opentrons 8.6.0__py3-none-any.whl

opentrons/protocol_runner/legacy_command_mapper.py

@@ -0,0 +1,852 @@
+"""Translate events from a legacy ``ProtocolContext`` into Protocol Engine commands."""
+
+from collections import defaultdict
+from datetime import datetime
+from typing import Dict, List, Optional, Tuple, Union
+
+from opentrons.hardware_control.modules.types import (
+    ModuleModel as HardwareModuleModel,
+    TemperatureModuleModel,
+    MagneticModuleModel,
+    ThermocyclerModuleModel,
+    HeaterShakerModuleModel,
+    AbsorbanceReaderModel,
+    FlexStackerModuleModel,
+)
+from opentrons_shared_data.pipette.types import PipetteNameType
+from opentrons.types import MountType, DeckSlotName, Location
+from opentrons.legacy_commands import types as legacy_command_types
+from opentrons.protocol_api import InstrumentContext
+from opentrons.protocol_api.core.legacy.deck import FIXED_TRASH_ID
+from opentrons.protocol_api.core.legacy.load_info import (
+    LoadInfo as LegacyLoadInfo,
+    LabwareLoadInfo as LegacyLabwareLoadInfo,
+    InstrumentLoadInfo as LegacyInstrumentLoadInfo,
+    ModuleLoadInfo as LegacyModuleLoadInfo,
+)
+from opentrons.protocol_engine import (
+    ProtocolEngineError,
+    actions as pe_actions,
+    commands as pe_commands,
+    types as pe_types,
+)
+from opentrons.protocol_engine.error_recovery_policy import ErrorRecoveryType
+from opentrons.protocol_engine.resources import (
+    ModelUtils,
+    ModuleDataProvider,
+    pipette_data_provider,
+)
+from opentrons.protocol_engine.state.update_types import (
+    StateUpdate,
+)
+
+from opentrons_shared_data.labware.labware_definition import (
+    labware_definition_type_adapter,
+)
+from opentrons_shared_data.errors import ErrorCodes, EnumeratedError, PythonException
+
+
+class LegacyCommandParams(pe_commands.CustomParams):
+    """Custom command data payload for mapped legacy commands."""
+
+    legacyCommandType: str
+    legacyCommandText: str
+
+
+class LegacyContextCommandError(ProtocolEngineError):
+    """An error returned when a PAPIv2 ProtocolContext command fails."""
+
+    def __init__(self, wrapping_exc: BaseException) -> None:
+        if isinstance(wrapping_exc, EnumeratedError):
+            super().__init__(
+                wrapping_exc.code,
+                wrapping_exc.message,
+                wrapping_exc.detail,
+                wrapping_exc.wrapping,
+            )
+        else:
+            super().__init__(
+                code=ErrorCodes.GENERAL_ERROR,
+                message=str(wrapping_exc),
+                wrapping=[PythonException(wrapping_exc)],
+            )
+
+
+_HARDWARE_TO_PE_MODULE: Dict[HardwareModuleModel, pe_types.ModuleModel] = {
+    MagneticModuleModel.MAGNETIC_V1: pe_types.ModuleModel.MAGNETIC_MODULE_V1,
+    MagneticModuleModel.MAGNETIC_V2: pe_types.ModuleModel.MAGNETIC_MODULE_V2,
+    TemperatureModuleModel.TEMPERATURE_V1: pe_types.ModuleModel.TEMPERATURE_MODULE_V1,
+    TemperatureModuleModel.TEMPERATURE_V2: pe_types.ModuleModel.TEMPERATURE_MODULE_V2,
+    ThermocyclerModuleModel.THERMOCYCLER_V1: pe_types.ModuleModel.THERMOCYCLER_MODULE_V1,
+    ThermocyclerModuleModel.THERMOCYCLER_V2: pe_types.ModuleModel.THERMOCYCLER_MODULE_V2,
+    HeaterShakerModuleModel.HEATER_SHAKER_V1: pe_types.ModuleModel.HEATER_SHAKER_MODULE_V1,
+    AbsorbanceReaderModel.ABSORBANCE_READER_V1: pe_types.ModuleModel.ABSORBANCE_READER_V1,
+    FlexStackerModuleModel.FLEX_STACKER_V1: pe_types.ModuleModel.FLEX_STACKER_MODULE_V1,
+}
+
+_HIGHER_ORDER_COMMAND_TYPES = {
+    legacy_command_types.MIX,
+    legacy_command_types.CONSOLIDATE,
+    legacy_command_types.DISTRIBUTE,
+    legacy_command_types.TRANSFER,
+    legacy_command_types.RETURN_TIP,
+    legacy_command_types.AIR_GAP,
+}
+
+
+class LegacyCommandMapper:
+    """Map broker commands to protocol engine commands.
