Optimize transaction writes

[jamesagnew]

No description provided.

[github-actions]

Formatting check succeeded!

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 83.45%. Comparing base (406db33) to head (959dfc2).
Report is 99 commits behind head on master.

Additional details and impacted files

@@             Coverage Diff              @@
##             master    #6460      +/-   ##
============================================
- Coverage     83.54%   83.45%   -0.10%     
- Complexity    27432    27876     +444     
============================================
  Files          1707     1751      +44     
  Lines        106185   108085    +1900     
  Branches      13397    13564     +167     
============================================
+ Hits          88710    90197    +1487     
- Misses        11750    12056     +306     
- Partials       5725     5832     +107     

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

---

🚨 Try these New Features:

• Flaky Tests Detection - Detect and resolve failed and flaky tests
• JS Bundle Analysis - Avoid shipping oversized bundles

[michaelabuckley]

Mostly questions.

One correctness worry. One performance worry.

Correctness - with delete entries processed before creates, do we handle delete followed by conditional-create by identifier without flush?

Performance - we have two paths to resolve ids - numeric and string. It might be faster to only use numeric if all ids are numeric, since they will also work when resolving fhir_ids. Or they could be combined into a single query like

select res_id, forced_id, res_type from hfj_resource
where res_id in (?,?,?)
or fhir_id in (?,?,?)

[fil512]

Nice work. Really like how much it reduces the db calls.

[fil512]

A bunch of my comments went missing. I wonder if this will revive them...