✨ Use bulk_write for save operations, including referenced documents

[tiangolo]

✨ Use bulk_write for save operations, including referenced documents

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (a99a258) 99.38% compared to head (867bfaf) 99.39%.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #271   +/-   ##
=======================================
  Coverage   99.38%   99.39%           
=======================================
  Files          50       50           
  Lines        5081     5127   +46     
  Branches      485      491    +6     
=======================================
+ Hits         5050     5096   +46     
  Misses         31       31           

Flag	Coverage Δ
tests-3.10-4.4-standalone	93.91% <96.33%> (+0.01%)	⬆️
tests-3.10-5-standalone	93.91% <96.33%> (+0.01%)	⬆️
tests-3.10-6-standalone	93.91% <96.33%> (+0.01%)	⬆️
tests-3.7-4.4-standalone	92.31% <96.29%> (+0.02%)	⬆️
tests-3.7-5-standalone	92.31% <96.29%> (+0.02%)	⬆️
tests-3.7-6-standalone	92.31% <96.29%> (+0.02%)	⬆️
tests-3.8-4-replicaSet	98.38% <100.00%> (+0.01%)	⬆️
tests-3.8-4.2-sharded	93.03% <96.33%> (+0.02%)	⬆️
tests-3.8-4.4-standalone	93.83% <96.33%> (+0.01%)	⬆️
tests-3.8-5-standalone	93.83% <96.33%> (+0.01%)	⬆️
tests-3.8-6-standalone	93.83% <96.33%> (+0.01%)	⬆️
tests-3.9-4.4-standalone	93.83% <96.33%> (+0.01%)	⬆️
tests-3.9-5-standalone	93.83% <96.33%> (+0.01%)	⬆️
tests-3.9-6-standalone	93.83% <96.33%> (+0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

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

[art049]

Just a few details to fix but awesome work 🔥.
Also, really nice _get_collection_from_name refactor 😉

[tiangolo]

Thanks for the review! I added the changes. 🤓

There's only one that I don't really know how to do, I added a comment about it.

I also included/merged #308 in this PR to be able to run CI and check if I'm breaking anything else.

[codspeed-hq]

CodSpeed Performance Report

Merging #271 will improve performances by ×3.5

_{Comparing tiangolo:use-bulk-write (867bfaf) with master (a99a258)}

Summary

⚡ 3 improvements
✅ 12 untouched benchmarks

Benchmarks breakdown

	Benchmark	master	tiangolo:use-bulk-write	Change
⚡	test_write_small_bulk[100]	198.4 ms	57.2 ms	×3.5
⚡	test_write_small_bulk[50]	99.4 ms	29.6 ms	×3.4
⚡	test_write_small_bulk[10]	20.1 ms	7.4 ms	×2.7

[tiangolo]

I'm struggling to debug the current errors, I had never used tox outside this project. I managed to install and setup all the dependencies needed by tox (a python installation for each version, with the exact name, etc). And I was able to replicate the error running a single tox environment with:

$ tox -e python3.9-motor24-pydantic18

And it consistently errors out the same way as in CI.

Then I'm trying to replicate it outside of tox to be able to investigate and debug it, I installed a local python3.9, motor 2.4, pydantic 1.8, and ran the same test, and it always passed.

I'm not sure what else to try as it seems like something particular to the tox-specific environment that I can't access and debug. Or maybe I'm doing something wrong, but I ran out of ideas of what else to try to check it. 🤔 ...also to see if it's an actual new error or some quirk of the setup I have.

[tiangolo]

I made another PR that doesn't do anything, just to trigger CI, and it seems it's triggering the same errors: #373

So I suspect the current errors are not something I changed in this PR but something that is currently broken in CI, maybe a tox change or something?

Thinking of all that, would you consider GitHub Actions' integrated matrices instead of tox? https://docs.github.com/en/actions/using-jobs/using-a-matrix-for-your-jobs

[art049]

I think we can probably merge after the v1 release :)
Btw with the bench, the speedup is amazing; this is super exciting, thank you. 👏