Dufresne!)

Features:

• Improve documentation. (300, 307) (Thank you, Jon Dufresne, Tom Most,
  Will Kahn-Greene!)
• Add iframe seamless boolean attribute. (Thank you, Ritwik Gupta!)
• Add itemscope as a boolean attribute. (194) (Thank you, Jonathan Vanasco!)
• Support Python 3.6. (333) (Thank you, Jon Dufresne!)
• Add CI support for Windows using AppVeyor. (Thank you, John Vandenberg!)
• Improve testing and CI and add code coverage (323, 334), (Thank you, Jon
  Dufresne, John Vandenberg, Geoffrey Sneddon, Will Kahn-Greene!)
• Semver-compliant version number.

Bug fixes:

• Add support for setuptools < 18.5 to support environment markers. (Thank you,
  John Vandenberg!)
• Add explicit dependency for six >= 1.9. (Thank you, Eric Amorde!)
• Fix regexes to work with Python 3.7 regex adjustments. (318, 379) (Thank
  you, Benedikt Morbach, Ville Skyttä, Mark Vasilkov!)
• Fix alphabeticalattributes filter namespace bug. (324) (Thank you, Will
  Kahn-Greene!)
• Include license file in generated wheel package. (350) (Thank you, Jon
  Dufresne!)
• Fix annotation-xml typo. (339) (Thank you, Will Kahn-Greene!)
• Allow uppercase hex chararcters in CSS colour check. (377) (Thank you,
  Komal Dembla, Hugo!)

1.0

Released and unreleased on December 7, 2017. Badly packaged release.

1.0b3

Released on July 24, 2013

• Removed RecursiveTreeWalker from treewalkers._base. Any
  implementation using it should be moved to
  NonRecursiveTreeWalker, as everything bundled with html5lib has
  for years.

• Fix 67 so that BufferedStream to correctly returns a bytes
  object, thereby fixing any case where html5lib is passed a
  non-seekable RawIOBase-like object.

1.0b2

Released on June 27, 2013

• Removed reordering of attributes within the serializer. There is now
  an alphabetical_attributes option which preserves the previous
  behaviour through a new filter. This allows attribute order to be
  preserved through html5lib if the tree builder preserves order.

• Removed dom2sax from DOM treebuilders. It has been replaced by
  treeadapters.sax.to_sax which is generic and supports any
  treewalker; it also resolves all known bugs with dom2sax.

• Fix treewalker assertions on hitting bytes strings on
  Python 2. Previous to 1.0b1, treewalkers coped with mixed
  bytes/unicode data on Python 2; this reintroduces this prior
  behaviour on Python 2. Behaviour is unchanged on Python 3.

1.0b1

Released on May 17, 2013

• Implementation updated to implement the HTML specification <http://www.whatwg.org/specs/web-apps/current-work/>_ as of 5th May
  2013 (SVN <http://svn.whatwg.org/webapps/>_ revision r7867).

• Python 3.2+ supported in a single codebase using the six library.

• Removed support for Python 2.5 and older.

• Removed the deprecated Beautiful Soup 3 treebuilder.
  beautifulsoup4 can use html5lib as a parser instead. Note that
  since it doesn't support namespaces, foreign content like SVG and
  MathML is parsed incorrectly.

• Removed simpletree from the package. The default tree builder is
  now etree (using the xml.etree.cElementTree implementation if
  available, and xml.etree.ElementTree otherwise).

• Removed the XHTMLSerializer as it never actually guaranteed its
  output was well-formed XML, and hence provided little of use.

• Removed default DOM treebuilder, so html5lib.treebuilders.dom is no
  longer supported. html5lib.treebuilders.getTreeBuilder("dom") will
  return the default DOM treebuilder, which uses xml.dom.minidom.

• Optional heuristic character encoding detection now based on
  charade for Python 2.6 - 3.3 compatibility.

• Optional Genshi treewalker support fixed.

