#16507: Tests that alter ROOT_URLCONF should set DEBUG_PROPAGATE_EXCEPTION
----------------------------------+------------------------------------
     Reporter:  andymckay         |                    Owner:  nobody
         Type:  Bug               |                   Status:  reopened
    Component:  contrib.messages  |                  Version:  SVN
     Severity:  Normal            |               Resolution:
     Keywords:                    |             Triage Stage:  Accepted
    Has patch:  1                 |      Needs documentation:  0
  Needs tests:  0                 |  Patch needs improvement:  0
Easy pickings:  0                 |                    UI/UX:  0
----------------------------------+------------------------------------

Comment (by carljm):

 Based on #17113, I think that my comment above was bogus, just a
 misunderstanding of what was happening here based on looking at the
 failing-handler case. Thanks for confirming that, Claude - sorry for the
 misdirection.

 Andy, if you can write a test case demonstrating a situation where the
 wrong handler is actually used, please upload to #17113 and re-open.

 It seems to me that Claude is right that this bug is a simple case where
 both urlconfs use the default 500 handler and template, but that template
 tries to reverse a URL that doesn't exist in the temporary URLconf. It's
 tempting to say we should set DEBUG_PROPAGATE_EXCEPTIONS to True in the
 base `TestCase` class whenever a custom URLconf is used (or just set it
 for all test runs, as Andy suggests), but I think this is overly implicit
 and non-obvious behavior, and not backwards-compatible; it could cause
 changes in the behavior of existing tests. (And is there any other
 precedent where we force the default value of a setting implicitly, just
 for running tests? Other than DATABASES, I guess).

 Really this is not a general issue in the test suite, only in the contrib
 tests which might be run from someone's project that has a custom 500
 template. And if someone triggers it in their own tests with their own
 custom 500 template, they can deal with that. So in the end I think
 Claude's latest approach of fixing the specific test is probably the best
 option?

-- 
Ticket URL: <https://code.djangoproject.com/ticket/16507#comment:18>
Django <https://code.djangoproject.com/>
The Web framework for perfectionists with deadlines.

-- 
You received this message because you are subscribed to the Google Groups 
"Django updates" group.
To post to this group, send email to django-updates@googlegroups.com.
To unsubscribe from this group, send email to 
django-updates+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/django-updates?hl=en.

Reply via email to