+
+    Each protocol should use its own instance of this class.
+    """
+
+    def __init__(
+        self, module_data_provider: Optional[ModuleDataProvider] = None
+    ) -> None:
+        """Initialize the command mapper."""
+        # commands keyed by broker message ID
+        self._commands_by_broker_id: Dict[str, pe_commands.Command] = {}
+
+        # running count of each legacy command type, to construct IDs
+        self._command_count: Dict[str, int] = defaultdict(lambda: 0)
+
+        # equipment IDs by physical location
+        self._labware_id_by_slot: Dict[DeckSlotName, str] = {
+            DeckSlotName.FIXED_TRASH: FIXED_TRASH_ID
+        }
+        self._labware_id_by_module_id: Dict[str, str] = {}
+        self._pipette_id_by_mount: Dict[MountType, str] = {}
+        self._module_id_by_slot: Dict[DeckSlotName, str] = {}
+
+        # module definition state and provider depedency
+        self._module_definition_by_model: Dict[
+            pe_types.ModuleModel, pe_types.ModuleDefinition
+        ] = {}
+        self._module_data_provider = module_data_provider or ModuleDataProvider()
+
+    def map_command(  # noqa: C901
+        self,
+        command: legacy_command_types.CommandMessage,
+    ) -> List[pe_actions.Action]:
+        """Map a legacy Broker command to a ProtocolEngine command.
+
+        A "before" message from the Broker
+        is mapped to a ``RUNNING`` ProtocolEngine command.
+
+        An "after" message from the Broker
+        is mapped to a ``SUCCEEDED`` ProtocolEngine command.
+        It has the same ID as the original ``RUNNING`` command,
+        so when you send it to the ProtocolEngine, it will update the original
+        command's status in-place.
+        """
+        command_type = command["name"]
+
+        if command_type in _HIGHER_ORDER_COMMAND_TYPES:
+            return []
+
+        command_error = command["error"]
+        stage = command["$"]
+        # TODO(mc, 2021-12-08): use message ID as command ID directly once
+        # https://github.com/Opentrons/opentrons/issues/8986 is resolved
+        broker_id = command["id"]
+        now = ModelUtils.get_timestamp()
+
+        results: List[pe_actions.Action] = []
+
+        if stage == "before":
+            count = self._command_count[command_type]
+            command_id = f"{command_type}-{count}"
+            command_create, running_command = self._build_initial_command(
+                command, command_id, now
+            )
+
+            self._command_count[command_type] = count + 1
+            self._commands_by_broker_id[broker_id] = running_command
+
+            results.append(
+                pe_actions.QueueCommandAction(
+                    command_id=command_id,
+                    created_at=running_command.createdAt,
+                    request=command_create,
+                    request_hash=None,
+                )
+            )
+            assert running_command.startedAt is not None
+            results.append(
+                pe_actions.RunCommandAction(
+                    running_command.id, started_at=running_command.startedAt
+                )
+            )
+
+        elif stage == "after":
+            running_command = self._commands_by_broker_id[broker_id]
+            completed_command: pe_commands.Command
+            if command_error is None:
+                if isinstance(running_command, pe_commands.PickUpTip):
+                    completed_command = running_command.model_copy(
+                        update={
+                            "result": pe_commands.PickUpTipResult.model_construct(
+                                tipVolume=command["payload"]["location"].max_volume,  # type: ignore[typeddict-item]
+                                tipLength=command["payload"]["instrument"].hw_pipette[  # type: ignore[typeddict-item]
+                                    "tip_length"
+                                ],
+                                position=pe_types.DeckPoint(x=0, y=0, z=0),
+                            ),
+                            "status": pe_commands.CommandStatus.SUCCEEDED,
+                            "completedAt": now,
+                            "notes": [],
+                        }
+                    )
+                elif isinstance(running_command, pe_commands.DropTip):
+                    completed_command = running_command.model_copy(
+                        update={
+                            "result": pe_commands.DropTipResult.model_construct(
+                                position=pe_types.DeckPoint(x=0, y=0, z=0)
+                            ),
+                            "status": pe_commands.CommandStatus.SUCCEEDED,
+                            "completedAt": now,
+                            "notes": [],
+                        }
+                    )
+                elif isinstance(running_command, pe_commands.Aspirate):
+                    completed_command = running_command.model_copy(
+                        update={
+                            # Don't .model_construct() result, because we want to validate
+                            # volume.
