James Oakley

I had the same issue and it went away after upgrading to 1.6.1.

Hmm. Not sure why Travis results are no longer showing up here, but they are getting executed. Here are the results for this PR: https://travis-ci.org/github/jimfunk/django-postgresql-netfields/builds/765115494 Looks like an error for...

I've never seen anything like that. Seems like a postgres-level thing. Have you seen https://stackoverflow.com/questions/44952432/delete-query-in-postgres-hangs-indefinitely ?

Odd. Perhaps it has something to do with the how Django sets up the Gist index. That would not be part of this project.

I've never used custom natural keys like this, but I just looked at the docs and it appears you may need to handle the conversions manually, which makes sense since...

As someone who just ran into this, I would like to add my 2 cents. My problem with the approach taken is that it makes supporting older and newer versions...

@PierreF When I made my suggestion I didn't realize that MQTT5 changed the values of the reason codes. That is unfortunate and certainly makes the heuristic a bit more difficult....

As of right now the only way I can see is to use `extra(where=...)` as described at https://docs.djangoproject.com/en/5.1/ref/models/querysets/#extra To add support in `filter()` we would need to register array variants...

I just made a test to prepare to fix this: ``` def test_ipv6_strip_scope(self): instance = self.model.objects.create(field='2001:db8::1%foo') instance = self.model.objects.get(pk=instance.pk) self.assertEqual(str(instance.field), '2001:db8::1') ``` However, it is only failing on Python 3.8,...

OK, have you re-checked in your environments? There is a possibility that the original issue is handled in new psycopg3 or PostgreSQL. (I tested against 17)