Sanity check please

[Mike Dewhirst]

In the Django Admin I have a model central to a bunch of FKs and M:Ms plus a lot of processing on saving.

This is my solution which seems to work but frightens me a bit.

class Chemical(models.Model):

    # lots of fields

    def save(self, *args, **kwargs):

        self.process_stuff_based_on_field_values_pre_save()

        first = False

        if not self.id:

            first = True

            super().save(force_insert=True, *args, **kwargs)

        if self.id:

            self.create_or_update_a_bunch_of_related_records(first=first)

        super().save(force_insert=False, *args, **kwargs)

        self.create_or_update_a_bunch_of_m2m_records(first=first)

    def process_stuff_based_on_field_values_pre_save(self, first=False):
        ...



    def create_or_update_a_bunch_of_m2m_records(self, first=False):
        ...


    # lots of other methods

Thanks for any warnings, caveats

Cheers

Mike

-- 
Signed email is an absolute defence against phishing. This email has
been signed with my private key. If you import my public key you can
automatically decrypt my signature and be sure it came from me. Just
ask and I'll send it to you. Your email software can handle signing.

[Ryan Nowakowski]

Do you need to guarantee that any of this is atomic? If so then I usually wrap this kind of code in a "with transaction.atomic()".

If not, is all this slowing down your UX? If so, then you might consider doing some of this in an asynchronous task after the save using celery or something like it.  If not, then what you have is fine.

[Mike Dewhirst]

On 23/10/2022 11:55 pm, Ryan Nowakowski wrote:

Do you need to guarantee that any of this is atomic? If so then I usually wrap this kind of code in a "with transaction.atomic()".

If not, is all this slowing down your UX? If so, then you might consider doing some of this in an asynchronous task after the save using celery or something like it.  If not, then what you have is fine.

Thanks Ryan. For other reasons I'll definitely look at async.

It still seems to work fine in real life but my unit tests take exception to force_insert being a KeyError when *args, **kwargs are included in the super calls. I have now removed those *args, **kwargs with no apparent ill effects.

Cheers

Mike

--
You received this message because you are subscribed to the Google Groups "Django users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/django-users/D30805E6-C2F6-4878-829E-CC39A2D15C6B%40fattuba.com.

[Ryan Nowakowski]

On October 23, 2022 4:54:00 PM CDT, Mike Dewhirst <mi...@dewhirst.com.au> wrote:
>On 23/10/2022 11:55 pm, Ryan Nowakowski wrote:
>It still seems to work fine in real life but my unit tests take exception to force_insert being a KeyError when *args, **kwargs are included in the super calls. I have now removed those *args, **kwargs with no apparent ill effects.

Yup, I missed that when reading your code below. Typically if your specify a keyword arg like force_insert, it needs to be after *args.

>> To view this discussion on the web visit https://groups.google.com/d/msgid/django-users/D30805E6-C2F6-4878-829E-CC39A2D15C6B%40fattuba.com <https://groups.google.com/d/msgid/django-users/D30805E6-C2F6-4878-829E-CC39A2D15C6B%40fattuba.com?utm_medium=email&utm_source=footer>.
>
>

[Mike Dewhirst]

On 24/10/2022 12:59 pm, Ryan Nowakowski wrote:

On October 23, 2022 4:54:00 PM CDT, Mike Dewhirst <mi...@dewhirst.com.au> wrote:

On 23/10/2022 11:55 pm, Ryan Nowakowski wrote:
It still seems to work fine in real life but my unit tests take exception to force_insert being a KeyError when *args, **kwargs are included in the super calls. I have now removed those *args, **kwargs with no apparent ill effects.

Yup, I missed that when reading your code below. Typically if your specify a keyword arg like force_insert, it needs to be after *args.

That *is* interesting. I'll have to re-study arg sequencing.

Thanks Ryan

M

[Derek]

Link to a discussion on RealPython about args & kwargs:
https://realpython.com/python-kwargs-and-args/#ordering-arguments-in-a-function

[Mike Dewhirst]

On 24/10/2022 4:09 pm, Derek wrote:

Link to a discussion on RealPython about args & kwargs:
https://realpython.com/python-kwargs-and-args/#ordering-arguments-in-a-function

Thank you Derek. Yes that is how I originally understood it.

Turns out Django doesn't seem to like *args, **kwargs at all when calling

super().save(force_insert=False)  (or True)

Cheers

mike

To view this discussion on the web visit https://groups.google.com/d/msgid/django-users/b257523a-87ba-46f5-bf9a-3ed143b3d86bn%40googlegroups.com.