• Many bugfixes, including:

• 33: null in attribute value breaks XML AttValue;

• 4: nested, indirect descendant, <button> causes infinite loop;

• Google Code 215 <http://code.google.com/p/html5lib/issues/detail?id=215>_: Properly
  detect seekable streams;

• Google Code 206 <http://code.google.com/p/html5lib/issues/detail?id=206>_: add
  support for <video preload=...>, <audio preload=...>;

• Google Code 205 <http://code.google.com/p/html5lib/issues/detail?id=205>_: add
  support for <video poster=...>;

• Google Code 202 <http://code.google.com/p/html5lib/issues/detail?id=202>_: Unicode
  file breaks InputStream.

• Source code is now mostly PEP 8 compliant.

• Test harness has been improved and now depends on nose.

• Documentation updated and moved to https://html5lib.readthedocs.io/.

nose2-html-report 0.5.0 -> 0.6.0

0.6.0

• 5 add metadata to template context (artragis)

numba 0.35.0 -> 0.36.2

0.36.2

---

This is a bugfix release that provides minor changes to address:

• PR 2645: Avoid CPython bug with exec in older 2.7.x.
• PR 2652: Add support for CUDA 9.

0.36.1

---

This release continues to add new features to the work undertaken in partnership
with Intel on ParallelAccelerator technology. Other changes of note include the
compilation chain being updated to use LLVM 5.0 and the production of conda
packages using conda-build 3 and the new compilers that ship with it.

NOTE: A version 0.36.0 was tagged for internal use but not released.

ParallelAccelerator:

NOTE: The ParallelAccelerator technology is under active development and should
be considered experimental.

New features relating to ParallelAccelerator, from work undertaken with Intel,
include the addition of the stencil decorator for ease of implementation of
stencil-like computations, support for general reductions, and slice and
range fusion for parallel slice/bit-array assignments. Documentation on both the
use and implementation of the above has been added. Further, a new debug
environment variable NUMBA_DEBUG_ARRAY_OPT_STATS is made available to give
information about which operators/calls are converted to parallel for-loops.

ParallelAccelerator features:

• PR 2457: Stencil Computations in ParallelAccelerator
• PR 2548: Slice and range fusion, parallelizing bitarray and slice assignment
• PR 2516: Support general reductions in ParallelAccelerator

ParallelAccelerator fixes:

• PR 2540: Fix bug 2537
• PR 2566: Fix issue 2564.
• PR 2599: Fix nested multi-dimensional parfor type inference issue
• PR 2604: Fixes for stencil tests and cmath sin().
• PR 2605: Fixes issue 2603.

Additional features of note:

This release of Numba (and llvmlite) is updated to use LLVM version 5.0 as the
compiler back end, the main change to Numba to support this was the addition of
a custom symbol tracker to avoid the calls to LLVM's ExecutionEngine that was
crashing when asking for non-existent symbol addresses. Further, the conda
packages for this release of Numba are built using conda build version 3 and the
new compilers/recipe grammar that are present in that release.

• PR 2568: Update for LLVM 5
• PR 2607: Fixes abort when getting address to "nrt_unresolved_abort"
• PR 2615: Working towards conda build 3

Thanks to community feedback and bug reports, the following fixes were also
made.

Misc fixes/enhancements:

• PR 2534: Add tuple support to np.take.
• PR 2551: Rebranding fix
• PR 2552: relative doc links
• PR 2570: Fix issue 2561, handle missing successor on loop exit
• PR 2588: Fix 2555. Disable libpython.so linking on linux
• PR 2601: Update llvmlite version dependency.
• PR 2608: Fix potential cache file collision
• PR 2612: Fix NRT test failure due to increased overhead when running in coverage
• PR 2619: Fix dubious pthread_cond_signal not in lock
• PR 2622: Fix np.nanmedian for all NaN case.
• PR 2633: Fix markdown in CONTRIBUTING.md
• PR 2635: Make the dependency on compilers for AOT optional.

CUDA support fixes:

• PR 2523: Fix invalid cuda context in memory transfer calls in another thread
• PR 2575: Use CPU to initialize xoroshiro states for GPU RNG. Fixes 2573
• PR 2581: Fix cuda gufunc mishandling of scalar arg as array and out argument

numpy 1.13.3 -> 1.14.0

1.14.0

==========================

Numpy 1.14.0 is the result of seven months of work and contains a large number
of bug fixes and new features, along with several changes with potential
compatibility issues. The major change that users will notice are the
stylistic changes in the way numpy arrays and scalars are printed, a change
that will affect doctests. See below for details on how to preserve the
old style printing when needed.

