added the outline of the PR. need to discuss the caching format

src/blueapi/core/context.py

@@ -42,6 +42,7 @@ class BlueskyContext:
     )
     plans: dict[str, Plan] = field(default_factory=dict)
     devices: dict[str, Device] = field(default_factory=dict)
+    # todo add some format to keep the lazy vs non-lazy devices
     plan_functions: dict[str, PlanGenerator] = field(default_factory=dict)
 
     _reference_cache: dict[type, type] = field(default_factory=dict)
@@ -104,6 +105,11 @@ def with_device_module(self, module: ModuleType) -> None:
 
     def with_dodal_module(self, module: ModuleType, **kwargs) -> None:
         devices, exceptions = make_all_devices(module, **kwargs)
+        # todo modify here
+        # factories = get_device_factories(module)
+
+        # for non-lazy devices, we instantiate them
+        # for lazy devices we add to the context to do when the plan is run
 
         for device in devices.values():
             self.device(device)

src/blueapi/worker/task.py

@@ -26,6 +26,8 @@ def prepare_params(self, ctx: BlueskyContext) -> BaseModel:
     def do_task(self, ctx: BlueskyContext) -> None:
         LOGGER.info(f"Asked to run plan {self.name} with {self.params}")
 
+        # todo here call ensure_connected on alpl the devices marked in the context as needing that
+        # ensure_connected(ctx.lazy_devices.values())
         func = ctx.plan_functions[self.name]
         prepared_params = self.prepare_params(ctx)
         ctx.run_engine(func(**prepared_params.dict()))