{{{
try:
instance = CustomModel.objects.get(pk=pk)
except CustomModel.DoesNotExist:
instance = CustomModel()
form = CustomModelForm(instance=instance, data=request.POST,
files=request.FILES))
if form.is_valid():
if not instance.pk:
instance.save() # saves the model while generating a pk
instance = form.save()
}}}
To my disappointment, after a few wasted hours, I figured out that binding
data to a model instance modifies that instance, so when I am calling
instance.save() it already contains data from the form, before even
calling form.save() method.
After finding this ticket: #14885 I also noticed that it may be a partial
update of the model instance. Despite closed ticket I couldn't find it in
current documentation.
In my opinion it is completely illogical. Model instance should only be
modified only upon form.save() call, this would also give a real use case
for the existence of commit=False argument: "apply cleaned form data to
model instance, but do not commit to database."
Even [https://docs.djangoproject.com/en/1.6/topics/forms/modelforms/#the-
save-method documentation] follows the logic I am proposing: "A subclass
of ModelForm can accept an existing model instance as the keyword argument
instance; if this is supplied, save() will update that instance"
Overall it sounds to me like a side effect or a bug. If it truly is an
unavoidable side-effect, I believe, we should get at least a really good
explanation of it in the documentation.
--
Ticket URL: <https://code.djangoproject.com/ticket/22346>
Django <https://code.djangoproject.com/>
The Web framework for perfectionists with deadlines.
* status: new => closed
* needs_better_patch: => 0
* resolution: => invalid
* needs_tests: => 0
* needs_docs: => 0
Comment:
Never mind, I just found a Warning in Documentation, and it actually makes
sense, since model fields need to be validate as well.
--
Ticket URL: <https://code.djangoproject.com/ticket/22346#comment:1>