* perf: optimize getBusyTimesForLimitChecks with batched parallel queries and raw SQL
Co-Authored-By: udit@cal.com <udit222001@gmail.com>
* test: add unit tests for getBusyTimesForLimitChecks batched query optimization
Co-Authored-By: udit@cal.com <udit222001@gmail.com>
* refactor: use Prisma findMany instead of raw SQL for better testability
The batching (50 user chunks) and parallel execution (Promise.all) provide
the main performance optimization. Using findMany instead of $queryRaw
maintains the same performance characteristics while being compatible
with prismock for testing.
Co-Authored-By: udit@cal.com <udit222001@gmail.com>
* fix: type error
* refactor: add integration test
* chore: remove log
* chore: remove log
* fix: avoid logging PII (userId) in busyTimes debug log
Replace JSON.stringify(busyTimes) with count to avoid exposing userId
in logs. Addresses Cubic AI review feedback (confidence 9/10).
Co-Authored-By: unknown <>
* chore: undo
---------
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