+                            "result": pe_commands.AspirateResult(
+                                volume=running_command.params.volume,
+                                position=pe_types.DeckPoint(x=0, y=0, z=0),
+                            ),
+                            "status": pe_commands.CommandStatus.SUCCEEDED,
+                            "completedAt": now,
+                            "notes": [],
+                        }
+                    )
+                elif isinstance(running_command, pe_commands.Dispense):
+                    completed_command = running_command.model_copy(
+                        update={
+                            # Don't .model_construct() result, because we want to validate
+                            # volume.
+                            "result": pe_commands.DispenseResult(
+                                volume=running_command.params.volume,
+                                position=pe_types.DeckPoint(x=0, y=0, z=0),
+                            ),
+                            "status": pe_commands.CommandStatus.SUCCEEDED,
+                            "completedAt": now,
+                            "notes": [],
+                        }
+                    )
+                elif isinstance(running_command, pe_commands.BlowOut):
+                    completed_command = running_command.model_copy(
+                        update={
+                            "result": pe_commands.BlowOutResult.model_construct(
+                                position=pe_types.DeckPoint(x=0, y=0, z=0)
+                            ),
+                            "status": pe_commands.CommandStatus.SUCCEEDED,
+                            "completedAt": now,
+                            "notes": [],
+                        }
+                    )
+                elif isinstance(running_command, pe_commands.Comment):
+                    completed_command = running_command.model_copy(
+                        update={
+                            "result": pe_commands.CommentResult.model_construct(),
+                            "status": pe_commands.CommandStatus.SUCCEEDED,
+                            "completedAt": now,
+                            "notes": [],
+                        }
+                    )
+                elif isinstance(running_command, pe_commands.Custom):
+                    completed_command = running_command.model_copy(
+                        update={
+                            "result": pe_commands.CustomResult.model_construct(),
+                            "status": pe_commands.CommandStatus.SUCCEEDED,
+                            "completedAt": now,
+                            "notes": [],
+                        }
+                    )
+                else:
+                    # TODO(mm, 2024-06-13): This looks potentially wrong.
+                    # We're creating a `SUCCEEDED` command that does not have a `result`,
+                    # which is not normally possible.
+                    completed_command = running_command.model_copy(
+                        update={
+                            "status": pe_commands.CommandStatus.SUCCEEDED,
+                            "completedAt": now,
+                            "notes": [],
+                        }
+                    )
+                results.append(
+                    pe_actions.SucceedCommandAction(
+                        completed_command,
+                        state_update=StateUpdate(),
+                    )
+                )
+
+                if isinstance(completed_command, pe_commands.WaitForResume):
+                    results.append(
+                        pe_actions.PauseAction(source=pe_actions.PauseSource.PROTOCOL)
+                    )
+
+            else:
+                results.append(
+                    pe_actions.FailCommandAction(
+                        command_id=running_command.id,
+                        running_command=running_command,
+                        error_id=ModelUtils.generate_id(),
+                        failed_at=now,
+                        error=LegacyContextCommandError(command_error),
+                        notes=[],
+                        # For legacy protocols, we don't attempt to support any kind
+                        # of error recovery at the Protocol Engine level.
+                        # These protocols only run on the OT-2, which doesn't have
+                        # any recoverable errors, anyway.
+                        type=ErrorRecoveryType.FAIL_RUN,
+                    )
+                )
+
+        return results
+
+    def map_equipment_load(self, load_info: LegacyLoadInfo) -> List[pe_actions.Action]:
+        """Map a labware, instrument (pipette), or module load to a PE command."""
