Add fixtures for user request factories, clients

[tony]

trafficstars

Fixes #565

This is based on conversations in issues #553, #554, and #284

• Introduces a django_user fixture, which is a user with no additional priveleges. Similar to admin_user.

• A user_client, based on django_user. Similar to admin_client.

• 3 new RequestFactory based fixtures, for unauthenticated, authenticated, and admin users:

  • rf_unauth: relies on AnonymousUser, similar to django docs example.
  • rf_admin: relies on admin_user fixture
  • rf_user: relies on new django_user fixture

  In addition, these 3 fixtures differ from rf in that they:

  • Mimic AuthenticationMiddleware by add the user attribute to the request object
  • Mimic SessionMiddleware by adding the 'session' attribute to the request object. This is an in-memory session store object from django.contrib.sessions.backends.base.SessionBase.

This fills the gaps left over by not having normal and unauthenticated user objects, clients, and request factories (when admin ones were available), and also covers a common case likely present in Django projects - anonymous (not logged in), authenticated, and admin users.

[codecov-io]

Codecov Report

Merging #568 into master will increase coverage by 0.37%. The diff coverage is 97.53%.

@@            Coverage Diff            @@
##           master    #568      +/-   ##
=========================================
+ Coverage   91.92%   92.3%   +0.37%     
=========================================
  Files          33      33              
  Lines        1660    1741      +81     
  Branches      143     144       +1     
=========================================
+ Hits         1526    1607      +81     
  Misses         95      95              
  Partials       39      39

Flag	Coverage Δ
#dj110	84.66% <97.53%> (+0.74%)	:arrow_up:
#dj111	86.73% <97.53%> (+0.64%)	:arrow_up:
#dj18	85.46% <97.53%> (+0.7%)	:arrow_up:
#dj19	84.54% <97.53%> (+0.75%)	:arrow_up:
#dj20	84.95% <97.53%> (+0.73%)	:arrow_up:
#djmaster	84.95% <97.53%> (+0.73%)	:arrow_up:
#mysql_innodb	84.95% <97.53%> (+0.73%)	:arrow_up:
#mysql_myisam	84.89% <97.53%> (+0.73%)	:arrow_up:
#postgres	88.16% <97.53%> (+0.57%)	:arrow_up:
#py27	89.71% <97.53%> (+0.5%)	:arrow_up:
#py34	84.54% <97.53%> (+0.75%)	:arrow_up:
#py35	84.66% <97.53%> (+0.74%)	:arrow_up:
#py36	85.41% <97.53%> (+0.71%)	:arrow_up:
#sqlite	86.61% <97.53%> (+0.65%)	:arrow_up:
#sqlite_file	84.54% <97.53%> (+0.75%)	:arrow_up:

Impacted Files	Coverage Δ
pytest_django/plugin.py	85.96% <100%> (+0.2%)	:arrow_up:
tests/test_fixtures.py	100% <100%> (ø)	:arrow_up:
pytest_django/fixtures.py	96.6% <95.74%> (-0.26%)	:arrow_down:
pytest_django_test/app/views.py	100% <0%> (+20%)	:arrow_up:

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 927d6c1...0f7f6e9. Read the comment docs.

[tony]

e.g. django_rf_user, django_rf_admin, django_rf_unauth?

Is it okay if i 1. rebase against master 2. squash 5fccb31 and 722feac into my commits?

[blueyed]

Yes, yes, and yes, please.. :)

[tony]

@blueyed Updates

• rebased against master
• renamed new fixtures use django_, including in tests and documentation
• squashed in typo fixes from 5fccb31 and 722feac