Support open archiefbeheer destruction in objects api

[CharString]

Fixes #708

[github-actions]

Bencher Report

Branch	708-support-open-archiefbeheer-destruction-in-objects-api
Testbed	ubuntu-latest

Click to view all benchmark results

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
performance_test/tests/test_objects_list.py::test_objects_api_list_filter_by_object_type	📈 view plot 🚷 view threshold	121.12 ms (-0.05%) Baseline: 121.17 ms	127.23 ms (95.19%)
performance_test/tests/test_objects_list.py::test_objects_api_list_filter_one_result	📈 view plot 🚷 view threshold	21.70 ms (-3.11%) Baseline: 22.39 ms	23.51 ms (92.27%)
performance_test/tests/test_objects_list.py::test_objects_api_list_large_page_size_page_1	📈 view plot 🚷 view threshold	302.67 ms (+2.31%) Baseline: 295.83 ms	310.62 ms (97.44%)
performance_test/tests/test_objects_list.py::test_objects_api_list_large_page_size_page_5	📈 view plot 🚷 view threshold	304.60 ms (+3.07%) Baseline: 295.52 ms	310.29 ms (98.17%)
performance_test/tests/test_objects_list.py::test_objects_api_list_small_page_size_page_20	📈 view plot 🚷 view threshold	131.06 ms (-1.00%) Baseline: 132.38 ms	139.00 ms (94.29%)

🐰 View full continuous benchmarking report in Bencher

[codecov-commenter]

Codecov Report

❌ Patch coverage is 94.06780% with 7 lines in your changes missing coverage. Please review.
✅ Project coverage is 84.80%. Comparing base (5779e78) to head (2e119cc).

Files with missing lines	Patch %	Lines
src/objects/api/v2/views.py	90.90%	2 Missing and 2 partials ⚠️
src/objects/conf/base.py	60.00%	1 Missing and 1 partial ⚠️
src/objects/core/models.py	93.75%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #719      +/-   ##
==========================================
+ Coverage   84.59%   84.80%   +0.21%     
==========================================
  Files         137      140       +3     
  Lines        2733     2837     +104     
  Branches      215      224       +9     
==========================================
+ Hits         2312     2406      +94     
- Misses        373      377       +4     
- Partials       48       54       +6     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[stevenbal]

@CharString could you look into the performance regressions for the list endpoint? There's probably a missing prefetch_related on references #719 (comment)

[CharString]

@CharString could you look into the performance regressions for the list endpoint? There's probably a missing prefetch_related on references #719 (comment)

Weird. The previous run was within bounds.
And now, after pushing OAS changes and a change to destruct that makes it faster, listing a deep data search is 10% slower?! 10% pure noise seems our benchmarking method is a bit off.

PS. Good enough to signal the missing prefetch. But bad enough to send me into the django-silk woods and postgres index docs. I did learn a lot about these fancy indices, but couldn't find a way to make the Sort in the plan any better. (The sort looks most expensive eye balling the query plan.)

[stevenbal]

Two minor things remaining

@CharString the benchmarking is far from ideal unfortunately, at the time of fixing performance problems I just needed to show at least something in CI, but it could definitely use some more attention

[stevenbal]

On second thought, let's add a separate flag to enable cloud events, same as Open Zaak: https://github.com/open-zaak/open-zaak/blob/1dc9637ee24368822fb74ab3a07631372a309ff0/src/openzaak/conf/includes/base.py#L532

[CharString]

@CharString the benchmarking is far from ideal unfortunately, at the time of fixing performance problems I just needed to show at least something in CI, but it could definitely use some more attention

@stevenbal Yeah. Could well be that it's just variance in GH runners.