+        if isinstance(load_info, LegacyLabwareLoadInfo):
+            return self._map_labware_load(load_info)
+        elif isinstance(load_info, LegacyInstrumentLoadInfo):
+            return self._map_instrument_load(load_info)
+        elif isinstance(load_info, LegacyModuleLoadInfo):
+            return self._map_module_load(load_info)
+
+    def _build_initial_command(
+        self,
+        command: legacy_command_types.CommandMessage,
+        command_id: str,
+        now: datetime,
+    ) -> Tuple[pe_commands.CommandCreate, pe_commands.Command]:
+        if command["name"] == legacy_command_types.PICK_UP_TIP:
+            return self._build_pick_up_tip(
+                command=command, command_id=command_id, now=now
+            )
+        elif command["name"] == legacy_command_types.DROP_TIP:
+            return self._build_drop_tip(command=command, command_id=command_id, now=now)
+
+        elif (
+            command["name"] == legacy_command_types.ASPIRATE
+            or command["name"] == legacy_command_types.DISPENSE
+        ):
+            return self._build_liquid_handling(
+                command=command, command_id=command_id, now=now
+            )
+        elif command["name"] == legacy_command_types.BLOW_OUT:
+            return self._build_blow_out(command=command, command_id=command_id, now=now)
+        elif command["name"] == legacy_command_types.PAUSE:
+            wait_for_resume_running = pe_commands.WaitForResume.model_construct(
+                id=command_id,
+                key=command_id,
+                status=pe_commands.CommandStatus.RUNNING,
+                createdAt=now,
+                startedAt=now,
+                params=pe_commands.WaitForResumeParams.model_construct(
+                    message=command["payload"]["userMessage"],
+                ),
+            )
+            wait_for_resume_create: pe_commands.CommandCreate = (
+                pe_commands.WaitForResumeCreate.model_construct(
+                    key=wait_for_resume_running.key,
+                    params=wait_for_resume_running.params,
+                )
+            )
+            return wait_for_resume_create, wait_for_resume_running
+        elif command["name"] == legacy_command_types.COMMENT:
+            comment_running = pe_commands.Comment.model_construct(
+                id=command_id,
+                key=command_id,
+                status=pe_commands.CommandStatus.RUNNING,
+                createdAt=now,
+                startedAt=now,
+                params=pe_commands.CommentParams.model_construct(
+                    message=command["payload"]["text"],
+                ),
+            )
+            comment_create = pe_commands.CommentCreate.model_construct(
+                key=comment_running.key, params=comment_running.params
+            )
+            return comment_create, comment_running
+        else:
+            custom_running = pe_commands.Custom.model_construct(
+                id=command_id,
+                key=command_id,
+                status=pe_commands.CommandStatus.RUNNING,
+                createdAt=now,
+                startedAt=now,
+                params=LegacyCommandParams.model_construct(
+                    legacyCommandType=command["name"],
+                    legacyCommandText=command["payload"]["text"],
+                ),
+            )
+            custom_create = pe_commands.CustomCreate.model_construct(
+                key=custom_running.key,
+                params=custom_running.params,
+            )
+            return custom_create, custom_running
+
+    def _build_drop_tip(
+        self,
+        command: legacy_command_types.DropTipMessage,
+        command_id: str,
+        now: datetime,
+    ) -> Tuple[pe_commands.CommandCreate, pe_commands.Command]:
+        pipette: InstrumentContext = command["payload"]["instrument"]
+        well = command["payload"]["location"]
+        mount = MountType(pipette.mount)
+        #   the following type checking suppression assumes the tiprack is not loaded on top of a module
+        slot = DeckSlotName.from_primitive(well.parent.parent)  # type: ignore[arg-type]
+        well_name = well.well_name
+        labware_id = self._labware_id_by_slot[slot]
+        pipette_id = self._pipette_id_by_mount[mount]
+
+        running = pe_commands.DropTip.model_construct(
+            id=command_id,
+            key=command_id,
+            status=pe_commands.CommandStatus.RUNNING,
+            createdAt=now,
+            startedAt=now,
+            params=pe_commands.DropTipParams.model_construct(
+                pipetteId=pipette_id,
+                labwareId=labware_id,
+                wellName=well_name,
+            ),
+        )
+        create = pe_commands.DropTipCreate.model_construct(
+            key=running.key,
+            params=running.params,
+        )
+        return create, running
+
+    def _build_pick_up_tip(
+        self,
+        command: legacy_command_types.PickUpTipMessage,
+        command_id: str,
+        now: datetime,
+    ) -> Tuple[pe_commands.CommandCreate, pe_commands.Command]:
+        pipette: InstrumentContext = command["payload"]["instrument"]
+        location = command["payload"]["location"]