A major decision affecting future development concerns the schedule for
dropping Python 2.7 support in the runup to 2020. The decision has been made to
support 2.7 for all releases made in 2018, with the last release being
designated a long term release with support for bug fixes extending through
2019. In 2019 support for 2.7 will be dropped in all new releases. More details
can be found in the relevant NEP_.

This release supports Python 2.7 and 3.4 - 3.6.

.. _NEP: https://github.com/numpy/numpy/blob/master/doc/neps/dropping-python2.7-proposal.rst

Highlights

• The np.einsum function uses BLAS when possible

• genfromtxt, loadtxt, fromregex and savetxt can now handle
  files with arbitrary Python supported encoding.

• Major improvements to printing of NumPy arrays and scalars.

New functions

• parametrize: decorator added to numpy.testing

• chebinterpolate: Interpolate function at Chebyshev points.

• format_float_positional and format_float_scientific : format
  floating-point scalars unambiguously with control of rounding and padding.

• PyArray_ResolveWritebackIfCopy and PyArray_SetWritebackIfCopyBase,
  new C-API functions useful in achieving PyPy compatibity.

Deprecations

• Using np.bool_ objects in place of integers is deprecated. Previously
  operator.index(np.bool_) was legal and allowed constructs such as
  [1, 2, 3][np.True_]. That was misleading, as it behaved differently from
  np.array([1, 2, 3])[np.True_].

• Truth testing of an empty array is deprecated. To check if an array is not
  empty, use array.size > 0.

• Calling np.bincount with minlength=None is deprecated.
  minlength=0 should be used instead.

• Calling np.fromstring with the default value of the sep argument is
  deprecated. When that argument is not provided, a broken version of
  np.frombuffer is used that silently accepts unicode strings and -- after
  encoding them as either utf-8 (python 3) or the default encoding
  (python 2) -- treats them as binary data. If reading binary data is
  desired, np.frombuffer should be used directly.

• The style option of array2string is deprecated in non-legacy printing mode.

• PyArray_SetUpdateIfCopyBase has been deprecated. For NumPy versions >= 1.14
  use PyArray_SetWritebackIfCopyBase instead, see C API changes below for
  more details.

• The use of UPDATEIFCOPY arrays is deprecated, see C API changes below
  for details. We will not be dropping support for those arrays, but they are
  not compatible with PyPy.

Future Changes

• np.issubdtype will stop downcasting dtype-like arguments.
  It might be expected that issubdtype(np.float32, 'float64') and
  issubdtype(np.float32, np.float64) mean the same thing - however, there
  was an undocumented special case that translated the former into
  issubdtype(np.float32, np.floating), giving the surprising result of True.

This translation now gives a warning that explains what translation is
occurring. In the future, the translation will be disabled, and the first
example will be made equivalent to the second.

• np.linalg.lstsq default for rcond will be changed. The rcond
  parameter to np.linalg.lstsq will change its default to machine precision
  times the largest of the input array dimensions. A FutureWarning is issued
  when rcond is not passed explicitly.

• a.flat.__array__() will return a writeable copy of a when a is
  non-contiguous. Previously it returned an UPDATEIFCOPY array when a was
  writeable. Currently it returns a non-writeable copy. See gh-7054 for a
  discussion of the issue.

• Unstructured void array's .item method will return a bytes object. In the
  future, calling .item() on arrays or scalars of np.void datatype will
  return a bytes object instead of a buffer or int array, the same as
  returned by bytes(void_scalar). This may affect code which assumed the
  return value was mutable, which will no longer be the case. A
  FutureWarning is now issued when this would occur.

