Camera actions are relatively easy to define because they only require one frame to complete.
All camera actions should be subclasses of the abstract class CameraAction. This class overrides get_ongoing_commands to return [] (because this function isn't used at all). In the default API, RotateCamera, LookAt, MoveCamera, and ResetCamera are subclasses of CameraAction.
Because camera actions require only one frame, you'll want to override get_initialization_commands to send the commands and set_status_after_initialization to immediately set self.status to something other than ActionStatus.ongoing, thereby ending the action.
This is an example subclass of CameraAction that sets the camera rotation to a target quaternion:
from typing import List, Dict
from magnebot.action_status import ActionStatus
from magnebot.magnebot_dynamic import MagnebotDynamic
from magnebot.magnebot_static import MagnebotStatic
from magnebot.image_frequency import ImageFrequency
from magnebot.actions.camera_action import CameraAction
class SetCameraQuaternion(CameraAction):
def __init__(self, quaternion: Dict[str, float]):
super().__init__()
self.quaternion: Dict[str, float] = quaternion
def get_initialization_commands(self, resp: List[bytes], static: MagnebotStatic, dynamic: MagnebotDynamic,
image_frequency: ImageFrequency) -> List[dict]:
commands = super().get_initialization_commands(resp=resp, static=static, dynamic=dynamic,
image_frequency=image_frequency)
commands.append({"$type": "rotate_sensor_container_to",
"rotation": self.quaternion,
"avatar_id": static.avatar_id})
return commands
def set_status_after_initialization(self) -> None:
self.status = ActionStatus.success