+        well = location
+        mount = MountType(pipette.mount)
+        #   the following type checking suppression assumes the tiprack is not loaded on top of a module
+        slot = DeckSlotName.from_primitive(well.parent.parent)  # type: ignore[arg-type]
+        well_name = well.well_name
+        labware_id = self._labware_id_by_slot[slot]
+        pipette_id = self._pipette_id_by_mount[mount]
+
+        running = pe_commands.PickUpTip.model_construct(
+            id=command_id,
+            key=command_id,
+            status=pe_commands.CommandStatus.RUNNING,
+            createdAt=now,
+            startedAt=now,
+            params=pe_commands.PickUpTipParams.model_construct(
+                pipetteId=pipette_id,
+                labwareId=labware_id,
+                wellName=well_name,
+            ),
+        )
+        create = pe_commands.PickUpTipCreate.model_construct(
+            key=running.key, params=running.params
+        )
+        return create, running
+
+    def _build_liquid_handling(
+        self,
+        command: Union[
+            legacy_command_types.AspirateMessage, legacy_command_types.DispenseMessage
+        ],
+        command_id: str,
+        now: datetime,
+    ) -> Tuple[pe_commands.CommandCreate, pe_commands.Command]:
+        pipette: InstrumentContext = command["payload"]["instrument"]
+        location = command["payload"]["location"]
+        volume = command["payload"]["volume"]
+        # TODO:(jr, 15.08.2022): aspirate and dispense commands with no specified labware
+        # get filtered into custom. Refactor this in followup legacy command mapping
+        if location.labware.is_well:
+            well = location.labware.as_well()
+            slot = DeckSlotName(location.labware.first_parent())
+            parent_module_id = self._module_id_by_slot.get(slot)
+            labware_id = (
+                self._labware_id_by_module_id[parent_module_id]
+                if parent_module_id is not None
+                else self._labware_id_by_slot[slot]
+            )
+            mount = MountType(pipette.mount)
+            well_name = well.well_name
+            pipette_id = self._pipette_id_by_mount[mount]
+
+            if volume == 0:
+                # In edge cases, it's possible for a Python protocol to do dispense()
+                # or aspirate() with a volume of 0, which behaves roughly like
+                # move_to(). Protocol Engine aspirate and dispense commands must have
+                # volume > 0, so we can't map into those.
+                #
+                # TODO(mm, 2024-03-22): I don't think this has been true since
+                # https://github.com/Opentrons/opentrons/pull/14211. Can we just use
+                # aspirate and dispense commands now?
+                move_to_well_running = pe_commands.MoveToWell.model_construct(
+                    id=command_id,
+                    key=command_id,
+                    status=pe_commands.CommandStatus.RUNNING,
+                    createdAt=now,
+                    startedAt=now,
+                    params=pe_commands.MoveToWellParams.model_construct(
+                        pipetteId=pipette_id,
+                        labwareId=labware_id,
+                        wellName=well_name,
+                    ),
+                )
+                move_to_well_create = pe_commands.MoveToWellCreate.model_construct(
+                    key=move_to_well_running.key, params=move_to_well_running.params
+                )
+                return move_to_well_create, move_to_well_running
+            elif command["name"] == legacy_command_types.ASPIRATE:
+                flow_rate = command["payload"]["rate"] * pipette.flow_rate.aspirate
+                aspirate_running = pe_commands.Aspirate.model_construct(
+                    id=command_id,
+                    key=command_id,
+                    status=pe_commands.CommandStatus.RUNNING,
+                    createdAt=now,
+                    startedAt=now,
+                    # Don't .model_construct() params, because we want to validate
+                    # volume and flowRate.
+                    params=pe_commands.AspirateParams(
+                        pipetteId=pipette_id,
+                        labwareId=labware_id,
+                        wellName=well_name,
+                        volume=volume,
+                        flowRate=flow_rate,
+                    ),
+                )
+                aspirate_create = pe_commands.AspirateCreate.model_construct(
+                    key=aspirate_running.key, params=aspirate_running.params
+                )
+                return aspirate_create, aspirate_running
+            else:
+                flow_rate = command["payload"]["rate"] * pipette.flow_rate.dispense
+                dispense_running = pe_commands.Dispense.model_construct(
+                    id=command_id,
+                    key=command_id,
+                    status=pe_commands.CommandStatus.RUNNING,
+                    createdAt=now,
+                    startedAt=now,
+                    # Don't .model_construct params, because we want to validate
+                    # volume and flowRate.
