#11716: Various methods in django.db.models.fields don't wrap ValueErrors and 
allow
them to escape
---------------------------------------------------+------------------------
          Reporter:  Leo                           |         Owner:  Leo
            Status:  assigned                      |     Milestone:  1.2
         Component:  Database layer (models, ORM)  |       Version:  SVN
        Resolution:                                |      Keywords:     
             Stage:  Accepted                      |     Has_patch:  1  
        Needs_docs:  0                             |   Needs_tests:  0  
Needs_better_patch:  0                             |  
---------------------------------------------------+------------------------
Comment (by kmtracey):

 In my view it is not a waste of time to ensure we don't blindside users
 with backwards-incompatible changes. Note we do attempt to document all
 backwards incompatible changes, so that people who might run into them may
 have a clue, if they read the release notes, that they could run into them
 when updating. There are no such notes included with this patch, nor was
 there even mention of the possibility of this introducing a backwards-
 incompatibility, until I brought it up. If this is to be done for 1.2, it
 needs documentation of the backwards incompatibility introduced.

 (Note claiming it's perfectly OK to add more backwards-incompatible
 changes into 1.2 because it already has so many is fallacious: each one is
 considered on its own merits. If anything I'd argue that the reasons needs
 to be even stronger as the list gets longer, not the other way around.)

 Alternatively, if there is a way to accomplish the bulk of what this
 ticket was aiming for without introducing a backwards-incompatibility,
 that would be even better, in my opinion. Hopefully I've clarified above
 what I mean by backwards-incompatible. (It is not true that every behavior
 change introduced by a bugfix introduces a backwards-incompatibility.)

 As for process, what's needed is consensus among the core devs. What we've
 got now is one (me) who has expressed concerns and none who have come
 forward to say my concerns are unfounded or overblown. While Russ accepted
 the ticket, I do not know if he was aware of the backwards-incompatibility
 introduced and did not think it enough of a problem to worry about, or if
 he did not notice it.

 I won't stand in the way if others on the core think this is a change
 worth making, even in light of the problems it may cause for existing
 code. I'll just strongly urge that it at least needs documentation of the
 backwards incompatibility.

-- 
Ticket URL: <http://code.djangoproject.com/ticket/11716#comment:19>
Django <http://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-upda...@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