bunny-lab/Borealis-Github-Replica
1
0
Fork 0
mirror of https://github.com/bunny-lab-io/Borealis.git synced 2025-10-26 20:01:57 -06:00
Code Issues Packages Projects Releases Wiki Activity

Add Engine scheduler service and job interfaces

Browse Source
This commit is contained in:
Nicole Rappe
2025-10-22 13:56:49 -06:00
parent 3524faa40f
commit d9f2a37b74
14 changed files with 1355 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
Data/Engine/README.md
View File

@@ -45,7 +45,7 @@ Step 9 introduces real-time handlers backed by the new service container:
- `Data/Engine/services/realtime/agent_registry.py` manages connected-agent state, last-seen persistence, collector updates, and screenshot caches without sharing globals with the legacy server.
- `Data/Engine/interfaces/ws/agents/events.py` ports the agent namespace, handling connect/disconnect logging, heartbeat reconciliation, screenshot relays, macro status broadcasts, and provisioning lookups through the realtime service.
- `Data/Engine/interfaces/ws/job_management/events.py` registers the job namespace; detailed scheduler coordination will arrive with the Step 10 migration.
- `Data/Engine/interfaces/ws/job_management/events.py` now forwards scheduler updates and responds to job status requests, keeping WebSocket clients informed as new runs are simulated.
The WebSocket factory (`Data/Engine/interfaces/ws/__init__.py`) now accepts the Engine service container so namespaces can resolve dependencies just like their HTTP counterparts.
@@ -69,3 +69,14 @@ Step 7 ports the first persistence adapters into the Engine:
- `Data/Engine/repositories/sqlite/enrollment_repository.py` surfaces enrollment install-code counters and device approval records so future services can operate without touching raw SQL.
Each repository accepts the shared `SQLiteConnectionFactory`, keeping all SQL execution confined to the Engine layer while services depend only on protocol interfaces.
## Job scheduling services
Step 10 migrates the foundational job scheduler into the Engine:
- `Data/Engine/builders/job_fabricator.py` transforms stored job definitions into immutable manifests, decoding scripts, resolving environment variables, and preparing execution metadata.
- `Data/Engine/repositories/sqlite/job_repository.py` encapsulates scheduled job persistence, run history, and status tracking in SQLite.
- `Data/Engine/services/jobs/scheduler_service.py` runs the background evaluation loop, emits Socket.IO lifecycle events, and exposes CRUD helpers for the HTTP and WebSocket interfaces.
- `Data/Engine/interfaces/http/job_management.py` mirrors the legacy REST surface for creating, updating, toggling, and inspecting scheduled jobs and their run history.
The scheduler service starts automatically from `Data/Engine/bootstrapper.py` once the Engine runtime builds the service container, ensuring a no-op scheduling loop executes independently of the legacy server.