+                    params=pe_commands.DispenseParams(
+                        pipetteId=pipette_id,
+                        labwareId=labware_id,
+                        wellName=well_name,
+                        volume=volume,
+                        flowRate=flow_rate,
+                    ),
+                )
+                dispense_create = pe_commands.DispenseCreate.model_construct(
+                    key=dispense_running.key, params=dispense_running.params
+                )
+                return dispense_create, dispense_running
+
+        else:
+            running = pe_commands.Custom.model_construct(
+                id=command_id,
+                key=command_id,
+                status=pe_commands.CommandStatus.RUNNING,
+                createdAt=now,
+                startedAt=now,
+                params=LegacyCommandParams.model_construct(
+                    legacyCommandType=command["name"],
+                    legacyCommandText=command["payload"]["text"],
+                ),
+            )
+            create = pe_commands.CustomCreate.model_construct(
+                key=running.key, params=running.params
+            )
+            return create, running
+
+    def _build_blow_out(
+        self,
+        command: legacy_command_types.BlowOutMessage,
+        command_id: str,
+        now: datetime,
+    ) -> Tuple[pe_commands.CommandCreate, pe_commands.Command]:
+        pipette: InstrumentContext = command["payload"]["instrument"]
+        location = command["payload"]["location"]
+        flow_rate = pipette.flow_rate.blow_out
+        #   TODO:(jr, 15.08.2022): blow_out commands with no specified labware get filtered
+        #   into custom. Remove location.labware.is_empty is False when refactor is complete
+        if isinstance(location, Location) and location.labware.is_well:
+            well = location.labware.as_well()
+            slot = DeckSlotName(location.labware.first_parent())
+            parent_module_id = self._module_id_by_slot.get(slot)
+            labware_id = (
+                self._labware_id_by_module_id[parent_module_id]
+                if parent_module_id is not None
+                else self._labware_id_by_slot[slot]
+            )
+            mount = MountType(pipette.mount)
+            well_name = well.well_name
+            pipette_id = self._pipette_id_by_mount[mount]
+
+            blow_out_running = pe_commands.BlowOut.model_construct(
+                id=command_id,
+                key=command_id,
+                status=pe_commands.CommandStatus.RUNNING,
+                createdAt=now,
+                startedAt=now,
+                # Don't .model_construct() params, because we want to validate flowRate.
+                params=pe_commands.BlowOutParams(
+                    pipetteId=pipette_id,
+                    labwareId=labware_id,
+                    wellName=well_name,
+                    flowRate=flow_rate,
+                ),
+            )
+            blow_out_create = pe_commands.BlowOutCreate.model_construct(
+                key=blow_out_running.key, params=blow_out_running.params
+            )
+            return blow_out_create, blow_out_running
+
+        #   TODO:(jr, 15.08.2022): blow_out commands with no specified labware get filtered
+        #   into custom. Refactor this in followup legacy command mapping
+        else:
+            custom_running = pe_commands.Custom.model_construct(
+                id=command_id,
+                key=command_id,
+                status=pe_commands.CommandStatus.RUNNING,
+                createdAt=now,
+                startedAt=now,
+                params=LegacyCommandParams.model_construct(
+                    legacyCommandType=command["name"],
+                    legacyCommandText=command["payload"]["text"],
+                ),
+            )
+            custom_create = pe_commands.CustomCreate.model_construct(
+                key=custom_running.key, params=custom_running.params
+            )
+            return custom_create, custom_running
+
+    def _map_labware_load(
+        self, labware_load_info: LegacyLabwareLoadInfo
+    ) -> List[pe_actions.Action]:
+        """Map a legacy labware load to a ProtocolEngine command."""
