Add excluded_field_kwargs to support custom OneToOneFields.

[tim-schilling]

Support custom OneToOneFields that specify additional arguments that don't exist in ForeignKey.__init__.

Description

These additional arguments can be specified via excluded_field_kwargs to allow these custom OneToOneFields be coerced into ForeignKeys on the historical model.

This adds the excluded_field_kwargs parameter to HistoricalRecords. It is a dictionary mapping fields to list of parameters that should be excluded when crafting the ForeignKey instance for the historical model.

Related Issue

Fixes #870

To reproduce, the following model can be added and then attempt to run makemigrations and migrate.

class CustomAttrNameOneToOneField(models.OneToOneField):
    def __init__(self, *args, **kwargs):
        self.attr_name = kwargs.pop("attr_name", None)
        super(CustomAttrNameOneToOneField, self).__init__(*args, **kwargs)

    def get_attname(self):
        return self.attr_name or super(CustomAttrNameOneToOneField, self).get_attname()

    def deconstruct(self):
        name, path, args, kwargs = super(
            CustomAttrNameOneToOneField, self
        ).deconstruct()
        if self.attr_name:
            kwargs["attr_name"] = self.attr_name
        return name, path, args, kwargs


class ModelWithCustomAttrOneToOneField(models.Model):
    poll = CustomAttrNameOneToOneField(Poll, models.CASCADE, attr_name="custom_poll")
    history = HistoricalRecords(excluded_field_kwargs={"poll": set(["attr_name"])})

Motivation and Context

Coercing OneToOneFields to be ForeignKeys on historical models causes issues when the OneToOneField is a custom subclass that includes additional arguments that do not exist on the ForeignKey.__init__ method.

How Has This Been Tested?

The problem was able to be reproduced by adding CustomAttrNameOneToOneField and ModelWithCustomAttrOneToOneField to the test app. There's an additional test that verifies the constraint of this change in that the additional arguments are not available on the ForeignKey field. This is done via TestCustomAttrOneToOneField.

Screenshots (if appropriate):

N/A

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist:

• I have run the make format command to format my code
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have added tests to cover my changes.
• I have added my name and/or github handle to AUTHORS.rst
• I have added my change to CHANGES.rst
• All new and existing tests passed.

[tim-schilling]

I didn't check the tests case because I did not run the entire testing matrix locally.

[tim-schilling]

This was a result of make format.

[tim-schilling]

This was a result of make format.

[tim-schilling]

And I named the branch incorrectly... It should have been 870, not 807.

[codecov]

Codecov Report

Merging #871 (7759125) into master (353a4ab) will decrease coverage by 0.19%.
The diff coverage is 96.96%.

❗ Current head 7759125 differs from pull request most recent head 2a5274c. Consider uploading reports for the commit 2a5274c to get more accurate results

@@            Coverage Diff             @@
##           master     #871      +/-   ##
==========================================
- Coverage   97.85%   97.65%   -0.20%     
==========================================
  Files          20       19       -1     
  Lines        1025     1023       -2     
  Branches      159      154       -5     
==========================================
- Hits         1003      999       -4     
  Misses         10       10              
- Partials       12       14       +2     

Impacted Files	Coverage Δ
...istory/registry_tests/migration_test_app/models.py	94.59% <92.85%> (-1.06%)	⬇️
simple_history/models.py	97.91% <100.00%> (-0.37%)	⬇️
...ronetoonefield_modelwithcustomattronetoonefield.py	100.00% <100.00%> (ø)
simple_history/registry_tests/tests.py	96.03% <100.00%> (+0.16%)	⬆️
simple_history/manager.py	97.64% <0.00%> (-0.03%)	⬇️
simple_history/utils.py	100.00% <0.00%> (ø)
...lmodelwithcustomattrforeignkey_options_and_more.py
..._test_app/migrations/0004_history_date_indexing.py

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 353a4ab...2a5274c. Read the comment docs.

[jeking3]

Documentation and unit tests - looks good. Looks like there's one additional branch that did not get covered in unit tests. Need to rebase and fix up the collision in CHANGES. If this gets in before my other two I will rebase them to fix the resulting migration collisions.

[jeking3]

@tim-schilling there's probably just one more branch that needs coverage here, and a rebase to fix up the changes file.