Compatibility notes

The mask of a masked array view is also a view rather than a copy

There was a FutureWarning about this change in NumPy 1.11.x. In short, it is
now the case that, when changing a view of a masked array, changes to the mask
are propagated to the original. That was not previously the case. This change
affects slices in particular. Note that this does not yet work properly if the
mask of the original array is nomask and the mask of the view is changed.
See gh-5580 for an extended discussion. The original behavior of having a copy
of the mask can be obtained by calling the unshare_mask method of the view.

np.ma.masked is no longer writeable

Attempts to mutate the masked constant now error, as the underlying arrays
are marked readonly. In the past, it was possible to get away with::

emulating a function that sometimes returns np.ma.masked

val = random.choice([np.ma.masked, 10])
var_arr = np.asarray(val)
val_arr += 1 now errors, previously changed np.ma.masked.data

np.ma functions producing fill_values have changed

Previously, np.ma.default_fill_value would return a 0d array, but
np.ma.minimum_fill_value and np.ma.maximum_fill_value would return a
tuple of the fields. Instead, all three methods return a structured np.void
object, which is what you would already find in the .fill_value attribute.

Additionally, the dtype guessing now matches that of np.array - so when
passing a python scalar x, maximum_fill_value(x) is always the same as
maximum_fill_value(np.array(x)). Previously x = long(1) on Python 2
violated this assumption.

a.flat.__array__() returns non-writeable arrays when a is non-contiguous

The intent is that the UPDATEIFCOPY array previously returned when a was
non-contiguous will be replaced by a writeable copy in the future. This
temporary measure is aimed to notify folks who expect the underlying array be
modified in this situation that that will no longer be the case. The most
likely places for this to be noticed is when expressions of the form
np.asarray(a.flat) are used, or when a.flat is passed as the out
parameter to a ufunc.

np.tensordot now returns zero array when contracting over 0-length dimension

Previously np.tensordot raised a ValueError when contracting over 0-length
dimension. Now it returns a zero array, which is consistent with the behaviour
of np.dot and np.einsum.

numpy.testing reorganized

This is not expected to cause problems, but possibly something has been left
out. If you experience an unexpected import problem using numpy.testing
let us know.

np.asfarray no longer accepts non-dtypes through the dtype argument

This previously would accept dtype=some_array, with the implied semantics
of dtype=some_array.dtype. This was undocumented, unique across the numpy
functions, and if used would likely correspond to a typo.

1D np.linalg.norm preserves float input types, even for arbitrary orders

Previously, this would promote to float64 when arbitrary orders were
passed, despite not doing so under the simple cases::

>>> f32 = np.float32([1, 2])
>>> np.linalg.norm(f32, 2.0).dtype
dtype('float32')
>>> np.linalg.norm(f32, 2.0001).dtype
dtype('float64') numpy 1.13
dtype('float32') numpy 1.14

This change affects only float32 and float16 arrays.

count_nonzero(arr, axis=()) now counts over no axes, not all axes

Elsewhere, axis==() is always understood as "no axes", but
count_nonzero had a special case to treat this as "all axes". This was
inconsistent and surprising. The correct way to count over all axes has always
been to pass axis == None.

__init__.py files added to test directories

This is for pytest compatibility in the case of duplicate test file names in
the different directories. As a result, run_module_suite no longer works,
i.e., python <path-to-test-file> results in an error.

.astype(bool) on unstructured void arrays now calls bool on each element

On Python 2, void_array.astype(bool) would always return an array of
True, unless the dtype is V0. On Python 3, this operation would usually
crash. Going forwards, astype matches the behavior of bool(np.void),
considering a buffer of all zeros as false, and anything else as true.
Checks for V0 can still be done with arr.dtype.itemsize == 0.

MaskedArray.squeeze never returns np.ma.masked

np.squeeze is documented as returning a view, but the masked variant would
sometimes return masked, which is not a view. This has been fixed, so that
the result is always a view on the original masked array.
This breaks any code that used masked_arr.squeeze() is np.ma.masked, but
fixes code that writes to the result of .squeeze().

Renamed first parameter of can_cast from from to from_

The previous parameter name from is a reserved keyword in Python, which made
it difficult to pass the argument by name. This has been fixed by renaming
the parameter to from_.

isnat raises TypeError when passed wrong type

The ufunc isnat used to raise a ValueError when it was not passed
variables of type datetime or timedelta. This has been changed to
raising a TypeError.

dtype.__getitem__ raises TypeError when passed wrong type

When indexed wi

[puruckertom]

kaibash