+        now = ModelUtils.get_timestamp()
+        count = self._command_count["LOAD_LABWARE"]
+        slot = labware_load_info.deck_slot
+        location: pe_types.LabwareLocation
+        location_sequence: pe_types.LabwareLocationSequence = []
+        if labware_load_info.on_module:
+            module_id = self._module_id_by_slot[slot]
+            location = pe_types.ModuleLocation.model_construct(moduleId=module_id)
+            location_sequence.append(
+                pe_types.OnModuleLocationSequenceComponent(moduleId=module_id)
+            )
+        else:
+            location = pe_types.DeckSlotLocation.model_construct(slotName=slot)
+
+        location_sequence.append(
+            pe_types.OnAddressableAreaLocationSequenceComponent(
+                addressableAreaName=slot.value
+            )
+        )
+        command_id = f"commands.LOAD_LABWARE-{count}"
+        labware_id = f"labware-{count}"
+        succeeded_command = pe_commands.LoadLabware(
+            id=command_id,
+            key=command_id,
+            status=pe_commands.CommandStatus.SUCCEEDED,
+            createdAt=now,
+            startedAt=now,
+            completedAt=now,
+            params=pe_commands.LoadLabwareParams.model_construct(
+                location=location,
+                loadName=labware_load_info.labware_load_name,
+                namespace=labware_load_info.labware_namespace,
+                version=labware_load_info.labware_version,
+                displayName=labware_load_info.labware_display_name,
+            ),
+            notes=[],
+            result=pe_commands.LoadLabwareResult.model_construct(
+                labwareId=labware_id,
+                definition=labware_definition_type_adapter.validate_python(
+                    labware_load_info.labware_definition
+                ),
+                offsetId=labware_load_info.offset_id,
+                locationSequence=location_sequence,
+            ),
+        )
+        queue_action = pe_actions.QueueCommandAction(
+            command_id=succeeded_command.id,
+            created_at=succeeded_command.createdAt,
+            request=pe_commands.LoadLabwareCreate.model_construct(
+                key=succeeded_command.key, params=succeeded_command.params
+            ),
+            request_hash=None,
+        )
+        run_action = pe_actions.RunCommandAction(
+            command_id=succeeded_command.id,
+            # We just set this above, so we know it's not None.
+            started_at=succeeded_command.startedAt,  # type: ignore[arg-type]
+        )
+        state_update = StateUpdate()
+        assert succeeded_command.result is not None
+        state_update.set_loaded_labware(
+            labware_id=labware_id,
+            definition=succeeded_command.result.definition,
+            display_name=labware_load_info.labware_display_name,
+            offset_id=labware_load_info.offset_id,
+            location=location,
+        )
+
+        succeed_action = pe_actions.SucceedCommandAction(
+            command=succeeded_command,
+            state_update=state_update,
+        )
+
+        self._command_count["LOAD_LABWARE"] = count + 1
+        if isinstance(location, pe_types.DeckSlotLocation):
+            self._labware_id_by_slot[location.slotName] = labware_id
+        elif isinstance(location, pe_types.ModuleLocation):
+            self._labware_id_by_module_id[location.moduleId] = labware_id
+
+        return [queue_action, run_action, succeed_action]
+
+    def _map_instrument_load(
+        self,
+        instrument_load_info: LegacyInstrumentLoadInfo,
+    ) -> List[pe_actions.Action]:
+        """Map a legacy instrument (pipette) load to a ProtocolEngine command.
+
+        Also creates a `AddPipetteConfigAction`, which is not necessary for the run,
+        but is needed for stop so tip geometry is in state for the HardwareStopper.
