Ignore null changeset on creation and deletion

[joeyjoejoejr]

Problem

I registered a nullable field of a model using include_fields. There was a a check for null changes in the log_update receiver, but not the log_create and log_delete ones. This meant that a log entry was created with null for the changes field whenever a model was created or updated, but didn't have a value for that field.

When the mixin method msg_short is called in the Audit Log admin index, it would error because it's expecting an object for changes.

What Changed:

• Made the SimpleIncludeModel label field nullable
• Added tests for creation and deletion of a model with a null label field
• Added checks to log_create and log_delete to make sure entries aren't created if the changes are None.

[codecov[bot]]

Codecov Report

Merging #282 (3aaada0) into master (31418d5) will decrease coverage by 0.18%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #282      +/-   ##
==========================================
- Coverage   82.20%   82.02%   -0.19%     
==========================================
  Files          18       18              
  Lines         489      495       +6     
==========================================
+ Hits          402      406       +4     
- Misses         87       89       +2     

Impacted Files	Coverage Δ
auditlog/diff.py	87.09% <100.00%> (ø)
auditlog/receivers.py	100.00% <100.00%> (ø)
auditlog/__init__.py	66.66% <0.00%> (-33.34%)	:arrow_down:
auditlog/admin.py	100.00% <0.00%> (ø)
auditlog/mixins.py	76.36% <0.00%> (ø)
auditlog/models.py	82.22% <0.00%> (ø)
auditlog/registry.py	91.07% <0.00%> (ø)
auditlog/middleware.py	68.42% <0.00%> (ø)
auditlog/management/commands/auditlogflush.py	0.00% <0.00%> (ø)

---

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 31418d5...3aaada0. Read the comment docs.

[hramezani]

Thanks @joeyjoejoejr for the patch :+1: Could you please rebase your patch?

[aqeelat]

@hramezani I think will be solved as a side-effect of #483. We might still need the tests but let's evaluate after we merge #483.