+        """
+        now = ModelUtils.get_timestamp()
+        count = self._command_count["LOAD_PIPETTE"]
+        command_id = f"commands.LOAD_PIPETTE-{count}"
+        pipette_id = f"pipette-{count}"
+        mount = MountType(str(instrument_load_info.mount).lower())
+
+        succeeded_command = pe_commands.LoadPipette.model_construct(
+            id=command_id,
+            key=command_id,
+            status=pe_commands.CommandStatus.SUCCEEDED,
+            createdAt=now,
+            startedAt=now,
+            completedAt=now,
+            params=pe_commands.LoadPipetteParams.model_construct(
+                pipetteName=PipetteNameType(instrument_load_info.instrument_load_name),
+                mount=mount,
+            ),
+            notes=[],
+            result=pe_commands.LoadPipetteResult.model_construct(pipetteId=pipette_id),
+        )
+        serial = instrument_load_info.pipette_dict.get("pipette_id", None) or ""
+        state_update = StateUpdate()
+        state_update.set_load_pipette(
+            pipette_id=pipette_id,
+            mount=succeeded_command.params.mount,
+            pipette_name=succeeded_command.params.pipetteName,
+            liquid_presence_detection=succeeded_command.params.liquidPresenceDetection,
+        )
+        state_update.update_pipette_config(
+            pipette_id=pipette_id,
+            serial_number=serial,
+            config=pipette_data_provider.get_pipette_static_config(
+                # Compatibility note - this is the version of tip overlap data, it stays at 0
+                # so protocol behavior does not change when you run a legacy JSON protocol
+                instrument_load_info.pipette_dict,
+                "v0",
+            ),
+        )
+        queue_action = pe_actions.QueueCommandAction(
+            command_id=succeeded_command.id,
+            created_at=succeeded_command.createdAt,
+            request=pe_commands.LoadPipetteCreate.model_construct(
+                key=succeeded_command.key, params=succeeded_command.params
+            ),
+            request_hash=None,
+        )
+        run_action = pe_actions.RunCommandAction(
+            command_id=succeeded_command.id,
+            # We just set this above, so we know it's not None.
+            started_at=succeeded_command.startedAt,  # type: ignore[arg-type]
+        )
+
+        succeed_action = pe_actions.SucceedCommandAction(
+            command=succeeded_command,
+            state_update=state_update,
+        )
+
+        self._command_count["LOAD_PIPETTE"] = count + 1
+        self._pipette_id_by_mount[mount] = pipette_id
+
+        return [queue_action, run_action, succeed_action]
+
+    def _map_module_load(
+        self, module_load_info: LegacyModuleLoadInfo
+    ) -> List[pe_actions.Action]:
+        """Map a legacy module load to a Protocol Engine command."""
+        now = ModelUtils.get_timestamp()
+
+        count = self._command_count["LOAD_MODULE"]
+        command_id = f"commands.LOAD_MODULE-{count}"
+        module_id = f"module-{count}"
+        requested_model = _HARDWARE_TO_PE_MODULE[module_load_info.requested_model]
+        loaded_model = _HARDWARE_TO_PE_MODULE[module_load_info.loaded_model]
+
+        # This will fetch a V2 definition only. PAPI < v2.3 use V1 definitions.
+        # When running a < v2.3 protocol, there will be a mismatch of definitions used
+        # during analysis+LPC (V2) and protocol execution (V1).
+        # But this shouldn't result in any problems since V2 and V1 definitions
+        # have similar info, with V2 having additional info fields.
+        loaded_definition = self._module_definition_by_model.get(
+            loaded_model
+        ) or self._module_data_provider.get_definition(loaded_model)
+
+        succeeded_command = pe_commands.LoadModule.model_construct(
+            id=command_id,
+            key=command_id,
+            status=pe_commands.CommandStatus.SUCCEEDED,
+            createdAt=now,
+            startedAt=now,
+            completedAt=now,
+            params=pe_commands.LoadModuleParams.model_construct(
+                model=requested_model,
+                location=pe_types.DeckSlotLocation(
+                    slotName=module_load_info.deck_slot,
+                ),
+                moduleId=module_id,
+            ),
+            notes=[],
+            result=pe_commands.LoadModuleResult.model_construct(
+                moduleId=module_id,
+                serialNumber=module_load_info.module_serial,
+                model=loaded_model,
+            ),
+        )
+        queue_action = pe_actions.QueueCommandAction(
+            command_id=succeeded_command.id,
+            created_at=succeeded_command.createdAt,
+            request=pe_commands.LoadModuleCreate.model_construct(
+                key=succeeded_command.key, params=succeeded_command.params
+            ),
+            request_hash=None,
+        )
+        run_action = pe_actions.RunCommandAction(
+            command_id=succeeded_command.id,
+            # We just set this above, so we know it's not None.
+            started_at=succeeded_command.startedAt,  # type: ignore[arg-type]
+        )
+        succeed_action = pe_actions.SucceedCommandAction(
+            command=succeeded_command, state_update=StateUpdate()
+        )
+
+        self._command_count["LOAD_MODULE"] = count + 1
+        self._module_id_by_slot[module_load_info.deck_slot] = module_id
+        self._module_definition_by_model[loaded_model] = loaded_definition
+
+        return [queue_action, run_action, succeed_action]